Article

Key clinical point: Patients with inflammatory bowel disease (IBD) and chronic migraine (CM) had higher serum levels of alpha-calcitonin gene-related peptide (CGRP) than patients with only IBD.

Major findings: The alpha-CGRP levels were significantly higher in patients with IBD (56.9 vs 37.2 pg/mL; P = .003) or CM (53.0 vs 37.2 pg/mL; P = .019) than healthy control participants without a history of IBD and CM. The levels of this peptide in the serum were further increased in patients with IBD and concomitant migraine compared with those with only IBD (70.9 vs 53.7 pg/mL; P = .046).

Study details: This cross-sectional study compared the serum CGRP levels in 96 patients with IBD, 50 patients with CM, and 50 healthy control participants without a history of IBD and CM.

Disclosures: This study was funded by Instituto de Salud Carlos III, Spain. The authors declared no conflicts of interest.

Source: Pascual-Mato M, Gárate G, González-Quintanilla V, et al. Differences in circulating alpha-calcitonin gene-related peptide levels in inflammatory bowel disease and its relation to migraine comorbidity: A cross-sectional study. Headache. 2024;64:849-858 (Jun 24). Doi: 10.1111/head.14768 Source

Key clinical point: Patients with inflammatory bowel disease (IBD) and chronic migraine (CM) had higher serum levels of alpha-calcitonin gene-related peptide (CGRP) than patients with only IBD.

Major findings: The alpha-CGRP levels were significantly higher in patients with IBD (56.9 vs 37.2 pg/mL; P = .003) or CM (53.0 vs 37.2 pg/mL; P = .019) than healthy control participants without a history of IBD and CM. The levels of this peptide in the serum were further increased in patients with IBD and concomitant migraine compared with those with only IBD (70.9 vs 53.7 pg/mL; P = .046).

Study details: This cross-sectional study compared the serum CGRP levels in 96 patients with IBD, 50 patients with CM, and 50 healthy control participants without a history of IBD and CM.

Disclosures: This study was funded by Instituto de Salud Carlos III, Spain. The authors declared no conflicts of interest.

Source: Pascual-Mato M, Gárate G, González-Quintanilla V, et al. Differences in circulating alpha-calcitonin gene-related peptide levels in inflammatory bowel disease and its relation to migraine comorbidity: A cross-sectional study. Headache. 2024;64:849-858 (Jun 24). Doi: 10.1111/head.14768 Source

Key clinical point: Patients with inflammatory bowel disease (IBD) and chronic migraine (CM) had higher serum levels of alpha-calcitonin gene-related peptide (CGRP) than patients with only IBD.

Major findings: The alpha-CGRP levels were significantly higher in patients with IBD (56.9 vs 37.2 pg/mL; P = .003) or CM (53.0 vs 37.2 pg/mL; P = .019) than healthy control participants without a history of IBD and CM. The levels of this peptide in the serum were further increased in patients with IBD and concomitant migraine compared with those with only IBD (70.9 vs 53.7 pg/mL; P = .046).

Study details: This cross-sectional study compared the serum CGRP levels in 96 patients with IBD, 50 patients with CM, and 50 healthy control participants without a history of IBD and CM.

Disclosures: This study was funded by Instituto de Salud Carlos III, Spain. The authors declared no conflicts of interest.

Source: Pascual-Mato M, Gárate G, González-Quintanilla V, et al. Differences in circulating alpha-calcitonin gene-related peptide levels in inflammatory bowel disease and its relation to migraine comorbidity: A cross-sectional study. Headache. 2024;64:849-858 (Jun 24). Doi: 10.1111/head.14768 Source

Key clinical point: Long-term eptinezumab treatment showed benefits in patients with chronic migraine (CM) who had concomitant medication-overuse headache (MOH).

Major findings: After eptinezumab treatment, the average number of headache days per 3 months reduced from 15.8 days (mean 47.5 headache days/month) at baseline to 3.8 days (mean 11.3 headache days/month) at 104 weeks, along with reductions observed in the Migraine Disability Assessment (mean change −51.9 points) and 6-item Headache Impact Test (mean change −9.7 points) scores. More than half (57.1%) the patients showed an improvement in their patient-identified most bothersome symptom at as early as 4 weeks.

Study details: This post hoc analysis of the PREVAIL study included 49 patients with CM and concomitant MOH.

Disclosures: This publication was supported by H. Lundbeck A/S, Copenhagen, Denmark. Two authors declared being employees of H. Lundbeck A/S. The other authors declared having ties with various sources, including H. Lundbeck A/S.

Source: Blumenfeld A, Kudrow D, McAllister P, et al. Long-term effectiveness of eptinezumab in the treatment of patients with chronic migraine and medication-overuse headache. Headache. 2024;64:738-749 (Jun 24). Doi: 10.1111/head.14767 Source

Key clinical point: Long-term eptinezumab treatment showed benefits in patients with chronic migraine (CM) who had concomitant medication-overuse headache (MOH).

Major findings: After eptinezumab treatment, the average number of headache days per 3 months reduced from 15.8 days (mean 47.5 headache days/month) at baseline to 3.8 days (mean 11.3 headache days/month) at 104 weeks, along with reductions observed in the Migraine Disability Assessment (mean change −51.9 points) and 6-item Headache Impact Test (mean change −9.7 points) scores. More than half (57.1%) the patients showed an improvement in their patient-identified most bothersome symptom at as early as 4 weeks.

Study details: This post hoc analysis of the PREVAIL study included 49 patients with CM and concomitant MOH.

Disclosures: This publication was supported by H. Lundbeck A/S, Copenhagen, Denmark. Two authors declared being employees of H. Lundbeck A/S. The other authors declared having ties with various sources, including H. Lundbeck A/S.

Source: Blumenfeld A, Kudrow D, McAllister P, et al. Long-term effectiveness of eptinezumab in the treatment of patients with chronic migraine and medication-overuse headache. Headache. 2024;64:738-749 (Jun 24). Doi: 10.1111/head.14767 Source

Key clinical point: Long-term eptinezumab treatment showed benefits in patients with chronic migraine (CM) who had concomitant medication-overuse headache (MOH).

Major findings: After eptinezumab treatment, the average number of headache days per 3 months reduced from 15.8 days (mean 47.5 headache days/month) at baseline to 3.8 days (mean 11.3 headache days/month) at 104 weeks, along with reductions observed in the Migraine Disability Assessment (mean change −51.9 points) and 6-item Headache Impact Test (mean change −9.7 points) scores. More than half (57.1%) the patients showed an improvement in their patient-identified most bothersome symptom at as early as 4 weeks.

Study details: This post hoc analysis of the PREVAIL study included 49 patients with CM and concomitant MOH.

Disclosures: This publication was supported by H. Lundbeck A/S, Copenhagen, Denmark. Two authors declared being employees of H. Lundbeck A/S. The other authors declared having ties with various sources, including H. Lundbeck A/S.

Source: Blumenfeld A, Kudrow D, McAllister P, et al. Long-term effectiveness of eptinezumab in the treatment of patients with chronic migraine and medication-overuse headache. Headache. 2024;64:738-749 (Jun 24). Doi: 10.1111/head.14767 Source

Key clinical point: Calcitonin gene-related peptide (CGRP)–induced migraine attacks did not affect therapeutic response to erenumab in patients with migraine.

Major finding: Overall, a similar proportion of patients who experienced vs did not experience CGRP-induced migraine attacks had a ≥50% reduction in monthly migraine days during weeks 13-24 following erenumab treatment (61% vs 52%; odds ratio [OR] 1.42; P = .498). No significant differences were seen between the two patient groups that achieved a ≥50% reduction in monthly migraine or headache days of moderate to severe intensity (OR 1.25; P = .625).

Study details: This interim analysis of the REFORM study included 124 patients with migraine who received CGRP infusion on a single experimental day and subsequently a 24-week treatment with erenumab.

Disclosures: This study was funded by Novartis Pharma and the Lundbeck Foundation. Two authors declared being employees of or holding shares in Novartis Pharma AG. Several authors declared receiving personal fees from various sources, including Novartis, outside of the submitted work. One author is an associate editor for Cephalalgia.

Source: Al-Khazali HM, Ashina H, Christensen RH, et al. Hypersensitivity to CGRP as a predictive biomarker of migraine prevention with erenumab. Cephalalgia. 2024;44(6):3331024241258734 (Jun 11). Doi: 10.1177/03331024241258734 Source

Key clinical point: Calcitonin gene-related peptide (CGRP)–induced migraine attacks did not affect therapeutic response to erenumab in patients with migraine.

Major finding: Overall, a similar proportion of patients who experienced vs did not experience CGRP-induced migraine attacks had a ≥50% reduction in monthly migraine days during weeks 13-24 following erenumab treatment (61% vs 52%; odds ratio [OR] 1.42; P = .498). No significant differences were seen between the two patient groups that achieved a ≥50% reduction in monthly migraine or headache days of moderate to severe intensity (OR 1.25; P = .625).

Study details: This interim analysis of the REFORM study included 124 patients with migraine who received CGRP infusion on a single experimental day and subsequently a 24-week treatment with erenumab.

Disclosures: This study was funded by Novartis Pharma and the Lundbeck Foundation. Two authors declared being employees of or holding shares in Novartis Pharma AG. Several authors declared receiving personal fees from various sources, including Novartis, outside of the submitted work. One author is an associate editor for Cephalalgia.

Source: Al-Khazali HM, Ashina H, Christensen RH, et al. Hypersensitivity to CGRP as a predictive biomarker of migraine prevention with erenumab. Cephalalgia. 2024;44(6):3331024241258734 (Jun 11). Doi: 10.1177/03331024241258734 Source

Key clinical point: Calcitonin gene-related peptide (CGRP)–induced migraine attacks did not affect therapeutic response to erenumab in patients with migraine.

Major finding: Overall, a similar proportion of patients who experienced vs did not experience CGRP-induced migraine attacks had a ≥50% reduction in monthly migraine days during weeks 13-24 following erenumab treatment (61% vs 52%; odds ratio [OR] 1.42; P = .498). No significant differences were seen between the two patient groups that achieved a ≥50% reduction in monthly migraine or headache days of moderate to severe intensity (OR 1.25; P = .625).

Study details: This interim analysis of the REFORM study included 124 patients with migraine who received CGRP infusion on a single experimental day and subsequently a 24-week treatment with erenumab.

Disclosures: This study was funded by Novartis Pharma and the Lundbeck Foundation. Two authors declared being employees of or holding shares in Novartis Pharma AG. Several authors declared receiving personal fees from various sources, including Novartis, outside of the submitted work. One author is an associate editor for Cephalalgia.

Source: Al-Khazali HM, Ashina H, Christensen RH, et al. Hypersensitivity to CGRP as a predictive biomarker of migraine prevention with erenumab. Cephalalgia. 2024;44(6):3331024241258734 (Jun 11). Doi: 10.1177/03331024241258734 Source

Key clinical point: In patients with episodic migraine (EM) or chronic migraine (CM), erenumab treatment interruption was associated with increased monthly migraine days (MMD) and worsening of migraine disability; these effects ameliorated on treatment restart.

