Cutis

Approximately 50% of melanomas contain BRAF mutations, which occur in a greater proportion of melanomas found on sites of intermittent sun exposure.¹BRAF-mutated melanomas have been associated with high levels of early-life ambient UV exposure, especially between ages 0 and 20 years.² In addition, studies have shown that BRAF-mutated melanomas commonly are found on the trunk and extremities.^1-3BRAF mutations also have been associated with younger age, superficial spreading subtype and low tumor thickness, absence of dermal melanocyte mitosis, low Ki-67 score, low phospho-histone H3 score, pigmented melanoma, advanced melanoma stage, and conjunctival melanoma.^4-7BRAF mutations are found more frequently in metastatic melanoma lesions than primary melanomas, suggesting that BRAF mutations may be acquired during metastasis.⁸ Studies have shown different conclusions on the effect of BRAF mutation on melanoma-related death.^5,9,10

The aim of this study was to identify trends in BRAF V600E–mutated melanoma according to age, sex, and melanoma-specific survival among Olmsted County, Minnesota, residents with a first diagnosis of melanoma at 18 to 60 years of age.

Methods

In total, 638 patients aged 18 to 60 years who resided in Olmsted County and had a first lifetime diagnosis of cutaneous melanoma between 1970 and 2009 were retrospectively identified as a part of the Rochester Epidemiology Project (REP). The REP is a health records linkage system that encompasses almost all sources of medical care available to the local population of Olmsted County.¹¹ This study was approved by the Mayo Clinic Institutional Review Board (Rochester, Minnesota).

Of the 638 individuals identified in the REP, 536 had been seen at Mayo Clinic and thus potentially had tissue blocks available for the study of BRAF mutation expression. Of these 536 patients, 156 did not have sufficient residual tissue available. As a result, 380 (60%) of the original 638 patients had available blocks with sufficient tissue for immunohistochemical analysis of BRAF expression. Only primary cutaneous melanomas were included in the present study.

All specimens were reviewed by a board-certified dermatopathologist (J.S.L.) for appropriateness of inclusion, which involved confirmation of the diagnosis of melanoma, histologic type of melanoma, and presence of sufficient residual tissue for immunohistochemical stains.

All specimens were originally diagnosed as malignant melanoma at the time of clinical care by at least 2 board-certified dermatopathologists. For the purposes of this study, all specimens were rereviewed for diagnostic accuracy. We required that specimens exhibit severe cytologic and architectural atypia as well as other features favoring melanoma, such as consumption of rete pegs, pagetosis, confluence of junctional melanocytes, evidence of regression, lack of maturation of melanocytes with descent into the dermis, or mitotic figures among the dermal melanocyte population.

The available tissue blocks were retrieved, sectioned, confirmed as melanoma, and stained with a mouse antihuman BRAF V600E monoclonal antibody (clone VE1; Spring Bioscience) to determine the presence of a BRAF V600E mutation. BRAF staining was evaluated in conjunction with a review of the associated slides stained with hematoxylin and eosin. Cytoplasmic staining of melanocytes for BRAF was graded as negative, focal or partial positive (<50% of tumor), or diffuse positive (>50% of tumor)(Figure 1). When a melanoma arose in association with a nevus, we considered only the melanoma component for BRAF staining. We categorized the histologic type as superficial spreading, nodular, or lentigo maligna, and the location as head and neck, trunk, or extremities.

 

Patient characteristics and survival outcomes were gathered through the health record and included age, Breslow thickness, location, decade of diagnosis, histologic type, stage (ie, noninvasive, invasive, or advanced), and follow-up. Pathologic stage 0 was considered noninvasive; stages IA and IB, invasive; and stages IIA or higher, advanced.

Statistical Analysis—Comparisons between the group of patients in the study (n=380) and the group of patients excluded for the reasons stated above (n=258) as well as associations of mutant BRAF status (positive [partial positive and diffuse positive] vs negative) with patient age (young adults [age range, 18–39 years] and middle-aged adults [age range, 40–60 years]), sex, decade of diagnosis, location, histologic type, and stage were evaluated with Wilcoxon rank sum, χ², Fisher exact, or Cochran-Armitage trend tests. Disease-specific survival and overall survival rates were estimated with the Kaplan-Meier method, and the duration of follow-up was calculated from the date of melanoma diagnosis to the date of death or the last follow-up. Associations of mutant BRAF expression status with death from melanoma and death from any cause were evaluated with Cox proportional hazard regression models and summarized with hazard ratio (HR) and 95% CI. Survival analyses were limited to patients with invasive or advanced disease. Statistical analyses were performed with SAS statistical software (SAS version 9.4). All tests were 2-sided, and P<.05 was considered statistically significant.

Results

Clinical and Tumor Characteristics—Of the 380 tissue specimens that underwent BRAF V600E analysis, 247 had negative staining; 106 had diffuse strong staining; and 27 had focal or partial staining. In total, 133 (35%) were positive, either partially or diffusely. The median age for patients who had negative staining was 45 years; for those with positive staining, it was 41 years (P=.07).

The patients who met inclusion criteria (n=380) were compared with those who were excluded (n=258)(eTable 1). The groups were similar on the basis of sex; age; and melanoma location, stage, and histologic subtype. However, some evidence showed that patients included in the study received the diagnosis of melanoma more recently (1970-1989, 13.2%; 1990-1999, 28.7%; 2000-2009, 58.2%) than those who were excluded (1970-1989, 24.7%; 1990-1999, 23.5%; 2000-2009, 51.8%)(P=.02).

BRAF V600E expression was more commonly found in superficial spreading (37.7%) and nodular melanomas (35.0%) than in situ melanomas (17.1%)(P=.01). Other characteristics of BRAF V600E expression are described in eTable 2. Overall, invasive and advanced melanomas were significantly more likely to harbor BRAF V600E expression than noninvasive melanomas (39.6% and 37.9%, respectively, vs 17.9%; P=.003). However, advanced melanomas more commonly expressed BRAF positivity among women, and invasive melanomas more commonly expressed BRAF positivity among men (eTable 2).

Survival—Survival analyses were limited to 297 patients with confirmed invasive or advanced disease. Of these, 180 (61%) had no BRAF V600E staining; 25 (8%) had partial staining; and 92 (31%) had diffuse positive staining. In total, 117 patients (39%) had a BRAF-mutated melanoma.

Among the patients still alive, the median (interquartile range [IQR]) duration of follow-up was 10.2 (7.0-16.8) years. Thirty-nine patients with invasive or advanced disease had died of any cause at a median (IQR) of 3.0 (1.3-10.2) years after diagnosis. In total, 26 patients died of melanoma at a median (IQR) follow-up of 2.5 (1.3-7.4) years after diagnosis. Eight women and 18 men died of malignant melanoma. Five deaths occurred because of malignant melanoma among patients aged 18 to 39 years, and 21 occurred among patients aged 40 to 60 years. In the 18- to 39-year-old group, all 5 deaths were among patients with a BRAF-positive melanoma. Estimated disease-specific survival rate (95% CI; number still at risk) at 5, 10, 15, and 20 years after diagnosis was 94% (91%-97%; 243), 91% (87%-95%; 142), 89% (85%-94%; 87), and 88% (83%-93%; 45), respectively.

 

In a univariable analysis, the HR for association of positive mutant BRAF expression with death of malignant melanoma was 1.84 (95% CI, 0.85-3.98; P=.12). No statistically significant interaction was observed between decade of diagnosis and BRAF expression (P=.60). However, the interaction between sex and BRAF expression was significant (P=.04), with increased risk of death from melanoma among women with BRAF-mutated melanoma (HR, 10.88; 95% CI, 1.34-88.41; P=.026) but not among men (HR 1.02; 95% CI, 0.40-2.64; P=.97)(Figures 2A and 2B). The HR for death from malignant melanoma among young adults aged 18 to 39 years with a BRAF-mutated melanoma was 16.4 (95% CI, 0.81-330.10; P=.068), whereas the HR among adults aged 40 to 60 years with a BRAF-mutated melanoma was 1.24 (95% CI, 0.52-2.98; P=.63)(Figures 2C and 2D).

BRAF V600E expression was not significantly associated with death from any cause (HR, 1.39; 95% CI, 0.74-2.61; P=.31) or with decade of diagnosis (P=.13). Similarly, BRAF expression was not associated with death from any cause according to sex (P=.31). However, a statistically significant interaction was seen between age at diagnosis and BRAF expression (P=.003). BRAF expression was significantly associated with death from any cause for adults aged 18 to 39 years (HR, 9.60; 95% CI, 1.15-80.00; P=.04). In comparison, no association of BRAF expression with death was observed for adults aged 40 to 60 years (HR, 0.99; 95% CI, 0.48-2.03; P=.98).

Comment

We found that melanomas with BRAF mutations were more likely in advanced and invasive melanoma. The frequency of BRAF mutations among melanomas that were considered advanced was higher in women than men. Although the number of deaths was limited, women with a melanoma with BRAF expression were more likely to die of melanoma, young adults with a BRAF-mutated melanoma had an almost 10-fold increased risk of dying from any cause, and middle-aged adults showed no increased risk of death. These findings suggest that young adults who are genetically prone to a BRAF-mutated melanoma could be at a disadvantage for all-cause mortality. Although this finding was significant, the 95% CI was large, and further studies would be warranted before sound conclusions could be made.

Melanoma has been increasing in incidence across all age groups in Olmsted County over the last 4 decades.^12-14 However, our results show that the percentage of BRAF-mutated melanomas in this population has been stable over time, with no statistically significant difference by age or sex. Other confounding factors may have an influence, such as increased rates of early detection and diagnosis of melanoma in contemporary times. Our data suggest that patients included in the BRAF-mutation analysis study had received the diagnosis of melanoma more recently than those who were excluded from the study, which could be due to older melanomas being less likely to have adequate tissue specimens available for immunohistochemical staining/evaluation.

Prior research has shown that BRAF-mutated melanomas typically occur on the trunk and are more likely in individuals with more than 14 nevi on the back.² In the present cohort, BRAF-positive melanomas had a predisposition toward the trunk but also were found on the head, neck, and extremities—areas that are more likely to have long-term sun damage. One suggestion is that 2 distinct pathways for melanoma development exist: one associated with a large number of melanocytic nevi (that is more prone to genetic mutations in melanocytes) and the other associated with long-term sun exposure.^15,16 The combination of these hypotheses suggests that individuals who are prone to the development of large numbers of nevi may require sun exposure for the initial insult, but the development of melanoma may be carried out by other factors after this initial sun exposure insult, whereas individuals without large numbers of nevi who may have less genetic risk may require continued long-term sun exposure for melanoma to develop.¹⁷

Our study had limitations, including the small numbers of deaths overall and cause-specific deaths of metastatic melanoma, which limited our ability to conduct more extensive multivariable modeling. Also, the retrospective nature and time frame of looking back 4 decades did not allow us to have information sufficient to categorize some patients as having dysplastic nevus syndrome or not, which would be a potentially interesting variable to include in the analysis. Because the number of deaths in the 18- to 39-year-old cohort was only 5, further statistical comparison regarding tumor type and other variables pertaining to BRAF positivity were not possible. In addition, our data were collected from patients residing in a single geographic county (Olmsted County, Minnesota), which may limit generalizability. Lastly, BRAF V600E mutations were identified through immunostaining only, not molecular data, so it is possible some patients had false-negative immunohistochemistry findings and thus were not identified.

Conclusion

BRAF-mutated melanomas were found in 35% of our cohort, with no significant change in the percentage of melanomas with BRAF V600E mutations over the last 4 decades in this population. In addition, no differences or significant trends existed according to sex and BRAF-mutated melanoma development. Women with BRAF-mutated melanomas were more likely to die of metastatic melanoma than men, and young adults with BRAF-mutated melanomas had a higher all-cause mortality risk. Further research is needed to decipher what effect BRAF-mutated melanomas have on metastasis and cause-specific death in women as well as all-cause mortality in young adults.

Acknowledgment—The authors are indebted to Scientific Publications, Mayo Clinic (Rochester, Minnesota).

Approximately 50% of melanomas contain BRAF mutations, which occur in a greater proportion of melanomas found on sites of intermittent sun exposure.¹BRAF-mutated melanomas have been associated with high levels of early-life ambient UV exposure, especially between ages 0 and 20 years.² In addition, studies have shown that BRAF-mutated melanomas commonly are found on the trunk and extremities.^1-3BRAF mutations also have been associated with younger age, superficial spreading subtype and low tumor thickness, absence of dermal melanocyte mitosis, low Ki-67 score, low phospho-histone H3 score, pigmented melanoma, advanced melanoma stage, and conjunctival melanoma.^4-7BRAF mutations are found more frequently in metastatic melanoma lesions than primary melanomas, suggesting that BRAF mutations may be acquired during metastasis.⁸ Studies have shown different conclusions on the effect of BRAF mutation on melanoma-related death.^5,9,10

The aim of this study was to identify trends in BRAF V600E–mutated melanoma according to age, sex, and melanoma-specific survival among Olmsted County, Minnesota, residents with a first diagnosis of melanoma at 18 to 60 years of age.

Methods

In total, 638 patients aged 18 to 60 years who resided in Olmsted County and had a first lifetime diagnosis of cutaneous melanoma between 1970 and 2009 were retrospectively identified as a part of the Rochester Epidemiology Project (REP). The REP is a health records linkage system that encompasses almost all sources of medical care available to the local population of Olmsted County.¹¹ This study was approved by the Mayo Clinic Institutional Review Board (Rochester, Minnesota).

Of the 638 individuals identified in the REP, 536 had been seen at Mayo Clinic and thus potentially had tissue blocks available for the study of BRAF mutation expression. Of these 536 patients, 156 did not have sufficient residual tissue available. As a result, 380 (60%) of the original 638 patients had available blocks with sufficient tissue for immunohistochemical analysis of BRAF expression. Only primary cutaneous melanomas were included in the present study.

All specimens were reviewed by a board-certified dermatopathologist (J.S.L.) for appropriateness of inclusion, which involved confirmation of the diagnosis of melanoma, histologic type of melanoma, and presence of sufficient residual tissue for immunohistochemical stains.

All specimens were originally diagnosed as malignant melanoma at the time of clinical care by at least 2 board-certified dermatopathologists. For the purposes of this study, all specimens were rereviewed for diagnostic accuracy. We required that specimens exhibit severe cytologic and architectural atypia as well as other features favoring melanoma, such as consumption of rete pegs, pagetosis, confluence of junctional melanocytes, evidence of regression, lack of maturation of melanocytes with descent into the dermis, or mitotic figures among the dermal melanocyte population.

The available tissue blocks were retrieved, sectioned, confirmed as melanoma, and stained with a mouse antihuman BRAF V600E monoclonal antibody (clone VE1; Spring Bioscience) to determine the presence of a BRAF V600E mutation. BRAF staining was evaluated in conjunction with a review of the associated slides stained with hematoxylin and eosin. Cytoplasmic staining of melanocytes for BRAF was graded as negative, focal or partial positive (<50% of tumor), or diffuse positive (>50% of tumor)(Figure 1). When a melanoma arose in association with a nevus, we considered only the melanoma component for BRAF staining. We categorized the histologic type as superficial spreading, nodular, or lentigo maligna, and the location as head and neck, trunk, or extremities.

 

Patient characteristics and survival outcomes were gathered through the health record and included age, Breslow thickness, location, decade of diagnosis, histologic type, stage (ie, noninvasive, invasive, or advanced), and follow-up. Pathologic stage 0 was considered noninvasive; stages IA and IB, invasive; and stages IIA or higher, advanced.

Statistical Analysis—Comparisons between the group of patients in the study (n=380) and the group of patients excluded for the reasons stated above (n=258) as well as associations of mutant BRAF status (positive [partial positive and diffuse positive] vs negative) with patient age (young adults [age range, 18–39 years] and middle-aged adults [age range, 40–60 years]), sex, decade of diagnosis, location, histologic type, and stage were evaluated with Wilcoxon rank sum, χ², Fisher exact, or Cochran-Armitage trend tests. Disease-specific survival and overall survival rates were estimated with the Kaplan-Meier method, and the duration of follow-up was calculated from the date of melanoma diagnosis to the date of death or the last follow-up. Associations of mutant BRAF expression status with death from melanoma and death from any cause were evaluated with Cox proportional hazard regression models and summarized with hazard ratio (HR) and 95% CI. Survival analyses were limited to patients with invasive or advanced disease. Statistical analyses were performed with SAS statistical software (SAS version 9.4). All tests were 2-sided, and P<.05 was considered statistically significant.

Results

Clinical and Tumor Characteristics—Of the 380 tissue specimens that underwent BRAF V600E analysis, 247 had negative staining; 106 had diffuse strong staining; and 27 had focal or partial staining. In total, 133 (35%) were positive, either partially or diffusely. The median age for patients who had negative staining was 45 years; for those with positive staining, it was 41 years (P=.07).

The patients who met inclusion criteria (n=380) were compared with those who were excluded (n=258)(eTable 1). The groups were similar on the basis of sex; age; and melanoma location, stage, and histologic subtype. However, some evidence showed that patients included in the study received the diagnosis of melanoma more recently (1970-1989, 13.2%; 1990-1999, 28.7%; 2000-2009, 58.2%) than those who were excluded (1970-1989, 24.7%; 1990-1999, 23.5%; 2000-2009, 51.8%)(P=.02).

BRAF V600E expression was more commonly found in superficial spreading (37.7%) and nodular melanomas (35.0%) than in situ melanomas (17.1%)(P=.01). Other characteristics of BRAF V600E expression are described in eTable 2. Overall, invasive and advanced melanomas were significantly more likely to harbor BRAF V600E expression than noninvasive melanomas (39.6% and 37.9%, respectively, vs 17.9%; P=.003). However, advanced melanomas more commonly expressed BRAF positivity among women, and invasive melanomas more commonly expressed BRAF positivity among men (eTable 2).

Survival—Survival analyses were limited to 297 patients with confirmed invasive or advanced disease. Of these, 180 (61%) had no BRAF V600E staining; 25 (8%) had partial staining; and 92 (31%) had diffuse positive staining. In total, 117 patients (39%) had a BRAF-mutated melanoma.

Among the patients still alive, the median (interquartile range [IQR]) duration of follow-up was 10.2 (7.0-16.8) years. Thirty-nine patients with invasive or advanced disease had died of any cause at a median (IQR) of 3.0 (1.3-10.2) years after diagnosis. In total, 26 patients died of melanoma at a median (IQR) follow-up of 2.5 (1.3-7.4) years after diagnosis. Eight women and 18 men died of malignant melanoma. Five deaths occurred because of malignant melanoma among patients aged 18 to 39 years, and 21 occurred among patients aged 40 to 60 years. In the 18- to 39-year-old group, all 5 deaths were among patients with a BRAF-positive melanoma. Estimated disease-specific survival rate (95% CI; number still at risk) at 5, 10, 15, and 20 years after diagnosis was 94% (91%-97%; 243), 91% (87%-95%; 142), 89% (85%-94%; 87), and 88% (83%-93%; 45), respectively.

 

In a univariable analysis, the HR for association of positive mutant BRAF expression with death of malignant melanoma was 1.84 (95% CI, 0.85-3.98; P=.12). No statistically significant interaction was observed between decade of diagnosis and BRAF expression (P=.60). However, the interaction between sex and BRAF expression was significant (P=.04), with increased risk of death from melanoma among women with BRAF-mutated melanoma (HR, 10.88; 95% CI, 1.34-88.41; P=.026) but not among men (HR 1.02; 95% CI, 0.40-2.64; P=.97)(Figures 2A and 2B). The HR for death from malignant melanoma among young adults aged 18 to 39 years with a BRAF-mutated melanoma was 16.4 (95% CI, 0.81-330.10; P=.068), whereas the HR among adults aged 40 to 60 years with a BRAF-mutated melanoma was 1.24 (95% CI, 0.52-2.98; P=.63)(Figures 2C and 2D).

