Case Letter

To the Editor:

Liposomal doxorubicin hydrochloride is an anthracycline topoisomerase inhibitor indicated for ovarian cancer, AIDS-related Kaposi sarcoma, and multiple myeloma.¹ It also has been used with limited success in a clinical trial of previously treated patients with endometrial cancer.² The most common adverse reactions include asthenia, fatigue, fever, anorexia, nausea, vomiting, stomatitis, diarrhea, constipation, hand-and-foot syndrome, rash, neutropenia, thrombocytopenia, and anemia.¹

A 58-year-old woman with a history of stage IIIA endometrial cancer underwent a total abdominal hysterectomy and bilateral salpingo-oophorectomy soon after diagnosis. She then completed 5 high-dose-rate brachytherapy treatments and 6 cycles of paclitaxel and carboplatin. Follow-up imaging revealed pulmonary metastasis. The patient was then enrolled in a clinical trial but was switched to 40 mg/m² liposomal doxorubicin given once every 28 days for 5 cycles after progression of disease.

After each dose of doxorubicin, she developed redness of the palms and soles. Following the third cycle of doxorubicin, a painful rash involving the thighs and axilla appeared with some desquamation in the left axilla. Three weeks after the fourth dose of doxorubicin, she presented with severe worsening of the rash to involve the extensor elbows (Figure 1), back, and lower legs with bilateral axillary desquamation. The bilateral medial thighs were erythematous with maceration that was tender and blanchable (Figure 2). The total affected body surface area was 10% to 15%. There was no involvement of the mucosa. She was treated with hydrogel sheet dressings and silver sulfadiazine cream 1%.

Figure 1. Extensor surface of the elbow with an erythematous patch 3 weeks after the fourth dose of doxorubicin.

Figure 2. Medial thigh with erythema and maceration.

The patient’s rash was thought to be due to doxorubicin toxicity; however, a 4-mm punch biopsy specimen from the left thigh was taken for culture and hemotoxylin and eosin stain to rule out other possibilities. Biopsy was consistent with a drug reaction, revealing superficial perivascular dermatitis with keratinocyte atypia of the epidermis. Doxorubicin was discontinued and the rash resolved completely within 2 weeks, except for some thickening of the skin on the palms, soles, and thighs. After a delay of approximately 1 week, doxorubicin was resumed at a lower dose of 30 mg/m². No dermatologic symptoms followed treatment at this dose.

Four clinical patterns of doxorubicin toxicity are recognized. The most common pattern is acral erythema, also known as hand-and-foot syndrome, which is followed by desquamation of the palms and soles, occurring in approximately 50% of patients. Ten percent of patients experience a diffuse follicular rash with mild, diffuse, scaly erythema and follicular accentuation that often occurs over the lateral limbs but also may occur over the trunk. New melanotic macules may appear on the trunk or extremities including palms and soles.³ Finally, an intertrigolike eruption exacerbated by friction with erythematous patches over skin folds or in areas of friction also has been described.^3-5 Our patient presented with a combination of dermatologic toxicities including acral erythema and intertrigolike eruption. Acral erythema occurred in 24 of 60 patients and intertrigolike eruption occurred in 5 of 60 patients in one study.³ Another report documented both occurring together.⁵

Treatment of doxorubicin skin toxicity consists of reduction of the dose of doxorubicin, supportive care, and patient education. Specific treatments include topical wound care, emollient creams, and pain management with analgesics. Other interventions include wearing loose clothing, avoiding vigorous exercise, and sitting on padded surfaces.⁶

Doxorubicin skin toxicity presents in several clinical patterns. Although acral erythema is the most common pattern, severe intertrigolike eruptions similar to our case may occur. Physicians caring for patients receiving doxorubicin should be aware of the variety of presentations of skin toxicity and the possible need for dose reduction to decrease symptoms.

To the Editor:

Liposomal doxorubicin hydrochloride is an anthracycline topoisomerase inhibitor indicated for ovarian cancer, AIDS-related Kaposi sarcoma, and multiple myeloma.¹ It also has been used with limited success in a clinical trial of previously treated patients with endometrial cancer.² The most common adverse reactions include asthenia, fatigue, fever, anorexia, nausea, vomiting, stomatitis, diarrhea, constipation, hand-and-foot syndrome, rash, neutropenia, thrombocytopenia, and anemia.¹

A 58-year-old woman with a history of stage IIIA endometrial cancer underwent a total abdominal hysterectomy and bilateral salpingo-oophorectomy soon after diagnosis. She then completed 5 high-dose-rate brachytherapy treatments and 6 cycles of paclitaxel and carboplatin. Follow-up imaging revealed pulmonary metastasis. The patient was then enrolled in a clinical trial but was switched to 40 mg/m² liposomal doxorubicin given once every 28 days for 5 cycles after progression of disease.

After each dose of doxorubicin, she developed redness of the palms and soles. Following the third cycle of doxorubicin, a painful rash involving the thighs and axilla appeared with some desquamation in the left axilla. Three weeks after the fourth dose of doxorubicin, she presented with severe worsening of the rash to involve the extensor elbows (Figure 1), back, and lower legs with bilateral axillary desquamation. The bilateral medial thighs were erythematous with maceration that was tender and blanchable (Figure 2). The total affected body surface area was 10% to 15%. There was no involvement of the mucosa. She was treated with hydrogel sheet dressings and silver sulfadiazine cream 1%.

Figure 1. Extensor surface of the elbow with an erythematous patch 3 weeks after the fourth dose of doxorubicin.

Figure 2. Medial thigh with erythema and maceration.

The patient’s rash was thought to be due to doxorubicin toxicity; however, a 4-mm punch biopsy specimen from the left thigh was taken for culture and hemotoxylin and eosin stain to rule out other possibilities. Biopsy was consistent with a drug reaction, revealing superficial perivascular dermatitis with keratinocyte atypia of the epidermis. Doxorubicin was discontinued and the rash resolved completely within 2 weeks, except for some thickening of the skin on the palms, soles, and thighs. After a delay of approximately 1 week, doxorubicin was resumed at a lower dose of 30 mg/m². No dermatologic symptoms followed treatment at this dose.

Four clinical patterns of doxorubicin toxicity are recognized. The most common pattern is acral erythema, also known as hand-and-foot syndrome, which is followed by desquamation of the palms and soles, occurring in approximately 50% of patients. Ten percent of patients experience a diffuse follicular rash with mild, diffuse, scaly erythema and follicular accentuation that often occurs over the lateral limbs but also may occur over the trunk. New melanotic macules may appear on the trunk or extremities including palms and soles.³ Finally, an intertrigolike eruption exacerbated by friction with erythematous patches over skin folds or in areas of friction also has been described.^3-5 Our patient presented with a combination of dermatologic toxicities including acral erythema and intertrigolike eruption. Acral erythema occurred in 24 of 60 patients and intertrigolike eruption occurred in 5 of 60 patients in one study.³ Another report documented both occurring together.⁵

Treatment of doxorubicin skin toxicity consists of reduction of the dose of doxorubicin, supportive care, and patient education. Specific treatments include topical wound care, emollient creams, and pain management with analgesics. Other interventions include wearing loose clothing, avoiding vigorous exercise, and sitting on padded surfaces.⁶

Doxorubicin skin toxicity presents in several clinical patterns. Although acral erythema is the most common pattern, severe intertrigolike eruptions similar to our case may occur. Physicians caring for patients receiving doxorubicin should be aware of the variety of presentations of skin toxicity and the possible need for dose reduction to decrease symptoms.

To the Editor:

Liposomal doxorubicin hydrochloride is an anthracycline topoisomerase inhibitor indicated for ovarian cancer, AIDS-related Kaposi sarcoma, and multiple myeloma.¹ It also has been used with limited success in a clinical trial of previously treated patients with endometrial cancer.² The most common adverse reactions include asthenia, fatigue, fever, anorexia, nausea, vomiting, stomatitis, diarrhea, constipation, hand-and-foot syndrome, rash, neutropenia, thrombocytopenia, and anemia.¹

A 58-year-old woman with a history of stage IIIA endometrial cancer underwent a total abdominal hysterectomy and bilateral salpingo-oophorectomy soon after diagnosis. She then completed 5 high-dose-rate brachytherapy treatments and 6 cycles of paclitaxel and carboplatin. Follow-up imaging revealed pulmonary metastasis. The patient was then enrolled in a clinical trial but was switched to 40 mg/m² liposomal doxorubicin given once every 28 days for 5 cycles after progression of disease.

After each dose of doxorubicin, she developed redness of the palms and soles. Following the third cycle of doxorubicin, a painful rash involving the thighs and axilla appeared with some desquamation in the left axilla. Three weeks after the fourth dose of doxorubicin, she presented with severe worsening of the rash to involve the extensor elbows (Figure 1), back, and lower legs with bilateral axillary desquamation. The bilateral medial thighs were erythematous with maceration that was tender and blanchable (Figure 2). The total affected body surface area was 10% to 15%. There was no involvement of the mucosa. She was treated with hydrogel sheet dressings and silver sulfadiazine cream 1%.

Figure 1. Extensor surface of the elbow with an erythematous patch 3 weeks after the fourth dose of doxorubicin.

Figure 2. Medial thigh with erythema and maceration.

The patient’s rash was thought to be due to doxorubicin toxicity; however, a 4-mm punch biopsy specimen from the left thigh was taken for culture and hemotoxylin and eosin stain to rule out other possibilities. Biopsy was consistent with a drug reaction, revealing superficial perivascular dermatitis with keratinocyte atypia of the epidermis. Doxorubicin was discontinued and the rash resolved completely within 2 weeks, except for some thickening of the skin on the palms, soles, and thighs. After a delay of approximately 1 week, doxorubicin was resumed at a lower dose of 30 mg/m². No dermatologic symptoms followed treatment at this dose.

Four clinical patterns of doxorubicin toxicity are recognized. The most common pattern is acral erythema, also known as hand-and-foot syndrome, which is followed by desquamation of the palms and soles, occurring in approximately 50% of patients. Ten percent of patients experience a diffuse follicular rash with mild, diffuse, scaly erythema and follicular accentuation that often occurs over the lateral limbs but also may occur over the trunk. New melanotic macules may appear on the trunk or extremities including palms and soles.³ Finally, an intertrigolike eruption exacerbated by friction with erythematous patches over skin folds or in areas of friction also has been described.^3-5 Our patient presented with a combination of dermatologic toxicities including acral erythema and intertrigolike eruption. Acral erythema occurred in 24 of 60 patients and intertrigolike eruption occurred in 5 of 60 patients in one study.³ Another report documented both occurring together.⁵

Treatment of doxorubicin skin toxicity consists of reduction of the dose of doxorubicin, supportive care, and patient education. Specific treatments include topical wound care, emollient creams, and pain management with analgesics. Other interventions include wearing loose clothing, avoiding vigorous exercise, and sitting on padded surfaces.⁶

Doxorubicin skin toxicity presents in several clinical patterns. Although acral erythema is the most common pattern, severe intertrigolike eruptions similar to our case may occur. Physicians caring for patients receiving doxorubicin should be aware of the variety of presentations of skin toxicity and the possible need for dose reduction to decrease symptoms.

To the Editor:

Hereditary leiomyomatosis and renal cell carcinoma syndrome (HLRCCS) is a rare, highly penetrant, autosomal-dominant disorder that has been reported in approximately 200 families worldwide.^1,2 More than 90% of patients with HLRCCS develop multiple cutaneous leiomyomata, frequently in a segmental distribution, that increase in number and size with age. The extent of skin lesions is variable, even within the same family. Approximately 90% of female family members also have symptomatic uterine leiomyomata; 10% to 16% of these patients develop aggressive renal cell carcinomas,³ with more than 50% dying of metastatic disease within 5 years of diagnosis. Clinical diagnosis is established by the presence of multiple cutaneous leiomyomata, at least 1 of which should be histologically confirmed, or by a single leiomyoma in the presence of a positive family history.⁴

Mutations of fumarate hydratase (FH), a Krebs cycle enzyme that interconverts fumarate and malate, have been implicated in this syndrome.⁵ The homotetrameric 50 kDa protein exists in the mitochondrial matrix and the cytoplasm. Diagnosis is confirmed by molecular genetic testing for FH mutations or rarely by demonstrating reduced activity of FH enzyme. So far, at least 155 variations in DNA sequence of FH have been identified in HLRCCS. However, no definite genotype-phenotype correlations have been established yet. We present the case of a sporadic form of HLRCCS, which is rare.

A 27-year-old man presented with multiple slowly growing, painful lesions on the chest and back of 11 years’ duration. Physical examination revealed approximately twenty 2- to 4-mm pink-tan papules on the left side of the chest and several 2- to 7-mm tan-pink papules on the upper back (Figure 1A). The lesions were tender to touch, pressure, and cold temperatures. Microscopic examination of one of the lesions on the back showed benign smooth muscle proliferation expanding the reticular dermis, consistent with a cutaneous leiomyoma (Figure 1B).

