Figure 1. Clinical pattern. A, Pruritic skin eruption classified as eczematous and photographed 18 days after onset. B, Closer view of the same eruption.
Figure 2. Histological pattern. A, Spongiotic dermatitis with superficial perivascular lymphocytic infiltrate and patchy lymphocyte exocytosis (hematoxylin-eosin, original magnification ×200). B, Spongiotic dermatitis with superficial dermal lymphocytic infiltrate and focus of vacuolar interface inflammation (hematoxylin-eosin, original magnification ×400).
Jean-Claude Roujeau, Maja Mockenhaupt, Steven R. Tahan, Joshua Henshaw, Emily C. Martin, Matt Harding, Ben van Baelen, Leif Bengtsson, Priya Singhal, Robert S. Kauffman, Robert S. Stern. Telaprevir-Related Dermatitis. JAMA Dermatol. 2013;149(2):152–158. doi:10.1001/jamadermatol.2013.938
Author Affiliations: Department of Dermaology, Université Paris-Est, Créteil, France (Dr Roujeau); Dokumentationszentrum schwerer Hautreaktionen, Department of Dermatology, University Medical Center, Freiburg, Germany (Dr Mockenhaupt); Division of Dermatopathology, Department of Pathology (Dr Tahan), and Department of Dermatology (Dr Stern), Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts; Vertex Pharmaceuticals, Cambridge, Massachusetts (Drs Henshaw, Martin, Harding, Singhal, and Kauffman and Mr Bengtsson); and Janssen Pharma, Beerse, Belgium (Dr van Baelen).
Objective To evaluate the incidence, type, and severity of telaprevir-associated skin reactions.
Design Three dermatologists assessed available information including photographs, biopsy results, and clinical summaries of all cases with skin eruptions reported as moderate or severe during the telaprevir clinical development program. For cases from placebo-controlled trials, they were masked to exposure.
Settings Phase 1 to 3 studies of telaprevir combination therapy for hepatitis C.
Patients All patients with skin eruptions enrolled in telaprevir clinical trials prior to 2011
Main Outcome Measures Incidence, diagnosis, morphologic features, extent, and severity of skin eruption.
Results Skin eruptions were more frequent in patients who received telaprevir as part of hepatitis C treatment compared with pegylated interferon (peginterferon) and ribavirin alone (56% vs 34% overall; 3.7% vs 0.4% severe). Occurring at any time during the 12 weeks of telaprevir combination regimen, in more than 90% of cases, this eruption is pruritic eczematous dermatitis. None of the clinical or genetic factors examined were substantial risk factors for dermatitis. Three cases of Stevens-Johnson Syndrome (SJS), and 11 cases of Drug Reaction with Eosinophilia and Systemic Symptoms (DRESS) were suspected, with 2 SJS and 3 DRESS cases considered likely.
Conclusions Telaprevir-related dermatitis occurs in a majority of telaprevir-treated patients. It is an eczematous dermatitis that differs in timing and appearance from the eruptions usually associated with drug reactions. The strong signal for an increased risk of DRESS or SJS requires particular vigilance in telaprevir-treated patients.
Skin manifestations have been previously reported in association with pegylated interferon (peginterferon) and ribavirin treatment of hepatitis C virus (HCV). Usually pruritic and described as “dermatitis” or “eczematiform,” skin eruptions occur in up to 30% of patients, are most often mild or moderate, and seldom lead to discontinuation of treatment.1- 10
The HCV protease inhibitor telaprevir, administered in addition to peginterferon and ribavirin, significantly increases the rate of sustained viral response in patients with genotype 1 chronic HCV infection.11- 17 During phase 2 studies, 41% to 61% of patients treated with telaprevir plus peginterferon and ribavirin developed skin eruptions, and 5% to 8% of patients discontinued therapy owing to skin complaint.11- 14 The rates of skin eruptions were far higher than those observed in control groups of patients using placebo with peginterferon and ribavirin. As a result, the phase 3 development program included special efforts to better evaluate the incidence, clinical characteristics, severity, and risk factors for skin reactions.