BRAF V600E expression was not significantly associated with death from any cause (HR, 1.39; 95% CI, 0.74-2.61; P=.31) or with decade of diagnosis (P=.13). Similarly, BRAF expression was not associated with death from any cause according to sex (P=.31). However, a statistically significant interaction was seen between age at diagnosis and BRAF expression (P=.003). BRAF expression was significantly associated with death from any cause for adults aged 18 to 39 years (HR, 9.60; 95% CI, 1.15-80.00; P=.04). In comparison, no association of BRAF expression with death was observed for adults aged 40 to 60 years (HR, 0.99; 95% CI, 0.48-2.03; P=.98).

Comment

We found that melanomas with BRAF mutations were more likely in advanced and invasive melanoma. The frequency of BRAF mutations among melanomas that were considered advanced was higher in women than men. Although the number of deaths was limited, women with a melanoma with BRAF expression were more likely to die of melanoma, young adults with a BRAF-mutated melanoma had an almost 10-fold increased risk of dying from any cause, and middle-aged adults showed no increased risk of death. These findings suggest that young adults who are genetically prone to a BRAF-mutated melanoma could be at a disadvantage for all-cause mortality. Although this finding was significant, the 95% CI was large, and further studies would be warranted before sound conclusions could be made.

Melanoma has been increasing in incidence across all age groups in Olmsted County over the last 4 decades.^12-14 However, our results show that the percentage of BRAF-mutated melanomas in this population has been stable over time, with no statistically significant difference by age or sex. Other confounding factors may have an influence, such as increased rates of early detection and diagnosis of melanoma in contemporary times. Our data suggest that patients included in the BRAF-mutation analysis study had received the diagnosis of melanoma more recently than those who were excluded from the study, which could be due to older melanomas being less likely to have adequate tissue specimens available for immunohistochemical staining/evaluation.

Prior research has shown that BRAF-mutated melanomas typically occur on the trunk and are more likely in individuals with more than 14 nevi on the back.² In the present cohort, BRAF-positive melanomas had a predisposition toward the trunk but also were found on the head, neck, and extremities—areas that are more likely to have long-term sun damage. One suggestion is that 2 distinct pathways for melanoma development exist: one associated with a large number of melanocytic nevi (that is more prone to genetic mutations in melanocytes) and the other associated with long-term sun exposure.^15,16 The combination of these hypotheses suggests that individuals who are prone to the development of large numbers of nevi may require sun exposure for the initial insult, but the development of melanoma may be carried out by other factors after this initial sun exposure insult, whereas individuals without large numbers of nevi who may have less genetic risk may require continued long-term sun exposure for melanoma to develop.¹⁷

Our study had limitations, including the small numbers of deaths overall and cause-specific deaths of metastatic melanoma, which limited our ability to conduct more extensive multivariable modeling. Also, the retrospective nature and time frame of looking back 4 decades did not allow us to have information sufficient to categorize some patients as having dysplastic nevus syndrome or not, which would be a potentially interesting variable to include in the analysis. Because the number of deaths in the 18- to 39-year-old cohort was only 5, further statistical comparison regarding tumor type and other variables pertaining to BRAF positivity were not possible. In addition, our data were collected from patients residing in a single geographic county (Olmsted County, Minnesota), which may limit generalizability. Lastly, BRAF V600E mutations were identified through immunostaining only, not molecular data, so it is possible some patients had false-negative immunohistochemistry findings and thus were not identified.

Conclusion

BRAF-mutated melanomas were found in 35% of our cohort, with no significant change in the percentage of melanomas with BRAF V600E mutations over the last 4 decades in this population. In addition, no differences or significant trends existed according to sex and BRAF-mutated melanoma development. Women with BRAF-mutated melanomas were more likely to die of metastatic melanoma than men, and young adults with BRAF-mutated melanomas had a higher all-cause mortality risk. Further research is needed to decipher what effect BRAF-mutated melanomas have on metastasis and cause-specific death in women as well as all-cause mortality in young adults.

Acknowledgment—The authors are indebted to Scientific Publications, Mayo Clinic (Rochester, Minnesota).

Approximately 50% of melanomas contain BRAF mutations, which occur in a greater proportion of melanomas found on sites of intermittent sun exposure.¹BRAF-mutated melanomas have been associated with high levels of early-life ambient UV exposure, especially between ages 0 and 20 years.² In addition, studies have shown that BRAF-mutated melanomas commonly are found on the trunk and extremities.^1-3BRAF mutations also have been associated with younger age, superficial spreading subtype and low tumor thickness, absence of dermal melanocyte mitosis, low Ki-67 score, low phospho-histone H3 score, pigmented melanoma, advanced melanoma stage, and conjunctival melanoma.^4-7BRAF mutations are found more frequently in metastatic melanoma lesions than primary melanomas, suggesting that BRAF mutations may be acquired during metastasis.⁸ Studies have shown different conclusions on the effect of BRAF mutation on melanoma-related death.^5,9,10

The aim of this study was to identify trends in BRAF V600E–mutated melanoma according to age, sex, and melanoma-specific survival among Olmsted County, Minnesota, residents with a first diagnosis of melanoma at 18 to 60 years of age.

Methods

In total, 638 patients aged 18 to 60 years who resided in Olmsted County and had a first lifetime diagnosis of cutaneous melanoma between 1970 and 2009 were retrospectively identified as a part of the Rochester Epidemiology Project (REP). The REP is a health records linkage system that encompasses almost all sources of medical care available to the local population of Olmsted County.¹¹ This study was approved by the Mayo Clinic Institutional Review Board (Rochester, Minnesota).

Of the 638 individuals identified in the REP, 536 had been seen at Mayo Clinic and thus potentially had tissue blocks available for the study of BRAF mutation expression. Of these 536 patients, 156 did not have sufficient residual tissue available. As a result, 380 (60%) of the original 638 patients had available blocks with sufficient tissue for immunohistochemical analysis of BRAF expression. Only primary cutaneous melanomas were included in the present study.

All specimens were reviewed by a board-certified dermatopathologist (J.S.L.) for appropriateness of inclusion, which involved confirmation of the diagnosis of melanoma, histologic type of melanoma, and presence of sufficient residual tissue for immunohistochemical stains.

All specimens were originally diagnosed as malignant melanoma at the time of clinical care by at least 2 board-certified dermatopathologists. For the purposes of this study, all specimens were rereviewed for diagnostic accuracy. We required that specimens exhibit severe cytologic and architectural atypia as well as other features favoring melanoma, such as consumption of rete pegs, pagetosis, confluence of junctional melanocytes, evidence of regression, lack of maturation of melanocytes with descent into the dermis, or mitotic figures among the dermal melanocyte population.

The available tissue blocks were retrieved, sectioned, confirmed as melanoma, and stained with a mouse antihuman BRAF V600E monoclonal antibody (clone VE1; Spring Bioscience) to determine the presence of a BRAF V600E mutation. BRAF staining was evaluated in conjunction with a review of the associated slides stained with hematoxylin and eosin. Cytoplasmic staining of melanocytes for BRAF was graded as negative, focal or partial positive (<50% of tumor), or diffuse positive (>50% of tumor)(Figure 1). When a melanoma arose in association with a nevus, we considered only the melanoma component for BRAF staining. We categorized the histologic type as superficial spreading, nodular, or lentigo maligna, and the location as head and neck, trunk, or extremities.

 

Patient characteristics and survival outcomes were gathered through the health record and included age, Breslow thickness, location, decade of diagnosis, histologic type, stage (ie, noninvasive, invasive, or advanced), and follow-up. Pathologic stage 0 was considered noninvasive; stages IA and IB, invasive; and stages IIA or higher, advanced.

Statistical Analysis—Comparisons between the group of patients in the study (n=380) and the group of patients excluded for the reasons stated above (n=258) as well as associations of mutant BRAF status (positive [partial positive and diffuse positive] vs negative) with patient age (young adults [age range, 18–39 years] and middle-aged adults [age range, 40–60 years]), sex, decade of diagnosis, location, histologic type, and stage were evaluated with Wilcoxon rank sum, χ², Fisher exact, or Cochran-Armitage trend tests. Disease-specific survival and overall survival rates were estimated with the Kaplan-Meier method, and the duration of follow-up was calculated from the date of melanoma diagnosis to the date of death or the last follow-up. Associations of mutant BRAF expression status with death from melanoma and death from any cause were evaluated with Cox proportional hazard regression models and summarized with hazard ratio (HR) and 95% CI. Survival analyses were limited to patients with invasive or advanced disease. Statistical analyses were performed with SAS statistical software (SAS version 9.4). All tests were 2-sided, and P<.05 was considered statistically significant.

Results

Clinical and Tumor Characteristics—Of the 380 tissue specimens that underwent BRAF V600E analysis, 247 had negative staining; 106 had diffuse strong staining; and 27 had focal or partial staining. In total, 133 (35%) were positive, either partially or diffusely. The median age for patients who had negative staining was 45 years; for those with positive staining, it was 41 years (P=.07).

The patients who met inclusion criteria (n=380) were compared with those who were excluded (n=258)(eTable 1). The groups were similar on the basis of sex; age; and melanoma location, stage, and histologic subtype. However, some evidence showed that patients included in the study received the diagnosis of melanoma more recently (1970-1989, 13.2%; 1990-1999, 28.7%; 2000-2009, 58.2%) than those who were excluded (1970-1989, 24.7%; 1990-1999, 23.5%; 2000-2009, 51.8%)(P=.02).

BRAF V600E expression was more commonly found in superficial spreading (37.7%) and nodular melanomas (35.0%) than in situ melanomas (17.1%)(P=.01). Other characteristics of BRAF V600E expression are described in eTable 2. Overall, invasive and advanced melanomas were significantly more likely to harbor BRAF V600E expression than noninvasive melanomas (39.6% and 37.9%, respectively, vs 17.9%; P=.003). However, advanced melanomas more commonly expressed BRAF positivity among women, and invasive melanomas more commonly expressed BRAF positivity among men (eTable 2).

Survival—Survival analyses were limited to 297 patients with confirmed invasive or advanced disease. Of these, 180 (61%) had no BRAF V600E staining; 25 (8%) had partial staining; and 92 (31%) had diffuse positive staining. In total, 117 patients (39%) had a BRAF-mutated melanoma.

Among the patients still alive, the median (interquartile range [IQR]) duration of follow-up was 10.2 (7.0-16.8) years. Thirty-nine patients with invasive or advanced disease had died of any cause at a median (IQR) of 3.0 (1.3-10.2) years after diagnosis. In total, 26 patients died of melanoma at a median (IQR) follow-up of 2.5 (1.3-7.4) years after diagnosis. Eight women and 18 men died of malignant melanoma. Five deaths occurred because of malignant melanoma among patients aged 18 to 39 years, and 21 occurred among patients aged 40 to 60 years. In the 18- to 39-year-old group, all 5 deaths were among patients with a BRAF-positive melanoma. Estimated disease-specific survival rate (95% CI; number still at risk) at 5, 10, 15, and 20 years after diagnosis was 94% (91%-97%; 243), 91% (87%-95%; 142), 89% (85%-94%; 87), and 88% (83%-93%; 45), respectively.

 

In a univariable analysis, the HR for association of positive mutant BRAF expression with death of malignant melanoma was 1.84 (95% CI, 0.85-3.98; P=.12). No statistically significant interaction was observed between decade of diagnosis and BRAF expression (P=.60). However, the interaction between sex and BRAF expression was significant (P=.04), with increased risk of death from melanoma among women with BRAF-mutated melanoma (HR, 10.88; 95% CI, 1.34-88.41; P=.026) but not among men (HR 1.02; 95% CI, 0.40-2.64; P=.97)(Figures 2A and 2B). The HR for death from malignant melanoma among young adults aged 18 to 39 years with a BRAF-mutated melanoma was 16.4 (95% CI, 0.81-330.10; P=.068), whereas the HR among adults aged 40 to 60 years with a BRAF-mutated melanoma was 1.24 (95% CI, 0.52-2.98; P=.63)(Figures 2C and 2D).

BRAF V600E expression was not significantly associated with death from any cause (HR, 1.39; 95% CI, 0.74-2.61; P=.31) or with decade of diagnosis (P=.13). Similarly, BRAF expression was not associated with death from any cause according to sex (P=.31). However, a statistically significant interaction was seen between age at diagnosis and BRAF expression (P=.003). BRAF expression was significantly associated with death from any cause for adults aged 18 to 39 years (HR, 9.60; 95% CI, 1.15-80.00; P=.04). In comparison, no association of BRAF expression with death was observed for adults aged 40 to 60 years (HR, 0.99; 95% CI, 0.48-2.03; P=.98).

Comment

We found that melanomas with BRAF mutations were more likely in advanced and invasive melanoma. The frequency of BRAF mutations among melanomas that were considered advanced was higher in women than men. Although the number of deaths was limited, women with a melanoma with BRAF expression were more likely to die of melanoma, young adults with a BRAF-mutated melanoma had an almost 10-fold increased risk of dying from any cause, and middle-aged adults showed no increased risk of death. These findings suggest that young adults who are genetically prone to a BRAF-mutated melanoma could be at a disadvantage for all-cause mortality. Although this finding was significant, the 95% CI was large, and further studies would be warranted before sound conclusions could be made.

Melanoma has been increasing in incidence across all age groups in Olmsted County over the last 4 decades.^12-14 However, our results show that the percentage of BRAF-mutated melanomas in this population has been stable over time, with no statistically significant difference by age or sex. Other confounding factors may have an influence, such as increased rates of early detection and diagnosis of melanoma in contemporary times. Our data suggest that patients included in the BRAF-mutation analysis study had received the diagnosis of melanoma more recently than those who were excluded from the study, which could be due to older melanomas being less likely to have adequate tissue specimens available for immunohistochemical staining/evaluation.

Prior research has shown that BRAF-mutated melanomas typically occur on the trunk and are more likely in individuals with more than 14 nevi on the back.² In the present cohort, BRAF-positive melanomas had a predisposition toward the trunk but also were found on the head, neck, and extremities—areas that are more likely to have long-term sun damage. One suggestion is that 2 distinct pathways for melanoma development exist: one associated with a large number of melanocytic nevi (that is more prone to genetic mutations in melanocytes) and the other associated with long-term sun exposure.^15,16 The combination of these hypotheses suggests that individuals who are prone to the development of large numbers of nevi may require sun exposure for the initial insult, but the development of melanoma may be carried out by other factors after this initial sun exposure insult, whereas individuals without large numbers of nevi who may have less genetic risk may require continued long-term sun exposure for melanoma to develop.¹⁷

Our study had limitations, including the small numbers of deaths overall and cause-specific deaths of metastatic melanoma, which limited our ability to conduct more extensive multivariable modeling. Also, the retrospective nature and time frame of looking back 4 decades did not allow us to have information sufficient to categorize some patients as having dysplastic nevus syndrome or not, which would be a potentially interesting variable to include in the analysis. Because the number of deaths in the 18- to 39-year-old cohort was only 5, further statistical comparison regarding tumor type and other variables pertaining to BRAF positivity were not possible. In addition, our data were collected from patients residing in a single geographic county (Olmsted County, Minnesota), which may limit generalizability. Lastly, BRAF V600E mutations were identified through immunostaining only, not molecular data, so it is possible some patients had false-negative immunohistochemistry findings and thus were not identified.

Conclusion

BRAF-mutated melanomas were found in 35% of our cohort, with no significant change in the percentage of melanomas with BRAF V600E mutations over the last 4 decades in this population. In addition, no differences or significant trends existed according to sex and BRAF-mutated melanoma development. Women with BRAF-mutated melanomas were more likely to die of metastatic melanoma than men, and young adults with BRAF-mutated melanomas had a higher all-cause mortality risk. Further research is needed to decipher what effect BRAF-mutated melanomas have on metastasis and cause-specific death in women as well as all-cause mortality in young adults.

Acknowledgment—The authors are indebted to Scientific Publications, Mayo Clinic (Rochester, Minnesota).

The COVID-19 pandemic has brought about unprecedented changes and challenges to medical practice, including new public health measure legislation, local and national medical authority recommendations, nursing home and other ancillary health center protocols, and novel clinical decision-making considerations.^1-3 In July 2020, the American Academy of Dermatology (AAD) addressed the changing landscape in dermatologic surgery, in part, by publishing recommendations on practice protocols during the COVID-19 pandemic.⁴ The guidelines recommended deferred treatment of superficial basal cell carcinomas (BCCs) for 6 months and all other BCC subtypes for 3 to 6 months. Furthermore, the guidelines recommended deferring treatment of all actinic keratoses and squamous cell carcinomas (SCCs) in situ “for now.” Squamous cell carcinoma treatment was to be guided by prognostic variables, such as location, size, depth, differentiation, perineural or lymphovascular invasion, recurrence, and immunosuppression. The guidelines recommended melanoma in situ (MIS) treatment be deferred for 3 months and invasive melanoma with histologic clearance obtained on excisional biopsy for 3 months. Other general recommendations included triaging clinics, rebooking according to clinical priority, using telehealth where possible, screening patients for COVID-19 signs and symptoms, staggering appointment times, spacing patient chairs, limiting support persons to 1, removing possible sources of infection in the waiting room, ensuring all patients sanitized their hands on arrival, rationing personal protective equipment, considering N95 masks for periorificial surgery, and using dissolving sutures to minimize multiple presentations.⁴

The American College of Mohs Surgery (ACMS), with guidance from its sister societies and the National Comprehensive Cancer Network, also communicated COVID-19–related recommendations to its members via intermittent newsletters during the initial peak of the pandemic in March and June 2020.⁵ General social distancing and office recommendations were similar to those released by the AAD. Recommendations for skin cancer treatment included deferring all BCCs for up to 3 months, with exceptions for highly symptomatic cancers and those with potential for substantial rapid growth. Squamous cell carcinoma in situ and small, well-differentiated SCCs were deferred, with priority placed on SCCs that were rapidly enlarging, poorly differentiated, demonstrated perineural invasion, were ulcerated, or were symptomatic. Patients with major risk factors were prioritized for treatment. Melanoma in situ was deferred for 2 to 3 months.⁵

State-level guidance from the Texas Dermatological Society (TDS) communicated in April 2020 stated that skin cancers with a potential for rapid progression and metastasis, such as melanoma and SCC, may require treatment as determined by the physician.⁶ The potential risk of serious adverse medical outcomes from not treating these cancers should be carefully documented. General practice measures for preventing the spread of COVID-19 were also recommended.⁶

In the setting of emerging novel recommendations, the practice of Mohs micrographic surgery (MMS) was notably impacted by the COVID-19 pandemic. According to one survey study from the United Kingdom conducted in April and May 2020, 49% of MMS services ceased and 36% were reduced during the infancy of the COVID-19 pandemic.⁷ Mohs micrographic surgery was largely suspended because of a lack of personal protective equipment and safety concerns, according to respondents. Additionally, respondents reported 77% of departments experienced redeployment of physicians and nurses to intensive care and medical wards. Thirty-five percent reported a reduction in the proportion of flaps/grafts to primary closures performed, 74% reported a decrease in outside referrals for repair by other specialties, 81% reported increased usage of dissolvable sutures, and 29% reported an increase in prophylactic antibiotic prescriptions.⁷ Another study from Italy reported a 46.5% reduction in dermatologic surgeries performed during the initial lockdown of the COVID-19 pandemic. Patients canceled 52.9% of procedures, and 12.5% were cancelled because of confirmed or suspected COVID-19 infection.⁸ Patient perceptions of MMS have also been impacted by the COVID-19 pandemic. According to a survey study of patients in the United Kingdom undergoing MMS during the pandemic, 47% were worried the hospital would cancel their surgery, 54% were anxious about using public transportation to attend their appointment, 30% were concerned about transmitting COVID-19 to household or family members, and 19% were worried about their ability to socially distance in the hospital.⁹

Evidence is also emerging that suggests the potential negative impact of the COVID-19 pandemic on morbidity and mortality outcomes in patients with skin cancer. One European study found an increase in Breslow thickness in primary melanomas diagnosed following the initial COVID-19 lockdown (0.88-mm average thickness prelockdown vs 1.96-mm average thickness postlockdown).¹⁰ An Italian study observed similar results—an increase in median Breslow thickness during the initial COVID-19 lockdown period of 0.5 mm from 0.4 mm during the prelockdown time period.¹¹ Also providing evidence for potentially poor patient outcomes, one study modeled the impact of backlog in cutaneous melanoma referrals in the United Kingdom on patient survival and predicted 138 attributable lives lost for a 1-month delay and 1171 lives lost for a 6-month delay. The model further predicted a 3.1% to 12.5% reduction in 10-year net survival incurred from a 3-month delay in melanoma treatment, with the largest reduction seen in the patient population older than 80 years.¹²

Although the COVID-19 pandemic has been observed to impact MMS practice, patient perceptions, and clinical outcomes, it is unknown how the COVID-19 pandemic and corresponding rapidly evolving recommendations in dermatologic surgery have impacted the characteristics of cutaneous tumors treated by MMS.