Major finding: After erenumab treatment interruption for 3 months, the number of monthly migraine days (MMD) increased from 6.1 to 10.9 days for patients with EM and from 11.4 to 16.8 days for patients with CM; the modified Migraine Disability Assessment scores (mMIDAS) also worsened during this period. Both MMD and mMIDAS scores improved upon restarting treatment.

Study details: This interim analysis of the 24-month, multicentric, non-interventional observational study included 172 patients with CM or EM who received erenumab and underwent treatment interruptions on average 11.2 months after initiation, with interruptions lasting for a mean duration of 4 months.

Disclosures: This study was funded by Novartis Pharma Schweiz AG. Four authors declared being full-time employees of Novartis Pharma Schweiz AG, and others declared having ties with various sources, including Novartis.

Source: Gantenbein AR, Bonvin C, Kamm CP, et al. Implications of therapy interruption on monthly migraine days and modified migraine disability assessment in patients treated with erenumab for chronic and episodic migraine: SQUARE study interim results. J Neurol. 2024 (Jun 13). Doi: 10.1007/s00415-024-12470-6 Source

Key clinical point: In patients with episodic migraine (EM) or chronic migraine (CM), erenumab treatment interruption was associated with increased monthly migraine days (MMD) and worsening of migraine disability; these effects ameliorated on treatment restart.

Major finding: After erenumab treatment interruption for 3 months, the number of monthly migraine days (MMD) increased from 6.1 to 10.9 days for patients with EM and from 11.4 to 16.8 days for patients with CM; the modified Migraine Disability Assessment scores (mMIDAS) also worsened during this period. Both MMD and mMIDAS scores improved upon restarting treatment.

Study details: This interim analysis of the 24-month, multicentric, non-interventional observational study included 172 patients with CM or EM who received erenumab and underwent treatment interruptions on average 11.2 months after initiation, with interruptions lasting for a mean duration of 4 months.

Disclosures: This study was funded by Novartis Pharma Schweiz AG. Four authors declared being full-time employees of Novartis Pharma Schweiz AG, and others declared having ties with various sources, including Novartis.

Source: Gantenbein AR, Bonvin C, Kamm CP, et al. Implications of therapy interruption on monthly migraine days and modified migraine disability assessment in patients treated with erenumab for chronic and episodic migraine: SQUARE study interim results. J Neurol. 2024 (Jun 13). Doi: 10.1007/s00415-024-12470-6 Source

Key clinical point: In patients with episodic migraine (EM) or chronic migraine (CM), erenumab treatment interruption was associated with increased monthly migraine days (MMD) and worsening of migraine disability; these effects ameliorated on treatment restart.

Major finding: After erenumab treatment interruption for 3 months, the number of monthly migraine days (MMD) increased from 6.1 to 10.9 days for patients with EM and from 11.4 to 16.8 days for patients with CM; the modified Migraine Disability Assessment scores (mMIDAS) also worsened during this period. Both MMD and mMIDAS scores improved upon restarting treatment.

Study details: This interim analysis of the 24-month, multicentric, non-interventional observational study included 172 patients with CM or EM who received erenumab and underwent treatment interruptions on average 11.2 months after initiation, with interruptions lasting for a mean duration of 4 months.

Disclosures: This study was funded by Novartis Pharma Schweiz AG. Four authors declared being full-time employees of Novartis Pharma Schweiz AG, and others declared having ties with various sources, including Novartis.

Source: Gantenbein AR, Bonvin C, Kamm CP, et al. Implications of therapy interruption on monthly migraine days and modified migraine disability assessment in patients treated with erenumab for chronic and episodic migraine: SQUARE study interim results. J Neurol. 2024 (Jun 13). Doi: 10.1007/s00415-024-12470-6 Source

Key clinical point: Patients with migraine, migraine with aura (MA), or migraine without aura (MO) faced a significantly higher risk for retinal vascular occlusion.

Major findings: Compared with control individuals without migraine, those with migraine (adjusted hazard ratio [aHR] 1.69; 95% CI 1.57-1.83), MA (aHR 1.77; 95% CI 1.58-1.98), or MO (aHR 1.92; 95% CI 1.64-2.25; P < .001 for all) had a significantly higher risk for retinal vascular occlusion. The risk was, however, reduced in the migraine population that received nonsteroidal anti-inflammatory drugs (aHR 0.19; 95% CI 0.16-0.22), propranolol (aHR 0.73; 95% CI 0.62-0.86), or flunarizine (aHR 0.84; 95% CI 0.76-0.93; P < .001 for all).

Study details: This population-based retrospective cohort study included 628,760 patients with migraine and 628,760 control individuals without migraine.

Disclosures: This study was supported in part by the Taiwan Ministry of Health and Welfare Clinical Trial Center, China Medical University Hospital, and National Science and Technology Council. The authors declared no conflicts of interest.

Source: Ho K-Y, Lin C-D, Hsu T-J, et al. Increased risks of retinal vascular occlusion in patients with migraine and the protective effects of migraine treatment: A population-based retrospective cohort study. Sci Rep. 2024;14:15429 (Jul 4). Doi: 10.1038/s41598-024-66363-9 Source

Key clinical point: Patients with migraine, migraine with aura (MA), or migraine without aura (MO) faced a significantly higher risk for retinal vascular occlusion.

Major findings: Compared with control individuals without migraine, those with migraine (adjusted hazard ratio [aHR] 1.69; 95% CI 1.57-1.83), MA (aHR 1.77; 95% CI 1.58-1.98), or MO (aHR 1.92; 95% CI 1.64-2.25; P < .001 for all) had a significantly higher risk for retinal vascular occlusion. The risk was, however, reduced in the migraine population that received nonsteroidal anti-inflammatory drugs (aHR 0.19; 95% CI 0.16-0.22), propranolol (aHR 0.73; 95% CI 0.62-0.86), or flunarizine (aHR 0.84; 95% CI 0.76-0.93; P < .001 for all).

Study details: This population-based retrospective cohort study included 628,760 patients with migraine and 628,760 control individuals without migraine.

Disclosures: This study was supported in part by the Taiwan Ministry of Health and Welfare Clinical Trial Center, China Medical University Hospital, and National Science and Technology Council. The authors declared no conflicts of interest.

Source: Ho K-Y, Lin C-D, Hsu T-J, et al. Increased risks of retinal vascular occlusion in patients with migraine and the protective effects of migraine treatment: A population-based retrospective cohort study. Sci Rep. 2024;14:15429 (Jul 4). Doi: 10.1038/s41598-024-66363-9 Source

Key clinical point: Patients with migraine, migraine with aura (MA), or migraine without aura (MO) faced a significantly higher risk for retinal vascular occlusion.

Major findings: Compared with control individuals without migraine, those with migraine (adjusted hazard ratio [aHR] 1.69; 95% CI 1.57-1.83), MA (aHR 1.77; 95% CI 1.58-1.98), or MO (aHR 1.92; 95% CI 1.64-2.25; P < .001 for all) had a significantly higher risk for retinal vascular occlusion. The risk was, however, reduced in the migraine population that received nonsteroidal anti-inflammatory drugs (aHR 0.19; 95% CI 0.16-0.22), propranolol (aHR 0.73; 95% CI 0.62-0.86), or flunarizine (aHR 0.84; 95% CI 0.76-0.93; P < .001 for all).

Study details: This population-based retrospective cohort study included 628,760 patients with migraine and 628,760 control individuals without migraine.

Disclosures: This study was supported in part by the Taiwan Ministry of Health and Welfare Clinical Trial Center, China Medical University Hospital, and National Science and Technology Council. The authors declared no conflicts of interest.

Source: Ho K-Y, Lin C-D, Hsu T-J, et al. Increased risks of retinal vascular occlusion in patients with migraine and the protective effects of migraine treatment: A population-based retrospective cohort study. Sci Rep. 2024;14:15429 (Jul 4). Doi: 10.1038/s41598-024-66363-9 Source

Key clinical point: Atogepant was effective in reducing monthly migraine days (MMD), monthly headache days (MHD), and acute medication use days in patients with chronic migraine (CM), irrespective of acute medication overuse.

Major findings: Patients with acute medication overuse receiving 30 mg or 60 mg atogepant vs placebo had a significantly greater reduction in MMD (least squares mean difference [LSMD] −2.7 and −1.9, respectively), MHD (LSMD −2.8 and −2.1, respectively), and monthly acute medication use days (LSMD −2.8 and −2.6, respectively). Similar reductions were observed in patients without acute medication overuse.

Study details: This subgroup analysis of the PROGRESS trial included 755 patients with CM with or without acute medication overuse who were randomly assigned to receive atogepant (30 mg or 60 mg) or placebo.

Disclosures: This study was funded by AbbVie. Six authors declared being employees or stockholders of AbbVie. Several authors declared having ties with various sources, including AbbVie.

Source: Goadsby PJ, Friedman DI, Holle-Lee D, et al. Efficacy of atogepant in chronic migraine with and without acute medication overuse in the randomized, double-blind, phase 3 PROGRESS trial. Neurology. 2024;103(2):e209584 (July 23). Doi: 10.1212/WNL.0000000000209584 Source

Key clinical point: Atogepant was effective in reducing monthly migraine days (MMD), monthly headache days (MHD), and acute medication use days in patients with chronic migraine (CM), irrespective of acute medication overuse.

Major findings: Patients with acute medication overuse receiving 30 mg or 60 mg atogepant vs placebo had a significantly greater reduction in MMD (least squares mean difference [LSMD] −2.7 and −1.9, respectively), MHD (LSMD −2.8 and −2.1, respectively), and monthly acute medication use days (LSMD −2.8 and −2.6, respectively). Similar reductions were observed in patients without acute medication overuse.

Study details: This subgroup analysis of the PROGRESS trial included 755 patients with CM with or without acute medication overuse who were randomly assigned to receive atogepant (30 mg or 60 mg) or placebo.

Disclosures: This study was funded by AbbVie. Six authors declared being employees or stockholders of AbbVie. Several authors declared having ties with various sources, including AbbVie.

Source: Goadsby PJ, Friedman DI, Holle-Lee D, et al. Efficacy of atogepant in chronic migraine with and without acute medication overuse in the randomized, double-blind, phase 3 PROGRESS trial. Neurology. 2024;103(2):e209584 (July 23). Doi: 10.1212/WNL.0000000000209584 Source

Key clinical point: Atogepant was effective in reducing monthly migraine days (MMD), monthly headache days (MHD), and acute medication use days in patients with chronic migraine (CM), irrespective of acute medication overuse.