On the basis of photographic evaluations, a predominantly eczematous reaction pattern was identified. Demographic characteristics, HLA genotypes, pharmacokinetic parameters, and mutations of MDR1 that may affect telaprevir metabolism were compared between subgroups of patients with and without skin eruptions during telaprevir therapy.18- 20
Data from 3 international phase 3 studies of telaprevir in combination with interferon and ribavirin were assessed. Study investigators graded reactions by the extent and morphologic features: grade 1 (mild) reactions were localized skin eruptions; grade 2 (moderate), diffuse skin eruptions involving up to 50% of body surface area; and grade 3 (severe), generalized eruptions involving 50% or more of body surface area or any eruption with bullae, vesicles, purpura, epidermal detachment, or mucous membrane erosions. A special search category that grouped related MedDRA (Medical Dictionary for Regulatory Activities) terms21 was used to capture possible cutaneous reactions.
In the largest phase 3 study, a prospective evaluation of all “rash events of special interest” (ESI) was implemented, defined as any skin eruption leading to treatment discontinuation and all grade 3 skin eruptions. Data collected for ESIs included an investigator-completed form that provided details on the extent, distribution, morphologic features of the eruption, and associated symptoms. The protocol included standardized dermatologic consultations, photographs, skin biopsies, and laboratory studies including serial complete blood cell counts, viral load, and liver function tests. In other phase 3 trials,16,17 similar forms were completed. Dermatologic evaluations and photographs were encouraged but not required.
A “Dermatology Expert Panel” ( J.C.R., M.M., and R.S.S.) assessed all patients with an ESI in phase 3 studies. The panel was masked to treatment group (telaprevir or placebo in addition to peginterferon and ribavirin). A single dermatopathologist (S.R.T.) reviewed 53 histopathologic specimens available from 84 skin biopsies.
Logistic regression analyses were performed on data from 1346 patients treated with telaprevir plus peginterferon and ribavirin and 764 patients treated with placebo plus peginterferon and ribavirin from phase 2 and 3 studies to evaluate predictors of skin eruption. For each of the demographic and baseline disease factors (sex, age, body mass index, region, ethnic group, prior treatment status, and fibrosis category), a bivariate logistic regression analysis was performed including treatment (telaprevir or placebo) and the evaluated factor.
Logistic regression analyses were also performed on 556 patients assigned to 12 weeks of telaprevir-based regimen in combination with peginterferon and ribavirin to determine the correlation between skin reaction and plasma concentration of telaprevir, peginterferon, and ribavirin at steady state (week 4 or 8). Univariate logistic regression analysis was performed on each of the exposure factors.
Genomic DNA was isolated from peripheral blood mononuclear cells, and quality control was performed according to established standard laboratory procedures. For MDR1 genotyping, polymerase chain reaction (PCR) reagents for single-nucleotide polymorphism assays were obtained from Applied Biosystems Inc and are part of their validated Assay on Demand portfolio. The single-nucleotide polymorphism assay classifies samples as homozygotes (having only allele 1 or 2 [eg, 3435 CC or TT]) or heterozygotes (having both alleles 1 and 2 [eg, 3435 CT]). For HLA typing, DNA from a convenience sample of phase 2 and 3 patients were sent to the HLA Services Laboratory, American Red Cross–Northeast Division. Typing was performed by a high-resolution PCR-based method designed to type patients across 5319 distinct HLA-A, HLA-B, HLA-CW, HLA-DRB, and HLA-DQB loci. Each HLA allele was tested for an association with skin eruption using a 2-sided Wald test for logistic regression with significance level of .01. The odds ratio (OR) with 95% confidence interval and a P value for a Wald test based on the OR were reported for each allele. The Sidak method was used to correct multiple comparisons.22 All studies included in these analyses were approved by an institutional review board or human ethics committee.