Our study sought to determine the characteristics of skin cancers treated by MMS during the peak of government-mandated medical practice restrictions and business shutdowns in response to the COVID-19 pandemic and to compare them with characteristics of skin cancers treated during a prepandemic control period.

 

Methods

A retrospective chart review was conducted with approval from our institutional review board at the University of Texas Medical Branch (Galveston, Texas). Included in the chart review were all cutaneous malignancies treated by MMS at our outpatient, office-based surgical center from March 15, 2020, to April 30, 2020; this period corresponded to the peak of the COVID-19–related government-mandated medical and business shutdowns in our geographic region (southeast Texas). All cases performed were in compliance with national- and state-level guidance. Data were also collected for all cutaneous malignancies treated by MMS at our office from March 15, 2019, to April 30, 2019, as well as March 15, 2018, to April 30, 2018; these periods represented prepandemic control periods.

Data were collected for 516 surgeries performed on 458 patients and included patient age, preoperative clinical size, postoperative defect size, number of Mohs stages to achieve clearance, MMS appropriate use criteria (AUC) location (categorized as high-, medium-, or low-risk tumor location),¹³ and tumor type (categorized as BCC, SCC, or MIS). All variables were examined for unusual or missing values. Five patients with rare tumor types were observed and removed from the data set.

Statistical Analysis—An a priori power analysis for a power set at 0.85 determined sample sizes of 105 per group. Bivariate analyses were performed to compare variables for patients undergoing MMS during the pandemic vs prepandemic periods. Continuous outcome variables—Mohs stages, preoperative size, postoperative size, and patient age—were categorized for the analysis. Preoperative tumor size was dichotomized, with less than 2 cm² as the referent category vs 2 cm² or greater, and postoperative defect size was dichotomized with less than 3.6 cm² as the referent category vs 3.6 cm² or greater. Mohs stage was dichotomized as 1 stage (referent) vs more than 1 stage, and patient age was dichotomized as younger than 65 years (referent) vs 65 years or older.

Multivariate analyses were also performed to compare preoperative and postoperative sizes for patients undergoing MMS during the pandemic vs prepandemic periods, controlling for Mohs AUC location. Bivariate unadjusted and multivariate analyses were performed using a GENMOD logistic regression procedure in SAS (SAS Institute) to account for correlation in clustered data because a patient could be included for more than 1 surgery in the data set. Data were analyzed using SAS 9.4 for Windows. Because outcome variables tended to be skewed and not distributed normally, outcome variables were recorded as medians with interquartile ranges where possible to give a more accurate representation of the data than could be demonstrated with means with standard deviations.

Results

One hundred thirty-eight skin cancers were treated during the COVID-19 pandemic from March 15, 2020, to April 30, 2020, and 378 skin cancers were treated during the prepandemic control periods of March 15, 2019, to April 30, 2019, and March 15, 2018, to April 30, 2018. Tumor type treated during the pandemic period was more likely to be SCC or MIS (representing generally more severe tumor types) vs BCC when compared with the prepandemic periods, with an odds ratio (OR) of 1.763 (95% CI, 1.17-2.66). This outcome was statistically significant (P=.01).

Tumors treated during the pandemic period were more likely to have necessitated more than one Mohs stage for clearance compared to the prepandemic periods, though this difference was not statistically significant (OR, 1.461; 95% CI, 0.97-2.19; P=.056). Neither AUC location of treated tumors nor age were significantly different between prepandemic and pandemic periods (P=.58 and P=.84, respectively). Table 1 includes all bivariate analysis results.

Additionally, although mean preoperative and postoperative sizes were larger for each AUC location during the pandemic vs prepandemic periods, these differences did not reach statistical significance on multivariate analysis (P=.71 and P=.50, respectively)(Table 2).

 

 

Comment

Our practice has followed best practice guidelines dictated by our governing professional societies during the COVID-19 pandemic in the treatment of skin cancers by MMS, specifically highly symptomatic BCCs (in accordance with ACMS guidance), SCCs with high-risk features (in accordance with AAD, ACMS, and TDS guidance), and tumors with high risk for progression and metastasis such as melanomas (in accordance with TDS guidance). Melanoma in situ was also treated during the COVID-19 pandemic in accordance with the latter TDS guidance, particularly in light of the potential for upstaging to melanoma following resection (a phenomenon demonstrated to occur in 5%–29% of biopsied MIS lesions).¹⁴

In following best practice guidelines, our results suggested tumors treated by MMS were more severe, as evidenced by a statistically significant higher proportion of SCC and MIS tumors (representing more severe tumor types) vs BCC when compared to the prepandemic period. Supporting this conclusion, we observed larger pretreatment and posttreatment tumor sizes for all AUC locations and more tumors necessitating 2 or more stages for clearance during the pandemic vs prepandemic periods, though these differences did not reach statistical significance. We postulate these findings may be attributed to allocation of finite medical resources to the treatment of larger and more aggressive skin cancers. Additionally, these findings may be explained, in part, by limitations on patient case load imposed by social distancing measures and governing body regulations in effect during the study period, including those put forth by the AAD, ACMS, and TDS. Of note, our practice observed no hospitalizations or 911 calls during the studied period. This suggests no allocation of precious hospital resources away from patients with COVID-19 in our treatment of high-risk skin cancers.

The changing characteristics of cutaneous tumors treated by MMS during the pandemic are of clinical relevance. Larger postoperative wound sizes as observed during the pandemic, albeit not statistically significant, presumably affect reconstructive decisions. With larger wounds tending to necessitate repair by techniques higher on the reconstructive ladder, greater patient morbidity and cost are expected.¹⁵ As the cost-effectiveness of dermatology services remains a critical issue, this is an area ripe for future follow-up research. Furthermore, our observation that tumors tended to necessitate 2 or more stages for clearance during the pandemic more often than prepandemic periods, though not statistically significant, presumably affected operating times. Longer operating times during the pandemic may be of importance when making clinical decisions for patients for whom limiting health care exposure may be of particular concern. With more SCC and MIS tumors being treated relative to BCCs during the pandemic, one might expect greater size and severity of the BCCs we observe in the proceeding months to years.

As the ongoing COVID-19 pandemic continues to impact the landscape of cutaneous oncology, the need for adaptability is imperative. With 3- and 6-month skin cancer treatment deferrals lapsed, uncertainty surrounds ideal management of existing and new skin cancers arising during the pandemic. This study adds to a growing body of literature elucidating the impact of the COVID-19 pandemic on MMS practice; however, further studies and a tincture of time are needed to guide future best practice standards.

Acknowledgment—The authors acknowledge Gwen Baillargeon, MS (Galveston, Texas), who was the statistician for this article.

The COVID-19 pandemic has brought about unprecedented changes and challenges to medical practice, including new public health measure legislation, local and national medical authority recommendations, nursing home and other ancillary health center protocols, and novel clinical decision-making considerations.^1-3 In July 2020, the American Academy of Dermatology (AAD) addressed the changing landscape in dermatologic surgery, in part, by publishing recommendations on practice protocols during the COVID-19 pandemic.⁴ The guidelines recommended deferred treatment of superficial basal cell carcinomas (BCCs) for 6 months and all other BCC subtypes for 3 to 6 months. Furthermore, the guidelines recommended deferring treatment of all actinic keratoses and squamous cell carcinomas (SCCs) in situ “for now.” Squamous cell carcinoma treatment was to be guided by prognostic variables, such as location, size, depth, differentiation, perineural or lymphovascular invasion, recurrence, and immunosuppression. The guidelines recommended melanoma in situ (MIS) treatment be deferred for 3 months and invasive melanoma with histologic clearance obtained on excisional biopsy for 3 months. Other general recommendations included triaging clinics, rebooking according to clinical priority, using telehealth where possible, screening patients for COVID-19 signs and symptoms, staggering appointment times, spacing patient chairs, limiting support persons to 1, removing possible sources of infection in the waiting room, ensuring all patients sanitized their hands on arrival, rationing personal protective equipment, considering N95 masks for periorificial surgery, and using dissolving sutures to minimize multiple presentations.⁴

The American College of Mohs Surgery (ACMS), with guidance from its sister societies and the National Comprehensive Cancer Network, also communicated COVID-19–related recommendations to its members via intermittent newsletters during the initial peak of the pandemic in March and June 2020.⁵ General social distancing and office recommendations were similar to those released by the AAD. Recommendations for skin cancer treatment included deferring all BCCs for up to 3 months, with exceptions for highly symptomatic cancers and those with potential for substantial rapid growth. Squamous cell carcinoma in situ and small, well-differentiated SCCs were deferred, with priority placed on SCCs that were rapidly enlarging, poorly differentiated, demonstrated perineural invasion, were ulcerated, or were symptomatic. Patients with major risk factors were prioritized for treatment. Melanoma in situ was deferred for 2 to 3 months.⁵

State-level guidance from the Texas Dermatological Society (TDS) communicated in April 2020 stated that skin cancers with a potential for rapid progression and metastasis, such as melanoma and SCC, may require treatment as determined by the physician.⁶ The potential risk of serious adverse medical outcomes from not treating these cancers should be carefully documented. General practice measures for preventing the spread of COVID-19 were also recommended.⁶

In the setting of emerging novel recommendations, the practice of Mohs micrographic surgery (MMS) was notably impacted by the COVID-19 pandemic. According to one survey study from the United Kingdom conducted in April and May 2020, 49% of MMS services ceased and 36% were reduced during the infancy of the COVID-19 pandemic.⁷ Mohs micrographic surgery was largely suspended because of a lack of personal protective equipment and safety concerns, according to respondents. Additionally, respondents reported 77% of departments experienced redeployment of physicians and nurses to intensive care and medical wards. Thirty-five percent reported a reduction in the proportion of flaps/grafts to primary closures performed, 74% reported a decrease in outside referrals for repair by other specialties, 81% reported increased usage of dissolvable sutures, and 29% reported an increase in prophylactic antibiotic prescriptions.⁷ Another study from Italy reported a 46.5% reduction in dermatologic surgeries performed during the initial lockdown of the COVID-19 pandemic. Patients canceled 52.9% of procedures, and 12.5% were cancelled because of confirmed or suspected COVID-19 infection.⁸ Patient perceptions of MMS have also been impacted by the COVID-19 pandemic. According to a survey study of patients in the United Kingdom undergoing MMS during the pandemic, 47% were worried the hospital would cancel their surgery, 54% were anxious about using public transportation to attend their appointment, 30% were concerned about transmitting COVID-19 to household or family members, and 19% were worried about their ability to socially distance in the hospital.⁹

Evidence is also emerging that suggests the potential negative impact of the COVID-19 pandemic on morbidity and mortality outcomes in patients with skin cancer. One European study found an increase in Breslow thickness in primary melanomas diagnosed following the initial COVID-19 lockdown (0.88-mm average thickness prelockdown vs 1.96-mm average thickness postlockdown).¹⁰ An Italian study observed similar results—an increase in median Breslow thickness during the initial COVID-19 lockdown period of 0.5 mm from 0.4 mm during the prelockdown time period.¹¹ Also providing evidence for potentially poor patient outcomes, one study modeled the impact of backlog in cutaneous melanoma referrals in the United Kingdom on patient survival and predicted 138 attributable lives lost for a 1-month delay and 1171 lives lost for a 6-month delay. The model further predicted a 3.1% to 12.5% reduction in 10-year net survival incurred from a 3-month delay in melanoma treatment, with the largest reduction seen in the patient population older than 80 years.¹²

Although the COVID-19 pandemic has been observed to impact MMS practice, patient perceptions, and clinical outcomes, it is unknown how the COVID-19 pandemic and corresponding rapidly evolving recommendations in dermatologic surgery have impacted the characteristics of cutaneous tumors treated by MMS.

Our study sought to determine the characteristics of skin cancers treated by MMS during the peak of government-mandated medical practice restrictions and business shutdowns in response to the COVID-19 pandemic and to compare them with characteristics of skin cancers treated during a prepandemic control period.

 

Methods

A retrospective chart review was conducted with approval from our institutional review board at the University of Texas Medical Branch (Galveston, Texas). Included in the chart review were all cutaneous malignancies treated by MMS at our outpatient, office-based surgical center from March 15, 2020, to April 30, 2020; this period corresponded to the peak of the COVID-19–related government-mandated medical and business shutdowns in our geographic region (southeast Texas). All cases performed were in compliance with national- and state-level guidance. Data were also collected for all cutaneous malignancies treated by MMS at our office from March 15, 2019, to April 30, 2019, as well as March 15, 2018, to April 30, 2018; these periods represented prepandemic control periods.

Data were collected for 516 surgeries performed on 458 patients and included patient age, preoperative clinical size, postoperative defect size, number of Mohs stages to achieve clearance, MMS appropriate use criteria (AUC) location (categorized as high-, medium-, or low-risk tumor location),¹³ and tumor type (categorized as BCC, SCC, or MIS). All variables were examined for unusual or missing values. Five patients with rare tumor types were observed and removed from the data set.

Statistical Analysis—An a priori power analysis for a power set at 0.85 determined sample sizes of 105 per group. Bivariate analyses were performed to compare variables for patients undergoing MMS during the pandemic vs prepandemic periods. Continuous outcome variables—Mohs stages, preoperative size, postoperative size, and patient age—were categorized for the analysis. Preoperative tumor size was dichotomized, with less than 2 cm² as the referent category vs 2 cm² or greater, and postoperative defect size was dichotomized with less than 3.6 cm² as the referent category vs 3.6 cm² or greater. Mohs stage was dichotomized as 1 stage (referent) vs more than 1 stage, and patient age was dichotomized as younger than 65 years (referent) vs 65 years or older.

Multivariate analyses were also performed to compare preoperative and postoperative sizes for patients undergoing MMS during the pandemic vs prepandemic periods, controlling for Mohs AUC location. Bivariate unadjusted and multivariate analyses were performed using a GENMOD logistic regression procedure in SAS (SAS Institute) to account for correlation in clustered data because a patient could be included for more than 1 surgery in the data set. Data were analyzed using SAS 9.4 for Windows. Because outcome variables tended to be skewed and not distributed normally, outcome variables were recorded as medians with interquartile ranges where possible to give a more accurate representation of the data than could be demonstrated with means with standard deviations.

Results

One hundred thirty-eight skin cancers were treated during the COVID-19 pandemic from March 15, 2020, to April 30, 2020, and 378 skin cancers were treated during the prepandemic control periods of March 15, 2019, to April 30, 2019, and March 15, 2018, to April 30, 2018. Tumor type treated during the pandemic period was more likely to be SCC or MIS (representing generally more severe tumor types) vs BCC when compared with the prepandemic periods, with an odds ratio (OR) of 1.763 (95% CI, 1.17-2.66). This outcome was statistically significant (P=.01).

Tumors treated during the pandemic period were more likely to have necessitated more than one Mohs stage for clearance compared to the prepandemic periods, though this difference was not statistically significant (OR, 1.461; 95% CI, 0.97-2.19; P=.056). Neither AUC location of treated tumors nor age were significantly different between prepandemic and pandemic periods (P=.58 and P=.84, respectively). Table 1 includes all bivariate analysis results.

Additionally, although mean preoperative and postoperative sizes were larger for each AUC location during the pandemic vs prepandemic periods, these differences did not reach statistical significance on multivariate analysis (P=.71 and P=.50, respectively)(Table 2).

 

 

Comment

Our practice has followed best practice guidelines dictated by our governing professional societies during the COVID-19 pandemic in the treatment of skin cancers by MMS, specifically highly symptomatic BCCs (in accordance with ACMS guidance), SCCs with high-risk features (in accordance with AAD, ACMS, and TDS guidance), and tumors with high risk for progression and metastasis such as melanomas (in accordance with TDS guidance). Melanoma in situ was also treated during the COVID-19 pandemic in accordance with the latter TDS guidance, particularly in light of the potential for upstaging to melanoma following resection (a phenomenon demonstrated to occur in 5%–29% of biopsied MIS lesions).¹⁴

In following best practice guidelines, our results suggested tumors treated by MMS were more severe, as evidenced by a statistically significant higher proportion of SCC and MIS tumors (representing more severe tumor types) vs BCC when compared to the prepandemic period. Supporting this conclusion, we observed larger pretreatment and posttreatment tumor sizes for all AUC locations and more tumors necessitating 2 or more stages for clearance during the pandemic vs prepandemic periods, though these differences did not reach statistical significance. We postulate these findings may be attributed to allocation of finite medical resources to the treatment of larger and more aggressive skin cancers. Additionally, these findings may be explained, in part, by limitations on patient case load imposed by social distancing measures and governing body regulations in effect during the study period, including those put forth by the AAD, ACMS, and TDS. Of note, our practice observed no hospitalizations or 911 calls during the studied period. This suggests no allocation of precious hospital resources away from patients with COVID-19 in our treatment of high-risk skin cancers.

The changing characteristics of cutaneous tumors treated by MMS during the pandemic are of clinical relevance. Larger postoperative wound sizes as observed during the pandemic, albeit not statistically significant, presumably affect reconstructive decisions. With larger wounds tending to necessitate repair by techniques higher on the reconstructive ladder, greater patient morbidity and cost are expected.¹⁵ As the cost-effectiveness of dermatology services remains a critical issue, this is an area ripe for future follow-up research. Furthermore, our observation that tumors tended to necessitate 2 or more stages for clearance during the pandemic more often than prepandemic periods, though not statistically significant, presumably affected operating times. Longer operating times during the pandemic may be of importance when making clinical decisions for patients for whom limiting health care exposure may be of particular concern. With more SCC and MIS tumors being treated relative to BCCs during the pandemic, one might expect greater size and severity of the BCCs we observe in the proceeding months to years.