Figure 1. Cluster of slow-growing, 2- to 7-mm, slightly erythematous papules on the upper back (A). Shave biopsy showed an unencapsulated dermal proliferation composed of interlacing fascicles of smooth muscle bundles with bland morphology, cigar-shaped nuclei, and lack of mitotic activity, compatible with cutaneous leiomyoma (B)(H&E, original magnification ×40).

Based on the clinical presentation, the possibility of HLRCCS was raised. Subsequently, the FH gene was sequenced from the peripheral blood revealing a heterozygous 4-base pair frameshift deletion mutation (TGAA deleted at positions 1083 through 1086 [complementary DNA][c.1083_1086delTGAA]), confirming the diagnosis (Figure 2). There was no family history of leiomyomata of the skin or uterus or renal tumors. Therefore, this case represents sporadic HLRCCS. Magnetic resonance imaging revealed only a 0.4-cm renal cortical cyst for which he was monitored for approximately a year but was lost to follow-up.

Figure 2. Sequencing analysis of the fumarate hydratase gene. DNA chromatograms: top, wild-type (WT) control; middle, patient (PT); bottom, comparison of WT and mutant DNA and protein sequences. Each gene located on autosomes has 2 copies, both of which are amplified during DNA sequencing. The height of peaks in the chromatograms represents the sum of nucleotides from both the copies. In this case (PT), there is a heterozygous c.1083_1086delTGAA 4-base pair deletion (TGAA deleted at positions 1083 through 1086 [complementary DNA]) in one copy and therefore the respective peak heights are reduced by approximately half compared to the WT. This deletion (underlined in bottom panel) leads to a frameshift in the coding sequence, resulting in altered amino acid sequence and a premature stop codon 10 codons downstream of the deletion, and thus a truncated protein.

The molecular mechanism of tumorigenesis in HLRCCS is poorly understood.⁶ Under normal circumstances, hypoxia-inducible factor (HIF) is hydroxylated by HIF prolyl hydroxylase after which it is targeted for an ubiquitin-mediated degradation (Figure 3 [top panel]). In the absence of FH, there is accumulation of fumarate, an inhibitor of HIF prolyl hydroxylase, leading to an increase in intracellular levels of unhydroxylated and undegradable HIF (Figure 3 [bottom panel]). Because of insufficient malate levels, the glucose metabolism through Krebs cycle shifts toward anaerobic glycolysis, even when sufficient oxygen is present to support respiration, creating a pseudohypoxic milieu that is similar to the Warburg effect. This environment leads to further stabilization of HIF, which is a transcription factor, that upregulates the expression of angiogenic factors (eg, vascular endothelial growth factor), growth factors (eg, erythropoietin, transforming growth factor a, platelet-derived growth factor), glucose transporters (eg, glucose transporter 1), and glycolytic enzymes (eg, phosphokinase mutase 1, lactate dehydrogenase A). These alterations may favor tumor growth by increasing the availability of biosynthetic intermediates needed for cellular proliferation and survival.

Figure 3.  Proposed mechanism of tumorigenesis in hereditary leiomyomatosis and renal cell carcinoma syndrome. In the presence of functional fumarate hydratase (FH), hypoxia-inducible factor (HIF) is degraded, resulting in normoxia (top panel). In the absence of functional FH, there is accumulation of fumarate, while malate levels decrease, and the glucose metabolism through Krebs cycle shifts toward anaerobic glycolysis, even when sufficient oxygen is present to support respiration (bottom panel). Increased fumarate inhibits HIF prolyl hydroxylase (HPH), which leads to stabilization of HIF, a transcription factor, that enhances anaerobic glycolysis, cellular proliferation, and angiogenesis, leading to tumor growth.

Patients with renal tumor–associated hereditary syndromes may present initially to dermatologists; therefore, it is important to recognize the cutaneous manifestations of these conditions because early diagnosis of renal cancer may prove to be lifesaving.

To the Editor:

Hereditary leiomyomatosis and renal cell carcinoma syndrome (HLRCCS) is a rare, highly penetrant, autosomal-dominant disorder that has been reported in approximately 200 families worldwide.^1,2 More than 90% of patients with HLRCCS develop multiple cutaneous leiomyomata, frequently in a segmental distribution, that increase in number and size with age. The extent of skin lesions is variable, even within the same family. Approximately 90% of female family members also have symptomatic uterine leiomyomata; 10% to 16% of these patients develop aggressive renal cell carcinomas,³ with more than 50% dying of metastatic disease within 5 years of diagnosis. Clinical diagnosis is established by the presence of multiple cutaneous leiomyomata, at least 1 of which should be histologically confirmed, or by a single leiomyoma in the presence of a positive family history.⁴

Mutations of fumarate hydratase (FH), a Krebs cycle enzyme that interconverts fumarate and malate, have been implicated in this syndrome.⁵ The homotetrameric 50 kDa protein exists in the mitochondrial matrix and the cytoplasm. Diagnosis is confirmed by molecular genetic testing for FH mutations or rarely by demonstrating reduced activity of FH enzyme. So far, at least 155 variations in DNA sequence of FH have been identified in HLRCCS. However, no definite genotype-phenotype correlations have been established yet. We present the case of a sporadic form of HLRCCS, which is rare.

A 27-year-old man presented with multiple slowly growing, painful lesions on the chest and back of 11 years’ duration. Physical examination revealed approximately twenty 2- to 4-mm pink-tan papules on the left side of the chest and several 2- to 7-mm tan-pink papules on the upper back (Figure 1A). The lesions were tender to touch, pressure, and cold temperatures. Microscopic examination of one of the lesions on the back showed benign smooth muscle proliferation expanding the reticular dermis, consistent with a cutaneous leiomyoma (Figure 1B).

Figure 1. Cluster of slow-growing, 2- to 7-mm, slightly erythematous papules on the upper back (A). Shave biopsy showed an unencapsulated dermal proliferation composed of interlacing fascicles of smooth muscle bundles with bland morphology, cigar-shaped nuclei, and lack of mitotic activity, compatible with cutaneous leiomyoma (B)(H&E, original magnification ×40).

Based on the clinical presentation, the possibility of HLRCCS was raised. Subsequently, the FH gene was sequenced from the peripheral blood revealing a heterozygous 4-base pair frameshift deletion mutation (TGAA deleted at positions 1083 through 1086 [complementary DNA][c.1083_1086delTGAA]), confirming the diagnosis (Figure 2). There was no family history of leiomyomata of the skin or uterus or renal tumors. Therefore, this case represents sporadic HLRCCS. Magnetic resonance imaging revealed only a 0.4-cm renal cortical cyst for which he was monitored for approximately a year but was lost to follow-up.

Figure 2. Sequencing analysis of the fumarate hydratase gene. DNA chromatograms: top, wild-type (WT) control; middle, patient (PT); bottom, comparison of WT and mutant DNA and protein sequences. Each gene located on autosomes has 2 copies, both of which are amplified during DNA sequencing. The height of peaks in the chromatograms represents the sum of nucleotides from both the copies. In this case (PT), there is a heterozygous c.1083_1086delTGAA 4-base pair deletion (TGAA deleted at positions 1083 through 1086 [complementary DNA]) in one copy and therefore the respective peak heights are reduced by approximately half compared to the WT. This deletion (underlined in bottom panel) leads to a frameshift in the coding sequence, resulting in altered amino acid sequence and a premature stop codon 10 codons downstream of the deletion, and thus a truncated protein.

The molecular mechanism of tumorigenesis in HLRCCS is poorly understood.⁶ Under normal circumstances, hypoxia-inducible factor (HIF) is hydroxylated by HIF prolyl hydroxylase after which it is targeted for an ubiquitin-mediated degradation (Figure 3 [top panel]). In the absence of FH, there is accumulation of fumarate, an inhibitor of HIF prolyl hydroxylase, leading to an increase in intracellular levels of unhydroxylated and undegradable HIF (Figure 3 [bottom panel]). Because of insufficient malate levels, the glucose metabolism through Krebs cycle shifts toward anaerobic glycolysis, even when sufficient oxygen is present to support respiration, creating a pseudohypoxic milieu that is similar to the Warburg effect. This environment leads to further stabilization of HIF, which is a transcription factor, that upregulates the expression of angiogenic factors (eg, vascular endothelial growth factor), growth factors (eg, erythropoietin, transforming growth factor a, platelet-derived growth factor), glucose transporters (eg, glucose transporter 1), and glycolytic enzymes (eg, phosphokinase mutase 1, lactate dehydrogenase A). These alterations may favor tumor growth by increasing the availability of biosynthetic intermediates needed for cellular proliferation and survival.

Figure 3.  Proposed mechanism of tumorigenesis in hereditary leiomyomatosis and renal cell carcinoma syndrome. In the presence of functional fumarate hydratase (FH), hypoxia-inducible factor (HIF) is degraded, resulting in normoxia (top panel). In the absence of functional FH, there is accumulation of fumarate, while malate levels decrease, and the glucose metabolism through Krebs cycle shifts toward anaerobic glycolysis, even when sufficient oxygen is present to support respiration (bottom panel). Increased fumarate inhibits HIF prolyl hydroxylase (HPH), which leads to stabilization of HIF, a transcription factor, that enhances anaerobic glycolysis, cellular proliferation, and angiogenesis, leading to tumor growth.

Patients with renal tumor–associated hereditary syndromes may present initially to dermatologists; therefore, it is important to recognize the cutaneous manifestations of these conditions because early diagnosis of renal cancer may prove to be lifesaving.

To the Editor:

Hereditary leiomyomatosis and renal cell carcinoma syndrome (HLRCCS) is a rare, highly penetrant, autosomal-dominant disorder that has been reported in approximately 200 families worldwide.^1,2 More than 90% of patients with HLRCCS develop multiple cutaneous leiomyomata, frequently in a segmental distribution, that increase in number and size with age. The extent of skin lesions is variable, even within the same family. Approximately 90% of female family members also have symptomatic uterine leiomyomata; 10% to 16% of these patients develop aggressive renal cell carcinomas,³ with more than 50% dying of metastatic disease within 5 years of diagnosis. Clinical diagnosis is established by the presence of multiple cutaneous leiomyomata, at least 1 of which should be histologically confirmed, or by a single leiomyoma in the presence of a positive family history.⁴

Mutations of fumarate hydratase (FH), a Krebs cycle enzyme that interconverts fumarate and malate, have been implicated in this syndrome.⁵ The homotetrameric 50 kDa protein exists in the mitochondrial matrix and the cytoplasm. Diagnosis is confirmed by molecular genetic testing for FH mutations or rarely by demonstrating reduced activity of FH enzyme. So far, at least 155 variations in DNA sequence of FH have been identified in HLRCCS. However, no definite genotype-phenotype correlations have been established yet. We present the case of a sporadic form of HLRCCS, which is rare.

A 27-year-old man presented with multiple slowly growing, painful lesions on the chest and back of 11 years’ duration. Physical examination revealed approximately twenty 2- to 4-mm pink-tan papules on the left side of the chest and several 2- to 7-mm tan-pink papules on the upper back (Figure 1A). The lesions were tender to touch, pressure, and cold temperatures. Microscopic examination of one of the lesions on the back showed benign smooth muscle proliferation expanding the reticular dermis, consistent with a cutaneous leiomyoma (Figure 1B).

Figure 1. Cluster of slow-growing, 2- to 7-mm, slightly erythematous papules on the upper back (A). Shave biopsy showed an unencapsulated dermal proliferation composed of interlacing fascicles of smooth muscle bundles with bland morphology, cigar-shaped nuclei, and lack of mitotic activity, compatible with cutaneous leiomyoma (B)(H&E, original magnification ×40).

Based on the clinical presentation, the possibility of HLRCCS was raised. Subsequently, the FH gene was sequenced from the peripheral blood revealing a heterozygous 4-base pair frameshift deletion mutation (TGAA deleted at positions 1083 through 1086 [complementary DNA][c.1083_1086delTGAA]), confirming the diagnosis (Figure 2). There was no family history of leiomyomata of the skin or uterus or renal tumors. Therefore, this case represents sporadic HLRCCS. Magnetic resonance imaging revealed only a 0.4-cm renal cortical cyst for which he was monitored for approximately a year but was lost to follow-up.

Figure 2. Sequencing analysis of the fumarate hydratase gene. DNA chromatograms: top, wild-type (WT) control; middle, patient (PT); bottom, comparison of WT and mutant DNA and protein sequences. Each gene located on autosomes has 2 copies, both of which are amplified during DNA sequencing. The height of peaks in the chromatograms represents the sum of nucleotides from both the copies. In this case (PT), there is a heterozygous c.1083_1086delTGAA 4-base pair deletion (TGAA deleted at positions 1083 through 1086 [complementary DNA]) in one copy and therefore the respective peak heights are reduced by approximately half compared to the WT. This deletion (underlined in bottom panel) leads to a frameshift in the coding sequence, resulting in altered amino acid sequence and a premature stop codon 10 codons downstream of the deletion, and thus a truncated protein.