Major findings: Patients with acute medication overuse receiving 30 mg or 60 mg atogepant vs placebo had a significantly greater reduction in MMD (least squares mean difference [LSMD] −2.7 and −1.9, respectively), MHD (LSMD −2.8 and −2.1, respectively), and monthly acute medication use days (LSMD −2.8 and −2.6, respectively). Similar reductions were observed in patients without acute medication overuse.

Study details: This subgroup analysis of the PROGRESS trial included 755 patients with CM with or without acute medication overuse who were randomly assigned to receive atogepant (30 mg or 60 mg) or placebo.

Disclosures: This study was funded by AbbVie. Six authors declared being employees or stockholders of AbbVie. Several authors declared having ties with various sources, including AbbVie.

Source: Goadsby PJ, Friedman DI, Holle-Lee D, et al. Efficacy of atogepant in chronic migraine with and without acute medication overuse in the randomized, double-blind, phase 3 PROGRESS trial. Neurology. 2024;103(2):e209584 (July 23). Doi: 10.1212/WNL.0000000000209584 Source

To the Editor:

Immune checkpoint inhibitors have revolutionized the treatment of advanced-stage melanoma, with remarkably improved progression-free survival.¹ Anti–programmed death receptor 1 (anti–PD-1) therapies, such as nivolumab and pembrolizumab, are a class of checkpoint inhibitors that have been approved by the US Food and Drug Administration for unresectable metastatic melanoma. Anti–PD-1 agents block the interaction of programmed death-ligand 1 (PD-L1) found on tumor cells with the PD-1 receptor on T cells, facilitating a positive immune response.²

Although these therapies have demonstrated notable antitumor efficacy, they also give rise to numerous immune-related adverse events (irAEs). As many as 70% of patients treated with PD-1/PD-L1 inhibitors experience some type of organ system irAE, of which 30% to 40% are cutaneous.^3-6 Dermatologic adverse events are the most common irAEs, specifically spongiotic dermatitis, lichenoid dermatitis, pruritus, and vitiligo.⁷ Bullous pemphigoid (BP), an autoimmune bullous skin disorder caused by autoantibodies to basement membrane zone antigens, is a rare but potentially serious cutaneous irAE.⁸ Systemic corticosteroids commonly are used to treat immune checkpoint inhibitor–induced BP; other options include tetracyclines for maintenance therapy and rituximab for corticosteroid-refractory BP associated with anti-PD-1.⁹ We present a case of recalcitrant BP secondary to nivolumab therapy in a patient with metastatic melanoma who had near-complete resolution of BP following 2 months of cyclosporine.

A 41-year-old man presented with a generalized papular skin eruption of 1 month’s duration. He had a history of stage IIIC malignant melanoma of the lower right leg with positive sentinel lymph node biopsy. The largest lymph node deposit was 0.03 mm without extracapsular extension. Whole-body positron emission tomography–computed tomography showed no evidence of distant disease. The patient was treated with wide local excision with clear surgical margins plus 12 cycles of nivolumab, which was discontinued due to colitis. Four months after the final cycle of nivolumab, the patient developed widespread erythematous papules with hemorrhagic yellow crusting and no mucosal involvement. He was referred to dermatology by his primary oncologist for further evaluation.

A punch biopsy from the abdomen showed para­keratosis with leukocytoclasis and a superficial dermal infiltrate of neutrophils and eosinophils (Figure 1). Direct immunofluorescence revealed linear basement membrane deposits of IgG and C3, consistent with subepidermal blistering disease. Indirect immunofluorescence demonstrated trace IgG and IgG4 antibodies localized to the epidermal roof of salt-split skin and was negative for IgA antibodies. An enzyme-linked immunoassay was positive for BP antigen 2 (BP180) antibodies (98.4 U/mL [positive, ≥9 U/mL]) and negative for BP antigen 1 (BP230) antibodies (4.3 U/mL [positive, ≥9 U/mL]). Overall, these findings were consistent with a diagnosis of BP.

The patient was treated with prednisone 60 mg daily with initial response; however, there was disease recurrence with tapering. Doxycycline 100 mg twice daily and nicotinamide 500 mg twice daily were added as steroid-sparing agents, as prednisone was discontinued due to mood changes. Three months after the prednisone taper, the patient continued to develop new blisters. He completed treatment with doxycycline and nicotinamide. Rituximab 375 mg weekly was then initiated for 4 weeks.

At 2-week follow-up after completing the rituximab course, the patient reported worsening symptoms and presented with new bullae on the abdomen and upper extremities (Figure 2). Because of the recent history of mood changes while taking prednisone, a trial of cyclosporine 100 mg twice daily (1.37 mg/kg/d) was initiated, with notable improvement within 2 weeks of treatment. After 2 months of cyclosporine, approximately 90% of the rash had resolved with a few tense bullae remaining on the left frontal scalp but no new flares (Figure 3). One month after treatment ended, the patient remained clear of lesions without relapse.

Programmed death receptor 1 inhibitors have shown dramatic efficacy for a growing number of solid and hematologic malignancies, especially malignant melanoma. However, their use is accompanied by nonspecific activation of the immune system, resulting in a variety of adverse events, many manifesting on the skin. Several cases of BP in patients treated with PD-1/PD-L1 inhibitors have been reported.⁹ Cutaneous irAEs usually manifest within 3 weeks of initiation of PD-1 inhibitor therapy; however, the onset of BP typically occurs later at approximately 21 weeks.^4,9 Our patient developed cutaneous manifestations 4 months after cessation of nivolumab.

To the Editor:

Immune checkpoint inhibitors have revolutionized the treatment of advanced-stage melanoma, with remarkably improved progression-free survival.¹ Anti–programmed death receptor 1 (anti–PD-1) therapies, such as nivolumab and pembrolizumab, are a class of checkpoint inhibitors that have been approved by the US Food and Drug Administration for unresectable metastatic melanoma. Anti–PD-1 agents block the interaction of programmed death-ligand 1 (PD-L1) found on tumor cells with the PD-1 receptor on T cells, facilitating a positive immune response.²

Although these therapies have demonstrated notable antitumor efficacy, they also give rise to numerous immune-related adverse events (irAEs). As many as 70% of patients treated with PD-1/PD-L1 inhibitors experience some type of organ system irAE, of which 30% to 40% are cutaneous.^3-6 Dermatologic adverse events are the most common irAEs, specifically spongiotic dermatitis, lichenoid dermatitis, pruritus, and vitiligo.⁷ Bullous pemphigoid (BP), an autoimmune bullous skin disorder caused by autoantibodies to basement membrane zone antigens, is a rare but potentially serious cutaneous irAE.⁸ Systemic corticosteroids commonly are used to treat immune checkpoint inhibitor–induced BP; other options include tetracyclines for maintenance therapy and rituximab for corticosteroid-refractory BP associated with anti-PD-1.⁹ We present a case of recalcitrant BP secondary to nivolumab therapy in a patient with metastatic melanoma who had near-complete resolution of BP following 2 months of cyclosporine.

A 41-year-old man presented with a generalized papular skin eruption of 1 month’s duration. He had a history of stage IIIC malignant melanoma of the lower right leg with positive sentinel lymph node biopsy. The largest lymph node deposit was 0.03 mm without extracapsular extension. Whole-body positron emission tomography–computed tomography showed no evidence of distant disease. The patient was treated with wide local excision with clear surgical margins plus 12 cycles of nivolumab, which was discontinued due to colitis. Four months after the final cycle of nivolumab, the patient developed widespread erythematous papules with hemorrhagic yellow crusting and no mucosal involvement. He was referred to dermatology by his primary oncologist for further evaluation.

A punch biopsy from the abdomen showed para­keratosis with leukocytoclasis and a superficial dermal infiltrate of neutrophils and eosinophils (Figure 1). Direct immunofluorescence revealed linear basement membrane deposits of IgG and C3, consistent with subepidermal blistering disease. Indirect immunofluorescence demonstrated trace IgG and IgG4 antibodies localized to the epidermal roof of salt-split skin and was negative for IgA antibodies. An enzyme-linked immunoassay was positive for BP antigen 2 (BP180) antibodies (98.4 U/mL [positive, ≥9 U/mL]) and negative for BP antigen 1 (BP230) antibodies (4.3 U/mL [positive, ≥9 U/mL]). Overall, these findings were consistent with a diagnosis of BP.

The patient was treated with prednisone 60 mg daily with initial response; however, there was disease recurrence with tapering. Doxycycline 100 mg twice daily and nicotinamide 500 mg twice daily were added as steroid-sparing agents, as prednisone was discontinued due to mood changes. Three months after the prednisone taper, the patient continued to develop new blisters. He completed treatment with doxycycline and nicotinamide. Rituximab 375 mg weekly was then initiated for 4 weeks.

At 2-week follow-up after completing the rituximab course, the patient reported worsening symptoms and presented with new bullae on the abdomen and upper extremities (Figure 2). Because of the recent history of mood changes while taking prednisone, a trial of cyclosporine 100 mg twice daily (1.37 mg/kg/d) was initiated, with notable improvement within 2 weeks of treatment. After 2 months of cyclosporine, approximately 90% of the rash had resolved with a few tense bullae remaining on the left frontal scalp but no new flares (Figure 3). One month after treatment ended, the patient remained clear of lesions without relapse.

Programmed death receptor 1 inhibitors have shown dramatic efficacy for a growing number of solid and hematologic malignancies, especially malignant melanoma. However, their use is accompanied by nonspecific activation of the immune system, resulting in a variety of adverse events, many manifesting on the skin. Several cases of BP in patients treated with PD-1/PD-L1 inhibitors have been reported.⁹ Cutaneous irAEs usually manifest within 3 weeks of initiation of PD-1 inhibitor therapy; however, the onset of BP typically occurs later at approximately 21 weeks.^4,9 Our patient developed cutaneous manifestations 4 months after cessation of nivolumab.

To the Editor:

Immune checkpoint inhibitors have revolutionized the treatment of advanced-stage melanoma, with remarkably improved progression-free survival.¹ Anti–programmed death receptor 1 (anti–PD-1) therapies, such as nivolumab and pembrolizumab, are a class of checkpoint inhibitors that have been approved by the US Food and Drug Administration for unresectable metastatic melanoma. Anti–PD-1 agents block the interaction of programmed death-ligand 1 (PD-L1) found on tumor cells with the PD-1 receptor on T cells, facilitating a positive immune response.²

Although these therapies have demonstrated notable antitumor efficacy, they also give rise to numerous immune-related adverse events (irAEs). As many as 70% of patients treated with PD-1/PD-L1 inhibitors experience some type of organ system irAE, of which 30% to 40% are cutaneous.^3-6 Dermatologic adverse events are the most common irAEs, specifically spongiotic dermatitis, lichenoid dermatitis, pruritus, and vitiligo.⁷ Bullous pemphigoid (BP), an autoimmune bullous skin disorder caused by autoantibodies to basement membrane zone antigens, is a rare but potentially serious cutaneous irAE.⁸ Systemic corticosteroids commonly are used to treat immune checkpoint inhibitor–induced BP; other options include tetracyclines for maintenance therapy and rituximab for corticosteroid-refractory BP associated with anti-PD-1.⁹ We present a case of recalcitrant BP secondary to nivolumab therapy in a patient with metastatic melanoma who had near-complete resolution of BP following 2 months of cyclosporine.