As the ongoing COVID-19 pandemic continues to impact the landscape of cutaneous oncology, the need for adaptability is imperative. With 3- and 6-month skin cancer treatment deferrals lapsed, uncertainty surrounds ideal management of existing and new skin cancers arising during the pandemic. This study adds to a growing body of literature elucidating the impact of the COVID-19 pandemic on MMS practice; however, further studies and a tincture of time are needed to guide future best practice standards.

Acknowledgment—The authors acknowledge Gwen Baillargeon, MS (Galveston, Texas), who was the statistician for this article.

The COVID-19 pandemic has brought about unprecedented changes and challenges to medical practice, including new public health measure legislation, local and national medical authority recommendations, nursing home and other ancillary health center protocols, and novel clinical decision-making considerations.^1-3 In July 2020, the American Academy of Dermatology (AAD) addressed the changing landscape in dermatologic surgery, in part, by publishing recommendations on practice protocols during the COVID-19 pandemic.⁴ The guidelines recommended deferred treatment of superficial basal cell carcinomas (BCCs) for 6 months and all other BCC subtypes for 3 to 6 months. Furthermore, the guidelines recommended deferring treatment of all actinic keratoses and squamous cell carcinomas (SCCs) in situ “for now.” Squamous cell carcinoma treatment was to be guided by prognostic variables, such as location, size, depth, differentiation, perineural or lymphovascular invasion, recurrence, and immunosuppression. The guidelines recommended melanoma in situ (MIS) treatment be deferred for 3 months and invasive melanoma with histologic clearance obtained on excisional biopsy for 3 months. Other general recommendations included triaging clinics, rebooking according to clinical priority, using telehealth where possible, screening patients for COVID-19 signs and symptoms, staggering appointment times, spacing patient chairs, limiting support persons to 1, removing possible sources of infection in the waiting room, ensuring all patients sanitized their hands on arrival, rationing personal protective equipment, considering N95 masks for periorificial surgery, and using dissolving sutures to minimize multiple presentations.⁴

The American College of Mohs Surgery (ACMS), with guidance from its sister societies and the National Comprehensive Cancer Network, also communicated COVID-19–related recommendations to its members via intermittent newsletters during the initial peak of the pandemic in March and June 2020.⁵ General social distancing and office recommendations were similar to those released by the AAD. Recommendations for skin cancer treatment included deferring all BCCs for up to 3 months, with exceptions for highly symptomatic cancers and those with potential for substantial rapid growth. Squamous cell carcinoma in situ and small, well-differentiated SCCs were deferred, with priority placed on SCCs that were rapidly enlarging, poorly differentiated, demonstrated perineural invasion, were ulcerated, or were symptomatic. Patients with major risk factors were prioritized for treatment. Melanoma in situ was deferred for 2 to 3 months.⁵

State-level guidance from the Texas Dermatological Society (TDS) communicated in April 2020 stated that skin cancers with a potential for rapid progression and metastasis, such as melanoma and SCC, may require treatment as determined by the physician.⁶ The potential risk of serious adverse medical outcomes from not treating these cancers should be carefully documented. General practice measures for preventing the spread of COVID-19 were also recommended.⁶

In the setting of emerging novel recommendations, the practice of Mohs micrographic surgery (MMS) was notably impacted by the COVID-19 pandemic. According to one survey study from the United Kingdom conducted in April and May 2020, 49% of MMS services ceased and 36% were reduced during the infancy of the COVID-19 pandemic.⁷ Mohs micrographic surgery was largely suspended because of a lack of personal protective equipment and safety concerns, according to respondents. Additionally, respondents reported 77% of departments experienced redeployment of physicians and nurses to intensive care and medical wards. Thirty-five percent reported a reduction in the proportion of flaps/grafts to primary closures performed, 74% reported a decrease in outside referrals for repair by other specialties, 81% reported increased usage of dissolvable sutures, and 29% reported an increase in prophylactic antibiotic prescriptions.⁷ Another study from Italy reported a 46.5% reduction in dermatologic surgeries performed during the initial lockdown of the COVID-19 pandemic. Patients canceled 52.9% of procedures, and 12.5% were cancelled because of confirmed or suspected COVID-19 infection.⁸ Patient perceptions of MMS have also been impacted by the COVID-19 pandemic. According to a survey study of patients in the United Kingdom undergoing MMS during the pandemic, 47% were worried the hospital would cancel their surgery, 54% were anxious about using public transportation to attend their appointment, 30% were concerned about transmitting COVID-19 to household or family members, and 19% were worried about their ability to socially distance in the hospital.⁹

Evidence is also emerging that suggests the potential negative impact of the COVID-19 pandemic on morbidity and mortality outcomes in patients with skin cancer. One European study found an increase in Breslow thickness in primary melanomas diagnosed following the initial COVID-19 lockdown (0.88-mm average thickness prelockdown vs 1.96-mm average thickness postlockdown).¹⁰ An Italian study observed similar results—an increase in median Breslow thickness during the initial COVID-19 lockdown period of 0.5 mm from 0.4 mm during the prelockdown time period.¹¹ Also providing evidence for potentially poor patient outcomes, one study modeled the impact of backlog in cutaneous melanoma referrals in the United Kingdom on patient survival and predicted 138 attributable lives lost for a 1-month delay and 1171 lives lost for a 6-month delay. The model further predicted a 3.1% to 12.5% reduction in 10-year net survival incurred from a 3-month delay in melanoma treatment, with the largest reduction seen in the patient population older than 80 years.¹²

Although the COVID-19 pandemic has been observed to impact MMS practice, patient perceptions, and clinical outcomes, it is unknown how the COVID-19 pandemic and corresponding rapidly evolving recommendations in dermatologic surgery have impacted the characteristics of cutaneous tumors treated by MMS.

Our study sought to determine the characteristics of skin cancers treated by MMS during the peak of government-mandated medical practice restrictions and business shutdowns in response to the COVID-19 pandemic and to compare them with characteristics of skin cancers treated during a prepandemic control period.

 

Methods

A retrospective chart review was conducted with approval from our institutional review board at the University of Texas Medical Branch (Galveston, Texas). Included in the chart review were all cutaneous malignancies treated by MMS at our outpatient, office-based surgical center from March 15, 2020, to April 30, 2020; this period corresponded to the peak of the COVID-19–related government-mandated medical and business shutdowns in our geographic region (southeast Texas). All cases performed were in compliance with national- and state-level guidance. Data were also collected for all cutaneous malignancies treated by MMS at our office from March 15, 2019, to April 30, 2019, as well as March 15, 2018, to April 30, 2018; these periods represented prepandemic control periods.

Data were collected for 516 surgeries performed on 458 patients and included patient age, preoperative clinical size, postoperative defect size, number of Mohs stages to achieve clearance, MMS appropriate use criteria (AUC) location (categorized as high-, medium-, or low-risk tumor location),¹³ and tumor type (categorized as BCC, SCC, or MIS). All variables were examined for unusual or missing values. Five patients with rare tumor types were observed and removed from the data set.

Statistical Analysis—An a priori power analysis for a power set at 0.85 determined sample sizes of 105 per group. Bivariate analyses were performed to compare variables for patients undergoing MMS during the pandemic vs prepandemic periods. Continuous outcome variables—Mohs stages, preoperative size, postoperative size, and patient age—were categorized for the analysis. Preoperative tumor size was dichotomized, with less than 2 cm² as the referent category vs 2 cm² or greater, and postoperative defect size was dichotomized with less than 3.6 cm² as the referent category vs 3.6 cm² or greater. Mohs stage was dichotomized as 1 stage (referent) vs more than 1 stage, and patient age was dichotomized as younger than 65 years (referent) vs 65 years or older.

Multivariate analyses were also performed to compare preoperative and postoperative sizes for patients undergoing MMS during the pandemic vs prepandemic periods, controlling for Mohs AUC location. Bivariate unadjusted and multivariate analyses were performed using a GENMOD logistic regression procedure in SAS (SAS Institute) to account for correlation in clustered data because a patient could be included for more than 1 surgery in the data set. Data were analyzed using SAS 9.4 for Windows. Because outcome variables tended to be skewed and not distributed normally, outcome variables were recorded as medians with interquartile ranges where possible to give a more accurate representation of the data than could be demonstrated with means with standard deviations.

Results

One hundred thirty-eight skin cancers were treated during the COVID-19 pandemic from March 15, 2020, to April 30, 2020, and 378 skin cancers were treated during the prepandemic control periods of March 15, 2019, to April 30, 2019, and March 15, 2018, to April 30, 2018. Tumor type treated during the pandemic period was more likely to be SCC or MIS (representing generally more severe tumor types) vs BCC when compared with the prepandemic periods, with an odds ratio (OR) of 1.763 (95% CI, 1.17-2.66). This outcome was statistically significant (P=.01).

Tumors treated during the pandemic period were more likely to have necessitated more than one Mohs stage for clearance compared to the prepandemic periods, though this difference was not statistically significant (OR, 1.461; 95% CI, 0.97-2.19; P=.056). Neither AUC location of treated tumors nor age were significantly different between prepandemic and pandemic periods (P=.58 and P=.84, respectively). Table 1 includes all bivariate analysis results.

Additionally, although mean preoperative and postoperative sizes were larger for each AUC location during the pandemic vs prepandemic periods, these differences did not reach statistical significance on multivariate analysis (P=.71 and P=.50, respectively)(Table 2).

 

 

Comment

Our practice has followed best practice guidelines dictated by our governing professional societies during the COVID-19 pandemic in the treatment of skin cancers by MMS, specifically highly symptomatic BCCs (in accordance with ACMS guidance), SCCs with high-risk features (in accordance with AAD, ACMS, and TDS guidance), and tumors with high risk for progression and metastasis such as melanomas (in accordance with TDS guidance). Melanoma in situ was also treated during the COVID-19 pandemic in accordance with the latter TDS guidance, particularly in light of the potential for upstaging to melanoma following resection (a phenomenon demonstrated to occur in 5%–29% of biopsied MIS lesions).¹⁴

In following best practice guidelines, our results suggested tumors treated by MMS were more severe, as evidenced by a statistically significant higher proportion of SCC and MIS tumors (representing more severe tumor types) vs BCC when compared to the prepandemic period. Supporting this conclusion, we observed larger pretreatment and posttreatment tumor sizes for all AUC locations and more tumors necessitating 2 or more stages for clearance during the pandemic vs prepandemic periods, though these differences did not reach statistical significance. We postulate these findings may be attributed to allocation of finite medical resources to the treatment of larger and more aggressive skin cancers. Additionally, these findings may be explained, in part, by limitations on patient case load imposed by social distancing measures and governing body regulations in effect during the study period, including those put forth by the AAD, ACMS, and TDS. Of note, our practice observed no hospitalizations or 911 calls during the studied period. This suggests no allocation of precious hospital resources away from patients with COVID-19 in our treatment of high-risk skin cancers.

The changing characteristics of cutaneous tumors treated by MMS during the pandemic are of clinical relevance. Larger postoperative wound sizes as observed during the pandemic, albeit not statistically significant, presumably affect reconstructive decisions. With larger wounds tending to necessitate repair by techniques higher on the reconstructive ladder, greater patient morbidity and cost are expected.¹⁵ As the cost-effectiveness of dermatology services remains a critical issue, this is an area ripe for future follow-up research. Furthermore, our observation that tumors tended to necessitate 2 or more stages for clearance during the pandemic more often than prepandemic periods, though not statistically significant, presumably affected operating times. Longer operating times during the pandemic may be of importance when making clinical decisions for patients for whom limiting health care exposure may be of particular concern. With more SCC and MIS tumors being treated relative to BCCs during the pandemic, one might expect greater size and severity of the BCCs we observe in the proceeding months to years.

As the ongoing COVID-19 pandemic continues to impact the landscape of cutaneous oncology, the need for adaptability is imperative. With 3- and 6-month skin cancer treatment deferrals lapsed, uncertainty surrounds ideal management of existing and new skin cancers arising during the pandemic. This study adds to a growing body of literature elucidating the impact of the COVID-19 pandemic on MMS practice; however, further studies and a tincture of time are needed to guide future best practice standards.

Acknowledgment—The authors acknowledge Gwen Baillargeon, MS (Galveston, Texas), who was the statistician for this article.

Melanomas have been designated as small melanomas or micromelanomas according to their long-axis diameter (<6 mm and ≤3 mm, respectively).^1-3 Because small-diameter melanomas also have the potential to metastasize, particularly if nodular, early diagnosis can be highly rewarding. Deep melanomas with small diameters may have the same potential for metastasis as large-diameter melanomas. In this context, dermoscopy, digital dermoscopic monitoring, and total-body photography are useful in clinical practice. However, these techniques are of limited utility for small, dermoscopic feature–poor melanomas. Conversely, less than 10% of changing lesions, which are spotted via digital dermoscopic surveillance, turn out to be melanomas; therefore, simply removing all changing lesions may result in many unnecessary excisions of benign lesions.⁴

In vivo reflectance confocal microscopy (RCM) is an advanced technique that allows recognition of the architectural and cellular details of pigmented lesions. Reflectance confocal microscopy has the potential to reduce the rate of unnecessary excisions and to diminish the risk for missing a melanoma.^5-7 In meta-analyses, RCM sensitivity was reported as 90% to 93% and specificity was reported as 78% to 82% in detecting melanoma.^8,9

We describe a case that highlights the potential role of RCM in the diagnosis of small-diameter melanomas.

Case Report

A 57-year-old man with Fitzpatrick skin type III presented to the dermato-oncology unit for evaluation of multiple nevi. He was otherwise healthy and denied a history of skin cancer. Total-body skin examination with dermoscopy was performed, and several mildly atypical lesions were identified. We decided to perform digital dermoscopic monitoring. The patient’s 6-month monitoring appointment had been scheduled, but he did not arrive for the follow-up visit until 10 months after the initial examination. A lesion on the left arm, which initially was 1.5 mm in diameter, had enlarged. It was now a dark brown–gray papule with a 2.5-mm diameter (Figure 1). Dermoscopy revealed grayish globules/dots at the center of the lesion, reticular gray-blue areas, and few milialike cysts; at the periphery, a narrow rim of brownish delicate pigment network also was seen (Figure 2). The clinical and dermoscopic differential diagnosis was either an atypical nevus or an early melanoma. For a more precise diagnosis before excision, the lesion was evaluated with RCM, which takes 10 to 15 minutes to perform.

Under RCM at the epidermis level, there was a cobblestone pattern that showed a focus with mild disarrangement and few small, roundish, nucleated cells (Figure 3). A mosaic image, akin to low-magnification microscopy that enables overview of the entire lesion, at the level of the dermoepidermal junction (DEJ) showed an overall irregular meshwork pattern. Higher-magnification optical sections showed marked and diffuse (extending >10% of lesion area) architectural disorder with confluent junctional nests that were irregular to bizarre in shape and uneven in size and spacing as well as edged and nonedged papillae. At the superficial dermal level, atypical bright nucleated cells (>5 cells/mm²) were observed (Figure 4). Bright dots and/or plump bright cells within papillae also were observed. These RCM findings were highly suggestive for melanoma.

Histopathology showed an asymmetric, junctional, lentiginous, and nested proliferation of atypical epithelioid melanocytes, with few melanocytes in a pagetoid spread. There were small nests of atypical epithelioid melanocytes at the superficial dermis extending to a depth of 0.3 mm. The atypical epithelioid melanocytes displayed angulated hyperchromatic nuclei with conspicuous nucleoli and dusty brown cytoplasm. There was notable inflammation and pigment incontinence at the dermis. There was no evidence of ulceration or mitosis at the dermal component. The diagnosis of a pT1a malignant melanoma was reported (Figure 5).

Comment

A small but enlarging dark gray papule with reticular gray-blue areas under dermoscopy in a 57-year-old man is obviously suspicious for melanoma. In daily practice, this type of small-diameter melanoma is difficult to diagnose with high confidence. We balance our aim to diagnose melanomas early with the need to reduce unnecessary excisions. Reflectance confocal microscopy may allow the clinician to arrive at the correct diagnosis and management decision with confidence before excision of the lesion.

 

The distinction of a small-diameter melanoma from a nevus via RCM relies on evaluation of the architectural and cellular features. Findings on RCM in small-diameter melanomas have been scarcely reported in the literature; Pupelli et al¹⁰ evaluated small melanomas with a diameter of 2 to 5 mm. Among these small-diameter melanomas, the RCM features suggestive for melanomas were the presence of cytologic atypia with cellular pleomorphism, architectural disorder with irregular nests, at least 5 pagetoid cells/mm², dendrites or tangled lines (ie, short fine lines with no visible nucleus interlacing with the adjacent keratinocytes) within the epidermis, and atypical roundish cells at the DEJ.¹⁰

The distinction between an atypical nevus and a small-diameter melanoma using RCM occasionally may be challenging.¹¹ Pellacani et al¹² reported an algorithm to distinguish melanoma from atypical nevi. According to this algorithm, when at least 1 of the architectural atypia features (irregular junctional nests, short interconnections between junctional nests, and nonhomogeneous cellularity within junctional nests) and at least 1 of the cytologic atypia features (round pagetoid cells or atypical cells at the DEJ) are observed simultaneously, the lesion is diagnosed as a dysplastic nevus or a melanoma in the first step. In the second step, the RCM diagnosis of melanoma requires at least 1 of 3 parameters: roundish pagetoid cells encompassing at least 50% of the lesional area at the spinous layer, atypical cells involving at least 50% of the lesional area at the DEJ level, and nonedged papillae involving at least 10% of the lesional area.¹² Accordingly, our case corresponded with these RCM criteria for a melanoma, given that there were irregular junctional nests, atypical cells at the DEJ, and nonedged papillae involving at least 10% of the lesion.

The current limitations of RCM are the high cost of the device (approximately $58,125–$139,400 for different models), the amount of time needed to train staff in RCM units (seminars, congresses, and special courses organized by the International Confocal Working Group), and the amount of time needed for evaluation of individual lesions (15–20 minutes). However, RCM can be valuable in the clinical diagnosis of difficult lesions, as seen in our case.

Conclusion

Our case highlights the benefit of RCM in allowing the confident diagnosis and correct management of a small-diameter melanoma that turned out to be a melanoma with 0.3-mm Breslow thickness. Even so, histopathologic evaluation remains the gold standard for the diagnosis of melanoma.

Melanomas have been designated as small melanomas or micromelanomas according to their long-axis diameter (<6 mm and ≤3 mm, respectively).^1-3 Because small-diameter melanomas also have the potential to metastasize, particularly if nodular, early diagnosis can be highly rewarding. Deep melanomas with small diameters may have the same potential for metastasis as large-diameter melanomas. In this context, dermoscopy, digital dermoscopic monitoring, and total-body photography are useful in clinical practice. However, these techniques are of limited utility for small, dermoscopic feature–poor melanomas. Conversely, less than 10% of changing lesions, which are spotted via digital dermoscopic surveillance, turn out to be melanomas; therefore, simply removing all changing lesions may result in many unnecessary excisions of benign lesions.⁴

In vivo reflectance confocal microscopy (RCM) is an advanced technique that allows recognition of the architectural and cellular details of pigmented lesions. Reflectance confocal microscopy has the potential to reduce the rate of unnecessary excisions and to diminish the risk for missing a melanoma.^5-7 In meta-analyses, RCM sensitivity was reported as 90% to 93% and specificity was reported as 78% to 82% in detecting melanoma.^8,9

We describe a case that highlights the potential role of RCM in the diagnosis of small-diameter melanomas.