The molecular mechanism of tumorigenesis in HLRCCS is poorly understood.⁶ Under normal circumstances, hypoxia-inducible factor (HIF) is hydroxylated by HIF prolyl hydroxylase after which it is targeted for an ubiquitin-mediated degradation (Figure 3 [top panel]). In the absence of FH, there is accumulation of fumarate, an inhibitor of HIF prolyl hydroxylase, leading to an increase in intracellular levels of unhydroxylated and undegradable HIF (Figure 3 [bottom panel]). Because of insufficient malate levels, the glucose metabolism through Krebs cycle shifts toward anaerobic glycolysis, even when sufficient oxygen is present to support respiration, creating a pseudohypoxic milieu that is similar to the Warburg effect. This environment leads to further stabilization of HIF, which is a transcription factor, that upregulates the expression of angiogenic factors (eg, vascular endothelial growth factor), growth factors (eg, erythropoietin, transforming growth factor a, platelet-derived growth factor), glucose transporters (eg, glucose transporter 1), and glycolytic enzymes (eg, phosphokinase mutase 1, lactate dehydrogenase A). These alterations may favor tumor growth by increasing the availability of biosynthetic intermediates needed for cellular proliferation and survival.

Figure 3.  Proposed mechanism of tumorigenesis in hereditary leiomyomatosis and renal cell carcinoma syndrome. In the presence of functional fumarate hydratase (FH), hypoxia-inducible factor (HIF) is degraded, resulting in normoxia (top panel). In the absence of functional FH, there is accumulation of fumarate, while malate levels decrease, and the glucose metabolism through Krebs cycle shifts toward anaerobic glycolysis, even when sufficient oxygen is present to support respiration (bottom panel). Increased fumarate inhibits HIF prolyl hydroxylase (HPH), which leads to stabilization of HIF, a transcription factor, that enhances anaerobic glycolysis, cellular proliferation, and angiogenesis, leading to tumor growth.

Patients with renal tumor–associated hereditary syndromes may present initially to dermatologists; therefore, it is important to recognize the cutaneous manifestations of these conditions because early diagnosis of renal cancer may prove to be lifesaving.

To the Editor:

Orificial tuberculosis (OT) constitutes 2% of cutaneous tuberculosis cases and 0.01% to 1% of all clinical presentations of tuberculosis.¹ It is clinically classified as primary or secondary OT. In primary OT, the oral mucosa is the initial site of the infection without any internal organ involvement.² This form is more prevalent among men and young adults.^1,3 Secondary OT is the cutaneous tuberculosis type that occurs in patients with internal organ tuberculosis from autoinoculation of bacilli to the orificial area. It is more common and usually affects elderly patients.^2-4 We present the development of primary OT in an immunocompetent woman.

A 51-year-old woman was admitted with painful enlarging oral ulcers of 1 year’s duration. There was no history of tuberculosis infection, dental trauma, or smoking habit prior to the development of oral ulcers, and no family history of tuberculosis. On dermatological examination white-yellow indurated ulcers with 1×1.5-cm irregular margins located on the hard palate and gingiva were observed (Figures 1A and 1B). Oral hygiene was good. There was no regional lym-phadenopathy on palpation. Physical findings were normal. The histopathology of the biopsy from the gingival ulcer revealed noncaseating granulomatous inflammation in the dermis (Figure 2). Ziehl-Neelsen and periodic acid–Schiff stains were negative for acid-fast bacilli and fungi, respectively.

Figure 1. White-yellow indurated ulcers with 1×1.5-cm irregular margins located on the hard palate (A) and gingiva (B). Resolution of lesions on the hard palate (C) and the gingiva (D) after 9 months of antituberculosis therapy.

Laboratory results from blood chemistry, complete blood cell count, erythrocyte sedimentation rate, C-reactive protein level, and urine analysis, as well as titers of serum immunoglobulin, antineutrophil cytoplasmic antibodies, and antinuclear antibodies, were within reference range. Human immunodeficiency virus serology was negative. Chest radiography and ultrasonography of the abdomen revealed no abnormalities. A purified protein derivative (tuberculin) test showed an induration of 20 mm. Mycobacterium tuberculosis grew on the culture of the tissue specimen.

Figure 2. Noncaseating granulomatous inflammation in the dermis (A)(H&E, original magnification ×200). Giant cells within the granuloma (B)(H&E, original magnification ×400).

The patient was diagnosed with primary OT and treated with isoniazid (300 mg daily), rifampin (600 mg daily), ethambutol (1500 mg daily), and pyrazinamide (2000 mg daily). At 2 months of therapy the lesions started to heal and showed complete resolution at the end of 9 months of treatment (Figures 1C and 1D). There was no recurrence at 2-year follow-up.

Orificial tuberculosis is a rare form of cutaneous tuberculosis. Therefore, it is regarded as a “forgotten disease” in the literature.⁵ The pathogenesis of OT has not been clearly defined. The intact mucous membrane, thickened squamous epithelium, cleansing and antibacterial function of saliva, and presence of saprophytes act as protective mechanisms against penetration of mycobacteria. Some factors such as poor oral hygiene, smoking, local trauma, or presence of a dental cyst or an abscess may predispose the direct mucosal inoculation of the tuberculosis bacilli.^1,2,6 Orificial tuberculosis may develop as an opportunistic infection in 1.33% of human immunodeficiency virus patients.⁷ However, our patient had no prior trauma or known predisposing factors.

The common presentation of OT is an ulcerative lesion with irregular and well-delineated margins with a yellowish granular base. A vesicle, papule, or nodule may precede the ulcers. The presence of yellow satellite nodules around the lesion is characteristic for OT.^1,2,6 Although there were no satellite nodules around the ulcerations in our case, the lesions had a yellowish granular base and irregular, well-delineated margins.

The tongue, gingiva, lips, tonsils, and epiglottis are the most common sites of involvement.^2,5,8 Hard palate involvement rarely has been reported, even in immunosuppressed patients.⁷ Enlarged painful cervical lymph nodes may accompany the disease. Most patients with OT have been reported to have active pulmonary tuberculosis at the time of diagnosis.^6,8 Thus, internal organ involvement, particularly the pulmonary system, should be checked in patients with OT. In our patient, the hard palate was involved together with the gingiva despite the absence of immunosuppression. No internal organ involvement was found in the systemic evaluation.

Orificial tuberculosis is a form of cutaneous tuberculosis that is difficult to diagnose because of the varying nature of clinical features, failure of growth of M tuberculosis on culture, and rarity of the disease.^1,2,6 As in our case, biopsies may not always exhibit a caseous necrosis, which is specific to tuberculosis.^6,7 Thus, it may be difficult to distinguish oral cavity tuberculosis from conditions demonstrating oral ulcers such as bullous diseases, trauma, fungal diseases, syphilis, sarcoidosis, or squamous cell carcinoma by evaluating only signs and symptoms.⁶ Clinical suspicion is the first and foremost step in the diagnostic process of OT. In our patient, OT was suspected in the differential diagnosis because the resistant oral ulcerations showed the most common presentation of OT: irregular and well-delineated margins and a yellowish granular base. By considering tuberculosis within the differential diagnosis in our patient, microbiologic cultivation was performed from the oral mucosa and accurate diagnosis was established by determination of the pathogen’s growth in the culture.

Because of the increased incidence of tuberculosis and unusual manifestations, clinicians may easily overlook OT. It should be considered in the differential diagnosis of resistant nodules or ulcers of the oral cavity.

To the Editor:

Orificial tuberculosis (OT) constitutes 2% of cutaneous tuberculosis cases and 0.01% to 1% of all clinical presentations of tuberculosis.¹ It is clinically classified as primary or secondary OT. In primary OT, the oral mucosa is the initial site of the infection without any internal organ involvement.² This form is more prevalent among men and young adults.^1,3 Secondary OT is the cutaneous tuberculosis type that occurs in patients with internal organ tuberculosis from autoinoculation of bacilli to the orificial area. It is more common and usually affects elderly patients.^2-4 We present the development of primary OT in an immunocompetent woman.

A 51-year-old woman was admitted with painful enlarging oral ulcers of 1 year’s duration. There was no history of tuberculosis infection, dental trauma, or smoking habit prior to the development of oral ulcers, and no family history of tuberculosis. On dermatological examination white-yellow indurated ulcers with 1×1.5-cm irregular margins located on the hard palate and gingiva were observed (Figures 1A and 1B). Oral hygiene was good. There was no regional lym-phadenopathy on palpation. Physical findings were normal. The histopathology of the biopsy from the gingival ulcer revealed noncaseating granulomatous inflammation in the dermis (Figure 2). Ziehl-Neelsen and periodic acid–Schiff stains were negative for acid-fast bacilli and fungi, respectively.

Figure 1. White-yellow indurated ulcers with 1×1.5-cm irregular margins located on the hard palate (A) and gingiva (B). Resolution of lesions on the hard palate (C) and the gingiva (D) after 9 months of antituberculosis therapy.

Laboratory results from blood chemistry, complete blood cell count, erythrocyte sedimentation rate, C-reactive protein level, and urine analysis, as well as titers of serum immunoglobulin, antineutrophil cytoplasmic antibodies, and antinuclear antibodies, were within reference range. Human immunodeficiency virus serology was negative. Chest radiography and ultrasonography of the abdomen revealed no abnormalities. A purified protein derivative (tuberculin) test showed an induration of 20 mm. Mycobacterium tuberculosis grew on the culture of the tissue specimen.

Figure 2. Noncaseating granulomatous inflammation in the dermis (A)(H&E, original magnification ×200). Giant cells within the granuloma (B)(H&E, original magnification ×400).

The patient was diagnosed with primary OT and treated with isoniazid (300 mg daily), rifampin (600 mg daily), ethambutol (1500 mg daily), and pyrazinamide (2000 mg daily). At 2 months of therapy the lesions started to heal and showed complete resolution at the end of 9 months of treatment (Figures 1C and 1D). There was no recurrence at 2-year follow-up.

Orificial tuberculosis is a rare form of cutaneous tuberculosis. Therefore, it is regarded as a “forgotten disease” in the literature.⁵ The pathogenesis of OT has not been clearly defined. The intact mucous membrane, thickened squamous epithelium, cleansing and antibacterial function of saliva, and presence of saprophytes act as protective mechanisms against penetration of mycobacteria. Some factors such as poor oral hygiene, smoking, local trauma, or presence of a dental cyst or an abscess may predispose the direct mucosal inoculation of the tuberculosis bacilli.^1,2,6 Orificial tuberculosis may develop as an opportunistic infection in 1.33% of human immunodeficiency virus patients.⁷ However, our patient had no prior trauma or known predisposing factors.

The common presentation of OT is an ulcerative lesion with irregular and well-delineated margins with a yellowish granular base. A vesicle, papule, or nodule may precede the ulcers. The presence of yellow satellite nodules around the lesion is characteristic for OT.^1,2,6 Although there were no satellite nodules around the ulcerations in our case, the lesions had a yellowish granular base and irregular, well-delineated margins.

The tongue, gingiva, lips, tonsils, and epiglottis are the most common sites of involvement.^2,5,8 Hard palate involvement rarely has been reported, even in immunosuppressed patients.⁷ Enlarged painful cervical lymph nodes may accompany the disease. Most patients with OT have been reported to have active pulmonary tuberculosis at the time of diagnosis.^6,8 Thus, internal organ involvement, particularly the pulmonary system, should be checked in patients with OT. In our patient, the hard palate was involved together with the gingiva despite the absence of immunosuppression. No internal organ involvement was found in the systemic evaluation.

Orificial tuberculosis is a form of cutaneous tuberculosis that is difficult to diagnose because of the varying nature of clinical features, failure of growth of M tuberculosis on culture, and rarity of the disease.^1,2,6 As in our case, biopsies may not always exhibit a caseous necrosis, which is specific to tuberculosis.^6,7 Thus, it may be difficult to distinguish oral cavity tuberculosis from conditions demonstrating oral ulcers such as bullous diseases, trauma, fungal diseases, syphilis, sarcoidosis, or squamous cell carcinoma by evaluating only signs and symptoms.⁶ Clinical suspicion is the first and foremost step in the diagnostic process of OT. In our patient, OT was suspected in the differential diagnosis because the resistant oral ulcerations showed the most common presentation of OT: irregular and well-delineated margins and a yellowish granular base. By considering tuberculosis within the differential diagnosis in our patient, microbiologic cultivation was performed from the oral mucosa and accurate diagnosis was established by determination of the pathogen’s growth in the culture.

Because of the increased incidence of tuberculosis and unusual manifestations, clinicians may easily overlook OT. It should be considered in the differential diagnosis of resistant nodules or ulcers of the oral cavity.

To the Editor:

Orificial tuberculosis (OT) constitutes 2% of cutaneous tuberculosis cases and 0.01% to 1% of all clinical presentations of tuberculosis.¹ It is clinically classified as primary or secondary OT. In primary OT, the oral mucosa is the initial site of the infection without any internal organ involvement.² This form is more prevalent among men and young adults.^1,3 Secondary OT is the cutaneous tuberculosis type that occurs in patients with internal organ tuberculosis from autoinoculation of bacilli to the orificial area. It is more common and usually affects elderly patients.^2-4 We present the development of primary OT in an immunocompetent woman.