In phase 3 studies, adverse skin events occurred significantly more frequently with telaprevir plus peginterferon and ribavirin (56%) than with placebo plus peginterferon and ribavirin (34%; 95% CI of risk difference [RD], 17% to 27%) (Table 1). Severe skin reactions were also more frequent in telaprevir-treated patients (3.7% vs 0.4%; 95% CI of RD, 2.0% to 4.3%), as was discontinuation of telaprevir or placebo because of skin complaints (6.4% vs 0.4%; 95% CI of RD, 4.6% to 7.3%). Only 0.8% of telaprevir-treated and 0.4% of placebo-treated patients discontinued all 3 drugs owing to skin eruption (95% CI of RD, −0.6% to 1.1%).
The panel reviewed a total of 221 cases with skin eruptions, including 151 (68%) with photographs, 131 (59%) with a local dermatologist's evaluation, and 84 (38%) with skin biopsy results. Quiz Ref IDIn the largest controlled study,15 an ESI affected the skin in 56 of 727 patients who received telaprevir (7.7%) and 3 of 361 who received placebo (0.8%; 95% CI of RD, 4.6% to 9.2%). Photographs were available for 45 (80%) and biopsy specimens for 36 (64%) of the 56 telaprevir-related cases.
Nearly all photographs showed an eczematous component (39 of 41 [95%]) (Figure 1), which we call telaprevir-related dermatitis. Seven (17%) also exhibited maculopapular features, and 4 (10%), papular lichenoid features. Pruritus was reported in 53 of the 56 telaprevir-treated cases (95%). Xerosis, excoriations, and/or lichenification were commonly observed.
Quiz Ref IDOf 36 biopsy specimens obtained from patients treated with telaprevir-based regimen, 34 (95%) showed a spongiotic dermatitic pattern, which most often included epidermal spongiosis and superficial perivascular predominantly lymphocytic infiltrate (Figure 2). Eosinophils were only occasionally present except for 1 case with eosinophil-rich infiltrate. Of the 34 biopsy specimens, 13 (38%) also showed focal, patchy, low-intensity vacuolar interface dermatitis, characterized by a focal, low-density lymphocytic infiltrate along the dermoepidermal interface. No keratinocyte apoptosis or necrosis was seen. These findings were consistent with the eczematous clinical appearance and support the denomination of telaprevir-related dermatitis.
One case (Table 2) with a clinical diagnosis of Stevens-Johnson Syndrome (SJS)15 exhibited a well-developed interface dermatitis with numerous apoptotic and necrotic keratinocytes (erythema multiforme/SJS pattern).
Telaprevir-related dermatitis occurred at any time after initiating therapy (median, 15 days; interquartile range, 4-41 days). Of the 869 telaprevir plus peginterferon and ribavirin–treated and 182 placebo plus peginterferon and ribavirin–treated patients with an initially limited extent of skin involvement, 69 (7.9%) and 4 (1.9%) cases, respectively, progressed to moderate and 18 (2.1%) and 0 cases, respectively, progressed to severe. Of 178 telaprevir plus peginterferon and ribavirin–treated and 31 placebo plus peginterferon and ribavirin–treated patients with an initially moderate extent of skin involvement, 12 (0.7%) and 1 (0.3%) cases, respectively, progressed to severe. Resolution of dermatitis was slow (median time to resolution, 44 days [range, 1-504 days]), even among patients who discontinued telaprevir treatment.
Most patients were prescribed emollients and/or topical corticosteroids. In the absence of systematic data collection, assessment of their efficacy on telaprevir-related dermatitis was not possible. Use of systemic steroids required study drug discontinuation per protocol.
Using the European Registry of Severe Cutaneous Adverse Reactions (RegiSCAR)23 scoring criteria, we assessed the 221 ESI cases to determine the likelihood that they were severe cutaneous adverse reaction (SCAR) cases. There was 1 definite, 1 probable, and 1 possible case of Stevens-Johnson syndrome (SJS) (Table 2). The definite case occurred 11 weeks after the last dose of telaprevir and was considered to be unlikely related to telaprevir. One probable and 1 possible case of SJS occurred during the course of the telaprevir combination regimen.