A 41-year-old man presented with a generalized papular skin eruption of 1 month’s duration. He had a history of stage IIIC malignant melanoma of the lower right leg with positive sentinel lymph node biopsy. The largest lymph node deposit was 0.03 mm without extracapsular extension. Whole-body positron emission tomography–computed tomography showed no evidence of distant disease. The patient was treated with wide local excision with clear surgical margins plus 12 cycles of nivolumab, which was discontinued due to colitis. Four months after the final cycle of nivolumab, the patient developed widespread erythematous papules with hemorrhagic yellow crusting and no mucosal involvement. He was referred to dermatology by his primary oncologist for further evaluation.

A punch biopsy from the abdomen showed para­keratosis with leukocytoclasis and a superficial dermal infiltrate of neutrophils and eosinophils (Figure 1). Direct immunofluorescence revealed linear basement membrane deposits of IgG and C3, consistent with subepidermal blistering disease. Indirect immunofluorescence demonstrated trace IgG and IgG4 antibodies localized to the epidermal roof of salt-split skin and was negative for IgA antibodies. An enzyme-linked immunoassay was positive for BP antigen 2 (BP180) antibodies (98.4 U/mL [positive, ≥9 U/mL]) and negative for BP antigen 1 (BP230) antibodies (4.3 U/mL [positive, ≥9 U/mL]). Overall, these findings were consistent with a diagnosis of BP.

The patient was treated with prednisone 60 mg daily with initial response; however, there was disease recurrence with tapering. Doxycycline 100 mg twice daily and nicotinamide 500 mg twice daily were added as steroid-sparing agents, as prednisone was discontinued due to mood changes. Three months after the prednisone taper, the patient continued to develop new blisters. He completed treatment with doxycycline and nicotinamide. Rituximab 375 mg weekly was then initiated for 4 weeks.

At 2-week follow-up after completing the rituximab course, the patient reported worsening symptoms and presented with new bullae on the abdomen and upper extremities (Figure 2). Because of the recent history of mood changes while taking prednisone, a trial of cyclosporine 100 mg twice daily (1.37 mg/kg/d) was initiated, with notable improvement within 2 weeks of treatment. After 2 months of cyclosporine, approximately 90% of the rash had resolved with a few tense bullae remaining on the left frontal scalp but no new flares (Figure 3). One month after treatment ended, the patient remained clear of lesions without relapse.

Programmed death receptor 1 inhibitors have shown dramatic efficacy for a growing number of solid and hematologic malignancies, especially malignant melanoma. However, their use is accompanied by nonspecific activation of the immune system, resulting in a variety of adverse events, many manifesting on the skin. Several cases of BP in patients treated with PD-1/PD-L1 inhibitors have been reported.⁹ Cutaneous irAEs usually manifest within 3 weeks of initiation of PD-1 inhibitor therapy; however, the onset of BP typically occurs later at approximately 21 weeks.^4,9 Our patient developed cutaneous manifestations 4 months after cessation of nivolumab.

Recognizing cutaneous drug eruptions is important for treatment and prevention of recurrence. Fixed drug eruptions (FDEs) typically are harmless but can have major negative cosmetic consequences for patients. In its more severe forms, patients are at risk for widespread epithelial necrosis with accompanying complications. We report 1 patient with generalized FDE and 2 with generalized bullous FDE. We also discuss the recognition and treatment of the condition. Two patients previously had been diagnosed with systemic lupus erythematosus (SLE).

Case Series

Patient 1—A 60-year-old woman presented to dermatology with a rash on the trunk and groin folds of 4 days’ duration. She had a history of SLE and cutaneous lupus treated with hydroxychloroquine 200 mg twice daily and topical corticosteroids. She had started sulfamethoxazole-trimethoprim for a urinary tract infection with a rash appearing 1 day later. She reported burning skin pain with progression to blisters that “sloughed” off. She denied any known history of allergy to sulfa drugs. Prior to evaluation by dermatology, she visited an urgent care facility and was prescribed hydroxyzine and intramuscular corticosteroids. At presentation to dermatology 3 days after taking sulfamethoxazole-trimethoprim, she had annular flaccid bullae and superficial erosions with dusky borders on the right posterior thigh, right side of the chest, left inframammary fold, and right inguinal fold (Figure 1). She had no ocular, oral, or vaginal erosions. A diagnosis of generalized bullous FDE was favored over erythema multiforme or Stevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN). Shave biopsies from lesions on the right posterior thigh and right inguinal fold demonstrated interface dermatitis with epidermal necrosis, pigment incontinence, and numerous eosinophils. Direct immunofluorescence of the perilesional skin was negative for immunoprotein deposition. These findings were consistent with the clinical impression of generalized bullous FDE. Prior to receiving the histopathology report, the patient was initiated on a regimen of cyclosporine 5 mg/kg/d in the setting of normal renal function and followed until the eruption resolved completely. Cyclosporine was tapered at 2 weeks and discontinued at 3 weeks.

Patient 2—A 32-year-old woman presented for follow-up management of discoid lupus erythematosus. She had a history of systemic and cutaneous lupus, juvenile rheumatoid arthritis, and mixed connective tissue disease managed with prednisone, hydroxychloroquine, azathioprine, and belimumab. Physical examination revealed scarring alopecia with dyspigmentation and active inflammation consistent with uncontrolled cutaneous lupus. However, she also had oval-shaped hyperpigmented patches over the left breast, clavicle, and anterior chest consistent with a generalized FDE (Figure 2). The patient did not recall a history of similar lesions and could not identify a possible trigger. She was counseled on possible culprits and advised to avoid unnecessary medications. She had an unremarkable clinical course; therefore, no further intervention was necessary.

 

Patient 3—A 33-year-old man presented to the emergency department with a painful rash on the chest and back of 2 days’ duration that began 1 hour after taking naproxen (dosage unknown) for back pain. He had no notable medical history. The patient stated that the rash had slowly worsened and started to develop blisters. He visited an urgent care facility 1 day prior to the current presentation and was started on a 5-day course of prednisone 40 mg daily; the first 2 doses did not help. He denied any mucosal involvement apart from a tender lesion on the penis. He reported a history of an allergic reaction to penicillin. Physical examination revealed extensive dusky violaceous annular plaques with erythematous borders across the anterior and posterior trunk (Figure 3). Multiple flaccid bullae developed within these plaques, involving 15% of the body surface area. He was diagnosed with generalized bullous FDE based on the clinical history and histopathology. He was admitted to the burn intensive care unit and treated with cyclosporine 3 mg/kg/d with subsequent resolution of the eruption.

Comment

Presentation of FDEs—A fixed drug eruption manifests with 1 or more well-demarcated, red or violaceous, annular patches that resolve with postinflammatory hyperpigmentation; it occasionally may manifest with bullae. Initial eruptions may occur up to 2 weeks following medication exposure, but recurrent eruptions usually happen within minutes to hours later. They often are in the same location as prior lesions. A fixed drug eruption can be solitary, scattered, or generalized; a generalized FDE typically demonstrates multiple bilateral lesions that may itch, burn, or cause no symptoms. Patients can experience an FDE at any age, though the median age is reported as 35 to 60 years of age.¹ A fixed drug eruption usually occurs after ingestion of oral medications, though there have been a few reports with iodinated contrast.² Well-known culprits include antibiotics (eg, sulfamethoxazole-trimethoprim, tetracyclines, penicillins/cephalosporins, quinolones, dapsone), nonsteroidal anti-inflammatory drugs, acetaminophen (eg, paracetamol), barbiturates, antimalarials, and anticonvulsants. It also can occur with vaccines or with certain foods (fixed food eruption).^3,4 Clinicians may try an oral drug challenge to identify the cause of an FDE, but in patients with a history of a generalized FDE, the risk for developing an increasingly severe reaction with repeated exposure to the medication is too high.⁵

Histopathology—Patch testing at the site of prior eruption with suspected drug culprits may be useful.⁶ Histopathology of FDE typically demonstrates vacuolar changes at the dermoepidermal junction with a lichenoid lymphocytic infiltrate. Early lesions often show a predominance of eosinophils. Subepidermal clefting is a feature of the bullous variant. In an active lesion, there are large numbers of CD8+ T lymphocytes expressing natural killer cell–associated molecules.⁷ The pathologic mechanism is not well understood, though it has been hypothesized that memory CD8+ cells are maintained in specific regions of the epidermis by IL-15 produced in the microenvironment and are activated upon rechallenge.⁷Considerations in Generalized Bullous FDE—Generalized FDE is defined in the literature as an FDE with involvement of 3 of 6 body areas: head, neck, trunk, upper limbs, lower limbs, and genital area. It may cover more or less than 10% of the body surface area.^8-10 Although an isolated FDE frequently is asymptomatic and may not be cause for alarm, recurring drug eruptions increase the risk for development of generalized bullous FDE. Generalized bullous FDE is a rare subset. It is frequently misdiagnosed, and data on its incidence are uncertain.¹¹ Of note, several pathologies causing bullous lesions may be in the differential diagnosis, including bullous pemphigoid; pemphigus vulgaris; bullous SLE; or bullae from cutaneous lupus, staphylococcal scalded skin syndrome, erythema multiforme, or SJS/TEN.¹² When matched for body surface area involvement with SJS/TEN, generalized bullous FDE shares nearly identical mortality rates¹⁰; therefore, these patients should be treated with the same level of urgency and admitted to a critical care or burn unit, as they are at serious risk for infection and other complications.¹³

Clinical history and presentation along with histopathologic findings help to narrow down the differential diagnosis. Clinically, generalized bullous FDE does not affect the surrounding skin and manifests sooner after drug exposure (1–24 hours) with less mucosal involvement than SJS/TEN.⁹ Additionally, SJS/TEN patients frequently have generalized malaise and/or fever, while generalized bullous FDE patients do not. Finally, patients with generalized bullous FDE may report a history of a cutaneous eruption similar in morphology or in the same location.

Histopathologically, generalized bullous FDE may be similar to FDE with the addition of a subepidermal blister. Generalized bullous FDE patients have greater eosinophil infiltration and dermal melanophages than patients with SJS/TEN.⁹ Cellular infiltrates in generalized bullous FDE include more dermal CD41 cells, such as Foxp31 regulatory T cells; fewer intraepidermal CD561 cells; and fewer intraepidermal cells with granulysin.⁹ Occasionally, generalized bullous FDE causes full-thickness necrosis. In those cases, generalized bullous FDE cannot reliably be distinguished from other conditions with epidermal necrolysis on histopathology.¹³

FDE Diagnostics—A cytotoxin produced by cytotoxic T lymphocytes, granulysin can be measured to aid in diagnosis of FDE, though this test may not be widely available. High levels of granulysin in the blister fluid and serum can be used to distinguish SJS/TEN, erythema multiforme, and localized and generalized bullous FDE from other non–cytotoxic T lymphocyte–mediated bullous skin disorders, such as bullous pemphigoid, pemphigus, and bullous SLE.¹⁴ Blister granulysin levels are notably lower in generalized bullous FDE than in SJS/TEN.^9,14 Chen et al¹⁴ also found that granulysin levels can be used to gauge disease progression given that the levels sharply decrease after patients have reached maximal skin detachment.