Case Report

A 57-year-old man with Fitzpatrick skin type III presented to the dermato-oncology unit for evaluation of multiple nevi. He was otherwise healthy and denied a history of skin cancer. Total-body skin examination with dermoscopy was performed, and several mildly atypical lesions were identified. We decided to perform digital dermoscopic monitoring. The patient’s 6-month monitoring appointment had been scheduled, but he did not arrive for the follow-up visit until 10 months after the initial examination. A lesion on the left arm, which initially was 1.5 mm in diameter, had enlarged. It was now a dark brown–gray papule with a 2.5-mm diameter (Figure 1). Dermoscopy revealed grayish globules/dots at the center of the lesion, reticular gray-blue areas, and few milialike cysts; at the periphery, a narrow rim of brownish delicate pigment network also was seen (Figure 2). The clinical and dermoscopic differential diagnosis was either an atypical nevus or an early melanoma. For a more precise diagnosis before excision, the lesion was evaluated with RCM, which takes 10 to 15 minutes to perform.

Under RCM at the epidermis level, there was a cobblestone pattern that showed a focus with mild disarrangement and few small, roundish, nucleated cells (Figure 3). A mosaic image, akin to low-magnification microscopy that enables overview of the entire lesion, at the level of the dermoepidermal junction (DEJ) showed an overall irregular meshwork pattern. Higher-magnification optical sections showed marked and diffuse (extending >10% of lesion area) architectural disorder with confluent junctional nests that were irregular to bizarre in shape and uneven in size and spacing as well as edged and nonedged papillae. At the superficial dermal level, atypical bright nucleated cells (>5 cells/mm²) were observed (Figure 4). Bright dots and/or plump bright cells within papillae also were observed. These RCM findings were highly suggestive for melanoma.

Histopathology showed an asymmetric, junctional, lentiginous, and nested proliferation of atypical epithelioid melanocytes, with few melanocytes in a pagetoid spread. There were small nests of atypical epithelioid melanocytes at the superficial dermis extending to a depth of 0.3 mm. The atypical epithelioid melanocytes displayed angulated hyperchromatic nuclei with conspicuous nucleoli and dusty brown cytoplasm. There was notable inflammation and pigment incontinence at the dermis. There was no evidence of ulceration or mitosis at the dermal component. The diagnosis of a pT1a malignant melanoma was reported (Figure 5).

Comment

A small but enlarging dark gray papule with reticular gray-blue areas under dermoscopy in a 57-year-old man is obviously suspicious for melanoma. In daily practice, this type of small-diameter melanoma is difficult to diagnose with high confidence. We balance our aim to diagnose melanomas early with the need to reduce unnecessary excisions. Reflectance confocal microscopy may allow the clinician to arrive at the correct diagnosis and management decision with confidence before excision of the lesion.

 

The distinction of a small-diameter melanoma from a nevus via RCM relies on evaluation of the architectural and cellular features. Findings on RCM in small-diameter melanomas have been scarcely reported in the literature; Pupelli et al¹⁰ evaluated small melanomas with a diameter of 2 to 5 mm. Among these small-diameter melanomas, the RCM features suggestive for melanomas were the presence of cytologic atypia with cellular pleomorphism, architectural disorder with irregular nests, at least 5 pagetoid cells/mm², dendrites or tangled lines (ie, short fine lines with no visible nucleus interlacing with the adjacent keratinocytes) within the epidermis, and atypical roundish cells at the DEJ.¹⁰

The distinction between an atypical nevus and a small-diameter melanoma using RCM occasionally may be challenging.¹¹ Pellacani et al¹² reported an algorithm to distinguish melanoma from atypical nevi. According to this algorithm, when at least 1 of the architectural atypia features (irregular junctional nests, short interconnections between junctional nests, and nonhomogeneous cellularity within junctional nests) and at least 1 of the cytologic atypia features (round pagetoid cells or atypical cells at the DEJ) are observed simultaneously, the lesion is diagnosed as a dysplastic nevus or a melanoma in the first step. In the second step, the RCM diagnosis of melanoma requires at least 1 of 3 parameters: roundish pagetoid cells encompassing at least 50% of the lesional area at the spinous layer, atypical cells involving at least 50% of the lesional area at the DEJ level, and nonedged papillae involving at least 10% of the lesional area.¹² Accordingly, our case corresponded with these RCM criteria for a melanoma, given that there were irregular junctional nests, atypical cells at the DEJ, and nonedged papillae involving at least 10% of the lesion.

The current limitations of RCM are the high cost of the device (approximately $58,125–$139,400 for different models), the amount of time needed to train staff in RCM units (seminars, congresses, and special courses organized by the International Confocal Working Group), and the amount of time needed for evaluation of individual lesions (15–20 minutes). However, RCM can be valuable in the clinical diagnosis of difficult lesions, as seen in our case.

Conclusion

Our case highlights the benefit of RCM in allowing the confident diagnosis and correct management of a small-diameter melanoma that turned out to be a melanoma with 0.3-mm Breslow thickness. Even so, histopathologic evaluation remains the gold standard for the diagnosis of melanoma.

Melanomas have been designated as small melanomas or micromelanomas according to their long-axis diameter (<6 mm and ≤3 mm, respectively).^1-3 Because small-diameter melanomas also have the potential to metastasize, particularly if nodular, early diagnosis can be highly rewarding. Deep melanomas with small diameters may have the same potential for metastasis as large-diameter melanomas. In this context, dermoscopy, digital dermoscopic monitoring, and total-body photography are useful in clinical practice. However, these techniques are of limited utility for small, dermoscopic feature–poor melanomas. Conversely, less than 10% of changing lesions, which are spotted via digital dermoscopic surveillance, turn out to be melanomas; therefore, simply removing all changing lesions may result in many unnecessary excisions of benign lesions.⁴

In vivo reflectance confocal microscopy (RCM) is an advanced technique that allows recognition of the architectural and cellular details of pigmented lesions. Reflectance confocal microscopy has the potential to reduce the rate of unnecessary excisions and to diminish the risk for missing a melanoma.^5-7 In meta-analyses, RCM sensitivity was reported as 90% to 93% and specificity was reported as 78% to 82% in detecting melanoma.^8,9

We describe a case that highlights the potential role of RCM in the diagnosis of small-diameter melanomas.

Case Report

A 57-year-old man with Fitzpatrick skin type III presented to the dermato-oncology unit for evaluation of multiple nevi. He was otherwise healthy and denied a history of skin cancer. Total-body skin examination with dermoscopy was performed, and several mildly atypical lesions were identified. We decided to perform digital dermoscopic monitoring. The patient’s 6-month monitoring appointment had been scheduled, but he did not arrive for the follow-up visit until 10 months after the initial examination. A lesion on the left arm, which initially was 1.5 mm in diameter, had enlarged. It was now a dark brown–gray papule with a 2.5-mm diameter (Figure 1). Dermoscopy revealed grayish globules/dots at the center of the lesion, reticular gray-blue areas, and few milialike cysts; at the periphery, a narrow rim of brownish delicate pigment network also was seen (Figure 2). The clinical and dermoscopic differential diagnosis was either an atypical nevus or an early melanoma. For a more precise diagnosis before excision, the lesion was evaluated with RCM, which takes 10 to 15 minutes to perform.

Under RCM at the epidermis level, there was a cobblestone pattern that showed a focus with mild disarrangement and few small, roundish, nucleated cells (Figure 3). A mosaic image, akin to low-magnification microscopy that enables overview of the entire lesion, at the level of the dermoepidermal junction (DEJ) showed an overall irregular meshwork pattern. Higher-magnification optical sections showed marked and diffuse (extending >10% of lesion area) architectural disorder with confluent junctional nests that were irregular to bizarre in shape and uneven in size and spacing as well as edged and nonedged papillae. At the superficial dermal level, atypical bright nucleated cells (>5 cells/mm²) were observed (Figure 4). Bright dots and/or plump bright cells within papillae also were observed. These RCM findings were highly suggestive for melanoma.

Histopathology showed an asymmetric, junctional, lentiginous, and nested proliferation of atypical epithelioid melanocytes, with few melanocytes in a pagetoid spread. There were small nests of atypical epithelioid melanocytes at the superficial dermis extending to a depth of 0.3 mm. The atypical epithelioid melanocytes displayed angulated hyperchromatic nuclei with conspicuous nucleoli and dusty brown cytoplasm. There was notable inflammation and pigment incontinence at the dermis. There was no evidence of ulceration or mitosis at the dermal component. The diagnosis of a pT1a malignant melanoma was reported (Figure 5).

Comment

A small but enlarging dark gray papule with reticular gray-blue areas under dermoscopy in a 57-year-old man is obviously suspicious for melanoma. In daily practice, this type of small-diameter melanoma is difficult to diagnose with high confidence. We balance our aim to diagnose melanomas early with the need to reduce unnecessary excisions. Reflectance confocal microscopy may allow the clinician to arrive at the correct diagnosis and management decision with confidence before excision of the lesion.

 

The distinction of a small-diameter melanoma from a nevus via RCM relies on evaluation of the architectural and cellular features. Findings on RCM in small-diameter melanomas have been scarcely reported in the literature; Pupelli et al¹⁰ evaluated small melanomas with a diameter of 2 to 5 mm. Among these small-diameter melanomas, the RCM features suggestive for melanomas were the presence of cytologic atypia with cellular pleomorphism, architectural disorder with irregular nests, at least 5 pagetoid cells/mm², dendrites or tangled lines (ie, short fine lines with no visible nucleus interlacing with the adjacent keratinocytes) within the epidermis, and atypical roundish cells at the DEJ.¹⁰

The distinction between an atypical nevus and a small-diameter melanoma using RCM occasionally may be challenging.¹¹ Pellacani et al¹² reported an algorithm to distinguish melanoma from atypical nevi. According to this algorithm, when at least 1 of the architectural atypia features (irregular junctional nests, short interconnections between junctional nests, and nonhomogeneous cellularity within junctional nests) and at least 1 of the cytologic atypia features (round pagetoid cells or atypical cells at the DEJ) are observed simultaneously, the lesion is diagnosed as a dysplastic nevus or a melanoma in the first step. In the second step, the RCM diagnosis of melanoma requires at least 1 of 3 parameters: roundish pagetoid cells encompassing at least 50% of the lesional area at the spinous layer, atypical cells involving at least 50% of the lesional area at the DEJ level, and nonedged papillae involving at least 10% of the lesional area.¹² Accordingly, our case corresponded with these RCM criteria for a melanoma, given that there were irregular junctional nests, atypical cells at the DEJ, and nonedged papillae involving at least 10% of the lesion.

The current limitations of RCM are the high cost of the device (approximately $58,125–$139,400 for different models), the amount of time needed to train staff in RCM units (seminars, congresses, and special courses organized by the International Confocal Working Group), and the amount of time needed for evaluation of individual lesions (15–20 minutes). However, RCM can be valuable in the clinical diagnosis of difficult lesions, as seen in our case.

Conclusion

Our case highlights the benefit of RCM in allowing the confident diagnosis and correct management of a small-diameter melanoma that turned out to be a melanoma with 0.3-mm Breslow thickness. Even so, histopathologic evaluation remains the gold standard for the diagnosis of melanoma.

Practice Gap

An essential part of training for a micrographic surgery and dermatologic oncology fellowship and scope of practice involves planning and execution of reconstructive surgery for Mohs defects. Recently, a surgical pearl presented by Rickstrew and colleagues¹ highlighted the use of different colored surgical marking pens and their benefit in a trainee-based environment.

Delineating multiple options for reconstruction with different colored markers on live patients allows fellows in-training to participate in surgical planning but introduces more markings or drawings that need to be wiped off during or after surgery, potentially prolonging operative time. Furthermore, the Rickstrew approach has the potential to (1) cause unnecessary emotional distress for the patient during surgical planning and (2) add to the cost of surgery with the purchase of various colors of surgical markers.

Technique

To improve patient experience and trainee education, we propose fine-tuning the colored marker approach by utilizing a digital drawing program for surgical planning prior to the procedure. We recommend Snip & Sketch—a free, readily accessible digital annotating application that runs on the Microsoft Windows 10 operating system (https://www.microsoft.com/en-us/p/snip-sketch/9mz95kl8mr0l#activetab=pivot:overviewtab)—to mark up screenshot photographs of postoperative Mohs defects from the electronic medical record.

Using Snip & Sketch, the fellow in-training can then use, for example, a green “digital pen” to draw on the captured image and plan their surgical repairs (Figure 1) without input from the attending physician. Different colored pens can be used to highlight nerves, vessels, relaxed skin tension lines, and tension vectors associated with flap movement.

Subsequently, the attending physician, using a different color digital pen—say, blue—can design alternative reconstructive options (Figure 1). Suture lines also can be drawn to outline the predicted appearance of surgical scars (Figure 2).

Then, the attending physician and fellow in-training brainstorm and discuss the advantages and disadvantages of each reconstructive option to determine the optimal approach to repairing the Mohs defect.

Advantages and Disadvantages

The main advantage of using a digital drawing program is that it is time-saving and cost-efficient. Digital planning also spares the patient undue anxiety from listening to the discussion on each repair option.

 

The primary downside of digital surgical planning is that it is 2-dimensional, thus providing an incomplete representation of a 3-dimensional cutaneous structure. In addition, skin laxity, flap mobility, and free-margin distortion cannot be fully appreciated on a 2-dimensional image.

Despite these drawbacks, digital surgical planning provides trainees with an active learning experience through a more collaborative and comprehensive discussion of reconstructive options.

Practice Implications

Active learning using an electronic device has been validated as a beneficial addition to Mohs micrographic surgery training.² Utilizing a digitized annotating program for surgical planning increases the independence of trainees and allows immediate feedback from the attending physician. The synergy of digital technology and collaborative learning helps cultivate the next generation of confident and competent Mohs surgeons.

Practice Gap

An essential part of training for a micrographic surgery and dermatologic oncology fellowship and scope of practice involves planning and execution of reconstructive surgery for Mohs defects. Recently, a surgical pearl presented by Rickstrew and colleagues¹ highlighted the use of different colored surgical marking pens and their benefit in a trainee-based environment.

Delineating multiple options for reconstruction with different colored markers on live patients allows fellows in-training to participate in surgical planning but introduces more markings or drawings that need to be wiped off during or after surgery, potentially prolonging operative time. Furthermore, the Rickstrew approach has the potential to (1) cause unnecessary emotional distress for the patient during surgical planning and (2) add to the cost of surgery with the purchase of various colors of surgical markers.

Technique

To improve patient experience and trainee education, we propose fine-tuning the colored marker approach by utilizing a digital drawing program for surgical planning prior to the procedure. We recommend Snip & Sketch—a free, readily accessible digital annotating application that runs on the Microsoft Windows 10 operating system (https://www.microsoft.com/en-us/p/snip-sketch/9mz95kl8mr0l#activetab=pivot:overviewtab)—to mark up screenshot photographs of postoperative Mohs defects from the electronic medical record.

Using Snip & Sketch, the fellow in-training can then use, for example, a green “digital pen” to draw on the captured image and plan their surgical repairs (Figure 1) without input from the attending physician. Different colored pens can be used to highlight nerves, vessels, relaxed skin tension lines, and tension vectors associated with flap movement.

Subsequently, the attending physician, using a different color digital pen—say, blue—can design alternative reconstructive options (Figure 1). Suture lines also can be drawn to outline the predicted appearance of surgical scars (Figure 2).

Then, the attending physician and fellow in-training brainstorm and discuss the advantages and disadvantages of each reconstructive option to determine the optimal approach to repairing the Mohs defect.

Advantages and Disadvantages

The main advantage of using a digital drawing program is that it is time-saving and cost-efficient. Digital planning also spares the patient undue anxiety from listening to the discussion on each repair option.

 

The primary downside of digital surgical planning is that it is 2-dimensional, thus providing an incomplete representation of a 3-dimensional cutaneous structure. In addition, skin laxity, flap mobility, and free-margin distortion cannot be fully appreciated on a 2-dimensional image.

Despite these drawbacks, digital surgical planning provides trainees with an active learning experience through a more collaborative and comprehensive discussion of reconstructive options.

Practice Implications

Active learning using an electronic device has been validated as a beneficial addition to Mohs micrographic surgery training.² Utilizing a digitized annotating program for surgical planning increases the independence of trainees and allows immediate feedback from the attending physician. The synergy of digital technology and collaborative learning helps cultivate the next generation of confident and competent Mohs surgeons.

Practice Gap

An essential part of training for a micrographic surgery and dermatologic oncology fellowship and scope of practice involves planning and execution of reconstructive surgery for Mohs defects. Recently, a surgical pearl presented by Rickstrew and colleagues¹ highlighted the use of different colored surgical marking pens and their benefit in a trainee-based environment.

Delineating multiple options for reconstruction with different colored markers on live patients allows fellows in-training to participate in surgical planning but introduces more markings or drawings that need to be wiped off during or after surgery, potentially prolonging operative time. Furthermore, the Rickstrew approach has the potential to (1) cause unnecessary emotional distress for the patient during surgical planning and (2) add to the cost of surgery with the purchase of various colors of surgical markers.

Technique

To improve patient experience and trainee education, we propose fine-tuning the colored marker approach by utilizing a digital drawing program for surgical planning prior to the procedure. We recommend Snip & Sketch—a free, readily accessible digital annotating application that runs on the Microsoft Windows 10 operating system (https://www.microsoft.com/en-us/p/snip-sketch/9mz95kl8mr0l#activetab=pivot:overviewtab)—to mark up screenshot photographs of postoperative Mohs defects from the electronic medical record.

Using Snip & Sketch, the fellow in-training can then use, for example, a green “digital pen” to draw on the captured image and plan their surgical repairs (Figure 1) without input from the attending physician. Different colored pens can be used to highlight nerves, vessels, relaxed skin tension lines, and tension vectors associated with flap movement.

Subsequently, the attending physician, using a different color digital pen—say, blue—can design alternative reconstructive options (Figure 1). Suture lines also can be drawn to outline the predicted appearance of surgical scars (Figure 2).

Then, the attending physician and fellow in-training brainstorm and discuss the advantages and disadvantages of each reconstructive option to determine the optimal approach to repairing the Mohs defect.

Advantages and Disadvantages

The main advantage of using a digital drawing program is that it is time-saving and cost-efficient. Digital planning also spares the patient undue anxiety from listening to the discussion on each repair option.

 

The primary downside of digital surgical planning is that it is 2-dimensional, thus providing an incomplete representation of a 3-dimensional cutaneous structure. In addition, skin laxity, flap mobility, and free-margin distortion cannot be fully appreciated on a 2-dimensional image.

Despite these drawbacks, digital surgical planning provides trainees with an active learning experience through a more collaborative and comprehensive discussion of reconstructive options.

Practice Implications

Active learning using an electronic device has been validated as a beneficial addition to Mohs micrographic surgery training.² Utilizing a digitized annotating program for surgical planning increases the independence of trainees and allows immediate feedback from the attending physician. The synergy of digital technology and collaborative learning helps cultivate the next generation of confident and competent Mohs surgeons.

The chances are good that at least one patient you saw today could have been provided a better environment to foster your patient-physician relationship. A 2020 Gallup poll revealed that an estimated 5.6% of US adults identified as lesbian, gay, bisexual, and transgender (LGBT).¹ Based on the estimated US population of 331.7 million individuals on December 3, 2020, this means that approximately 18.6 million identified as LGBT and could potentially require health care services.² These numbers highlight the increasing need within the medical community to provide quality and accessible care to the LGBT community, and dermatologists have a role to play. They treat conditions that are apparent to the patient and others around them, attracting those that may not be motivated to see different physicians. They can not only help with skin diseases that affect all patients but also can train other physicians to screen for some dermatologic diseases that may have a higher prevalence within the LGBT community. Dermatologists have a unique opportunity to help patients better reflect themselves through both surgical and nonsurgical modalities.