There were 1 definite, 2 probable (including 1 published case24), and 8 possible cases of DRESS (Table 3). Of these 11 patients, 7 were hospitalized. Resolution was documented for 10 of the 11 cases. The remaining patient was lost to follow-up. Of the 11 cases of suspected DRESS, 6 had data available on telaprevir pharmacokinetics. Telaprevir plasma concentrations in suspected DRESS cases did not differ substantially from that of the study population (data not shown). No case of angioedema, anaphylaxis, or vasculitis was observed.
In both univariate and multivariate analyses, the incidence of telaprevir-related dermatitis was significantly higher (P ≤ .03) with age above 45 years, body mass index below 30 (calculated as weight in kilograms divided by height in meters squared), white race, and if receiving therapy for HCV for the first time. However, the impact of these factors on the risk were modest (differences in incidence of telaprevir-related dermatitis between subgroups was always below 20% and most often below 10%)
Pharmacokinetics studies found no statistically significant relationship between the occurrence of extensive skin reactions during the telaprevir or placebo phase and plasma concentration of telaprevir (P = .48), peginterferon (P = .54), or ribavirin (P = .56). Table 4 gives telaprevir exposures by the extent of skin eruption.
A total of 187 patients were included in the HLA analyses. They were treatment-naive white subjects treated with a telaprevir-based regimen in a phase 2 or 3 study, who consented to HLA typing and did not experience SCAR. In the primary analysis, 114 patients with skin reaction of any severity were compared with 73 controls without a skin reaction. No association remained significant at the .05 level when P values were corrected for the 143 allele-specific comparisons.
Since HLA associations have been mainly described in severe reactions,25,26 in our secondary analysis, patients with an extensive eruption (n = 59) were compared with those with less-extensive skin eruptions (n = 55) and no eruption (n = 73). Five alleles were significant at the .05 level based on uncorrected P values: 2 alleles (B*4402 [OR, 2.43]; DQB1*0202 [OR, 2.01]) were risk factors for severe skin reactions. After correction for multiple comparisons, none remained significantly associated.
More than 50 single-nucleotide polymorphisms have been identified in the human MDR1 gene, especially at positions C3435T and C1236T, in approximately 40% of the white population.18- 20MDR1 C3435T and C1236T genotypes were determined in patients treated with telaprevir plus peginterferon and ribavirin, who had severe reactions (n = 12), mild or moderate reactions (n = 11), or no reactions (n = 10), and in patients treated only with peginterferon and ribavirin without a skin eruption (n = 11). As given in Table 5, no trend was discerned for the presence of a particular genotype and the severity of skin eruption. Although a homozygous 1236T/3435T haplotype might be a risk factor, most patients were heterozygous (CT) or homozygous (CC) for the 1236 and 3435 alleles in all subgroups independent of the presence, absence, or severity grade of the skin reaction.
Treatment regimens for HCV that include telaprevir in combination with peginterferon and ribavirin frequently result in a pruritic eczematous dermatitis with a spongiotic pattern as the predominant histologic finding, known as telaprevir-related dermatitis. The incidence of dermatitis in telaprevir-treated patients (>50%) is 20% higher than with peginterferon and ribavirin alone. Approximately 5% of patients treated with telaprevir developed extensive eruptions that are infrequent with peginterferon and ribavirin alone. The addition of telaprevir increases the incidence and the severity but does not appear to change the nature of the eruption most commonly observed with peginterferon and ribavirin.
In addition to the frequent occurrence of telaprevir-related dermatitis, cases of severe cutaneous adverse reactions were observed, including 3 suspected cases of SJS and 11 suspected cases of DRESS. Five cases of severe cutaneous adverse reactions (2 SJS and 3 DRESS) were assessed as likely. Given the exposed population (n = 3819), it is likely that the risk of SCAR is increased in patients receiving telaprevir combination therapy. A recent report of 2 possible cases of DRESS observed at a single institution reinforces this concern.27
These observations support the need for education of prescribing physicians, patient monitoring to detect possible SCAR cases early, and prompt treatment withdrawal in suspected cases. The strengths of this evaluation include the prospective collection of data using a standardized questionnaire, the frequent availability of clinical photographs, dermatologic evaluations and skin biopsy specimens, and our being masked to treatment group in the placebo-controlled studies.