A 51-year-old woman was admitted with painful enlarging oral ulcers of 1 year’s duration. There was no history of tuberculosis infection, dental trauma, or smoking habit prior to the development of oral ulcers, and no family history of tuberculosis. On dermatological examination white-yellow indurated ulcers with 1×1.5-cm irregular margins located on the hard palate and gingiva were observed (Figures 1A and 1B). Oral hygiene was good. There was no regional lym-phadenopathy on palpation. Physical findings were normal. The histopathology of the biopsy from the gingival ulcer revealed noncaseating granulomatous inflammation in the dermis (Figure 2). Ziehl-Neelsen and periodic acid–Schiff stains were negative for acid-fast bacilli and fungi, respectively.

Figure 1. White-yellow indurated ulcers with 1×1.5-cm irregular margins located on the hard palate (A) and gingiva (B). Resolution of lesions on the hard palate (C) and the gingiva (D) after 9 months of antituberculosis therapy.

Laboratory results from blood chemistry, complete blood cell count, erythrocyte sedimentation rate, C-reactive protein level, and urine analysis, as well as titers of serum immunoglobulin, antineutrophil cytoplasmic antibodies, and antinuclear antibodies, were within reference range. Human immunodeficiency virus serology was negative. Chest radiography and ultrasonography of the abdomen revealed no abnormalities. A purified protein derivative (tuberculin) test showed an induration of 20 mm. Mycobacterium tuberculosis grew on the culture of the tissue specimen.

Figure 2. Noncaseating granulomatous inflammation in the dermis (A)(H&E, original magnification ×200). Giant cells within the granuloma (B)(H&E, original magnification ×400).

The patient was diagnosed with primary OT and treated with isoniazid (300 mg daily), rifampin (600 mg daily), ethambutol (1500 mg daily), and pyrazinamide (2000 mg daily). At 2 months of therapy the lesions started to heal and showed complete resolution at the end of 9 months of treatment (Figures 1C and 1D). There was no recurrence at 2-year follow-up.

Orificial tuberculosis is a rare form of cutaneous tuberculosis. Therefore, it is regarded as a “forgotten disease” in the literature.⁵ The pathogenesis of OT has not been clearly defined. The intact mucous membrane, thickened squamous epithelium, cleansing and antibacterial function of saliva, and presence of saprophytes act as protective mechanisms against penetration of mycobacteria. Some factors such as poor oral hygiene, smoking, local trauma, or presence of a dental cyst or an abscess may predispose the direct mucosal inoculation of the tuberculosis bacilli.^1,2,6 Orificial tuberculosis may develop as an opportunistic infection in 1.33% of human immunodeficiency virus patients.⁷ However, our patient had no prior trauma or known predisposing factors.

The common presentation of OT is an ulcerative lesion with irregular and well-delineated margins with a yellowish granular base. A vesicle, papule, or nodule may precede the ulcers. The presence of yellow satellite nodules around the lesion is characteristic for OT.^1,2,6 Although there were no satellite nodules around the ulcerations in our case, the lesions had a yellowish granular base and irregular, well-delineated margins.

The tongue, gingiva, lips, tonsils, and epiglottis are the most common sites of involvement.^2,5,8 Hard palate involvement rarely has been reported, even in immunosuppressed patients.⁷ Enlarged painful cervical lymph nodes may accompany the disease. Most patients with OT have been reported to have active pulmonary tuberculosis at the time of diagnosis.^6,8 Thus, internal organ involvement, particularly the pulmonary system, should be checked in patients with OT. In our patient, the hard palate was involved together with the gingiva despite the absence of immunosuppression. No internal organ involvement was found in the systemic evaluation.

Orificial tuberculosis is a form of cutaneous tuberculosis that is difficult to diagnose because of the varying nature of clinical features, failure of growth of M tuberculosis on culture, and rarity of the disease.^1,2,6 As in our case, biopsies may not always exhibit a caseous necrosis, which is specific to tuberculosis.^6,7 Thus, it may be difficult to distinguish oral cavity tuberculosis from conditions demonstrating oral ulcers such as bullous diseases, trauma, fungal diseases, syphilis, sarcoidosis, or squamous cell carcinoma by evaluating only signs and symptoms.⁶ Clinical suspicion is the first and foremost step in the diagnostic process of OT. In our patient, OT was suspected in the differential diagnosis because the resistant oral ulcerations showed the most common presentation of OT: irregular and well-delineated margins and a yellowish granular base. By considering tuberculosis within the differential diagnosis in our patient, microbiologic cultivation was performed from the oral mucosa and accurate diagnosis was established by determination of the pathogen’s growth in the culture.

Because of the increased incidence of tuberculosis and unusual manifestations, clinicians may easily overlook OT. It should be considered in the differential diagnosis of resistant nodules or ulcers of the oral cavity.

To the Editor:

In a recent Cutis^® article, Cesaretti et al¹ reported a case of cutaneous metastasis from primary gastric cancer that appeared on a resection scar 6 years after remission and without any relapse of the primary tumor. We report a case of a 68-year-old man who was referred to the dermatology clinic with a 15×20-cm nonpruritic, nonscaly, bruiselike lesion on the right forearm of 1 month’s duration. Approximately 1.5 years prior to presentation, the patient was diagnosed with gastric adenocarcinoma (stage IV: T4N3M1) with hepatic and lung metastasis. Following 6 months of chemotherapy with cisplatin and 5-fluorouracil, a positron emission tomography–computed tomography scan was performed and showed a reduction in metastasis but growth of the primitive tumor. After 1 year of chemotherapy, the new positron emission tomography–computed tomography scan showed no metastases. However, the primitive tumor had increased in size.

One month prior to presentation to the dermatology department, a traumatic blood sample on the right forearm left the patient with a persistent ecchymosis. The lesion was thought to be a healing ecchymosis and no biopsy was performed. One month later, the skin lesion had become much thicker and more erythematous (Figure) but not larger. A skin biopsy of this well-defined plaque was performed. Histologic examination showed neovascularization, proliferative epithelial cells, and cytokeratin markers AE1/AE3 and CK20, leading to the diagnosis of skin metastasis of the gastric adenocarcinoma. Chemotherapy was discontinued because of the patient’s altered general status and palliative care was given until he died the following month (2 months after presentation).

Iatrogenic dissemination of cancer cells has been described often on scars of tumor surgery,² and in malignant melanoma, bruises and hematoma revealing preexisting metastases have been reported.^3,4 Our report of secondary metastasis on an ecchymosis suggests that the traumatic blood sample performed before the development of the metastasis caused circulating tumor cells in the skin, which led to their local proliferation. The skin metastasis was the first sign of relapse and was followed by alteration of the general status and death.

Our patient is an example of the “soil and seed”⁵ hypothesis. Our case illustrates the abilities of tumor cells to colonize the skin under favorable conditions and emphasizes the importance of minimizing bleeding events and iatrogenic seeding of internal neoplasms in daily practice.

To the Editor:

In a recent Cutis^® article, Cesaretti et al¹ reported a case of cutaneous metastasis from primary gastric cancer that appeared on a resection scar 6 years after remission and without any relapse of the primary tumor. We report a case of a 68-year-old man who was referred to the dermatology clinic with a 15×20-cm nonpruritic, nonscaly, bruiselike lesion on the right forearm of 1 month’s duration. Approximately 1.5 years prior to presentation, the patient was diagnosed with gastric adenocarcinoma (stage IV: T4N3M1) with hepatic and lung metastasis. Following 6 months of chemotherapy with cisplatin and 5-fluorouracil, a positron emission tomography–computed tomography scan was performed and showed a reduction in metastasis but growth of the primitive tumor. After 1 year of chemotherapy, the new positron emission tomography–computed tomography scan showed no metastases. However, the primitive tumor had increased in size.

One month prior to presentation to the dermatology department, a traumatic blood sample on the right forearm left the patient with a persistent ecchymosis. The lesion was thought to be a healing ecchymosis and no biopsy was performed. One month later, the skin lesion had become much thicker and more erythematous (Figure) but not larger. A skin biopsy of this well-defined plaque was performed. Histologic examination showed neovascularization, proliferative epithelial cells, and cytokeratin markers AE1/AE3 and CK20, leading to the diagnosis of skin metastasis of the gastric adenocarcinoma. Chemotherapy was discontinued because of the patient’s altered general status and palliative care was given until he died the following month (2 months after presentation).

Iatrogenic dissemination of cancer cells has been described often on scars of tumor surgery,² and in malignant melanoma, bruises and hematoma revealing preexisting metastases have been reported.^3,4 Our report of secondary metastasis on an ecchymosis suggests that the traumatic blood sample performed before the development of the metastasis caused circulating tumor cells in the skin, which led to their local proliferation. The skin metastasis was the first sign of relapse and was followed by alteration of the general status and death.

Our patient is an example of the “soil and seed”⁵ hypothesis. Our case illustrates the abilities of tumor cells to colonize the skin under favorable conditions and emphasizes the importance of minimizing bleeding events and iatrogenic seeding of internal neoplasms in daily practice.

To the Editor:

In a recent Cutis^® article, Cesaretti et al¹ reported a case of cutaneous metastasis from primary gastric cancer that appeared on a resection scar 6 years after remission and without any relapse of the primary tumor. We report a case of a 68-year-old man who was referred to the dermatology clinic with a 15×20-cm nonpruritic, nonscaly, bruiselike lesion on the right forearm of 1 month’s duration. Approximately 1.5 years prior to presentation, the patient was diagnosed with gastric adenocarcinoma (stage IV: T4N3M1) with hepatic and lung metastasis. Following 6 months of chemotherapy with cisplatin and 5-fluorouracil, a positron emission tomography–computed tomography scan was performed and showed a reduction in metastasis but growth of the primitive tumor. After 1 year of chemotherapy, the new positron emission tomography–computed tomography scan showed no metastases. However, the primitive tumor had increased in size.

One month prior to presentation to the dermatology department, a traumatic blood sample on the right forearm left the patient with a persistent ecchymosis. The lesion was thought to be a healing ecchymosis and no biopsy was performed. One month later, the skin lesion had become much thicker and more erythematous (Figure) but not larger. A skin biopsy of this well-defined plaque was performed. Histologic examination showed neovascularization, proliferative epithelial cells, and cytokeratin markers AE1/AE3 and CK20, leading to the diagnosis of skin metastasis of the gastric adenocarcinoma. Chemotherapy was discontinued because of the patient’s altered general status and palliative care was given until he died the following month (2 months after presentation).

Iatrogenic dissemination of cancer cells has been described often on scars of tumor surgery,² and in malignant melanoma, bruises and hematoma revealing preexisting metastases have been reported.^3,4 Our report of secondary metastasis on an ecchymosis suggests that the traumatic blood sample performed before the development of the metastasis caused circulating tumor cells in the skin, which led to their local proliferation. The skin metastasis was the first sign of relapse and was followed by alteration of the general status and death.

Our patient is an example of the “soil and seed”⁵ hypothesis. Our case illustrates the abilities of tumor cells to colonize the skin under favorable conditions and emphasizes the importance of minimizing bleeding events and iatrogenic seeding of internal neoplasms in daily practice.

To the Editor:

Cutaneous aspergillosis mostly has been reported in immunosuppressed hosts and usually is caused by Aspergillus flavus or Aspergillus fumigatus. We report the occurrence of primary cutaneous aspergillosis (PCA) caused by a relatively rare species, Aspergillus nidulans, in a middle-aged patient without overt immunosuppression or history of trauma.

A 57-year-old woman was referred to the dermatology outpatient department for evaluation of a lesion on the right hand of 1 month's duration. On examination the lesion measured approximately 4×3 cm with central necrosis (Figure 1). Her medical history was unremarkable and routine laboratory test results were within reference range.

The patient was an agricultural worker with no history of trauma. Her history was unremarkable. A 20% potassium hydroxide mount of the tissue revealed septate, branched, hyaline hyphae. A soft, wooly, greenish brown growth was observed after 3 days of incubation on Sabouraud dextrose agar (Figure 2). No growth was observed on dermatophyte test medium. A lactophenol cotton blue mount revealed columnar conidial heads with brown, short, smooth-walled conidiophores (Figures 3–6). Vesicles were hemispheric and small (8–12 µm in diameter), with metulae and phialides occurring in the upper portion. Conidia were globose (3–4 µm) and rough. Based on these findings the fungus was identified as A nidulans. The patient did not respond to daily oral ketoconazole, and after 1 month of therapy the lesion did not regress. She was eventually treated with oral itraconazole and the lesion completely healed within 15 weeks.

An overwhelming majority of the cases of cutaneous aspergillosis have been reported either in immunocompromised hosts (ie, leukemia, cutaneous T-cell lymphoma, Hodgkin disease, human immunodeficiency virus/AIDS, solid-organ or hematopoietic stem cell transplant recipients) or in patients with contributing risk factors (ie, severe burns, diabetes mellitus, preterm or underweight neonates, elderly patients). Two outbreaks of this condition have been reported in neonatal intensive care units, with the source of contamination being linked to nonsterile disposable gloves, incubators, and humidity chambers.^1,2 However, PCA is a relatively rare condition and often is associated with disruption of dermal integrity by trauma or maceration, followed by colonization of the wound by Aspergillus spores that are ubiquitously present in soil and decomposed vegetation.^3-5 Our case was remarkable, as the patient was not immunosuppressed and did not have a history of trauma. However, we surmise that fungal inoculation might have inadvertently occurred through some trivial trauma sustained through her professional work.