Demographics and Definitions

To discuss this topic effectively, it is important to define LGBT terms (Table).³ As a disclaimer, language is fluid. Despite a word or term currently being used and accepted, it quickly can become obsolete. A clinician can always do research, follow the lead of the patient, and respectfully ask questions if there is ever confusion surrounding terminology. Patients do not expect every physician they encounter to be an expert in this subject. What is most important is that patients are approached with an open mind and humility with the goal of providing optimal care.

Although the federal government now uses the term sexual and gender minorities (SGM), the more specific terms lesbian, gay, bisexual, and transgender usually are preferred.^3,4 Other letters are at times added to the acronym LGBT, including Q for questioning or queer, I for intersex, and A for asexual; all of these letters are under the larger SGM umbrella. Because LGBT is the most commonly used acronym in the daily vernacular, it will be the default for this article.

A term describing sexual orientation does not necessarily describe sexual practices. A woman who identifies as straight may have sex with both men and women, and a gay man may not have sex at all. To be more descriptive regarding sexual practices, one may use the terms men who have sex with men or women who have sex with women.³ Because of this nuance, it is important to elicit a sexual history when speaking to all patients in a forward nonjudgmental manner.

The term transgender is used to describe people whose gender identity differs from the sex they were assigned at birth. Two examples of transgender individuals would be transgender women who were assigned male at birth and transgender men who were assigned female at birth. The term transgender is used in opposition to the term cisgender, which is applied to a person whose gender and sex assigned at birth align.³ When a transgender patient presents to a physician, they may want to discuss methods of gender affirmation or transitioning. These terms encompass any action a person may take to align their body or gender expression with that of the gender they identify with. This could be in the form of gender-affirming hormone therapy (ie, estrogen or testosterone treatment) or gender-affirming surgery (ie, “top” and “bottom” surgeries, in which someone surgically treats their chest or genitals, respectively).³

Creating a Safe Space

The physician is responsible for providing a safe space for patients to disclose medically pertinent information. It is then the job of the dermatologist to be cognizant of health concerns that directly affect the LGBT population and to be prepared if one of these concerns should arise. A safe space consists of both the physical location in which the patient encounter will occur and the people that will be conducting and assisting in the patient encounter. Safe spaces provide a patient with reassurance that they will receive care in a judgement-free location. To create a safe space, both the physical and interpersonal aspects must be addressed to provide an environment that strengthens the patient-physician alliance.

Dermatology residents often spend more time with patients than their attending physicians, providing them the opportunity to foster robust relationships with those served. Although they may not be able to change the physical environment, residents can advocate for patients in their departments and show solidarity in subtle ways. One way to show support for the LGBT community is to publicly display a symbol of solidarity, which could be done by wearing a symbol of support on a white coat lapel. Although there are many designs and styles to choose from, one example is the American Medical Student Association pins that combine the caduceus (a common symbol for medicine) with a rainbow design.⁵ Whichever symbol is chosen, this small gesture allows patients to immediately know that their physician is an ally. Residents also can encourage their department to add a rainbow flag, a pink triangle, or another symbol somewhere prominent in the check-in area that conveys a message of support.⁶ Many institutions require residents to perform quality improvement projects. The resident can make a substantial difference in their patients’ experiences by revising their office’s intake forms as a quality improvement project, which can be done by including a section on assigned sex at birth separate from gender.⁷ When inquiring about gender, in addition to “male” and “female,” a space can be left for people that do not identify with the traditional binary. When asking about sexual orientation, inclusive language options can be provided with additional space for self-identification. Finally, residents can incorporate pronouns below their name in their email signature to normalize this disclosure of information.⁸ These small changes can have a substantial impact on the health care experience of SGM patients.

 

 

Medical Problems Encountered

The previously described changes can be implemented by residents to provide better care to SGM patients, a group usually considered to be more burdened by physical and psychological diseases.⁹ Furthermore, dermatologists can provide care for these patients in ways that other physicians cannot. There are special considerations for LGBT patients, as some dermatologic conditions may be more common in this patient population.

Prior studies have shown that men who have sex with men have a higher rate of HIV and other sexually transmitted infections, methicillin-resistant Staphylococcus aureus skin infections, and potentially nonmelanoma skin cancer.^10-14 Transgender women also have been found to have higher rates of HIV, in addition to a higher incidence of anal human papillomavirus.^15,16 Women who have sex with women have been shown to see physicians less frequently and to be less up to date on their pertinent cancer-related screenings.^10,17 Although these associations should not dictate the patient encounter, awareness of them will lead to better patient care. Such awareness also can provide further motivation for dermatologists to discuss safe sexual practices, potential initiation of pre-exposure prophylactic antiretroviral therapy, sun-protective practices, and the importance of following up with a primary physician for examinations and age-specific cancer screening.

Transgender patients may present with unique dermatologic concerns. For transgender male patients, testosterone therapy can cause acne breakouts and androgenetic alopecia. Usually considered worse during the start of treatment, hormone-related acne can be managed with topical retinoids, topical and oral antibiotics, and isotretinoin (if severe).^18,19 The iPLEDGE system necessary for prescribing isotretinoin to patients in the United States recently has changed its language to “patients who can get pregnant” and “patients who cannot get pregnant,” following urging by the medical community for inclusivity and progress.^20,21 This change creates an inclusive space where registration is no longer centered around gender and instead focuses on the presence of anatomy. Although androgenetic alopecia is a side effect of hormone therapy, it may not be unwanted.¹⁸ Discussion about patient desires is important. If the alopecia is unwanted, the Endocrine Society recommends treating cisgender and transgender patients the same in terms of treatment modalities.²²

Transgender female patients also can experience dermatologic manifestations of gender-affirming hormone therapy. Melasma may develop secondary to estrogen replacement and can be treated with topical bleaching creams, lasers, and phototherapy.²³ Hair removal may be pursued for patients with refractory unwanted body hair, with laser hair removal being the most commonly pursued treatment. Patients also may desire cosmetic procedures, such as botulinum toxin or fillers, to augment their physical appearance.²⁴ Providing these services to patients may allow them to better express themselves and live authentically.

Final Thoughts

There is no way to summarize the experience of everyone within a community. Each person has different thoughts, values, and goals. It also is impossible to encompass every topic that is important for SGM patients. The goal of this article is to empower clinicians to be comfortable discussing issues related to sexuality and gender while also offering resources to learn more, allowing optimal care to be provided to this population. Thus, this article is not comprehensive. There are articles to provide further resources and education, such as the continuing medical education series by Yeung et al^10,25 in the Journal of the American Academy of Dermatology, as well as organizations within medicine, such as the GLMA: Health Professionals Advancing LGBTQ Equality (https://www.glma.org/), and in dermatology, such as GALDA, the Gay and Lesbian Dermatology Association (https://www.glderm.org/). By providing a safe space for our patients and learning about specific health-related risk factors, dermatologists can provide the best possible care to the LGBT community.

Acknowledgments—I thank Warren R. Heymann, MD (Camden, New Jersey), and Howa Yeung, MD, MSc (Atlanta, Georgia), for their guidance and mentorship in the creation of this article.

The chances are good that at least one patient you saw today could have been provided a better environment to foster your patient-physician relationship. A 2020 Gallup poll revealed that an estimated 5.6% of US adults identified as lesbian, gay, bisexual, and transgender (LGBT).¹ Based on the estimated US population of 331.7 million individuals on December 3, 2020, this means that approximately 18.6 million identified as LGBT and could potentially require health care services.² These numbers highlight the increasing need within the medical community to provide quality and accessible care to the LGBT community, and dermatologists have a role to play. They treat conditions that are apparent to the patient and others around them, attracting those that may not be motivated to see different physicians. They can not only help with skin diseases that affect all patients but also can train other physicians to screen for some dermatologic diseases that may have a higher prevalence within the LGBT community. Dermatologists have a unique opportunity to help patients better reflect themselves through both surgical and nonsurgical modalities.

Demographics and Definitions

To discuss this topic effectively, it is important to define LGBT terms (Table).³ As a disclaimer, language is fluid. Despite a word or term currently being used and accepted, it quickly can become obsolete. A clinician can always do research, follow the lead of the patient, and respectfully ask questions if there is ever confusion surrounding terminology. Patients do not expect every physician they encounter to be an expert in this subject. What is most important is that patients are approached with an open mind and humility with the goal of providing optimal care.

Although the federal government now uses the term sexual and gender minorities (SGM), the more specific terms lesbian, gay, bisexual, and transgender usually are preferred.^3,4 Other letters are at times added to the acronym LGBT, including Q for questioning or queer, I for intersex, and A for asexual; all of these letters are under the larger SGM umbrella. Because LGBT is the most commonly used acronym in the daily vernacular, it will be the default for this article.

A term describing sexual orientation does not necessarily describe sexual practices. A woman who identifies as straight may have sex with both men and women, and a gay man may not have sex at all. To be more descriptive regarding sexual practices, one may use the terms men who have sex with men or women who have sex with women.³ Because of this nuance, it is important to elicit a sexual history when speaking to all patients in a forward nonjudgmental manner.

The term transgender is used to describe people whose gender identity differs from the sex they were assigned at birth. Two examples of transgender individuals would be transgender women who were assigned male at birth and transgender men who were assigned female at birth. The term transgender is used in opposition to the term cisgender, which is applied to a person whose gender and sex assigned at birth align.³ When a transgender patient presents to a physician, they may want to discuss methods of gender affirmation or transitioning. These terms encompass any action a person may take to align their body or gender expression with that of the gender they identify with. This could be in the form of gender-affirming hormone therapy (ie, estrogen or testosterone treatment) or gender-affirming surgery (ie, “top” and “bottom” surgeries, in which someone surgically treats their chest or genitals, respectively).³

Creating a Safe Space

The physician is responsible for providing a safe space for patients to disclose medically pertinent information. It is then the job of the dermatologist to be cognizant of health concerns that directly affect the LGBT population and to be prepared if one of these concerns should arise. A safe space consists of both the physical location in which the patient encounter will occur and the people that will be conducting and assisting in the patient encounter. Safe spaces provide a patient with reassurance that they will receive care in a judgement-free location. To create a safe space, both the physical and interpersonal aspects must be addressed to provide an environment that strengthens the patient-physician alliance.

Dermatology residents often spend more time with patients than their attending physicians, providing them the opportunity to foster robust relationships with those served. Although they may not be able to change the physical environment, residents can advocate for patients in their departments and show solidarity in subtle ways. One way to show support for the LGBT community is to publicly display a symbol of solidarity, which could be done by wearing a symbol of support on a white coat lapel. Although there are many designs and styles to choose from, one example is the American Medical Student Association pins that combine the caduceus (a common symbol for medicine) with a rainbow design.⁵ Whichever symbol is chosen, this small gesture allows patients to immediately know that their physician is an ally. Residents also can encourage their department to add a rainbow flag, a pink triangle, or another symbol somewhere prominent in the check-in area that conveys a message of support.⁶ Many institutions require residents to perform quality improvement projects. The resident can make a substantial difference in their patients’ experiences by revising their office’s intake forms as a quality improvement project, which can be done by including a section on assigned sex at birth separate from gender.⁷ When inquiring about gender, in addition to “male” and “female,” a space can be left for people that do not identify with the traditional binary. When asking about sexual orientation, inclusive language options can be provided with additional space for self-identification. Finally, residents can incorporate pronouns below their name in their email signature to normalize this disclosure of information.⁸ These small changes can have a substantial impact on the health care experience of SGM patients.

 

 

Medical Problems Encountered

The previously described changes can be implemented by residents to provide better care to SGM patients, a group usually considered to be more burdened by physical and psychological diseases.⁹ Furthermore, dermatologists can provide care for these patients in ways that other physicians cannot. There are special considerations for LGBT patients, as some dermatologic conditions may be more common in this patient population.

Prior studies have shown that men who have sex with men have a higher rate of HIV and other sexually transmitted infections, methicillin-resistant Staphylococcus aureus skin infections, and potentially nonmelanoma skin cancer.^10-14 Transgender women also have been found to have higher rates of HIV, in addition to a higher incidence of anal human papillomavirus.^15,16 Women who have sex with women have been shown to see physicians less frequently and to be less up to date on their pertinent cancer-related screenings.^10,17 Although these associations should not dictate the patient encounter, awareness of them will lead to better patient care. Such awareness also can provide further motivation for dermatologists to discuss safe sexual practices, potential initiation of pre-exposure prophylactic antiretroviral therapy, sun-protective practices, and the importance of following up with a primary physician for examinations and age-specific cancer screening.

Transgender patients may present with unique dermatologic concerns. For transgender male patients, testosterone therapy can cause acne breakouts and androgenetic alopecia. Usually considered worse during the start of treatment, hormone-related acne can be managed with topical retinoids, topical and oral antibiotics, and isotretinoin (if severe).^18,19 The iPLEDGE system necessary for prescribing isotretinoin to patients in the United States recently has changed its language to “patients who can get pregnant” and “patients who cannot get pregnant,” following urging by the medical community for inclusivity and progress.^20,21 This change creates an inclusive space where registration is no longer centered around gender and instead focuses on the presence of anatomy. Although androgenetic alopecia is a side effect of hormone therapy, it may not be unwanted.¹⁸ Discussion about patient desires is important. If the alopecia is unwanted, the Endocrine Society recommends treating cisgender and transgender patients the same in terms of treatment modalities.²²

Transgender female patients also can experience dermatologic manifestations of gender-affirming hormone therapy. Melasma may develop secondary to estrogen replacement and can be treated with topical bleaching creams, lasers, and phototherapy.²³ Hair removal may be pursued for patients with refractory unwanted body hair, with laser hair removal being the most commonly pursued treatment. Patients also may desire cosmetic procedures, such as botulinum toxin or fillers, to augment their physical appearance.²⁴ Providing these services to patients may allow them to better express themselves and live authentically.

Final Thoughts

There is no way to summarize the experience of everyone within a community. Each person has different thoughts, values, and goals. It also is impossible to encompass every topic that is important for SGM patients. The goal of this article is to empower clinicians to be comfortable discussing issues related to sexuality and gender while also offering resources to learn more, allowing optimal care to be provided to this population. Thus, this article is not comprehensive. There are articles to provide further resources and education, such as the continuing medical education series by Yeung et al^10,25 in the Journal of the American Academy of Dermatology, as well as organizations within medicine, such as the GLMA: Health Professionals Advancing LGBTQ Equality (https://www.glma.org/), and in dermatology, such as GALDA, the Gay and Lesbian Dermatology Association (https://www.glderm.org/). By providing a safe space for our patients and learning about specific health-related risk factors, dermatologists can provide the best possible care to the LGBT community.

Acknowledgments—I thank Warren R. Heymann, MD (Camden, New Jersey), and Howa Yeung, MD, MSc (Atlanta, Georgia), for their guidance and mentorship in the creation of this article.

The chances are good that at least one patient you saw today could have been provided a better environment to foster your patient-physician relationship. A 2020 Gallup poll revealed that an estimated 5.6% of US adults identified as lesbian, gay, bisexual, and transgender (LGBT).¹ Based on the estimated US population of 331.7 million individuals on December 3, 2020, this means that approximately 18.6 million identified as LGBT and could potentially require health care services.² These numbers highlight the increasing need within the medical community to provide quality and accessible care to the LGBT community, and dermatologists have a role to play. They treat conditions that are apparent to the patient and others around them, attracting those that may not be motivated to see different physicians. They can not only help with skin diseases that affect all patients but also can train other physicians to screen for some dermatologic diseases that may have a higher prevalence within the LGBT community. Dermatologists have a unique opportunity to help patients better reflect themselves through both surgical and nonsurgical modalities.

Demographics and Definitions

To discuss this topic effectively, it is important to define LGBT terms (Table).³ As a disclaimer, language is fluid. Despite a word or term currently being used and accepted, it quickly can become obsolete. A clinician can always do research, follow the lead of the patient, and respectfully ask questions if there is ever confusion surrounding terminology. Patients do not expect every physician they encounter to be an expert in this subject. What is most important is that patients are approached with an open mind and humility with the goal of providing optimal care.

Although the federal government now uses the term sexual and gender minorities (SGM), the more specific terms lesbian, gay, bisexual, and transgender usually are preferred.^3,4 Other letters are at times added to the acronym LGBT, including Q for questioning or queer, I for intersex, and A for asexual; all of these letters are under the larger SGM umbrella. Because LGBT is the most commonly used acronym in the daily vernacular, it will be the default for this article.

A term describing sexual orientation does not necessarily describe sexual practices. A woman who identifies as straight may have sex with both men and women, and a gay man may not have sex at all. To be more descriptive regarding sexual practices, one may use the terms men who have sex with men or women who have sex with women.³ Because of this nuance, it is important to elicit a sexual history when speaking to all patients in a forward nonjudgmental manner.

The term transgender is used to describe people whose gender identity differs from the sex they were assigned at birth. Two examples of transgender individuals would be transgender women who were assigned male at birth and transgender men who were assigned female at birth. The term transgender is used in opposition to the term cisgender, which is applied to a person whose gender and sex assigned at birth align.³ When a transgender patient presents to a physician, they may want to discuss methods of gender affirmation or transitioning. These terms encompass any action a person may take to align their body or gender expression with that of the gender they identify with. This could be in the form of gender-affirming hormone therapy (ie, estrogen or testosterone treatment) or gender-affirming surgery (ie, “top” and “bottom” surgeries, in which someone surgically treats their chest or genitals, respectively).³

Creating a Safe Space

The physician is responsible for providing a safe space for patients to disclose medically pertinent information. It is then the job of the dermatologist to be cognizant of health concerns that directly affect the LGBT population and to be prepared if one of these concerns should arise. A safe space consists of both the physical location in which the patient encounter will occur and the people that will be conducting and assisting in the patient encounter. Safe spaces provide a patient with reassurance that they will receive care in a judgement-free location. To create a safe space, both the physical and interpersonal aspects must be addressed to provide an environment that strengthens the patient-physician alliance.

Dermatology residents often spend more time with patients than their attending physicians, providing them the opportunity to foster robust relationships with those served. Although they may not be able to change the physical environment, residents can advocate for patients in their departments and show solidarity in subtle ways. One way to show support for the LGBT community is to publicly display a symbol of solidarity, which could be done by wearing a symbol of support on a white coat lapel. Although there are many designs and styles to choose from, one example is the American Medical Student Association pins that combine the caduceus (a common symbol for medicine) with a rainbow design.⁵ Whichever symbol is chosen, this small gesture allows patients to immediately know that their physician is an ally. Residents also can encourage their department to add a rainbow flag, a pink triangle, or another symbol somewhere prominent in the check-in area that conveys a message of support.⁶ Many institutions require residents to perform quality improvement projects. The resident can make a substantial difference in their patients’ experiences by revising their office’s intake forms as a quality improvement project, which can be done by including a section on assigned sex at birth separate from gender.⁷ When inquiring about gender, in addition to “male” and “female,” a space can be left for people that do not identify with the traditional binary. When asking about sexual orientation, inclusive language options can be provided with additional space for self-identification. Finally, residents can incorporate pronouns below their name in their email signature to normalize this disclosure of information.⁸ These small changes can have a substantial impact on the health care experience of SGM patients.