Management—Avoidance of the inciting drug often is sufficient for patients with an FDE, as demonstrated in patient 2 in our case series. Clinicians also should counsel patients on avoidance of potential cross-reacting drugs. Symptomatic treatment for itch or pain is appropriate and may include antihistamines or topical steroids. Nonsteroidal anti-inflammatory drugs may exacerbate or be causative of FDE. For generalized bullous FDE, cyclosporine is favored in the literature^15,16 and was used to successfully treat both patients 1 and 3 in our case series. A short course of systemic corticosteroids or intravenous immunoglobulin also may be considered. Mild cases of generalized bullous FDE may be treated with close outpatient follow-up (patient 1), while severe cases require inpatient or even critical care monitoring with aggressive medical management to prevent the progression of skin desquamation (patient 3). Patients with severe oral lesions may require inpatient support for fluid maintenance.

Lupus History—Two patients in our case series had a history of lupus. Lupus itself can cause primary bullous lesions. Similar to FDE, bullous SLE can involve sun-exposed and nonexposed areas of the skin as well as the mucous membranes with a predilection for the lower vermilion lip.¹⁷ In bullous SLE, tense subepidermal blisters with a neutrophil-rich infiltrate form due to circulating antibodies to type VII collagen. These blisters have an erythematous or urticated base, most commonly on the face, upper trunk, and proximal extremities.¹⁸ In both SLE with skin manifestations and lupus limited to the skin, bullae may form due to extensive vacuolar degeneration. Similar to TEN, they can form rapidly in a widespread distribution.¹⁷ However, there is limited mucosal involvement, no clear drug association, and a better prognosis. Bullae caused by lupus will frequently demonstrate deposition of immunoproteins IgG, IgM, IgA, and complement component 3 at the basement membrane zone in perilesional skin on direct immunofluorescence. However, negative direct immunofluorescence does not rule out lupus.¹² At the same time, patients with lupus frequently have comorbidities requiring multiple medications; the need for these medications may predispose patients to higher rates of cutaneous drug eruptions.¹⁹ To our knowledge, there is no known association between FDE and lupus.

Conclusion

Patients with acute eruptions following the initiation of a new prescription or over-the-counter medication require urgent evaluation. Generalized bullous FDE requires timely diagnosis and intervention. Patients with lupus have an increased risk for cutaneous drug eruptions due to polypharmacy. Further investigation is necessary to determine if there is a pathophysiologic mechanism responsible for the development of FDE in lupus patients.

Recognizing cutaneous drug eruptions is important for treatment and prevention of recurrence. Fixed drug eruptions (FDEs) typically are harmless but can have major negative cosmetic consequences for patients. In its more severe forms, patients are at risk for widespread epithelial necrosis with accompanying complications. We report 1 patient with generalized FDE and 2 with generalized bullous FDE. We also discuss the recognition and treatment of the condition. Two patients previously had been diagnosed with systemic lupus erythematosus (SLE).

Case Series

Patient 1—A 60-year-old woman presented to dermatology with a rash on the trunk and groin folds of 4 days’ duration. She had a history of SLE and cutaneous lupus treated with hydroxychloroquine 200 mg twice daily and topical corticosteroids. She had started sulfamethoxazole-trimethoprim for a urinary tract infection with a rash appearing 1 day later. She reported burning skin pain with progression to blisters that “sloughed” off. She denied any known history of allergy to sulfa drugs. Prior to evaluation by dermatology, she visited an urgent care facility and was prescribed hydroxyzine and intramuscular corticosteroids. At presentation to dermatology 3 days after taking sulfamethoxazole-trimethoprim, she had annular flaccid bullae and superficial erosions with dusky borders on the right posterior thigh, right side of the chest, left inframammary fold, and right inguinal fold (Figure 1). She had no ocular, oral, or vaginal erosions. A diagnosis of generalized bullous FDE was favored over erythema multiforme or Stevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN). Shave biopsies from lesions on the right posterior thigh and right inguinal fold demonstrated interface dermatitis with epidermal necrosis, pigment incontinence, and numerous eosinophils. Direct immunofluorescence of the perilesional skin was negative for immunoprotein deposition. These findings were consistent with the clinical impression of generalized bullous FDE. Prior to receiving the histopathology report, the patient was initiated on a regimen of cyclosporine 5 mg/kg/d in the setting of normal renal function and followed until the eruption resolved completely. Cyclosporine was tapered at 2 weeks and discontinued at 3 weeks.

Patient 2—A 32-year-old woman presented for follow-up management of discoid lupus erythematosus. She had a history of systemic and cutaneous lupus, juvenile rheumatoid arthritis, and mixed connective tissue disease managed with prednisone, hydroxychloroquine, azathioprine, and belimumab. Physical examination revealed scarring alopecia with dyspigmentation and active inflammation consistent with uncontrolled cutaneous lupus. However, she also had oval-shaped hyperpigmented patches over the left breast, clavicle, and anterior chest consistent with a generalized FDE (Figure 2). The patient did not recall a history of similar lesions and could not identify a possible trigger. She was counseled on possible culprits and advised to avoid unnecessary medications. She had an unremarkable clinical course; therefore, no further intervention was necessary.

 

Patient 3—A 33-year-old man presented to the emergency department with a painful rash on the chest and back of 2 days’ duration that began 1 hour after taking naproxen (dosage unknown) for back pain. He had no notable medical history. The patient stated that the rash had slowly worsened and started to develop blisters. He visited an urgent care facility 1 day prior to the current presentation and was started on a 5-day course of prednisone 40 mg daily; the first 2 doses did not help. He denied any mucosal involvement apart from a tender lesion on the penis. He reported a history of an allergic reaction to penicillin. Physical examination revealed extensive dusky violaceous annular plaques with erythematous borders across the anterior and posterior trunk (Figure 3). Multiple flaccid bullae developed within these plaques, involving 15% of the body surface area. He was diagnosed with generalized bullous FDE based on the clinical history and histopathology. He was admitted to the burn intensive care unit and treated with cyclosporine 3 mg/kg/d with subsequent resolution of the eruption.

Comment

Presentation of FDEs—A fixed drug eruption manifests with 1 or more well-demarcated, red or violaceous, annular patches that resolve with postinflammatory hyperpigmentation; it occasionally may manifest with bullae. Initial eruptions may occur up to 2 weeks following medication exposure, but recurrent eruptions usually happen within minutes to hours later. They often are in the same location as prior lesions. A fixed drug eruption can be solitary, scattered, or generalized; a generalized FDE typically demonstrates multiple bilateral lesions that may itch, burn, or cause no symptoms. Patients can experience an FDE at any age, though the median age is reported as 35 to 60 years of age.¹ A fixed drug eruption usually occurs after ingestion of oral medications, though there have been a few reports with iodinated contrast.² Well-known culprits include antibiotics (eg, sulfamethoxazole-trimethoprim, tetracyclines, penicillins/cephalosporins, quinolones, dapsone), nonsteroidal anti-inflammatory drugs, acetaminophen (eg, paracetamol), barbiturates, antimalarials, and anticonvulsants. It also can occur with vaccines or with certain foods (fixed food eruption).^3,4 Clinicians may try an oral drug challenge to identify the cause of an FDE, but in patients with a history of a generalized FDE, the risk for developing an increasingly severe reaction with repeated exposure to the medication is too high.⁵

Histopathology—Patch testing at the site of prior eruption with suspected drug culprits may be useful.⁶ Histopathology of FDE typically demonstrates vacuolar changes at the dermoepidermal junction with a lichenoid lymphocytic infiltrate. Early lesions often show a predominance of eosinophils. Subepidermal clefting is a feature of the bullous variant. In an active lesion, there are large numbers of CD8+ T lymphocytes expressing natural killer cell–associated molecules.⁷ The pathologic mechanism is not well understood, though it has been hypothesized that memory CD8+ cells are maintained in specific regions of the epidermis by IL-15 produced in the microenvironment and are activated upon rechallenge.⁷Considerations in Generalized Bullous FDE—Generalized FDE is defined in the literature as an FDE with involvement of 3 of 6 body areas: head, neck, trunk, upper limbs, lower limbs, and genital area. It may cover more or less than 10% of the body surface area.^8-10 Although an isolated FDE frequently is asymptomatic and may not be cause for alarm, recurring drug eruptions increase the risk for development of generalized bullous FDE. Generalized bullous FDE is a rare subset. It is frequently misdiagnosed, and data on its incidence are uncertain.¹¹ Of note, several pathologies causing bullous lesions may be in the differential diagnosis, including bullous pemphigoid; pemphigus vulgaris; bullous SLE; or bullae from cutaneous lupus, staphylococcal scalded skin syndrome, erythema multiforme, or SJS/TEN.¹² When matched for body surface area involvement with SJS/TEN, generalized bullous FDE shares nearly identical mortality rates¹⁰; therefore, these patients should be treated with the same level of urgency and admitted to a critical care or burn unit, as they are at serious risk for infection and other complications.¹³

Clinical history and presentation along with histopathologic findings help to narrow down the differential diagnosis. Clinically, generalized bullous FDE does not affect the surrounding skin and manifests sooner after drug exposure (1–24 hours) with less mucosal involvement than SJS/TEN.⁹ Additionally, SJS/TEN patients frequently have generalized malaise and/or fever, while generalized bullous FDE patients do not. Finally, patients with generalized bullous FDE may report a history of a cutaneous eruption similar in morphology or in the same location.

Histopathologically, generalized bullous FDE may be similar to FDE with the addition of a subepidermal blister. Generalized bullous FDE patients have greater eosinophil infiltration and dermal melanophages than patients with SJS/TEN.⁹ Cellular infiltrates in generalized bullous FDE include more dermal CD41 cells, such as Foxp31 regulatory T cells; fewer intraepidermal CD561 cells; and fewer intraepidermal cells with granulysin.⁹ Occasionally, generalized bullous FDE causes full-thickness necrosis. In those cases, generalized bullous FDE cannot reliably be distinguished from other conditions with epidermal necrolysis on histopathology.¹³

FDE Diagnostics—A cytotoxin produced by cytotoxic T lymphocytes, granulysin can be measured to aid in diagnosis of FDE, though this test may not be widely available. High levels of granulysin in the blister fluid and serum can be used to distinguish SJS/TEN, erythema multiforme, and localized and generalized bullous FDE from other non–cytotoxic T lymphocyte–mediated bullous skin disorders, such as bullous pemphigoid, pemphigus, and bullous SLE.¹⁴ Blister granulysin levels are notably lower in generalized bullous FDE than in SJS/TEN.^9,14 Chen et al¹⁴ also found that granulysin levels can be used to gauge disease progression given that the levels sharply decrease after patients have reached maximal skin detachment.