The 2 species that have most commonly been associated with PCA are A flavus and A fumigatus.^6,7 There have been isolated reports of PCA caused by other organisms such as Aspergillus niger,^8,9 Aspergillus terreus,¹⁰Aspergillus ustus,¹¹ or Aspergillus calidoustus.¹² In a report of a neutropenic 56-year-old patient suffering from acute myeloblastic leukemia, PCA developed in association with a double-lumen Hickman catheter after a period of prolonged hospitalization.¹³ A study by the National Institutes of Health (1976-1997) revealed 6 life-threatening cases of A nidulans infection in patients with chronic granulomatous disease.¹⁴

We did not perform antifungal susceptibility testing on the isolate in our patient. However, we observed disease that was refractory to ketoconazole therapy but successfully resolved with oral itraconazole. Antifungal susceptibility was noted in a large number of reported cases of Aspergillus infections that were resistant to conventional treatment, such as voriconazole, itraconazole, and amphotericin B.¹⁵ Thus antifungal susceptibility testing is necessary before starting treatment. There also have been reports of recurrence of cutaneous aspergillosis following incomplete and irregular treatment.¹⁶ Our case of PCA also failed to respond to ketoconazole therapy, thus stressing the need for thorough mycological characterization, including the determination of an antifungal susceptibility profile, for successful and complete management of this condition.

Acknowledgment
The authors would like to thank Arunaloke Chakraborti, MD, Chandigarh, India, for the help extended for identification of the fungus.

To the Editor:

Cutaneous aspergillosis mostly has been reported in immunosuppressed hosts and usually is caused by Aspergillus flavus or Aspergillus fumigatus. We report the occurrence of primary cutaneous aspergillosis (PCA) caused by a relatively rare species, Aspergillus nidulans, in a middle-aged patient without overt immunosuppression or history of trauma.

A 57-year-old woman was referred to the dermatology outpatient department for evaluation of a lesion on the right hand of 1 month's duration. On examination the lesion measured approximately 4×3 cm with central necrosis (Figure 1). Her medical history was unremarkable and routine laboratory test results were within reference range.

The patient was an agricultural worker with no history of trauma. Her history was unremarkable. A 20% potassium hydroxide mount of the tissue revealed septate, branched, hyaline hyphae. A soft, wooly, greenish brown growth was observed after 3 days of incubation on Sabouraud dextrose agar (Figure 2). No growth was observed on dermatophyte test medium. A lactophenol cotton blue mount revealed columnar conidial heads with brown, short, smooth-walled conidiophores (Figures 3–6). Vesicles were hemispheric and small (8–12 µm in diameter), with metulae and phialides occurring in the upper portion. Conidia were globose (3–4 µm) and rough. Based on these findings the fungus was identified as A nidulans. The patient did not respond to daily oral ketoconazole, and after 1 month of therapy the lesion did not regress. She was eventually treated with oral itraconazole and the lesion completely healed within 15 weeks.

An overwhelming majority of the cases of cutaneous aspergillosis have been reported either in immunocompromised hosts (ie, leukemia, cutaneous T-cell lymphoma, Hodgkin disease, human immunodeficiency virus/AIDS, solid-organ or hematopoietic stem cell transplant recipients) or in patients with contributing risk factors (ie, severe burns, diabetes mellitus, preterm or underweight neonates, elderly patients). Two outbreaks of this condition have been reported in neonatal intensive care units, with the source of contamination being linked to nonsterile disposable gloves, incubators, and humidity chambers.^1,2 However, PCA is a relatively rare condition and often is associated with disruption of dermal integrity by trauma or maceration, followed by colonization of the wound by Aspergillus spores that are ubiquitously present in soil and decomposed vegetation.^3-5 Our case was remarkable, as the patient was not immunosuppressed and did not have a history of trauma. However, we surmise that fungal inoculation might have inadvertently occurred through some trivial trauma sustained through her professional work.

The 2 species that have most commonly been associated with PCA are A flavus and A fumigatus.^6,7 There have been isolated reports of PCA caused by other organisms such as Aspergillus niger,^8,9 Aspergillus terreus,¹⁰Aspergillus ustus,¹¹ or Aspergillus calidoustus.¹² In a report of a neutropenic 56-year-old patient suffering from acute myeloblastic leukemia, PCA developed in association with a double-lumen Hickman catheter after a period of prolonged hospitalization.¹³ A study by the National Institutes of Health (1976-1997) revealed 6 life-threatening cases of A nidulans infection in patients with chronic granulomatous disease.¹⁴

We did not perform antifungal susceptibility testing on the isolate in our patient. However, we observed disease that was refractory to ketoconazole therapy but successfully resolved with oral itraconazole. Antifungal susceptibility was noted in a large number of reported cases of Aspergillus infections that were resistant to conventional treatment, such as voriconazole, itraconazole, and amphotericin B.¹⁵ Thus antifungal susceptibility testing is necessary before starting treatment. There also have been reports of recurrence of cutaneous aspergillosis following incomplete and irregular treatment.¹⁶ Our case of PCA also failed to respond to ketoconazole therapy, thus stressing the need for thorough mycological characterization, including the determination of an antifungal susceptibility profile, for successful and complete management of this condition.

Acknowledgment
The authors would like to thank Arunaloke Chakraborti, MD, Chandigarh, India, for the help extended for identification of the fungus.

To the Editor:

Cutaneous aspergillosis mostly has been reported in immunosuppressed hosts and usually is caused by Aspergillus flavus or Aspergillus fumigatus. We report the occurrence of primary cutaneous aspergillosis (PCA) caused by a relatively rare species, Aspergillus nidulans, in a middle-aged patient without overt immunosuppression or history of trauma.

A 57-year-old woman was referred to the dermatology outpatient department for evaluation of a lesion on the right hand of 1 month's duration. On examination the lesion measured approximately 4×3 cm with central necrosis (Figure 1). Her medical history was unremarkable and routine laboratory test results were within reference range.

The patient was an agricultural worker with no history of trauma. Her history was unremarkable. A 20% potassium hydroxide mount of the tissue revealed septate, branched, hyaline hyphae. A soft, wooly, greenish brown growth was observed after 3 days of incubation on Sabouraud dextrose agar (Figure 2). No growth was observed on dermatophyte test medium. A lactophenol cotton blue mount revealed columnar conidial heads with brown, short, smooth-walled conidiophores (Figures 3–6). Vesicles were hemispheric and small (8–12 µm in diameter), with metulae and phialides occurring in the upper portion. Conidia were globose (3–4 µm) and rough. Based on these findings the fungus was identified as A nidulans. The patient did not respond to daily oral ketoconazole, and after 1 month of therapy the lesion did not regress. She was eventually treated with oral itraconazole and the lesion completely healed within 15 weeks.

An overwhelming majority of the cases of cutaneous aspergillosis have been reported either in immunocompromised hosts (ie, leukemia, cutaneous T-cell lymphoma, Hodgkin disease, human immunodeficiency virus/AIDS, solid-organ or hematopoietic stem cell transplant recipients) or in patients with contributing risk factors (ie, severe burns, diabetes mellitus, preterm or underweight neonates, elderly patients). Two outbreaks of this condition have been reported in neonatal intensive care units, with the source of contamination being linked to nonsterile disposable gloves, incubators, and humidity chambers.^1,2 However, PCA is a relatively rare condition and often is associated with disruption of dermal integrity by trauma or maceration, followed by colonization of the wound by Aspergillus spores that are ubiquitously present in soil and decomposed vegetation.^3-5 Our case was remarkable, as the patient was not immunosuppressed and did not have a history of trauma. However, we surmise that fungal inoculation might have inadvertently occurred through some trivial trauma sustained through her professional work.

The 2 species that have most commonly been associated with PCA are A flavus and A fumigatus.^6,7 There have been isolated reports of PCA caused by other organisms such as Aspergillus niger,^8,9 Aspergillus terreus,¹⁰Aspergillus ustus,¹¹ or Aspergillus calidoustus.¹² In a report of a neutropenic 56-year-old patient suffering from acute myeloblastic leukemia, PCA developed in association with a double-lumen Hickman catheter after a period of prolonged hospitalization.¹³ A study by the National Institutes of Health (1976-1997) revealed 6 life-threatening cases of A nidulans infection in patients with chronic granulomatous disease.¹⁴

We did not perform antifungal susceptibility testing on the isolate in our patient. However, we observed disease that was refractory to ketoconazole therapy but successfully resolved with oral itraconazole. Antifungal susceptibility was noted in a large number of reported cases of Aspergillus infections that were resistant to conventional treatment, such as voriconazole, itraconazole, and amphotericin B.¹⁵ Thus antifungal susceptibility testing is necessary before starting treatment. There also have been reports of recurrence of cutaneous aspergillosis following incomplete and irregular treatment.¹⁶ Our case of PCA also failed to respond to ketoconazole therapy, thus stressing the need for thorough mycological characterization, including the determination of an antifungal susceptibility profile, for successful and complete management of this condition.

Acknowledgment
The authors would like to thank Arunaloke Chakraborti, MD, Chandigarh, India, for the help extended for identification of the fungus.

To the Editor:
Onychomycosis is a fungal infection of the nail plate by dermatophytes, yeasts, and nondermatophyte molds. It is a common problem with a prevalence of 10% to 12% in the United States.^1,2 The clinical presentation of onychomycosis is shown in the Figure. Although some patients may have mild asymptomatic cases of onychomycosis and do not inquire about treatment, many will have more advanced cases, presenting with pain and discomfort, secondary infection, unattractive appearance, or problems performing everyday functions. The goal of onychomycosis treatment is to eliminate the fungus, if possible, which usually restores the nail to its normal state when it fully grows out. Patients should be counseled that it is a long process that may take 6 months or more for fingernails and 12 to 18 months for toenails. These estimates are based on a growth rate of 2 to 3 mm per month for fingernails and 1 to 2 mm per month for toenails.³ Nails grow fastest during the teenaged years and slow down with advancing age.⁴ It should be noted that advanced cases of onychomycosis affecting the nail matrix may cause permanent scarring; therefore, the nail unit may still appear dystrophic after the causative organism is eliminated. The US Food and Drug Administration (FDA) defines a complete cure as negative potassium hydroxide preparation and negative fungal culture plus a completely normal appearance of the nail.

Treatment of onychomycosis poses a number of challenges. First, hyperkeratosis and the fungal mass may limit the delivery of topical and systemic drugs to the source of the infection. In addition, high rates of relapse and reinfection after treatment may be due to residual hyphae or spores.⁵ Furthermore, the extended length of treatment limits patient adherence and many patients are unwilling to forego wearing nail cosmetics during the course of some of the treatments.

The clinical presentation of onychomycosis. The great toenail has yellow discoloration of the nail plate, ridging, and subungual hyperkeratosis.

There are 4 approved classes of antifungal drugs for the treatment of onychomycosis: allylamines, azoles, morpholines, and hydroxypyridinones.⁶ The allylamines (eg, terbinafine) inhibit squalene epoxidase.⁷ Oral terbinafine (250 mg daily) taken for 6 weeks for fingernails and 12 weeks for toenails is considered the current systemic treatment preference in onychomycosis therapy⁸ with complete cure rates in 12-week studies of approximately 38%⁹ and 49%.¹⁰

The second class of drugs is the azoles, which inhibit lanosterol 14a-demethylase, a step in the ergosterol biosynthesis pathway.⁶ Two members of this class that are widely used in treating onychomycosis are oral itraconazole¹¹ and off-label oral fluconazole.¹² The approved dose for oral itraconazole is 200 mg daily for 3 months (or an alternative pulse regimen) with a reported complete cure rate of 14%.¹¹ Although fluconazole is not FDA approved for the treatment of onychomycosis in the United States, it is used extensively in other countries and to some extent off label in the United States. In a study of 362 patients with onychomycosis treated with oral fluconazole, complete cure rates were 48% in patients who received 450 mg weekly, 46% in those who received 300 mg weekly, and 37% in those who received 150 mg weekly for up to 9 months.¹² It should be noted that several oral triazole antifungals, namely albaconazole,¹³ posaconazole,¹⁴ and ravuconazole,¹⁵ have undergone phase 1 and 2 studies for the treatment of onychomycosis and have shown some efficacy.