The limitations of this study are those inherent to observational studies. It was not possible to evaluate the effect of topical treatments such as emollients or topical corticosteroids that were often prescribed and may have altered the extent of telaprevir-related dermatitis. In addition several laboratory investigations of risk factors (eg, polymorphism of P-glycoprotein coding gene MDR1 and pharmacokinetics measures) were performed on a limited number of patients.
Quiz Ref IDTelaprevir-related dermatitis differs from common drug-induced eruptions in several ways. The incidence (50%-60%) is higher than seen with “drug allergy.”28 Nearly one-fourth of cases began within 4 days, and 46% after 4 weeks—very different timing than that usually observed with most drug-induced eruptions.28 Telaprevir-related dermatitis, which accounts for 95% of skin events in telaprevir-treated patients, is clinically and histologically eczematous and different from the classic “maculopapular” drug-induced eruptions. A recent case report of rapid recurrence of an eczematous eruption with reuse of telaprevir suggests that immunologic mechanisms may play a role, but this proposed mechanism requires further exploration.29 While some clinical factors were shown to be associated with an increased risk of eczematous dermatitis, the associations were not very strong and mainly included demographic characteristics that are unlikely to be useful for excluding patients from a treatment because of a small increase in the risk of dermatitis.
No specific HLA alleles or gene variants of MDR1 were associated with skin reactions of any severity. The 1236T/3435T haplotype was expected to result in higher telaprevir exposures.19,20 However, this was not observed. Exposures were similar in the 1236T/3435T subjects (minimum steady-state plasma concentration during a dosage interval [Cmin,ss] 2680 + 999 ng/mL) (n = 5) compared with subjects with CT or CC 1236/3435 haplotypes (Cmin,ss 2280 + 692 ng/mL) (n = 28). It is important to note that the genetic analyses were performed on a small number of patients, especially for MDR1 polymorphism.
In clinical practice, telaprevir-related dermatitis is frequent but is most often of limited extent. Quiz Ref IDOn the basis of our experience, we suggest that clinicians focus on the following 2 objectives: (1) vigilance for early signs of SCAR that would require immediate discontinuation of all treatment and (2) therapy to alleviate signs and symptoms of telaprevir-related dermatitis to help patients tolerate treatment.
Quiz Ref IDSymptoms and signs of early SJS include skin pain, mucous membrane involvement, vesicles, and a positive Nikolsky sign on dusky macules.30 Early diagnosis of DRESS may be challenging. Late-onset cutaneous lesions, extensive confluent eruption, facial edema, temperature above 38.5°C, or lymph node enlargement should prompt laboratory tests including complete blood cell count to detect eosinophilia and/or activated “atypical” lymphocytes, altered serum liver tests, and serum creatinine measurements to detect raised levels, which help confirm the likelihood of DRESS. If SJS or DRESS is suspected, all drug treatments suspected as possible causes, including telaprevir, should be discontinued.
Unfortunately no controlled study is available on optimal management of telaprevir-related dermatitis. However, the treatments often helpful for contact dermatitis or atopic dermatitis may help to control also the telaprevir-related dermatitis, which shares many clinical and histological characteristics of these conditions.
Correspondence: Robert S. Stern, MD, Department of Dermatology, Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA 02115 (email@example.com).
Accepted for Publication: August 8, 2012.