 

 

Medical Problems Encountered

The previously described changes can be implemented by residents to provide better care to SGM patients, a group usually considered to be more burdened by physical and psychological diseases.⁹ Furthermore, dermatologists can provide care for these patients in ways that other physicians cannot. There are special considerations for LGBT patients, as some dermatologic conditions may be more common in this patient population.

Prior studies have shown that men who have sex with men have a higher rate of HIV and other sexually transmitted infections, methicillin-resistant Staphylococcus aureus skin infections, and potentially nonmelanoma skin cancer.^10-14 Transgender women also have been found to have higher rates of HIV, in addition to a higher incidence of anal human papillomavirus.^15,16 Women who have sex with women have been shown to see physicians less frequently and to be less up to date on their pertinent cancer-related screenings.^10,17 Although these associations should not dictate the patient encounter, awareness of them will lead to better patient care. Such awareness also can provide further motivation for dermatologists to discuss safe sexual practices, potential initiation of pre-exposure prophylactic antiretroviral therapy, sun-protective practices, and the importance of following up with a primary physician for examinations and age-specific cancer screening.

Transgender patients may present with unique dermatologic concerns. For transgender male patients, testosterone therapy can cause acne breakouts and androgenetic alopecia. Usually considered worse during the start of treatment, hormone-related acne can be managed with topical retinoids, topical and oral antibiotics, and isotretinoin (if severe).^18,19 The iPLEDGE system necessary for prescribing isotretinoin to patients in the United States recently has changed its language to “patients who can get pregnant” and “patients who cannot get pregnant,” following urging by the medical community for inclusivity and progress.^20,21 This change creates an inclusive space where registration is no longer centered around gender and instead focuses on the presence of anatomy. Although androgenetic alopecia is a side effect of hormone therapy, it may not be unwanted.¹⁸ Discussion about patient desires is important. If the alopecia is unwanted, the Endocrine Society recommends treating cisgender and transgender patients the same in terms of treatment modalities.²²

Transgender female patients also can experience dermatologic manifestations of gender-affirming hormone therapy. Melasma may develop secondary to estrogen replacement and can be treated with topical bleaching creams, lasers, and phototherapy.²³ Hair removal may be pursued for patients with refractory unwanted body hair, with laser hair removal being the most commonly pursued treatment. Patients also may desire cosmetic procedures, such as botulinum toxin or fillers, to augment their physical appearance.²⁴ Providing these services to patients may allow them to better express themselves and live authentically.

Final Thoughts

There is no way to summarize the experience of everyone within a community. Each person has different thoughts, values, and goals. It also is impossible to encompass every topic that is important for SGM patients. The goal of this article is to empower clinicians to be comfortable discussing issues related to sexuality and gender while also offering resources to learn more, allowing optimal care to be provided to this population. Thus, this article is not comprehensive. There are articles to provide further resources and education, such as the continuing medical education series by Yeung et al^10,25 in the Journal of the American Academy of Dermatology, as well as organizations within medicine, such as the GLMA: Health Professionals Advancing LGBTQ Equality (https://www.glma.org/), and in dermatology, such as GALDA, the Gay and Lesbian Dermatology Association (https://www.glderm.org/). By providing a safe space for our patients and learning about specific health-related risk factors, dermatologists can provide the best possible care to the LGBT community.

Acknowledgments—I thank Warren R. Heymann, MD (Camden, New Jersey), and Howa Yeung, MD, MSc (Atlanta, Georgia), for their guidance and mentorship in the creation of this article.

Photographs courtesy of Richard P. Usatine, MD.

THE COMPARISON

A Acral lentiginous melanoma on the sole of the foot in a 30-year-old Black woman. The depth of the lesion was 2 mm with a positive sentinel lymph node biopsy.

B Nodular melanoma on the shoulder of a 63-year-old Hispanic woman. The depth of the lesion was 5.5 mm with a positive sentinel lymph node biopsy.

 

Melanoma occurs less frequently in individuals with darker skin types than in lighter skin types but is associated with higher rates of morbidity and mortality in this patient population.^1-7 In the cases shown here (A and B), both patients had advanced melanomas with large primary lesions and lymph node metastases.

Epidemiology

A systematic review by Higgins et al⁶ reported the following on the epidemiology of melanomas in patients with skin of color:

• African Americans have deeper tumors at the time of diagnosis, in addition to increased rates of regionally advanced and distant disease. Lesions generally are located on the lower extremities and have an increased propensity for ulceration. Acral lentiginous melanoma is the most common melanoma subtype found in African American patients.⁶
• In Hispanic individuals, superficial spreading melanoma is the most common melanoma subtype. Lower extremity lesions are more common relative to White individuals. Hispanic individuals have the highest rate of oral cavity melanomas across all ethnic groups.⁶
• In Asian individuals, acral and subungual sites are most common. Specifically, Pacific Islanders have the highest proportion of mucosal melanomas across all ethnic groups.⁶

Key clinical features in people with darker skin tones

Melanomas are found more often on the palms, soles, nail units, oral cavity, and mucosae.6 The melanomas have the same clinical and dermoscopic features found in individuals with lighter skin tones.

Worth noting

Factors that may contribute to the diagnosis of more advanced melanomas in racial/ethnic minorities in the United States include:

• decreased access to health care based on lack of health insurance and low socioeconomic status,
• less awareness of the risk of melanoma among patients and health care providers because melanoma is less common in persons of color, and
• lesions found in areas less likely to be seen in screening examinations, such as the soles of the feet and the oral and genital mucosae.

Health disparity highlight

• In a large US study of 96,953 patients with a diagnosis of cutaneous melanoma from 1992 to 2009, the proportion of later-stage melanoma—stages II to IV—was greater in Black patients compared to White patients.⁷
• Based on this same data set, White patients had the longest survival time (P<.05), followed by Hispanic (P<.05), Asian American/Native American/Pacific Islander (P<.05), and Black (P<.05) patients, respectively.⁷
• In Miami-Dade County, one study of 1690 melanoma cases found that 48% of Black patients had regional or distant disease at presentation compared to 22% of White patients (P=.015).⁵ Analysis of multiple factors found that only race was a significant predictor for late-stage melanoma (P<.001). Black patients in this study were 3 times more likely than others to be diagnosed with melanoma at a late stage (P=.07).⁵
• Black patients in the United States are more likely to have a delayed time from diagnosis to definitive surgery even when controlled for type of health insurance and stage of diagnosis.⁸

Final thoughts

Efforts are needed to overcome these disparities by:

• educating patients with skin of color and their health care providers about the risks of advanced melanoma with the goal of prevention and earlier diagnosis;
• breaking down barriers to care caused by poverty, lack of health insurance, and systemic racism; and
• eliminating factors that lead to delays from diagnosis to definitive surgery.

Photographs courtesy of Richard P. Usatine, MD.

THE COMPARISON

A Acral lentiginous melanoma on the sole of the foot in a 30-year-old Black woman. The depth of the lesion was 2 mm with a positive sentinel lymph node biopsy.

B Nodular melanoma on the shoulder of a 63-year-old Hispanic woman. The depth of the lesion was 5.5 mm with a positive sentinel lymph node biopsy.

 

Melanoma occurs less frequently in individuals with darker skin types than in lighter skin types but is associated with higher rates of morbidity and mortality in this patient population.^1-7 In the cases shown here (A and B), both patients had advanced melanomas with large primary lesions and lymph node metastases.

Epidemiology

A systematic review by Higgins et al⁶ reported the following on the epidemiology of melanomas in patients with skin of color:

• African Americans have deeper tumors at the time of diagnosis, in addition to increased rates of regionally advanced and distant disease. Lesions generally are located on the lower extremities and have an increased propensity for ulceration. Acral lentiginous melanoma is the most common melanoma subtype found in African American patients.⁶
• In Hispanic individuals, superficial spreading melanoma is the most common melanoma subtype. Lower extremity lesions are more common relative to White individuals. Hispanic individuals have the highest rate of oral cavity melanomas across all ethnic groups.⁶
• In Asian individuals, acral and subungual sites are most common. Specifically, Pacific Islanders have the highest proportion of mucosal melanomas across all ethnic groups.⁶

Key clinical features in people with darker skin tones

Melanomas are found more often on the palms, soles, nail units, oral cavity, and mucosae.6 The melanomas have the same clinical and dermoscopic features found in individuals with lighter skin tones.

Worth noting

Factors that may contribute to the diagnosis of more advanced melanomas in racial/ethnic minorities in the United States include:

• decreased access to health care based on lack of health insurance and low socioeconomic status,
• less awareness of the risk of melanoma among patients and health care providers because melanoma is less common in persons of color, and
• lesions found in areas less likely to be seen in screening examinations, such as the soles of the feet and the oral and genital mucosae.

Health disparity highlight

• In a large US study of 96,953 patients with a diagnosis of cutaneous melanoma from 1992 to 2009, the proportion of later-stage melanoma—stages II to IV—was greater in Black patients compared to White patients.⁷
• Based on this same data set, White patients had the longest survival time (P<.05), followed by Hispanic (P<.05), Asian American/Native American/Pacific Islander (P<.05), and Black (P<.05) patients, respectively.⁷
• In Miami-Dade County, one study of 1690 melanoma cases found that 48% of Black patients had regional or distant disease at presentation compared to 22% of White patients (P=.015).⁵ Analysis of multiple factors found that only race was a significant predictor for late-stage melanoma (P<.001). Black patients in this study were 3 times more likely than others to be diagnosed with melanoma at a late stage (P=.07).⁵
• Black patients in the United States are more likely to have a delayed time from diagnosis to definitive surgery even when controlled for type of health insurance and stage of diagnosis.⁸

Final thoughts

Efforts are needed to overcome these disparities by:

• educating patients with skin of color and their health care providers about the risks of advanced melanoma with the goal of prevention and earlier diagnosis;
• breaking down barriers to care caused by poverty, lack of health insurance, and systemic racism; and
• eliminating factors that lead to delays from diagnosis to definitive surgery.

Photographs courtesy of Richard P. Usatine, MD.

THE COMPARISON

A Acral lentiginous melanoma on the sole of the foot in a 30-year-old Black woman. The depth of the lesion was 2 mm with a positive sentinel lymph node biopsy.

B Nodular melanoma on the shoulder of a 63-year-old Hispanic woman. The depth of the lesion was 5.5 mm with a positive sentinel lymph node biopsy.

 

Melanoma occurs less frequently in individuals with darker skin types than in lighter skin types but is associated with higher rates of morbidity and mortality in this patient population.^1-7 In the cases shown here (A and B), both patients had advanced melanomas with large primary lesions and lymph node metastases.

Epidemiology

A systematic review by Higgins et al⁶ reported the following on the epidemiology of melanomas in patients with skin of color:

• African Americans have deeper tumors at the time of diagnosis, in addition to increased rates of regionally advanced and distant disease. Lesions generally are located on the lower extremities and have an increased propensity for ulceration. Acral lentiginous melanoma is the most common melanoma subtype found in African American patients.⁶
• In Hispanic individuals, superficial spreading melanoma is the most common melanoma subtype. Lower extremity lesions are more common relative to White individuals. Hispanic individuals have the highest rate of oral cavity melanomas across all ethnic groups.⁶
• In Asian individuals, acral and subungual sites are most common. Specifically, Pacific Islanders have the highest proportion of mucosal melanomas across all ethnic groups.⁶

Key clinical features in people with darker skin tones

Melanomas are found more often on the palms, soles, nail units, oral cavity, and mucosae.6 The melanomas have the same clinical and dermoscopic features found in individuals with lighter skin tones.

Worth noting

Factors that may contribute to the diagnosis of more advanced melanomas in racial/ethnic minorities in the United States include:

• decreased access to health care based on lack of health insurance and low socioeconomic status,
• less awareness of the risk of melanoma among patients and health care providers because melanoma is less common in persons of color, and
• lesions found in areas less likely to be seen in screening examinations, such as the soles of the feet and the oral and genital mucosae.

Health disparity highlight

• In a large US study of 96,953 patients with a diagnosis of cutaneous melanoma from 1992 to 2009, the proportion of later-stage melanoma—stages II to IV—was greater in Black patients compared to White patients.⁷
• Based on this same data set, White patients had the longest survival time (P<.05), followed by Hispanic (P<.05), Asian American/Native American/Pacific Islander (P<.05), and Black (P<.05) patients, respectively.⁷
• In Miami-Dade County, one study of 1690 melanoma cases found that 48% of Black patients had regional or distant disease at presentation compared to 22% of White patients (P=.015).⁵ Analysis of multiple factors found that only race was a significant predictor for late-stage melanoma (P<.001). Black patients in this study were 3 times more likely than others to be diagnosed with melanoma at a late stage (P=.07).⁵
• Black patients in the United States are more likely to have a delayed time from diagnosis to definitive surgery even when controlled for type of health insurance and stage of diagnosis.⁸

Final thoughts

Efforts are needed to overcome these disparities by:

• educating patients with skin of color and their health care providers about the risks of advanced melanoma with the goal of prevention and earlier diagnosis;
• breaking down barriers to care caused by poverty, lack of health insurance, and systemic racism; and
• eliminating factors that lead to delays from diagnosis to definitive surgery.

The Diagnosis: Condyloma Latum

A punch biopsy of the perianal mass revealed epidermal acanthosis with elongated slender rete ridges, scattered intraepidermal neutrophils, and a dense dermal inflammatory infiltrate (Figure, A) with a prominent plasma cell component (Figure, B). A treponemal immunohistochemical stain revealed numerous coiled spirochetes concentrated in the lower epidermis (Figure, C). Serologic test results including rapid plasma reagin (titer 1:1024) and Treponema pallidum antibody were reactive, confirming the diagnosis of secondary syphilis with condyloma latum. The patient was treated with intramuscular penicillin G with resolution of the lesion 2 weeks later.

Syphilis, a sexually transmitted infection caused by the spirochete T pallidum, reached historically low rates in the United States in the early 2000s due to the widespread use of penicillin and effective public health efforts.1 However, the rates of primary and secondary syphilis infections recently have markedly increased, resulting in the current epidemic of syphilis in the United States and Europe.^1,2 Its wide variety of clinical and histopathologic manifestations make recognition challenging and lend it the moniker “the great imitator.”

Secondary syphilis results from the systemic spread of T pallidum and classically is characterized by the triad of a skin rash that frequently involves the palms and soles, mucosal ulceration such as condyloma latum, and lymphadenopathy.^2,3 However, condyloma latum may represent the only manifestation of secondary syphilis in a subset of patients,⁴ as observed in our patient.

In the 2 months prior to diagnosis, our patient was evaluated at multiple emergency departments and primary care clinics, receiving diagnoses of condyloma acuminatum, genital herpes simplex virus, hemorrhoids, and suspicion for malignancy—entities that comprise the differential diagnosis for condyloma latum.^2,5 Despite some degree of overlap in patient populations, risk factors, and presentations between these diagnostic considerations, recognition of certain clinical features, in addition to histopathologic evaluation, may facilitate navigation of this differential diagnosis.

Primary and secondary syphilis infections have been predominantly observed in men, mostly men who have sex with men and/or those who are infected with HIV.¹ Condyloma acuminata, genital herpes simplex virus, and chancroid also are seen in younger individuals, more commonly in those with multiple sexual partners, but show a more even gender distribution and are not restricted to those partaking in anal intercourse. The clinical presentation of condyloma latum can be differentiated by its painless, flat, smooth, and commonly hypopigmented appearance, often with associated surface erosion and a gray exudate, in contrast to condyloma acuminatum, which typically presents as nontender, flesh-colored or hyperpigmented, exophytic papules that may coalesce into plaques.^2,3,6 Genital herpes simplex virus infection presents with multiple small papulovesicular lesions with ulceration, most commonly on the tip or shaft of the penis, though perianal lesions may be seen in men who have sex with men.⁷ Similarly, chancroid presents with painful necrotizing genital ulcers most commonly on the penis, though perianal lesions also may be seen.8 Hemorrhoids classically are seen in middle-aged adults with a history of constipation, present with rectal bleeding, and may be associated with pain in the setting of thrombosis or ulceration.⁹ Finally, perianal squamous cell carcinoma primarily occurs in older adults, typically in the sixth decade of life. Verrucous carcinoma most commonly arises in the oropharynx or anogenital region in sites of chronic irritation and presents as a slow-growing exophytic mass. Classic squamous cell carcinoma most commonly occurs in association with human papillomavirus infection and presents with scaly erythematous papules or plaques.¹⁰

Our case highlighted the clinical difficulty in recognizing condyloma latum, as this lesion remained undiagnosed for 2 months, and our patient presumptively was treated for multiple perianal pathologies prior to a biopsy being performed. Due to the clinical similarity of various perianal lesions, the diagnosis of condyloma latum should be considered, and serologic studies should be performed in fitting clinical contexts, especially in light of recently rising rates of syphilis infection.^1,2

The Diagnosis: Condyloma Latum

A punch biopsy of the perianal mass revealed epidermal acanthosis with elongated slender rete ridges, scattered intraepidermal neutrophils, and a dense dermal inflammatory infiltrate (Figure, A) with a prominent plasma cell component (Figure, B). A treponemal immunohistochemical stain revealed numerous coiled spirochetes concentrated in the lower epidermis (Figure, C). Serologic test results including rapid plasma reagin (titer 1:1024) and Treponema pallidum antibody were reactive, confirming the diagnosis of secondary syphilis with condyloma latum. The patient was treated with intramuscular penicillin G with resolution of the lesion 2 weeks later.

Syphilis, a sexually transmitted infection caused by the spirochete T pallidum, reached historically low rates in the United States in the early 2000s due to the widespread use of penicillin and effective public health efforts.1 However, the rates of primary and secondary syphilis infections recently have markedly increased, resulting in the current epidemic of syphilis in the United States and Europe.^1,2 Its wide variety of clinical and histopathologic manifestations make recognition challenging and lend it the moniker “the great imitator.”

Secondary syphilis results from the systemic spread of T pallidum and classically is characterized by the triad of a skin rash that frequently involves the palms and soles, mucosal ulceration such as condyloma latum, and lymphadenopathy.^2,3 However, condyloma latum may represent the only manifestation of secondary syphilis in a subset of patients,⁴ as observed in our patient.

In the 2 months prior to diagnosis, our patient was evaluated at multiple emergency departments and primary care clinics, receiving diagnoses of condyloma acuminatum, genital herpes simplex virus, hemorrhoids, and suspicion for malignancy—entities that comprise the differential diagnosis for condyloma latum.^2,5 Despite some degree of overlap in patient populations, risk factors, and presentations between these diagnostic considerations, recognition of certain clinical features, in addition to histopathologic evaluation, may facilitate navigation of this differential diagnosis.