Another class of antifungals are the morpholines including topical amorolfine, which is approved for use in Europe but not in North America.¹⁶ Amorolfine inhibits D14 reductase and D7-D8 isomerase, thus depleting ergosterol.¹⁷ In one randomized controlled study, the combination of amorolfine nail lacquer and oral terbinafine compared to oral terbinafine alone resulted in a higher clinical cure rate with the combination (59.2% vs 46%); complete cure rate was not reported.¹⁶

Finally, the hydroxypyridinone class includes topical ciclopirox, which has a poorly understood mechanism of action but may involve iron chelation or oxidative damage.^18,19 Ciclopirox nail lacquer 8% was approved by the FDA in 1999 and has reported complete cure rates of 5.5% to 8.5% with monthly nail debridement.²⁰

Based on the poor efficacy of many of the currently available treatments and time-consuming treatment courses, it is clear that there is a need for alternative and novel therapies. There has been a greater emphasis on topical agents due to their more favorable side-effect profile and lower risk for drug-drug interactions. Although there are many agents for the treatment of onychomycosis currently in development, many are in vitro studies or phase 1 and 2 studies. However, we will focus on drugs that are further along in phase 3 studies and those that were recently FDA approved.

Efinaconazole is a member of the azole class of drugs and has completed 2 phase 3 clinical trials (study 1, N=870; study 2, N=785).²¹ Patients in these 2 studies were randomized to receive either efinaconazole nail solution 10% or vehicle for 48 weeks followed by a 4-week washout period. Complete cure rates in the 2 studies were 17.8% and 15.2% in the treated group and 3.3% and 5.5% in the control group. The mycological cure rates were 55.2% and 53.4% in the treated group and 16.8% and 16.9% in the control group. The side-effect profile was minimal, with the most common adverse events being application-site dermatitis and vesiculation, which were not significantly higher in the treated group versus the control group.²¹ Efinaconazole received FDA approval for the treatment of toenail onychomycosis in June 2014.

There are some notable differences between ciclopirox and efinaconazole that may improve patient compliance with the latter. First, treatment with ciclopirox includes monthly nail debridement, which is not required with efinaconazole. Secondly, although ciclopirox lacquer must be removed weekly, efinaconazole is a solution, so no removal is necessary.

Terbinafine nail solution (TNS) is a member of the allylamine class and has completed phase 3 clinical trials.²² Three studies—2 vehicle controlled and 1 active comparator—were performed. The first compared TNS and vehicle, both applied daily for 24 weeks; the second study repeated the same for 48 weeks; and the third study compared TNS to amorolfine nail lacquer 5% daily for 48 weeks. The best results for complete cure were achieved with TNS for 48 weeks in the vehicle-controlled study with a rate of 2.2% versus 0%. The authors also concluded TNS was not more effective than amorolfine, as complete cure rates were 1.2% for TNS and 0.96% for amorolfine. The most common side effects were headache, nasopharyngitis, and influenza.²²

Tavaborole is a member of the new benzoxaborole class, which inhibits protein synthesis by forming an adduct with the aminoacyl–transfer RNA synthetase.²³ The topical solution was engineered to have improved penetration through the nail plate. In vitro studies showed better penetration than both ciclopirox and amorolfine.²⁴ Two identical phase 3 randomized, double-blind, vehicle-controlled studies were completed involving 1197 patients who were treated with tavaborole topical solution 5% daily compared to vehicle for 48 weeks followed by a 4-week washout period with promising results.²⁵ The incidence of treatment-related side effects was comparable to the vehicle. The most common adverse events were exfoliation, erythema, and dermatitis, all occurring at the application site.²⁵ Tavaborole was approved by the FDA for the treatment of toenail onychomycosis in July 2014.

Luliconazole is a member of the azole class and a phase 2b/3 clinical trial with a 10% solution involving 334 patients was completed in June 2013.²⁶ Results from this trial are expected in early 2015.

Lasers are a developing area for onychomycosis therapy and the appeal stems from their ability to selectively deliver energy to the target tissue, thus avoiding systemic side effects. Since 2010, the FDA has approved numerous laser devices for the temporary cosmetic improvement of onychomycosis, all of which are Nd:YAG 1064-nm lasers.^27,28 It was previously thought that the mechanism of action for the fungicidal effect was achieved with heat,²⁹ but newer in vitro studies have shown that the amount of time and level of heat required to kill Trichophyton rubrum would not be tolerable to patients.³⁰ Although the mechanism of action is poorly understood, some clinical trials have shown success using the Nd:YAG 1064-nm laser for treatment of onychomycosis. However, in a study of 8 patients treated with the Nd:YAG 1064-nm laser for 5 treatment sessions, none had a mycological or clinical cure and there was only mild clinical improvement. In addition, most patients had pain and burning during the treatments requiring many short breaks.³⁰ Although not yet FDA approved for the treatment of onychomycosis, other types of lasers are currently being studied, including CO₂, near-infrared diode, and femtosecond-infrared laser systems.³

Plasma therapy is a developing area for the treatment of onychomycosis. Plasma was shown to be fungicidal to T rubrum in an in vitro model (MOE Medical Devices, written communication, July 2012), and a clinical trial to evaluate the safety, tolerability, and efficacy of plasma in human subjects is ongoing (registered on March 22, 2013, at www.clinicaltrials.gov with the identifier NCT01819051).

Onychomycosis is a common problem that increases in prevalence with advancing age. Oral terbinafine is considered the first-line treatment at this point in time.³¹ Two new topical agents, efinaconazole and tavaborole, were recently FDA approved and may be used for the treatment of toenail onychomycosis without the need for nail debridement. The Nd:YAG laser has shown some promise in earlier clinical studies but was ineffective in a more recent study.

To the Editor:
Onychomycosis is a fungal infection of the nail plate by dermatophytes, yeasts, and nondermatophyte molds. It is a common problem with a prevalence of 10% to 12% in the United States.^1,2 The clinical presentation of onychomycosis is shown in the Figure. Although some patients may have mild asymptomatic cases of onychomycosis and do not inquire about treatment, many will have more advanced cases, presenting with pain and discomfort, secondary infection, unattractive appearance, or problems performing everyday functions. The goal of onychomycosis treatment is to eliminate the fungus, if possible, which usually restores the nail to its normal state when it fully grows out. Patients should be counseled that it is a long process that may take 6 months or more for fingernails and 12 to 18 months for toenails. These estimates are based on a growth rate of 2 to 3 mm per month for fingernails and 1 to 2 mm per month for toenails.³ Nails grow fastest during the teenaged years and slow down with advancing age.⁴ It should be noted that advanced cases of onychomycosis affecting the nail matrix may cause permanent scarring; therefore, the nail unit may still appear dystrophic after the causative organism is eliminated. The US Food and Drug Administration (FDA) defines a complete cure as negative potassium hydroxide preparation and negative fungal culture plus a completely normal appearance of the nail.

Treatment of onychomycosis poses a number of challenges. First, hyperkeratosis and the fungal mass may limit the delivery of topical and systemic drugs to the source of the infection. In addition, high rates of relapse and reinfection after treatment may be due to residual hyphae or spores.⁵ Furthermore, the extended length of treatment limits patient adherence and many patients are unwilling to forego wearing nail cosmetics during the course of some of the treatments.

The clinical presentation of onychomycosis. The great toenail has yellow discoloration of the nail plate, ridging, and subungual hyperkeratosis.

There are 4 approved classes of antifungal drugs for the treatment of onychomycosis: allylamines, azoles, morpholines, and hydroxypyridinones.⁶ The allylamines (eg, terbinafine) inhibit squalene epoxidase.⁷ Oral terbinafine (250 mg daily) taken for 6 weeks for fingernails and 12 weeks for toenails is considered the current systemic treatment preference in onychomycosis therapy⁸ with complete cure rates in 12-week studies of approximately 38%⁹ and 49%.¹⁰

The second class of drugs is the azoles, which inhibit lanosterol 14a-demethylase, a step in the ergosterol biosynthesis pathway.⁶ Two members of this class that are widely used in treating onychomycosis are oral itraconazole¹¹ and off-label oral fluconazole.¹² The approved dose for oral itraconazole is 200 mg daily for 3 months (or an alternative pulse regimen) with a reported complete cure rate of 14%.¹¹ Although fluconazole is not FDA approved for the treatment of onychomycosis in the United States, it is used extensively in other countries and to some extent off label in the United States. In a study of 362 patients with onychomycosis treated with oral fluconazole, complete cure rates were 48% in patients who received 450 mg weekly, 46% in those who received 300 mg weekly, and 37% in those who received 150 mg weekly for up to 9 months.¹² It should be noted that several oral triazole antifungals, namely albaconazole,¹³ posaconazole,¹⁴ and ravuconazole,¹⁵ have undergone phase 1 and 2 studies for the treatment of onychomycosis and have shown some efficacy.

Another class of antifungals are the morpholines including topical amorolfine, which is approved for use in Europe but not in North America.¹⁶ Amorolfine inhibits D14 reductase and D7-D8 isomerase, thus depleting ergosterol.¹⁷ In one randomized controlled study, the combination of amorolfine nail lacquer and oral terbinafine compared to oral terbinafine alone resulted in a higher clinical cure rate with the combination (59.2% vs 46%); complete cure rate was not reported.¹⁶

Finally, the hydroxypyridinone class includes topical ciclopirox, which has a poorly understood mechanism of action but may involve iron chelation or oxidative damage.^18,19 Ciclopirox nail lacquer 8% was approved by the FDA in 1999 and has reported complete cure rates of 5.5% to 8.5% with monthly nail debridement.²⁰

Based on the poor efficacy of many of the currently available treatments and time-consuming treatment courses, it is clear that there is a need for alternative and novel therapies. There has been a greater emphasis on topical agents due to their more favorable side-effect profile and lower risk for drug-drug interactions. Although there are many agents for the treatment of onychomycosis currently in development, many are in vitro studies or phase 1 and 2 studies. However, we will focus on drugs that are further along in phase 3 studies and those that were recently FDA approved.

Efinaconazole is a member of the azole class of drugs and has completed 2 phase 3 clinical trials (study 1, N=870; study 2, N=785).²¹ Patients in these 2 studies were randomized to receive either efinaconazole nail solution 10% or vehicle for 48 weeks followed by a 4-week washout period. Complete cure rates in the 2 studies were 17.8% and 15.2% in the treated group and 3.3% and 5.5% in the control group. The mycological cure rates were 55.2% and 53.4% in the treated group and 16.8% and 16.9% in the control group. The side-effect profile was minimal, with the most common adverse events being application-site dermatitis and vesiculation, which were not significantly higher in the treated group versus the control group.²¹ Efinaconazole received FDA approval for the treatment of toenail onychomycosis in June 2014.

There are some notable differences between ciclopirox and efinaconazole that may improve patient compliance with the latter. First, treatment with ciclopirox includes monthly nail debridement, which is not required with efinaconazole. Secondly, although ciclopirox lacquer must be removed weekly, efinaconazole is a solution, so no removal is necessary.

Terbinafine nail solution (TNS) is a member of the allylamine class and has completed phase 3 clinical trials.²² Three studies—2 vehicle controlled and 1 active comparator—were performed. The first compared TNS and vehicle, both applied daily for 24 weeks; the second study repeated the same for 48 weeks; and the third study compared TNS to amorolfine nail lacquer 5% daily for 48 weeks. The best results for complete cure were achieved with TNS for 48 weeks in the vehicle-controlled study with a rate of 2.2% versus 0%. The authors also concluded TNS was not more effective than amorolfine, as complete cure rates were 1.2% for TNS and 0.96% for amorolfine. The most common side effects were headache, nasopharyngitis, and influenza.²²

Tavaborole is a member of the new benzoxaborole class, which inhibits protein synthesis by forming an adduct with the aminoacyl–transfer RNA synthetase.²³ The topical solution was engineered to have improved penetration through the nail plate. In vitro studies showed better penetration than both ciclopirox and amorolfine.²⁴ Two identical phase 3 randomized, double-blind, vehicle-controlled studies were completed involving 1197 patients who were treated with tavaborole topical solution 5% daily compared to vehicle for 48 weeks followed by a 4-week washout period with promising results.²⁵ The incidence of treatment-related side effects was comparable to the vehicle. The most common adverse events were exfoliation, erythema, and dermatitis, all occurring at the application site.²⁵ Tavaborole was approved by the FDA for the treatment of toenail onychomycosis in July 2014.

Luliconazole is a member of the azole class and a phase 2b/3 clinical trial with a 10% solution involving 334 patients was completed in June 2013.²⁶ Results from this trial are expected in early 2015.

Lasers are a developing area for onychomycosis therapy and the appeal stems from their ability to selectively deliver energy to the target tissue, thus avoiding systemic side effects. Since 2010, the FDA has approved numerous laser devices for the temporary cosmetic improvement of onychomycosis, all of which are Nd:YAG 1064-nm lasers.^27,28 It was previously thought that the mechanism of action for the fungicidal effect was achieved with heat,²⁹ but newer in vitro studies have shown that the amount of time and level of heat required to kill Trichophyton rubrum would not be tolerable to patients.³⁰ Although the mechanism of action is poorly understood, some clinical trials have shown success using the Nd:YAG 1064-nm laser for treatment of onychomycosis. However, in a study of 8 patients treated with the Nd:YAG 1064-nm laser for 5 treatment sessions, none had a mycological or clinical cure and there was only mild clinical improvement. In addition, most patients had pain and burning during the treatments requiring many short breaks.³⁰ Although not yet FDA approved for the treatment of onychomycosis, other types of lasers are currently being studied, including CO₂, near-infrared diode, and femtosecond-infrared laser systems.³

Plasma therapy is a developing area for the treatment of onychomycosis. Plasma was shown to be fungicidal to T rubrum in an in vitro model (MOE Medical Devices, written communication, July 2012), and a clinical trial to evaluate the safety, tolerability, and efficacy of plasma in human subjects is ongoing (registered on March 22, 2013, at www.clinicaltrials.gov with the identifier NCT01819051).