Published Online: November 19, 2012. doi:10.1001/jamadermatol.2013.938
Authors Contributions: Drs Roujeau, Mockenhaupt, Tahan, Singhal, Kauffman, and Stern had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Roujeau, Mockenhaupt, Singhal, Kauffman, and Stern. Acquisition of data: Harding, Bengtsson, and Singhal. Analysis and interpretation of data: Roujeau, Mockenhaupt, Tahan, Henshaw, Martin, Harding, van Baelen, Singhal, Kauffman, and Stern. Drafting of the manuscript: Henshaw, Harding, Singhal, and Stern. Critical revision of the manuscript for important intellectual content: Roujeau, Mockenhaupt, Tahan, Henshaw, Martin, Harding, van Baelen, Bengtsson, Singhal, Kauffman, and Stern. Statistical analysis: Martin, van Baelen, and Bengtsson. Obtained funding: Singhal and Kauffman. Administrative, technical, and material support: Tahan and Henshaw. Study supervision: Singhal and Stern.
Conflict of Interest Disclosures: Drs Roujeau, Mockenhaupt, Tahan, and Stern served as consultants and received consulting fees from Vertex Pharmaceuticals. Drs Henshaw, Martin, Harding, Singhal, and Kauffman and Mr Bengtsson are employed by Vertex Pharmaceuticals and may hold Vertex Pharmaceuticals stocks and/or options. Dr van Baelen is employed by Janssen Pharma and may hold Janssen Pharma stocks and/or options. Dr Roujean has served on expert panels on severe cutaneous adverse reaction (SCAR) cases for AB Science, Menarini, and US Lawyers; has served on safety boards on SCAR for OM Pharma and Servier; and has received unrestricted research grants (RegiSCAR) from Novartis, GlaxoSmithKline, Boehringer-Ingelheim, OM Pharma, Astellas, and Servier. Dr Mockenhaupt is the coordinator of the international RegiSCAR-project, which is funded by grants from the European Commission (QLRT-2002-01738) and GIS-Institut des Maladies Rares and INSERM (4CH09G) in France and by a consortium of pharmaceutical companies (Bayer Vital, Boehringer-Ingelheim, Cephalon, GlaxoSmithKline, MSD Sharp and Dohme, Merck, Novartis, Pfizer, Roche, sanofi-aventis, Servier, and Tibotec); has received the Else Kröner Memorial Stipendium for support of clinical research through the Else Kröner-Fresenius Foundation; has been an expert in litigations concerning Stevens-Johnson syndrome and toxic epidermal necrolysis; and has served in expert panels and advisory boards coordinated by pharmaceutical companies (2009-2011: dermatology safety board Merck & Co). Currently, RegiSCAR Germany receives funding by the Ministry for Education and Research (Bundesministerium für Bildung und Forschung; grant 01KG1018). Dr Tahan served as a consultant for Tibotec-Jansen Pharmaceuticals. Dr Stern has received consulting fees for the assessment of skin reactions associated with drugs under development from InterMune, Johnson & Johnson, Boehringer Ingelheim, and Takeda and its Millennium Division; has served on a drug-safety panel for Takeda; and has served as an expert witness in product liability litigation relating to skin reactions for Johnson & Johnson and Mutual.
Funding/Support: This study was funded by Vertex Pharmaceuticals and Janssen Pharma.
Role of the Sponsors: All authors, including employees of the sponsors, contributed to the design and conduct of the study; the collection, management, analysis, and interpretation of the data; and the preparation, review, or approval of the manuscript.
Additional Contributions: Cherif Benattia, MD (an employee of Vertex Pharmaceuticals at the time this study was initiated), appointed the expert panel and provided thoughtful discussion; Frank Tomaka, MD (from Janssen Pharma), Ravi Ramachandran, PhD (from Vertex Pharmaceuticals), and Yuko Sakurai, MSc, and Ichimaro Yamada, PhD (both from Mitsubishi Tanabe Pharma Corporation), assisted with data collection; Mohammad Bsharat, PhD (from Vertex Pharmaceuticals), assisted with statistical analysis; Laura Bower, MD (from Vertex Pharmaceuticals), assisted with data compilation; and Valérie Philippon, PhD, and Elizabeth Dorn, PhD (from Vertex Pharmaceuticals), assisted with medical writing and provided editorial and coordination support. We would like to thank all patients, studies coordinators, nurses, and investigators of the telaprevir clinical trials program.