Primary and secondary syphilis infections have been predominantly observed in men, mostly men who have sex with men and/or those who are infected with HIV.¹ Condyloma acuminata, genital herpes simplex virus, and chancroid also are seen in younger individuals, more commonly in those with multiple sexual partners, but show a more even gender distribution and are not restricted to those partaking in anal intercourse. The clinical presentation of condyloma latum can be differentiated by its painless, flat, smooth, and commonly hypopigmented appearance, often with associated surface erosion and a gray exudate, in contrast to condyloma acuminatum, which typically presents as nontender, flesh-colored or hyperpigmented, exophytic papules that may coalesce into plaques.^2,3,6 Genital herpes simplex virus infection presents with multiple small papulovesicular lesions with ulceration, most commonly on the tip or shaft of the penis, though perianal lesions may be seen in men who have sex with men.⁷ Similarly, chancroid presents with painful necrotizing genital ulcers most commonly on the penis, though perianal lesions also may be seen.8 Hemorrhoids classically are seen in middle-aged adults with a history of constipation, present with rectal bleeding, and may be associated with pain in the setting of thrombosis or ulceration.⁹ Finally, perianal squamous cell carcinoma primarily occurs in older adults, typically in the sixth decade of life. Verrucous carcinoma most commonly arises in the oropharynx or anogenital region in sites of chronic irritation and presents as a slow-growing exophytic mass. Classic squamous cell carcinoma most commonly occurs in association with human papillomavirus infection and presents with scaly erythematous papules or plaques.¹⁰

Our case highlighted the clinical difficulty in recognizing condyloma latum, as this lesion remained undiagnosed for 2 months, and our patient presumptively was treated for multiple perianal pathologies prior to a biopsy being performed. Due to the clinical similarity of various perianal lesions, the diagnosis of condyloma latum should be considered, and serologic studies should be performed in fitting clinical contexts, especially in light of recently rising rates of syphilis infection.^1,2

The Diagnosis: Condyloma Latum

A punch biopsy of the perianal mass revealed epidermal acanthosis with elongated slender rete ridges, scattered intraepidermal neutrophils, and a dense dermal inflammatory infiltrate (Figure, A) with a prominent plasma cell component (Figure, B). A treponemal immunohistochemical stain revealed numerous coiled spirochetes concentrated in the lower epidermis (Figure, C). Serologic test results including rapid plasma reagin (titer 1:1024) and Treponema pallidum antibody were reactive, confirming the diagnosis of secondary syphilis with condyloma latum. The patient was treated with intramuscular penicillin G with resolution of the lesion 2 weeks later.

Syphilis, a sexually transmitted infection caused by the spirochete T pallidum, reached historically low rates in the United States in the early 2000s due to the widespread use of penicillin and effective public health efforts.1 However, the rates of primary and secondary syphilis infections recently have markedly increased, resulting in the current epidemic of syphilis in the United States and Europe.^1,2 Its wide variety of clinical and histopathologic manifestations make recognition challenging and lend it the moniker “the great imitator.”

Secondary syphilis results from the systemic spread of T pallidum and classically is characterized by the triad of a skin rash that frequently involves the palms and soles, mucosal ulceration such as condyloma latum, and lymphadenopathy.^2,3 However, condyloma latum may represent the only manifestation of secondary syphilis in a subset of patients,⁴ as observed in our patient.

In the 2 months prior to diagnosis, our patient was evaluated at multiple emergency departments and primary care clinics, receiving diagnoses of condyloma acuminatum, genital herpes simplex virus, hemorrhoids, and suspicion for malignancy—entities that comprise the differential diagnosis for condyloma latum.^2,5 Despite some degree of overlap in patient populations, risk factors, and presentations between these diagnostic considerations, recognition of certain clinical features, in addition to histopathologic evaluation, may facilitate navigation of this differential diagnosis.

Primary and secondary syphilis infections have been predominantly observed in men, mostly men who have sex with men and/or those who are infected with HIV.¹ Condyloma acuminata, genital herpes simplex virus, and chancroid also are seen in younger individuals, more commonly in those with multiple sexual partners, but show a more even gender distribution and are not restricted to those partaking in anal intercourse. The clinical presentation of condyloma latum can be differentiated by its painless, flat, smooth, and commonly hypopigmented appearance, often with associated surface erosion and a gray exudate, in contrast to condyloma acuminatum, which typically presents as nontender, flesh-colored or hyperpigmented, exophytic papules that may coalesce into plaques.^2,3,6 Genital herpes simplex virus infection presents with multiple small papulovesicular lesions with ulceration, most commonly on the tip or shaft of the penis, though perianal lesions may be seen in men who have sex with men.⁷ Similarly, chancroid presents with painful necrotizing genital ulcers most commonly on the penis, though perianal lesions also may be seen.8 Hemorrhoids classically are seen in middle-aged adults with a history of constipation, present with rectal bleeding, and may be associated with pain in the setting of thrombosis or ulceration.⁹ Finally, perianal squamous cell carcinoma primarily occurs in older adults, typically in the sixth decade of life. Verrucous carcinoma most commonly arises in the oropharynx or anogenital region in sites of chronic irritation and presents as a slow-growing exophytic mass. Classic squamous cell carcinoma most commonly occurs in association with human papillomavirus infection and presents with scaly erythematous papules or plaques.¹⁰

Our case highlighted the clinical difficulty in recognizing condyloma latum, as this lesion remained undiagnosed for 2 months, and our patient presumptively was treated for multiple perianal pathologies prior to a biopsy being performed. Due to the clinical similarity of various perianal lesions, the diagnosis of condyloma latum should be considered, and serologic studies should be performed in fitting clinical contexts, especially in light of recently rising rates of syphilis infection.^1,2

To the Editor:

A 57-year-old man with hypertension, dyslipidemia, and congestive heart failure presented with a disfiguring eruption comprised of asymptomatic papules on the face that appeared 12 months post–heart transplantation. Immunosuppressive medications included mycophenolic acid and tacrolimus ointment (FK506). The pinpoint papules spread from the central face to the ears, arms, and legs. Physical examination revealed multiple 0.5- to 1-mm flesh-colored papules over the glabella, nose, nasolabial folds, philtrum, chin, ears, arms, and legs sparing the trunk. The initial appearance of the facial rash resembled the surface of a nutmeg grater with central white spiny excrescences overlying fine papules (spinulosism)(Figure 1). In addition, eyebrow alopecia was present.

A 3-mm punch biopsy of a papule with a central spine was performed on the left thigh. Microscopic examination revealed marked dilatation of anagen hair follicles with a proliferation of haphazard inner root sheath cells replacing the follicular lumen. Hair shafts were absent, and plugged infundibula were observed (Figure 2). The inner root sheath keratinocytes were enlarged and dystrophic with deeply eosinophilic trichohyalin granules (Figure 3). The epidermis, outer root sheath epithelium, and eccrine structures were unremarkable.

Transmission electron microscopy (TEM) confirmed the presence of intranuclear viral inclusions within affected inner root sheath keratinocytes composed of nonenveloped icosahedral viral particles measuring 33 to 38 nm in diameter (Figure 4). These findings morphologically were consistent with a polyomavirus. No intracytoplasmic or extracellular viral particles were identified. The clinical history, physical examination, histopathology, and electron microscopy features strongly supported the diagnosis of trichodysplasia spinulosa (TS) despite insufficient material being retrieved for polymerase chain reaction identification.

Trichodysplasia spinulosa was first described by Haycox et al¹ in 1999. The authors suggested a viral etiology. Eleven years later, TS-associated polyomavirus (TSPyV) was identified by van der Meijden et al.² Follicular keratinocytes are the specific target for TSPyV.³ Evidence has been presented suggesting that TS is caused by a primary infection or reactivation of TSPyV in the setting of immunosuppression.^4,5

Patients with TS present with papular eruptions that appear on the central face with spiny excrescences and various degrees of alopecia involving the eyebrows or eyelashes. Histopathologic features include distended hair follicles with expansion of inner root sheath cells, eosinophilic trichohyalin granules, and the absence of hair shafts. The viral protein can be verified through immunohistochemistry TSPyV VP1 staining that demonstrates co-localization with trichohyalin. Viral particles also can be visualized as 35- to 38-nm intranuclear particles with an organized crystalloid morphology on TEM.^6,7 The negative polymerase chain reaction in our patient could be the result of suboptimal template DNA concentration extracted from the limited amount of tissue remaining in the block after hematoxylin and eosin staining.

The clinical differential diagnosis of central facial spinulosism includes the follicular spicules of multiple myeloma (FSMM). In fact, FSMM and TS can only be differentiated after obtaining a blood profile and bone marrow biopsy that excludes the diagnosis of FSMM. A history of immunosuppression typically suggests TS. Histopathology often is equivocal in FSMM⁸; however, TEM reveals viral particles (TSPyV) in TS. Transmission electron microscopy in FSMM demonstrates fibrillary structures arranged in a paracrystalline configuration with unknown significance instead of viral particles. Despite the absence of viral particles on TEM, a low mean copy number of Merkel cell polyomavirus was isolated from a patient with FSMM who responded dramatically to treatment with topical cidofovir gel 1%.⁸ In addition to treating the underlying multiple myeloma in FSMM, topical cidofovir gel 1% also may have a role in treatment of these patients, suggesting a possible viral rather than simply paraneoplastic etiology of FSMM. Therefore, polyomavirus infection should be considered in the initial workup of any patient with fine facial follicular spicules.

The most effective management of TS in transplant recipients is to reduce immunosuppression to the lowest level possible without jeopardizing the transplanted organ.⁹ In our case, reduction of immunosuppressive drugs was not possible. In fact, immunosuppression in our patient was increased following evidence of early rejection of the heart transplant. Although manual extraction of the keratin spicules resulted in considerable improvement in a similar facial eruption in a patient with pediatric pre–B-cell acute lymphoblastic leukemia developing TS,¹⁰ it is impossible to apply this approach to patients such as ours who have thousands of tiny lesions. Fortunately, custom-compounded cidofovir gel 1% applied twice daily to the patient’s face and ears for 4 weeks led to near-complete clearance at follow-up (Figure 5). Due to the high cost of the medication (approaching $700 for one tube), our patient applied this medication to the face only several times weekly with excellent improvement. Thus, it appears that it is possible to suppress this virus with topical medication alone.

Polyomavirus infection should be considered in patients presenting with fine follicular spiny papules, especially those who are immunosuppressed. The possibility of coexisting multiple myeloma should be excluded.

Acknowledgment—We sincerely thank Glenn A. Hoskins (Jackson, Mississippi), the electron microscopy technologist, for the detection of viral particles and the electron microscope photographs.

To the Editor:

A 57-year-old man with hypertension, dyslipidemia, and congestive heart failure presented with a disfiguring eruption comprised of asymptomatic papules on the face that appeared 12 months post–heart transplantation. Immunosuppressive medications included mycophenolic acid and tacrolimus ointment (FK506). The pinpoint papules spread from the central face to the ears, arms, and legs. Physical examination revealed multiple 0.5- to 1-mm flesh-colored papules over the glabella, nose, nasolabial folds, philtrum, chin, ears, arms, and legs sparing the trunk. The initial appearance of the facial rash resembled the surface of a nutmeg grater with central white spiny excrescences overlying fine papules (spinulosism)(Figure 1). In addition, eyebrow alopecia was present.

A 3-mm punch biopsy of a papule with a central spine was performed on the left thigh. Microscopic examination revealed marked dilatation of anagen hair follicles with a proliferation of haphazard inner root sheath cells replacing the follicular lumen. Hair shafts were absent, and plugged infundibula were observed (Figure 2). The inner root sheath keratinocytes were enlarged and dystrophic with deeply eosinophilic trichohyalin granules (Figure 3). The epidermis, outer root sheath epithelium, and eccrine structures were unremarkable.

Transmission electron microscopy (TEM) confirmed the presence of intranuclear viral inclusions within affected inner root sheath keratinocytes composed of nonenveloped icosahedral viral particles measuring 33 to 38 nm in diameter (Figure 4). These findings morphologically were consistent with a polyomavirus. No intracytoplasmic or extracellular viral particles were identified. The clinical history, physical examination, histopathology, and electron microscopy features strongly supported the diagnosis of trichodysplasia spinulosa (TS) despite insufficient material being retrieved for polymerase chain reaction identification.

Trichodysplasia spinulosa was first described by Haycox et al¹ in 1999. The authors suggested a viral etiology. Eleven years later, TS-associated polyomavirus (TSPyV) was identified by van der Meijden et al.² Follicular keratinocytes are the specific target for TSPyV.³ Evidence has been presented suggesting that TS is caused by a primary infection or reactivation of TSPyV in the setting of immunosuppression.^4,5

Patients with TS present with papular eruptions that appear on the central face with spiny excrescences and various degrees of alopecia involving the eyebrows or eyelashes. Histopathologic features include distended hair follicles with expansion of inner root sheath cells, eosinophilic trichohyalin granules, and the absence of hair shafts. The viral protein can be verified through immunohistochemistry TSPyV VP1 staining that demonstrates co-localization with trichohyalin. Viral particles also can be visualized as 35- to 38-nm intranuclear particles with an organized crystalloid morphology on TEM.^6,7 The negative polymerase chain reaction in our patient could be the result of suboptimal template DNA concentration extracted from the limited amount of tissue remaining in the block after hematoxylin and eosin staining.

The clinical differential diagnosis of central facial spinulosism includes the follicular spicules of multiple myeloma (FSMM). In fact, FSMM and TS can only be differentiated after obtaining a blood profile and bone marrow biopsy that excludes the diagnosis of FSMM. A history of immunosuppression typically suggests TS. Histopathology often is equivocal in FSMM⁸; however, TEM reveals viral particles (TSPyV) in TS. Transmission electron microscopy in FSMM demonstrates fibrillary structures arranged in a paracrystalline configuration with unknown significance instead of viral particles. Despite the absence of viral particles on TEM, a low mean copy number of Merkel cell polyomavirus was isolated from a patient with FSMM who responded dramatically to treatment with topical cidofovir gel 1%.⁸ In addition to treating the underlying multiple myeloma in FSMM, topical cidofovir gel 1% also may have a role in treatment of these patients, suggesting a possible viral rather than simply paraneoplastic etiology of FSMM. Therefore, polyomavirus infection should be considered in the initial workup of any patient with fine facial follicular spicules.

The most effective management of TS in transplant recipients is to reduce immunosuppression to the lowest level possible without jeopardizing the transplanted organ.⁹ In our case, reduction of immunosuppressive drugs was not possible. In fact, immunosuppression in our patient was increased following evidence of early rejection of the heart transplant. Although manual extraction of the keratin spicules resulted in considerable improvement in a similar facial eruption in a patient with pediatric pre–B-cell acute lymphoblastic leukemia developing TS,¹⁰ it is impossible to apply this approach to patients such as ours who have thousands of tiny lesions. Fortunately, custom-compounded cidofovir gel 1% applied twice daily to the patient’s face and ears for 4 weeks led to near-complete clearance at follow-up (Figure 5). Due to the high cost of the medication (approaching $700 for one tube), our patient applied this medication to the face only several times weekly with excellent improvement. Thus, it appears that it is possible to suppress this virus with topical medication alone.

Polyomavirus infection should be considered in patients presenting with fine follicular spiny papules, especially those who are immunosuppressed. The possibility of coexisting multiple myeloma should be excluded.

Acknowledgment—We sincerely thank Glenn A. Hoskins (Jackson, Mississippi), the electron microscopy technologist, for the detection of viral particles and the electron microscope photographs.

To the Editor:

A 57-year-old man with hypertension, dyslipidemia, and congestive heart failure presented with a disfiguring eruption comprised of asymptomatic papules on the face that appeared 12 months post–heart transplantation. Immunosuppressive medications included mycophenolic acid and tacrolimus ointment (FK506). The pinpoint papules spread from the central face to the ears, arms, and legs. Physical examination revealed multiple 0.5- to 1-mm flesh-colored papules over the glabella, nose, nasolabial folds, philtrum, chin, ears, arms, and legs sparing the trunk. The initial appearance of the facial rash resembled the surface of a nutmeg grater with central white spiny excrescences overlying fine papules (spinulosism)(Figure 1). In addition, eyebrow alopecia was present.

A 3-mm punch biopsy of a papule with a central spine was performed on the left thigh. Microscopic examination revealed marked dilatation of anagen hair follicles with a proliferation of haphazard inner root sheath cells replacing the follicular lumen. Hair shafts were absent, and plugged infundibula were observed (Figure 2). The inner root sheath keratinocytes were enlarged and dystrophic with deeply eosinophilic trichohyalin granules (Figure 3). The epidermis, outer root sheath epithelium, and eccrine structures were unremarkable.

Transmission electron microscopy (TEM) confirmed the presence of intranuclear viral inclusions within affected inner root sheath keratinocytes composed of nonenveloped icosahedral viral particles measuring 33 to 38 nm in diameter (Figure 4). These findings morphologically were consistent with a polyomavirus. No intracytoplasmic or extracellular viral particles were identified. The clinical history, physical examination, histopathology, and electron microscopy features strongly supported the diagnosis of trichodysplasia spinulosa (TS) despite insufficient material being retrieved for polymerase chain reaction identification.

Trichodysplasia spinulosa was first described by Haycox et al¹ in 1999. The authors suggested a viral etiology. Eleven years later, TS-associated polyomavirus (TSPyV) was identified by van der Meijden et al.² Follicular keratinocytes are the specific target for TSPyV.³ Evidence has been presented suggesting that TS is caused by a primary infection or reactivation of TSPyV in the setting of immunosuppression.^4,5

Patients with TS present with papular eruptions that appear on the central face with spiny excrescences and various degrees of alopecia involving the eyebrows or eyelashes. Histopathologic features include distended hair follicles with expansion of inner root sheath cells, eosinophilic trichohyalin granules, and the absence of hair shafts. The viral protein can be verified through immunohistochemistry TSPyV VP1 staining that demonstrates co-localization with trichohyalin. Viral particles also can be visualized as 35- to 38-nm intranuclear particles with an organized crystalloid morphology on TEM.^6,7 The negative polymerase chain reaction in our patient could be the result of suboptimal template DNA concentration extracted from the limited amount of tissue remaining in the block after hematoxylin and eosin staining.

The clinical differential diagnosis of central facial spinulosism includes the follicular spicules of multiple myeloma (FSMM). In fact, FSMM and TS can only be differentiated after obtaining a blood profile and bone marrow biopsy that excludes the diagnosis of FSMM. A history of immunosuppression typically suggests TS. Histopathology often is equivocal in FSMM⁸; however, TEM reveals viral particles (TSPyV) in TS. Transmission electron microscopy in FSMM demonstrates fibrillary structures arranged in a paracrystalline configuration with unknown significance instead of viral particles. Despite the absence of viral particles on TEM, a low mean copy number of Merkel cell polyomavirus was isolated from a patient with FSMM who responded dramatically to treatment with topical cidofovir gel 1%.⁸ In addition to treating the underlying multiple myeloma in FSMM, topical cidofovir gel 1% also may have a role in treatment of these patients, suggesting a possible viral rather than simply paraneoplastic etiology of FSMM. Therefore, polyomavirus infection should be considered in the initial workup of any patient with fine facial follicular spicules.

The most effective management of TS in transplant recipients is to reduce immunosuppression to the lowest level possible without jeopardizing the transplanted organ.⁹ In our case, reduction of immunosuppressive drugs was not possible. In fact, immunosuppression in our patient was increased following evidence of early rejection of the heart transplant. Although manual extraction of the keratin spicules resulted in considerable improvement in a similar facial eruption in a patient with pediatric pre–B-cell acute lymphoblastic leukemia developing TS,¹⁰ it is impossible to apply this approach to patients such as ours who have thousands of tiny lesions. Fortunately, custom-compounded cidofovir gel 1% applied twice daily to the patient’s face and ears for 4 weeks led to near-complete clearance at follow-up (Figure 5). Due to the high cost of the medication (approaching $700 for one tube), our patient applied this medication to the face only several times weekly with excellent improvement. Thus, it appears that it is possible to suppress this virus with topical medication alone.

Polyomavirus infection should be considered in patients presenting with fine follicular spiny papules, especially those who are immunosuppressed. The possibility of coexisting multiple myeloma should be excluded.

Acknowledgment—We sincerely thank Glenn A. Hoskins (Jackson, Mississippi), the electron microscopy technologist, for the detection of viral particles and the electron microscope photographs.