Management—Avoidance of the inciting drug often is sufficient for patients with an FDE, as demonstrated in patient 2 in our case series. Clinicians also should counsel patients on avoidance of potential cross-reacting drugs. Symptomatic treatment for itch or pain is appropriate and may include antihistamines or topical steroids. Nonsteroidal anti-inflammatory drugs may exacerbate or be causative of FDE. For generalized bullous FDE, cyclosporine is favored in the literature^15,16 and was used to successfully treat both patients 1 and 3 in our case series. A short course of systemic corticosteroids or intravenous immunoglobulin also may be considered. Mild cases of generalized bullous FDE may be treated with close outpatient follow-up (patient 1), while severe cases require inpatient or even critical care monitoring with aggressive medical management to prevent the progression of skin desquamation (patient 3). Patients with severe oral lesions may require inpatient support for fluid maintenance.

Lupus History—Two patients in our case series had a history of lupus. Lupus itself can cause primary bullous lesions. Similar to FDE, bullous SLE can involve sun-exposed and nonexposed areas of the skin as well as the mucous membranes with a predilection for the lower vermilion lip.¹⁷ In bullous SLE, tense subepidermal blisters with a neutrophil-rich infiltrate form due to circulating antibodies to type VII collagen. These blisters have an erythematous or urticated base, most commonly on the face, upper trunk, and proximal extremities.¹⁸ In both SLE with skin manifestations and lupus limited to the skin, bullae may form due to extensive vacuolar degeneration. Similar to TEN, they can form rapidly in a widespread distribution.¹⁷ However, there is limited mucosal involvement, no clear drug association, and a better prognosis. Bullae caused by lupus will frequently demonstrate deposition of immunoproteins IgG, IgM, IgA, and complement component 3 at the basement membrane zone in perilesional skin on direct immunofluorescence. However, negative direct immunofluorescence does not rule out lupus.¹² At the same time, patients with lupus frequently have comorbidities requiring multiple medications; the need for these medications may predispose patients to higher rates of cutaneous drug eruptions.¹⁹ To our knowledge, there is no known association between FDE and lupus.

Conclusion

Patients with acute eruptions following the initiation of a new prescription or over-the-counter medication require urgent evaluation. Generalized bullous FDE requires timely diagnosis and intervention. Patients with lupus have an increased risk for cutaneous drug eruptions due to polypharmacy. Further investigation is necessary to determine if there is a pathophysiologic mechanism responsible for the development of FDE in lupus patients.

Recognizing cutaneous drug eruptions is important for treatment and prevention of recurrence. Fixed drug eruptions (FDEs) typically are harmless but can have major negative cosmetic consequences for patients. In its more severe forms, patients are at risk for widespread epithelial necrosis with accompanying complications. We report 1 patient with generalized FDE and 2 with generalized bullous FDE. We also discuss the recognition and treatment of the condition. Two patients previously had been diagnosed with systemic lupus erythematosus (SLE).

Case Series

Patient 1—A 60-year-old woman presented to dermatology with a rash on the trunk and groin folds of 4 days’ duration. She had a history of SLE and cutaneous lupus treated with hydroxychloroquine 200 mg twice daily and topical corticosteroids. She had started sulfamethoxazole-trimethoprim for a urinary tract infection with a rash appearing 1 day later. She reported burning skin pain with progression to blisters that “sloughed” off. She denied any known history of allergy to sulfa drugs. Prior to evaluation by dermatology, she visited an urgent care facility and was prescribed hydroxyzine and intramuscular corticosteroids. At presentation to dermatology 3 days after taking sulfamethoxazole-trimethoprim, she had annular flaccid bullae and superficial erosions with dusky borders on the right posterior thigh, right side of the chest, left inframammary fold, and right inguinal fold (Figure 1). She had no ocular, oral, or vaginal erosions. A diagnosis of generalized bullous FDE was favored over erythema multiforme or Stevens-Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN). Shave biopsies from lesions on the right posterior thigh and right inguinal fold demonstrated interface dermatitis with epidermal necrosis, pigment incontinence, and numerous eosinophils. Direct immunofluorescence of the perilesional skin was negative for immunoprotein deposition. These findings were consistent with the clinical impression of generalized bullous FDE. Prior to receiving the histopathology report, the patient was initiated on a regimen of cyclosporine 5 mg/kg/d in the setting of normal renal function and followed until the eruption resolved completely. Cyclosporine was tapered at 2 weeks and discontinued at 3 weeks.

Patient 2—A 32-year-old woman presented for follow-up management of discoid lupus erythematosus. She had a history of systemic and cutaneous lupus, juvenile rheumatoid arthritis, and mixed connective tissue disease managed with prednisone, hydroxychloroquine, azathioprine, and belimumab. Physical examination revealed scarring alopecia with dyspigmentation and active inflammation consistent with uncontrolled cutaneous lupus. However, she also had oval-shaped hyperpigmented patches over the left breast, clavicle, and anterior chest consistent with a generalized FDE (Figure 2). The patient did not recall a history of similar lesions and could not identify a possible trigger. She was counseled on possible culprits and advised to avoid unnecessary medications. She had an unremarkable clinical course; therefore, no further intervention was necessary.

 

Patient 3—A 33-year-old man presented to the emergency department with a painful rash on the chest and back of 2 days’ duration that began 1 hour after taking naproxen (dosage unknown) for back pain. He had no notable medical history. The patient stated that the rash had slowly worsened and started to develop blisters. He visited an urgent care facility 1 day prior to the current presentation and was started on a 5-day course of prednisone 40 mg daily; the first 2 doses did not help. He denied any mucosal involvement apart from a tender lesion on the penis. He reported a history of an allergic reaction to penicillin. Physical examination revealed extensive dusky violaceous annular plaques with erythematous borders across the anterior and posterior trunk (Figure 3). Multiple flaccid bullae developed within these plaques, involving 15% of the body surface area. He was diagnosed with generalized bullous FDE based on the clinical history and histopathology. He was admitted to the burn intensive care unit and treated with cyclosporine 3 mg/kg/d with subsequent resolution of the eruption.

Comment

Presentation of FDEs—A fixed drug eruption manifests with 1 or more well-demarcated, red or violaceous, annular patches that resolve with postinflammatory hyperpigmentation; it occasionally may manifest with bullae. Initial eruptions may occur up to 2 weeks following medication exposure, but recurrent eruptions usually happen within minutes to hours later. They often are in the same location as prior lesions. A fixed drug eruption can be solitary, scattered, or generalized; a generalized FDE typically demonstrates multiple bilateral lesions that may itch, burn, or cause no symptoms. Patients can experience an FDE at any age, though the median age is reported as 35 to 60 years of age.¹ A fixed drug eruption usually occurs after ingestion of oral medications, though there have been a few reports with iodinated contrast.² Well-known culprits include antibiotics (eg, sulfamethoxazole-trimethoprim, tetracyclines, penicillins/cephalosporins, quinolones, dapsone), nonsteroidal anti-inflammatory drugs, acetaminophen (eg, paracetamol), barbiturates, antimalarials, and anticonvulsants. It also can occur with vaccines or with certain foods (fixed food eruption).^3,4 Clinicians may try an oral drug challenge to identify the cause of an FDE, but in patients with a history of a generalized FDE, the risk for developing an increasingly severe reaction with repeated exposure to the medication is too high.⁵

Histopathology—Patch testing at the site of prior eruption with suspected drug culprits may be useful.⁶ Histopathology of FDE typically demonstrates vacuolar changes at the dermoepidermal junction with a lichenoid lymphocytic infiltrate. Early lesions often show a predominance of eosinophils. Subepidermal clefting is a feature of the bullous variant. In an active lesion, there are large numbers of CD8+ T lymphocytes expressing natural killer cell–associated molecules.⁷ The pathologic mechanism is not well understood, though it has been hypothesized that memory CD8+ cells are maintained in specific regions of the epidermis by IL-15 produced in the microenvironment and are activated upon rechallenge.⁷Considerations in Generalized Bullous FDE—Generalized FDE is defined in the literature as an FDE with involvement of 3 of 6 body areas: head, neck, trunk, upper limbs, lower limbs, and genital area. It may cover more or less than 10% of the body surface area.^8-10 Although an isolated FDE frequently is asymptomatic and may not be cause for alarm, recurring drug eruptions increase the risk for development of generalized bullous FDE. Generalized bullous FDE is a rare subset. It is frequently misdiagnosed, and data on its incidence are uncertain.¹¹ Of note, several pathologies causing bullous lesions may be in the differential diagnosis, including bullous pemphigoid; pemphigus vulgaris; bullous SLE; or bullae from cutaneous lupus, staphylococcal scalded skin syndrome, erythema multiforme, or SJS/TEN.¹² When matched for body surface area involvement with SJS/TEN, generalized bullous FDE shares nearly identical mortality rates¹⁰; therefore, these patients should be treated with the same level of urgency and admitted to a critical care or burn unit, as they are at serious risk for infection and other complications.¹³

Clinical history and presentation along with histopathologic findings help to narrow down the differential diagnosis. Clinically, generalized bullous FDE does not affect the surrounding skin and manifests sooner after drug exposure (1–24 hours) with less mucosal involvement than SJS/TEN.⁹ Additionally, SJS/TEN patients frequently have generalized malaise and/or fever, while generalized bullous FDE patients do not. Finally, patients with generalized bullous FDE may report a history of a cutaneous eruption similar in morphology or in the same location.

Histopathologically, generalized bullous FDE may be similar to FDE with the addition of a subepidermal blister. Generalized bullous FDE patients have greater eosinophil infiltration and dermal melanophages than patients with SJS/TEN.⁹ Cellular infiltrates in generalized bullous FDE include more dermal CD41 cells, such as Foxp31 regulatory T cells; fewer intraepidermal CD561 cells; and fewer intraepidermal cells with granulysin.⁹ Occasionally, generalized bullous FDE causes full-thickness necrosis. In those cases, generalized bullous FDE cannot reliably be distinguished from other conditions with epidermal necrolysis on histopathology.¹³

FDE Diagnostics—A cytotoxin produced by cytotoxic T lymphocytes, granulysin can be measured to aid in diagnosis of FDE, though this test may not be widely available. High levels of granulysin in the blister fluid and serum can be used to distinguish SJS/TEN, erythema multiforme, and localized and generalized bullous FDE from other non–cytotoxic T lymphocyte–mediated bullous skin disorders, such as bullous pemphigoid, pemphigus, and bullous SLE.¹⁴ Blister granulysin levels are notably lower in generalized bullous FDE than in SJS/TEN.^9,14 Chen et al¹⁴ also found that granulysin levels can be used to gauge disease progression given that the levels sharply decrease after patients have reached maximal skin detachment.