Onychomycosis is a common problem that increases in prevalence with advancing age. Oral terbinafine is considered the first-line treatment at this point in time.³¹ Two new topical agents, efinaconazole and tavaborole, were recently FDA approved and may be used for the treatment of toenail onychomycosis without the need for nail debridement. The Nd:YAG laser has shown some promise in earlier clinical studies but was ineffective in a more recent study.

To the Editor:
Onychomycosis is a fungal infection of the nail plate by dermatophytes, yeasts, and nondermatophyte molds. It is a common problem with a prevalence of 10% to 12% in the United States.^1,2 The clinical presentation of onychomycosis is shown in the Figure. Although some patients may have mild asymptomatic cases of onychomycosis and do not inquire about treatment, many will have more advanced cases, presenting with pain and discomfort, secondary infection, unattractive appearance, or problems performing everyday functions. The goal of onychomycosis treatment is to eliminate the fungus, if possible, which usually restores the nail to its normal state when it fully grows out. Patients should be counseled that it is a long process that may take 6 months or more for fingernails and 12 to 18 months for toenails. These estimates are based on a growth rate of 2 to 3 mm per month for fingernails and 1 to 2 mm per month for toenails.³ Nails grow fastest during the teenaged years and slow down with advancing age.⁴ It should be noted that advanced cases of onychomycosis affecting the nail matrix may cause permanent scarring; therefore, the nail unit may still appear dystrophic after the causative organism is eliminated. The US Food and Drug Administration (FDA) defines a complete cure as negative potassium hydroxide preparation and negative fungal culture plus a completely normal appearance of the nail.

Treatment of onychomycosis poses a number of challenges. First, hyperkeratosis and the fungal mass may limit the delivery of topical and systemic drugs to the source of the infection. In addition, high rates of relapse and reinfection after treatment may be due to residual hyphae or spores.⁵ Furthermore, the extended length of treatment limits patient adherence and many patients are unwilling to forego wearing nail cosmetics during the course of some of the treatments.

The clinical presentation of onychomycosis. The great toenail has yellow discoloration of the nail plate, ridging, and subungual hyperkeratosis.

There are 4 approved classes of antifungal drugs for the treatment of onychomycosis: allylamines, azoles, morpholines, and hydroxypyridinones.⁶ The allylamines (eg, terbinafine) inhibit squalene epoxidase.⁷ Oral terbinafine (250 mg daily) taken for 6 weeks for fingernails and 12 weeks for toenails is considered the current systemic treatment preference in onychomycosis therapy⁸ with complete cure rates in 12-week studies of approximately 38%⁹ and 49%.¹⁰

The second class of drugs is the azoles, which inhibit lanosterol 14a-demethylase, a step in the ergosterol biosynthesis pathway.⁶ Two members of this class that are widely used in treating onychomycosis are oral itraconazole¹¹ and off-label oral fluconazole.¹² The approved dose for oral itraconazole is 200 mg daily for 3 months (or an alternative pulse regimen) with a reported complete cure rate of 14%.¹¹ Although fluconazole is not FDA approved for the treatment of onychomycosis in the United States, it is used extensively in other countries and to some extent off label in the United States. In a study of 362 patients with onychomycosis treated with oral fluconazole, complete cure rates were 48% in patients who received 450 mg weekly, 46% in those who received 300 mg weekly, and 37% in those who received 150 mg weekly for up to 9 months.¹² It should be noted that several oral triazole antifungals, namely albaconazole,¹³ posaconazole,¹⁴ and ravuconazole,¹⁵ have undergone phase 1 and 2 studies for the treatment of onychomycosis and have shown some efficacy.

Another class of antifungals are the morpholines including topical amorolfine, which is approved for use in Europe but not in North America.¹⁶ Amorolfine inhibits D14 reductase and D7-D8 isomerase, thus depleting ergosterol.¹⁷ In one randomized controlled study, the combination of amorolfine nail lacquer and oral terbinafine compared to oral terbinafine alone resulted in a higher clinical cure rate with the combination (59.2% vs 46%); complete cure rate was not reported.¹⁶

Finally, the hydroxypyridinone class includes topical ciclopirox, which has a poorly understood mechanism of action but may involve iron chelation or oxidative damage.^18,19 Ciclopirox nail lacquer 8% was approved by the FDA in 1999 and has reported complete cure rates of 5.5% to 8.5% with monthly nail debridement.²⁰

Based on the poor efficacy of many of the currently available treatments and time-consuming treatment courses, it is clear that there is a need for alternative and novel therapies. There has been a greater emphasis on topical agents due to their more favorable side-effect profile and lower risk for drug-drug interactions. Although there are many agents for the treatment of onychomycosis currently in development, many are in vitro studies or phase 1 and 2 studies. However, we will focus on drugs that are further along in phase 3 studies and those that were recently FDA approved.

Efinaconazole is a member of the azole class of drugs and has completed 2 phase 3 clinical trials (study 1, N=870; study 2, N=785).²¹ Patients in these 2 studies were randomized to receive either efinaconazole nail solution 10% or vehicle for 48 weeks followed by a 4-week washout period. Complete cure rates in the 2 studies were 17.8% and 15.2% in the treated group and 3.3% and 5.5% in the control group. The mycological cure rates were 55.2% and 53.4% in the treated group and 16.8% and 16.9% in the control group. The side-effect profile was minimal, with the most common adverse events being application-site dermatitis and vesiculation, which were not significantly higher in the treated group versus the control group.²¹ Efinaconazole received FDA approval for the treatment of toenail onychomycosis in June 2014.

There are some notable differences between ciclopirox and efinaconazole that may improve patient compliance with the latter. First, treatment with ciclopirox includes monthly nail debridement, which is not required with efinaconazole. Secondly, although ciclopirox lacquer must be removed weekly, efinaconazole is a solution, so no removal is necessary.

Terbinafine nail solution (TNS) is a member of the allylamine class and has completed phase 3 clinical trials.²² Three studies—2 vehicle controlled and 1 active comparator—were performed. The first compared TNS and vehicle, both applied daily for 24 weeks; the second study repeated the same for 48 weeks; and the third study compared TNS to amorolfine nail lacquer 5% daily for 48 weeks. The best results for complete cure were achieved with TNS for 48 weeks in the vehicle-controlled study with a rate of 2.2% versus 0%. The authors also concluded TNS was not more effective than amorolfine, as complete cure rates were 1.2% for TNS and 0.96% for amorolfine. The most common side effects were headache, nasopharyngitis, and influenza.²²

Tavaborole is a member of the new benzoxaborole class, which inhibits protein synthesis by forming an adduct with the aminoacyl–transfer RNA synthetase.²³ The topical solution was engineered to have improved penetration through the nail plate. In vitro studies showed better penetration than both ciclopirox and amorolfine.²⁴ Two identical phase 3 randomized, double-blind, vehicle-controlled studies were completed involving 1197 patients who were treated with tavaborole topical solution 5% daily compared to vehicle for 48 weeks followed by a 4-week washout period with promising results.²⁵ The incidence of treatment-related side effects was comparable to the vehicle. The most common adverse events were exfoliation, erythema, and dermatitis, all occurring at the application site.²⁵ Tavaborole was approved by the FDA for the treatment of toenail onychomycosis in July 2014.

Luliconazole is a member of the azole class and a phase 2b/3 clinical trial with a 10% solution involving 334 patients was completed in June 2013.²⁶ Results from this trial are expected in early 2015.

Lasers are a developing area for onychomycosis therapy and the appeal stems from their ability to selectively deliver energy to the target tissue, thus avoiding systemic side effects. Since 2010, the FDA has approved numerous laser devices for the temporary cosmetic improvement of onychomycosis, all of which are Nd:YAG 1064-nm lasers.^27,28 It was previously thought that the mechanism of action for the fungicidal effect was achieved with heat,²⁹ but newer in vitro studies have shown that the amount of time and level of heat required to kill Trichophyton rubrum would not be tolerable to patients.³⁰ Although the mechanism of action is poorly understood, some clinical trials have shown success using the Nd:YAG 1064-nm laser for treatment of onychomycosis. However, in a study of 8 patients treated with the Nd:YAG 1064-nm laser for 5 treatment sessions, none had a mycological or clinical cure and there was only mild clinical improvement. In addition, most patients had pain and burning during the treatments requiring many short breaks.³⁰ Although not yet FDA approved for the treatment of onychomycosis, other types of lasers are currently being studied, including CO₂, near-infrared diode, and femtosecond-infrared laser systems.³

Plasma therapy is a developing area for the treatment of onychomycosis. Plasma was shown to be fungicidal to T rubrum in an in vitro model (MOE Medical Devices, written communication, July 2012), and a clinical trial to evaluate the safety, tolerability, and efficacy of plasma in human subjects is ongoing (registered on March 22, 2013, at www.clinicaltrials.gov with the identifier NCT01819051).

Onychomycosis is a common problem that increases in prevalence with advancing age. Oral terbinafine is considered the first-line treatment at this point in time.³¹ Two new topical agents, efinaconazole and tavaborole, were recently FDA approved and may be used for the treatment of toenail onychomycosis without the need for nail debridement. The Nd:YAG laser has shown some promise in earlier clinical studies but was ineffective in a more recent study.

To the Editor:
A 62-year-old woman was referred to the dermatology clinic for papules on both eyelids of 6 months’ duration. She underwent surgery for a thyroid gland adenoma 3 years prior and subsequently experienced hypothyroidism. Levothyroxine sodium was administered daily (100 µg initially; 50 µg over the last 1.5 years). Papules occurred on both eyelids 6 months prior to presentation and gradually increased in number. The center of each papule was filled with a black keratinous plug. The skin lesions became raised after the patient ate fatty foods. The lesions remained entirely asymptomatic and there was no family history of a similar disorder.

Physical examinations showed no systemic abnormalities. Dermatologic examination showed clustered 3- to 4-mm flesh-colored papules on both upper eyelids; the centers of the papules were filled with 1- to 2-mm black keratinous plugs (Figure 1A). Several similar skin lesions existed on the lower eyelids, nasal root, and right side of the nasal dorsum. On laboratory examination, the results of routine blood, urine, and stool tests, as well as renal and hepatic functions, electrolytes, and blood sugar levels, were within reference range. Indicators including triglyceride of 2.50 mmol/L (reference range, 0.40–1.90 mmol/L), total cholesterol of 6.31 mmol/L (reference range, 3.00–5.70 mmol/L), serum total thyroxine (T₄) of 5.32 µg/dL (reference range, 6.09–12.23 µg/dL), total triiodothyronine (T₃) of 64 ng/dL (reference range, 87–178 ng/dL), serum free thyroxine (FT₄) of 0.41 ng/dL (reference range, 0.61–1.12 ng/dL), serum free triiodothyronine (FT₃) of 182 pg/dL (reference range, 250–390 pg/dL), and thyrotropin of 33.75 µIU/mL (reference range, 0.34–5.60 µIU/mL) were not within reference range; however, thyroperoxidase antibodies, thyrotropin receptor antibodies, thyroglobulin antibodies, thyroglobulin, and calcitonin were within reference range. Color ultrasonography indicated post–subtotal resection of the bilateral thyroid glands.

Figure 1. Clustered flesh-colored papules on both upper eyelids; the centers of the papules were filled with black keratinous plugs (A). Following treatment with levothyroxine sodium, most of the skin lesions had resolved (B).

Histopathologic analysis of the skin lesions showed that the epidermis became atrophic and thinner, and several atrophic and cyst-dilated follicular structures existed inside the dermis. Some structures opened through the epidermis; the walls were squamous epithelium and keratin filled the structures (Figure 2). The condition was diagnosed as nevus comedonicus (NC).

The patient was referred to the endocrinology department and treated with levothyroxine sodium (100 µg daily). At 5-month follow-up, the T₄, T₃, FT₄, FT₃, and thyrotropin levels were within reference range and most of the skin lesions had resolved (Figure 1B).

Nevus comedonicus is an unusual skin lesion with a predilection for the face, neck, shoulders, upper arms, and trunk. The clinical manifestations include comedonelike papules with centers that are characterized by large, black, solid keratinous plugs. When the plugs are peeled off, volcanic craterlike pits will be left. The skin lesions usually are ribbonlike and clustered on 1 side of the body. Pathologic examination often shows that the epidermis is pitted downward, and the dilated follicular ostia are plugged with keratin.^1,2 Paige and Mendelson³ divided NC into 2 types: inflammatory and noninflammatory. Approximately half of NC patients experience cysts, abscesses, fistulae, and scars.⁴

The exact etiology of NC is unclear. Some researchers believe that it is a congenital hair follicle deformity; more specifically, that it is caused by a developmental defect in the hair follicles in the embryonic stage (ie, abnormal differentiation of epithelial stem cells that differentiate into follicles). Most incidences of NC occur at birth or before growth and development. However, few studies have reported late-onset NC.⁵

Figure 2. Several atrophic and cyst-dilated follicular structures existed inside the dermis. Some of them opened through the epidermis and were filled with keratin (A and B)(both H&E, original magnifications ×4 and ×10).