Management—Avoidance of the inciting drug often is sufficient for patients with an FDE, as demonstrated in patient 2 in our case series. Clinicians also should counsel patients on avoidance of potential cross-reacting drugs. Symptomatic treatment for itch or pain is appropriate and may include antihistamines or topical steroids. Nonsteroidal anti-inflammatory drugs may exacerbate or be causative of FDE. For generalized bullous FDE, cyclosporine is favored in the literature^15,16 and was used to successfully treat both patients 1 and 3 in our case series. A short course of systemic corticosteroids or intravenous immunoglobulin also may be considered. Mild cases of generalized bullous FDE may be treated with close outpatient follow-up (patient 1), while severe cases require inpatient or even critical care monitoring with aggressive medical management to prevent the progression of skin desquamation (patient 3). Patients with severe oral lesions may require inpatient support for fluid maintenance.

Lupus History—Two patients in our case series had a history of lupus. Lupus itself can cause primary bullous lesions. Similar to FDE, bullous SLE can involve sun-exposed and nonexposed areas of the skin as well as the mucous membranes with a predilection for the lower vermilion lip.¹⁷ In bullous SLE, tense subepidermal blisters with a neutrophil-rich infiltrate form due to circulating antibodies to type VII collagen. These blisters have an erythematous or urticated base, most commonly on the face, upper trunk, and proximal extremities.¹⁸ In both SLE with skin manifestations and lupus limited to the skin, bullae may form due to extensive vacuolar degeneration. Similar to TEN, they can form rapidly in a widespread distribution.¹⁷ However, there is limited mucosal involvement, no clear drug association, and a better prognosis. Bullae caused by lupus will frequently demonstrate deposition of immunoproteins IgG, IgM, IgA, and complement component 3 at the basement membrane zone in perilesional skin on direct immunofluorescence. However, negative direct immunofluorescence does not rule out lupus.¹² At the same time, patients with lupus frequently have comorbidities requiring multiple medications; the need for these medications may predispose patients to higher rates of cutaneous drug eruptions.¹⁹ To our knowledge, there is no known association between FDE and lupus.

Conclusion

Patients with acute eruptions following the initiation of a new prescription or over-the-counter medication require urgent evaluation. Generalized bullous FDE requires timely diagnosis and intervention. Patients with lupus have an increased risk for cutaneous drug eruptions due to polypharmacy. Further investigation is necessary to determine if there is a pathophysiologic mechanism responsible for the development of FDE in lupus patients.

To the Editor:

A 48-year-old man presented with nodular lesions in a sporotrichoid pattern on the right hand and forearm of 3 months’ duration (Figure). There were no lymphadeno-pathies, and he had no notable medical history. He denied fever and other systemic symptoms. The patient recently had manipulated a warm water fish aquarium. Although he did not recall a clear injury, inadvertent mild trauma was a possibility. He denied other contact or trauma in relation to animals or vegetables.

Histopathology from a punch biopsy of the forearm revealed a granulomatous infiltrate with necrosis at the deep dermis level at the interface with the subcutaneous cellular tissue that was composed of mainly epithelioid cells with a few multinucleated giant cells. No acid-fast bacilli or fungi were observed with special stains.

A polymerase chain reaction assay for atypical mycobacteria was positive for Mycobacterium interjectum. The culture of the skin biopsy was negative for fungi and mycobacteria after long incubation (6 weeks) on 2 occasions, and an antibiogram was not available. Complementary tests including hemogram, HIV serology, and chest and upper extremity radiographs did not reveal any abnormalities.

To the Editor:

A 48-year-old man presented with nodular lesions in a sporotrichoid pattern on the right hand and forearm of 3 months’ duration (Figure). There were no lymphadeno-pathies, and he had no notable medical history. He denied fever and other systemic symptoms. The patient recently had manipulated a warm water fish aquarium. Although he did not recall a clear injury, inadvertent mild trauma was a possibility. He denied other contact or trauma in relation to animals or vegetables.

Histopathology from a punch biopsy of the forearm revealed a granulomatous infiltrate with necrosis at the deep dermis level at the interface with the subcutaneous cellular tissue that was composed of mainly epithelioid cells with a few multinucleated giant cells. No acid-fast bacilli or fungi were observed with special stains.

A polymerase chain reaction assay for atypical mycobacteria was positive for Mycobacterium interjectum. The culture of the skin biopsy was negative for fungi and mycobacteria after long incubation (6 weeks) on 2 occasions, and an antibiogram was not available. Complementary tests including hemogram, HIV serology, and chest and upper extremity radiographs did not reveal any abnormalities.

To the Editor:

A 48-year-old man presented with nodular lesions in a sporotrichoid pattern on the right hand and forearm of 3 months’ duration (Figure). There were no lymphadeno-pathies, and he had no notable medical history. He denied fever and other systemic symptoms. The patient recently had manipulated a warm water fish aquarium. Although he did not recall a clear injury, inadvertent mild trauma was a possibility. He denied other contact or trauma in relation to animals or vegetables.

Histopathology from a punch biopsy of the forearm revealed a granulomatous infiltrate with necrosis at the deep dermis level at the interface with the subcutaneous cellular tissue that was composed of mainly epithelioid cells with a few multinucleated giant cells. No acid-fast bacilli or fungi were observed with special stains.

A polymerase chain reaction assay for atypical mycobacteria was positive for Mycobacterium interjectum. The culture of the skin biopsy was negative for fungi and mycobacteria after long incubation (6 weeks) on 2 occasions, and an antibiogram was not available. Complementary tests including hemogram, HIV serology, and chest and upper extremity radiographs did not reveal any abnormalities.

Brazilian peppertree (Schinus terebinthifolia), a member of the Anacardiaceae family, is an internationally invasive plant that causes allergic contact dermatitis (ACD) in susceptible individuals. This noxious weed has settled into the landscape of the southern United States and continues to expand. Its key identifying features include its year-round white flowers as well as a peppery and turpentinelike aroma created by cracking its bright red berries. The ACD associated with contact—primarily with the plant’s sap—stems from known alkenyl phenols, cardol and cardanol. Treatment of Brazilian peppertree–associated ACD parallels that for poison ivy. As this pest increases its range, dermatologists living in endemic areas should familiarize themselves with Brazilian peppertree and its potential for harm.

Brazilian Peppertree Morphology and Geography

Plants in the Anacardiaceae family contribute to more ACD than any other family, and its 80 genera include most of the urushiol-containing plants, such as Toxicodendron (poison ivy, poison oak, poison sumac, Japanese lacquer tree), Anacardium (cashew tree), Mangifera (mango fruit), Semecarpus (India marking nut tree), and Schinus (Brazilian peppertree). Deciduous and evergreen tree members of the Anacardiaceae family grow primarily in tropical and subtropical locations and produce thick resins, 5-petalled flowers, and small fruit known as drupes. The genus name for Brazilian peppertree, Schinus, derives from Latin and Greek words meaning “mastic tree,” a relative of the pistachio tree that the Brazilian peppertree resembles.¹ Brazilian peppertree leaves look and smell similar to Pistacia terebinthus (turpentine tree or terebinth), from which the species name terebinthifolia derives.²

Brazilian peppertree originated in South America, particularly Brazil, Paraguay, and Argentina.³ Since the 1840s,⁴ it has been an invasive weed in the United States, notably in Florida, California, Hawaii, Alabama, Georgia,⁵ Arizona,⁶ Nevada,³ and Texas.^5,7 The plant also grows throughout the world, including parts of Africa, Asia, Central America, Europe,⁶ New Zealand,⁸ Australia, and various islands.⁹ The plant expertly outcompetes neighboring plants and has prompted control and eradication efforts in many locations.³

Identifying Features and Allergenic Plant Parts

Brazilian peppertree can be either a shrub or tree up to 30 feet tall.⁴ As an evergreen, it retains its leaves year-round. During fruiting seasons (primarily December through March⁷), bright red or pink (depending on the variety³) berries appear (Figure 1A) and contribute to its nickname “Florida holly.” Although generally considered an unwelcome guest in Florida, it does display white flowers (Figure 1B) year-round, especially from September to November.⁹ It characteristically exhibits 3 to 13 leaflets per leaf.¹⁰ The leaflets’ ovoid and ridged edges, netlike vasculature, shiny hue, and aroma can help identify the plant (Figure 2A). For decades, the sap of the Brazilian peppertree has been associated with skin ­irritation (Figure 2B).⁶ Although the sap of the plant serves as the main culprit of Brazilian peppertree–­associated ACD, it appears that other parts of the plant, including the fruit, can cause irritating effects to skin on contact.^11,12 The leaves, trunk, and fruit can be harmful to both humans and animals.⁶ Chemicals from flowers and crushed fruit also can lead to irritating effects in the respiratory tract if aspirated.¹³

Urushiol, an oily resin present in most plants of the Anacardiaceae family,¹⁴ contains many chemicals, including allergenic phenols, catechols, and resorcinols.¹⁵ Urushiol-allergic individuals develop dermatitis upon exposure to Brazilian peppertree sap.⁶ Alkenyl phenols found in Brazilian peppertree lead to the cutaneous manifestations in sensitized patients.^11,12 In 1983, Stahl et al¹¹ identified a phenol, cardanol (chemical name ­3-pentadecylphenol¹⁶) C15:1, in Brazilian peppertree fruit. The group further tested this compound’s effect on skin via patch testing, which showed an allergic response.¹¹ Cashew nut shells (Anacardium occidentale) contain cardanol, anacardic acid (a phenolic acid), and cardol (a phenol with the chemical name ­5-pentadecylresorcinol),^15,16 though Stahl et al¹¹ were unable to extract these 2 substances (if present) from Brazilian peppertree fruit. When exposed to cardol and anacardic acid, those allergic to poison ivy often develop ACD,¹⁵ and these 2 substances are more irritating than cardanol.¹¹ A later study did identify cardol in addition to cardanol in Brazilian peppertree.¹²

Cutaneous Manifestations

Brazilian peppertree–induced ACD appears similar to other plant-induced ACD with linear streaks of erythema, juicy papules, vesicles, coalescing erythematous plaques, and/or occasional edema and bullae accompanied by intense pruritus.