The relationship between NC and thyroid disease is unique. Clinical research has shown that hypothyroidism can result in hair loss and cracks.⁶ In animal experiments, hypothyroidism model mice often experienced degenerative changes of their hair follicles and hair papillae as well as changes in the telogen phase, such as thinning of the outer and inner root sheaths.⁷ Meanwhile, decreased cell proliferation activity in the hair follicles was observed. Therefore, it is reasonable to conclude that thyroid hormones have regulatory effects on the growth and development of hair follicles.⁷

A study on human hair follicles found that thyroid hormone receptor β1 is expressed in human hair follicles.⁸ Research on in vitro–cultured human hair follicles showed that thyroid hormones T₃ and T₄ upregulated the proliferation of hair matrix cells and downregulated their apoptosis. Thyroid hormones also prolonged the duration of the hair growth phase (anagen).⁹ Furthermore, expression of thyrotropin receptor was detected in human hair follicles. Because increased serum thyrotropin levels can lead to clinical hair loss, thyrotropin may inhibit the growth of hair follicles via thyrotropin receptor.¹⁰ In our patient, NC occurred on both eyelids when the patient experienced hypothyroidism following thyroid gland adenoma surgery. Following treatment with levothyroxine sodium, the T₄, T₃, FT₄, FT₃, and thyrotropin levels were within reference range and most of the skin lesions resolved. Therefore, the occurrence of NC may be related to hypothyroidism in this patient. The low thyroid hormone levels and elevated thyrotropin level possibly induced degenerative changes and injuries to the hair matrix cells, resulting in hair follicle obstruction and accumulation of keratin, which ultimately led to NC. However, the exact relationship between NC and thyroid diseases requires elucidation in future studies.

To the Editor:
A 62-year-old woman was referred to the dermatology clinic for papules on both eyelids of 6 months’ duration. She underwent surgery for a thyroid gland adenoma 3 years prior and subsequently experienced hypothyroidism. Levothyroxine sodium was administered daily (100 µg initially; 50 µg over the last 1.5 years). Papules occurred on both eyelids 6 months prior to presentation and gradually increased in number. The center of each papule was filled with a black keratinous plug. The skin lesions became raised after the patient ate fatty foods. The lesions remained entirely asymptomatic and there was no family history of a similar disorder.

Physical examinations showed no systemic abnormalities. Dermatologic examination showed clustered 3- to 4-mm flesh-colored papules on both upper eyelids; the centers of the papules were filled with 1- to 2-mm black keratinous plugs (Figure 1A). Several similar skin lesions existed on the lower eyelids, nasal root, and right side of the nasal dorsum. On laboratory examination, the results of routine blood, urine, and stool tests, as well as renal and hepatic functions, electrolytes, and blood sugar levels, were within reference range. Indicators including triglyceride of 2.50 mmol/L (reference range, 0.40–1.90 mmol/L), total cholesterol of 6.31 mmol/L (reference range, 3.00–5.70 mmol/L), serum total thyroxine (T₄) of 5.32 µg/dL (reference range, 6.09–12.23 µg/dL), total triiodothyronine (T₃) of 64 ng/dL (reference range, 87–178 ng/dL), serum free thyroxine (FT₄) of 0.41 ng/dL (reference range, 0.61–1.12 ng/dL), serum free triiodothyronine (FT₃) of 182 pg/dL (reference range, 250–390 pg/dL), and thyrotropin of 33.75 µIU/mL (reference range, 0.34–5.60 µIU/mL) were not within reference range; however, thyroperoxidase antibodies, thyrotropin receptor antibodies, thyroglobulin antibodies, thyroglobulin, and calcitonin were within reference range. Color ultrasonography indicated post–subtotal resection of the bilateral thyroid glands.

Figure 1. Clustered flesh-colored papules on both upper eyelids; the centers of the papules were filled with black keratinous plugs (A). Following treatment with levothyroxine sodium, most of the skin lesions had resolved (B).

Histopathologic analysis of the skin lesions showed that the epidermis became atrophic and thinner, and several atrophic and cyst-dilated follicular structures existed inside the dermis. Some structures opened through the epidermis; the walls were squamous epithelium and keratin filled the structures (Figure 2). The condition was diagnosed as nevus comedonicus (NC).

The patient was referred to the endocrinology department and treated with levothyroxine sodium (100 µg daily). At 5-month follow-up, the T₄, T₃, FT₄, FT₃, and thyrotropin levels were within reference range and most of the skin lesions had resolved (Figure 1B).

Nevus comedonicus is an unusual skin lesion with a predilection for the face, neck, shoulders, upper arms, and trunk. The clinical manifestations include comedonelike papules with centers that are characterized by large, black, solid keratinous plugs. When the plugs are peeled off, volcanic craterlike pits will be left. The skin lesions usually are ribbonlike and clustered on 1 side of the body. Pathologic examination often shows that the epidermis is pitted downward, and the dilated follicular ostia are plugged with keratin.^1,2 Paige and Mendelson³ divided NC into 2 types: inflammatory and noninflammatory. Approximately half of NC patients experience cysts, abscesses, fistulae, and scars.⁴

The exact etiology of NC is unclear. Some researchers believe that it is a congenital hair follicle deformity; more specifically, that it is caused by a developmental defect in the hair follicles in the embryonic stage (ie, abnormal differentiation of epithelial stem cells that differentiate into follicles). Most incidences of NC occur at birth or before growth and development. However, few studies have reported late-onset NC.⁵

Figure 2. Several atrophic and cyst-dilated follicular structures existed inside the dermis. Some of them opened through the epidermis and were filled with keratin (A and B)(both H&E, original magnifications ×4 and ×10).

The relationship between NC and thyroid disease is unique. Clinical research has shown that hypothyroidism can result in hair loss and cracks.⁶ In animal experiments, hypothyroidism model mice often experienced degenerative changes of their hair follicles and hair papillae as well as changes in the telogen phase, such as thinning of the outer and inner root sheaths.⁷ Meanwhile, decreased cell proliferation activity in the hair follicles was observed. Therefore, it is reasonable to conclude that thyroid hormones have regulatory effects on the growth and development of hair follicles.⁷

A study on human hair follicles found that thyroid hormone receptor β1 is expressed in human hair follicles.⁸ Research on in vitro–cultured human hair follicles showed that thyroid hormones T₃ and T₄ upregulated the proliferation of hair matrix cells and downregulated their apoptosis. Thyroid hormones also prolonged the duration of the hair growth phase (anagen).⁹ Furthermore, expression of thyrotropin receptor was detected in human hair follicles. Because increased serum thyrotropin levels can lead to clinical hair loss, thyrotropin may inhibit the growth of hair follicles via thyrotropin receptor.¹⁰ In our patient, NC occurred on both eyelids when the patient experienced hypothyroidism following thyroid gland adenoma surgery. Following treatment with levothyroxine sodium, the T₄, T₃, FT₄, FT₃, and thyrotropin levels were within reference range and most of the skin lesions resolved. Therefore, the occurrence of NC may be related to hypothyroidism in this patient. The low thyroid hormone levels and elevated thyrotropin level possibly induced degenerative changes and injuries to the hair matrix cells, resulting in hair follicle obstruction and accumulation of keratin, which ultimately led to NC. However, the exact relationship between NC and thyroid diseases requires elucidation in future studies.

To the Editor:
A 62-year-old woman was referred to the dermatology clinic for papules on both eyelids of 6 months’ duration. She underwent surgery for a thyroid gland adenoma 3 years prior and subsequently experienced hypothyroidism. Levothyroxine sodium was administered daily (100 µg initially; 50 µg over the last 1.5 years). Papules occurred on both eyelids 6 months prior to presentation and gradually increased in number. The center of each papule was filled with a black keratinous plug. The skin lesions became raised after the patient ate fatty foods. The lesions remained entirely asymptomatic and there was no family history of a similar disorder.

Physical examinations showed no systemic abnormalities. Dermatologic examination showed clustered 3- to 4-mm flesh-colored papules on both upper eyelids; the centers of the papules were filled with 1- to 2-mm black keratinous plugs (Figure 1A). Several similar skin lesions existed on the lower eyelids, nasal root, and right side of the nasal dorsum. On laboratory examination, the results of routine blood, urine, and stool tests, as well as renal and hepatic functions, electrolytes, and blood sugar levels, were within reference range. Indicators including triglyceride of 2.50 mmol/L (reference range, 0.40–1.90 mmol/L), total cholesterol of 6.31 mmol/L (reference range, 3.00–5.70 mmol/L), serum total thyroxine (T₄) of 5.32 µg/dL (reference range, 6.09–12.23 µg/dL), total triiodothyronine (T₃) of 64 ng/dL (reference range, 87–178 ng/dL), serum free thyroxine (FT₄) of 0.41 ng/dL (reference range, 0.61–1.12 ng/dL), serum free triiodothyronine (FT₃) of 182 pg/dL (reference range, 250–390 pg/dL), and thyrotropin of 33.75 µIU/mL (reference range, 0.34–5.60 µIU/mL) were not within reference range; however, thyroperoxidase antibodies, thyrotropin receptor antibodies, thyroglobulin antibodies, thyroglobulin, and calcitonin were within reference range. Color ultrasonography indicated post–subtotal resection of the bilateral thyroid glands.

Figure 1. Clustered flesh-colored papules on both upper eyelids; the centers of the papules were filled with black keratinous plugs (A). Following treatment with levothyroxine sodium, most of the skin lesions had resolved (B).

Histopathologic analysis of the skin lesions showed that the epidermis became atrophic and thinner, and several atrophic and cyst-dilated follicular structures existed inside the dermis. Some structures opened through the epidermis; the walls were squamous epithelium and keratin filled the structures (Figure 2). The condition was diagnosed as nevus comedonicus (NC).

The patient was referred to the endocrinology department and treated with levothyroxine sodium (100 µg daily). At 5-month follow-up, the T₄, T₃, FT₄, FT₃, and thyrotropin levels were within reference range and most of the skin lesions had resolved (Figure 1B).

Nevus comedonicus is an unusual skin lesion with a predilection for the face, neck, shoulders, upper arms, and trunk. The clinical manifestations include comedonelike papules with centers that are characterized by large, black, solid keratinous plugs. When the plugs are peeled off, volcanic craterlike pits will be left. The skin lesions usually are ribbonlike and clustered on 1 side of the body. Pathologic examination often shows that the epidermis is pitted downward, and the dilated follicular ostia are plugged with keratin.^1,2 Paige and Mendelson³ divided NC into 2 types: inflammatory and noninflammatory. Approximately half of NC patients experience cysts, abscesses, fistulae, and scars.⁴

The exact etiology of NC is unclear. Some researchers believe that it is a congenital hair follicle deformity; more specifically, that it is caused by a developmental defect in the hair follicles in the embryonic stage (ie, abnormal differentiation of epithelial stem cells that differentiate into follicles). Most incidences of NC occur at birth or before growth and development. However, few studies have reported late-onset NC.⁵

Figure 2. Several atrophic and cyst-dilated follicular structures existed inside the dermis. Some of them opened through the epidermis and were filled with keratin (A and B)(both H&E, original magnifications ×4 and ×10).

The relationship between NC and thyroid disease is unique. Clinical research has shown that hypothyroidism can result in hair loss and cracks.⁶ In animal experiments, hypothyroidism model mice often experienced degenerative changes of their hair follicles and hair papillae as well as changes in the telogen phase, such as thinning of the outer and inner root sheaths.⁷ Meanwhile, decreased cell proliferation activity in the hair follicles was observed. Therefore, it is reasonable to conclude that thyroid hormones have regulatory effects on the growth and development of hair follicles.⁷

A study on human hair follicles found that thyroid hormone receptor β1 is expressed in human hair follicles.⁸ Research on in vitro–cultured human hair follicles showed that thyroid hormones T₃ and T₄ upregulated the proliferation of hair matrix cells and downregulated their apoptosis. Thyroid hormones also prolonged the duration of the hair growth phase (anagen).⁹ Furthermore, expression of thyrotropin receptor was detected in human hair follicles. Because increased serum thyrotropin levels can lead to clinical hair loss, thyrotropin may inhibit the growth of hair follicles via thyrotropin receptor.¹⁰ In our patient, NC occurred on both eyelids when the patient experienced hypothyroidism following thyroid gland adenoma surgery. Following treatment with levothyroxine sodium, the T₄, T₃, FT₄, FT₃, and thyrotropin levels were within reference range and most of the skin lesions resolved. Therefore, the occurrence of NC may be related to hypothyroidism in this patient. The low thyroid hormone levels and elevated thyrotropin level possibly induced degenerative changes and injuries to the hair matrix cells, resulting in hair follicle obstruction and accumulation of keratin, which ultimately led to NC. However, the exact relationship between NC and thyroid diseases requires elucidation in future studies.