Treatment

Avoiding contact with Brazilian peppertree is the first line of defense, and treatment for a reaction associated with exposure is similar to that of poison ivy.¹⁷ Application of cool compresses, calamine lotion, and topical astringents offer symptom alleviation, and topical steroids (eg, clobetasol propionate 0.05% twice daily) can improve mild localized ACD when given prior to formation of blisters. For more severe and diffuse ACD, oral steroids (eg, prednisone 1 mg/kg/d tapered over 2–3 weeks) likely are necessary, though intramuscular options greatly alleviate discomfort in more severe cases (eg, intramuscular triamcinolone acetonide 1 mg/kg combined with betamethasone 0.1 mg/kg). Physicians should monitor sites for any signs of superimposed bacterial infection and initiate antibiotics as necessary.¹⁷

Brazilian peppertree (Schinus terebinthifolia), a member of the Anacardiaceae family, is an internationally invasive plant that causes allergic contact dermatitis (ACD) in susceptible individuals. This noxious weed has settled into the landscape of the southern United States and continues to expand. Its key identifying features include its year-round white flowers as well as a peppery and turpentinelike aroma created by cracking its bright red berries. The ACD associated with contact—primarily with the plant’s sap—stems from known alkenyl phenols, cardol and cardanol. Treatment of Brazilian peppertree–associated ACD parallels that for poison ivy. As this pest increases its range, dermatologists living in endemic areas should familiarize themselves with Brazilian peppertree and its potential for harm.

Brazilian Peppertree Morphology and Geography

Plants in the Anacardiaceae family contribute to more ACD than any other family, and its 80 genera include most of the urushiol-containing plants, such as Toxicodendron (poison ivy, poison oak, poison sumac, Japanese lacquer tree), Anacardium (cashew tree), Mangifera (mango fruit), Semecarpus (India marking nut tree), and Schinus (Brazilian peppertree). Deciduous and evergreen tree members of the Anacardiaceae family grow primarily in tropical and subtropical locations and produce thick resins, 5-petalled flowers, and small fruit known as drupes. The genus name for Brazilian peppertree, Schinus, derives from Latin and Greek words meaning “mastic tree,” a relative of the pistachio tree that the Brazilian peppertree resembles.¹ Brazilian peppertree leaves look and smell similar to Pistacia terebinthus (turpentine tree or terebinth), from which the species name terebinthifolia derives.²

Brazilian peppertree originated in South America, particularly Brazil, Paraguay, and Argentina.³ Since the 1840s,⁴ it has been an invasive weed in the United States, notably in Florida, California, Hawaii, Alabama, Georgia,⁵ Arizona,⁶ Nevada,³ and Texas.^5,7 The plant also grows throughout the world, including parts of Africa, Asia, Central America, Europe,⁶ New Zealand,⁸ Australia, and various islands.⁹ The plant expertly outcompetes neighboring plants and has prompted control and eradication efforts in many locations.³

Identifying Features and Allergenic Plant Parts

Brazilian peppertree can be either a shrub or tree up to 30 feet tall.⁴ As an evergreen, it retains its leaves year-round. During fruiting seasons (primarily December through March⁷), bright red or pink (depending on the variety³) berries appear (Figure 1A) and contribute to its nickname “Florida holly.” Although generally considered an unwelcome guest in Florida, it does display white flowers (Figure 1B) year-round, especially from September to November.⁹ It characteristically exhibits 3 to 13 leaflets per leaf.¹⁰ The leaflets’ ovoid and ridged edges, netlike vasculature, shiny hue, and aroma can help identify the plant (Figure 2A). For decades, the sap of the Brazilian peppertree has been associated with skin ­irritation (Figure 2B).⁶ Although the sap of the plant serves as the main culprit of Brazilian peppertree–­associated ACD, it appears that other parts of the plant, including the fruit, can cause irritating effects to skin on contact.^11,12 The leaves, trunk, and fruit can be harmful to both humans and animals.⁶ Chemicals from flowers and crushed fruit also can lead to irritating effects in the respiratory tract if aspirated.¹³

Urushiol, an oily resin present in most plants of the Anacardiaceae family,¹⁴ contains many chemicals, including allergenic phenols, catechols, and resorcinols.¹⁵ Urushiol-allergic individuals develop dermatitis upon exposure to Brazilian peppertree sap.⁶ Alkenyl phenols found in Brazilian peppertree lead to the cutaneous manifestations in sensitized patients.^11,12 In 1983, Stahl et al¹¹ identified a phenol, cardanol (chemical name ­3-pentadecylphenol¹⁶) C15:1, in Brazilian peppertree fruit. The group further tested this compound’s effect on skin via patch testing, which showed an allergic response.¹¹ Cashew nut shells (Anacardium occidentale) contain cardanol, anacardic acid (a phenolic acid), and cardol (a phenol with the chemical name ­5-pentadecylresorcinol),^15,16 though Stahl et al¹¹ were unable to extract these 2 substances (if present) from Brazilian peppertree fruit. When exposed to cardol and anacardic acid, those allergic to poison ivy often develop ACD,¹⁵ and these 2 substances are more irritating than cardanol.¹¹ A later study did identify cardol in addition to cardanol in Brazilian peppertree.¹²

Cutaneous Manifestations

Brazilian peppertree–induced ACD appears similar to other plant-induced ACD with linear streaks of erythema, juicy papules, vesicles, coalescing erythematous plaques, and/or occasional edema and bullae accompanied by intense pruritus.

Treatment

Avoiding contact with Brazilian peppertree is the first line of defense, and treatment for a reaction associated with exposure is similar to that of poison ivy.¹⁷ Application of cool compresses, calamine lotion, and topical astringents offer symptom alleviation, and topical steroids (eg, clobetasol propionate 0.05% twice daily) can improve mild localized ACD when given prior to formation of blisters. For more severe and diffuse ACD, oral steroids (eg, prednisone 1 mg/kg/d tapered over 2–3 weeks) likely are necessary, though intramuscular options greatly alleviate discomfort in more severe cases (eg, intramuscular triamcinolone acetonide 1 mg/kg combined with betamethasone 0.1 mg/kg). Physicians should monitor sites for any signs of superimposed bacterial infection and initiate antibiotics as necessary.¹⁷

Brazilian peppertree (Schinus terebinthifolia), a member of the Anacardiaceae family, is an internationally invasive plant that causes allergic contact dermatitis (ACD) in susceptible individuals. This noxious weed has settled into the landscape of the southern United States and continues to expand. Its key identifying features include its year-round white flowers as well as a peppery and turpentinelike aroma created by cracking its bright red berries. The ACD associated with contact—primarily with the plant’s sap—stems from known alkenyl phenols, cardol and cardanol. Treatment of Brazilian peppertree–associated ACD parallels that for poison ivy. As this pest increases its range, dermatologists living in endemic areas should familiarize themselves with Brazilian peppertree and its potential for harm.

Brazilian Peppertree Morphology and Geography

Plants in the Anacardiaceae family contribute to more ACD than any other family, and its 80 genera include most of the urushiol-containing plants, such as Toxicodendron (poison ivy, poison oak, poison sumac, Japanese lacquer tree), Anacardium (cashew tree), Mangifera (mango fruit), Semecarpus (India marking nut tree), and Schinus (Brazilian peppertree). Deciduous and evergreen tree members of the Anacardiaceae family grow primarily in tropical and subtropical locations and produce thick resins, 5-petalled flowers, and small fruit known as drupes. The genus name for Brazilian peppertree, Schinus, derives from Latin and Greek words meaning “mastic tree,” a relative of the pistachio tree that the Brazilian peppertree resembles.¹ Brazilian peppertree leaves look and smell similar to Pistacia terebinthus (turpentine tree or terebinth), from which the species name terebinthifolia derives.²

Brazilian peppertree originated in South America, particularly Brazil, Paraguay, and Argentina.³ Since the 1840s,⁴ it has been an invasive weed in the United States, notably in Florida, California, Hawaii, Alabama, Georgia,⁵ Arizona,⁶ Nevada,³ and Texas.^5,7 The plant also grows throughout the world, including parts of Africa, Asia, Central America, Europe,⁶ New Zealand,⁸ Australia, and various islands.⁹ The plant expertly outcompetes neighboring plants and has prompted control and eradication efforts in many locations.³

Identifying Features and Allergenic Plant Parts

Brazilian peppertree can be either a shrub or tree up to 30 feet tall.⁴ As an evergreen, it retains its leaves year-round. During fruiting seasons (primarily December through March⁷), bright red or pink (depending on the variety³) berries appear (Figure 1A) and contribute to its nickname “Florida holly.” Although generally considered an unwelcome guest in Florida, it does display white flowers (Figure 1B) year-round, especially from September to November.⁹ It characteristically exhibits 3 to 13 leaflets per leaf.¹⁰ The leaflets’ ovoid and ridged edges, netlike vasculature, shiny hue, and aroma can help identify the plant (Figure 2A). For decades, the sap of the Brazilian peppertree has been associated with skin ­irritation (Figure 2B).⁶ Although the sap of the plant serves as the main culprit of Brazilian peppertree–­associated ACD, it appears that other parts of the plant, including the fruit, can cause irritating effects to skin on contact.^11,12 The leaves, trunk, and fruit can be harmful to both humans and animals.⁶ Chemicals from flowers and crushed fruit also can lead to irritating effects in the respiratory tract if aspirated.¹³

Urushiol, an oily resin present in most plants of the Anacardiaceae family,¹⁴ contains many chemicals, including allergenic phenols, catechols, and resorcinols.¹⁵ Urushiol-allergic individuals develop dermatitis upon exposure to Brazilian peppertree sap.⁶ Alkenyl phenols found in Brazilian peppertree lead to the cutaneous manifestations in sensitized patients.^11,12 In 1983, Stahl et al¹¹ identified a phenol, cardanol (chemical name ­3-pentadecylphenol¹⁶) C15:1, in Brazilian peppertree fruit. The group further tested this compound’s effect on skin via patch testing, which showed an allergic response.¹¹ Cashew nut shells (Anacardium occidentale) contain cardanol, anacardic acid (a phenolic acid), and cardol (a phenol with the chemical name ­5-pentadecylresorcinol),^15,16 though Stahl et al¹¹ were unable to extract these 2 substances (if present) from Brazilian peppertree fruit. When exposed to cardol and anacardic acid, those allergic to poison ivy often develop ACD,¹⁵ and these 2 substances are more irritating than cardanol.¹¹ A later study did identify cardol in addition to cardanol in Brazilian peppertree.¹²

Cutaneous Manifestations

Brazilian peppertree–induced ACD appears similar to other plant-induced ACD with linear streaks of erythema, juicy papules, vesicles, coalescing erythematous plaques, and/or occasional edema and bullae accompanied by intense pruritus.

Treatment

Avoiding contact with Brazilian peppertree is the first line of defense, and treatment for a reaction associated with exposure is similar to that of poison ivy.¹⁷ Application of cool compresses, calamine lotion, and topical astringents offer symptom alleviation, and topical steroids (eg, clobetasol propionate 0.05% twice daily) can improve mild localized ACD when given prior to formation of blisters. For more severe and diffuse ACD, oral steroids (eg, prednisone 1 mg/kg/d tapered over 2–3 weeks) likely are necessary, though intramuscular options greatly alleviate discomfort in more severe cases (eg, intramuscular triamcinolone acetonide 1 mg/kg combined with betamethasone 0.1 mg/kg). Physicians should monitor sites for any signs of superimposed bacterial infection and initiate antibiotics as necessary.¹⁷