Comparison of Skin Toxic Effects Associated with Gefitinib, Erlotinib Hydrochloride, Or Afatinib Treatment

Letters

4. Frühauf J, Kröck S, Quehenberger F, et al. Mobile teledermatology helping patients with treatment courses extending at least 180 con- patients control high-need acne: a randomized controlled trial. J Eur Acad secutive days (Figure, A), we found that afatinib therapy Dermatol Venereol. 2015;29(5):919-924. resulted in an earlier onset of paronychia (Table) and more 5. UPMC eDermatology Web site. https://edermatology.upmc.com. Accessed dermatologic visits for any skin toxic effects within the first September 8, 2015. 180 days. However, this finding was attenuated in patients 6. DermatologistOnCall Web site. http://www.dermatologistoncall.com. Accessed September 8, 2015. who developed skin toxic effects (Figure, F).

Discussion | All of our patients who developed skin toxic ef- Comparison of Skin Toxic Effects Associated fects were diagnosed and managed by specialized dermatolo- With Gefitinib, Erlotinib, or Afatinib Treatment gists (K.-L.C., Y.-T.C., C.-W.Y., and C.-Y.C.) within integrated on- for Non–Small Cell Lung Cancer cology clinics based on the recommendations of experts.1 Epidermal growth factor receptor tyrosine kinase inhibitors Therefore, their diagnoses are more accurate than those in clini- (EGFR-TKIs) have been widely used to treat non–small cell cal trials. The earlier onset and higher incidence of paro- lung cancer. Four major skin toxic effects with different nychia in patients treated with afatinib, which remained sig- incidences have been reported from clinical studies, includ- nificant when considering individual variation or accumulation ing acneiform eruption (60%-94%), pruritus (16%-60%), effect, may be explained by 2 reasons. First, afatinib, which is xerosis (4%-38%), and paronychia (6%-12%).1,2 However, a an irreversible EGFR-TKI exhibiting strong affinity to wild- direct comparison of the incidences and severities of the 4 type EGFR,3 may result in greater skin inflammation. Sec- types of skin toxic effects for 3 different EGFR-TKIs in the ond, synergistic effects could occur with dual inhibition of same patient cohort has been lacking to date. EGFR and ERBB2 (formerly HER2 or HER2/neu) by afatinib,4 with 27.5% of patients receiving adjuvant trastuzumab (an Methods | This retrospective study was approved by the re- ERBB2 monoclonal antibody) manifesting nail toxicity.5 search ethics committee of National Taiwan University Hos- The increased dermatologic visits during the first 6 pital. We recruited patients within a named patient program months of treatment with afatinib seemed to be related to for compassionate use before registration who had ever re- higher incidences of skin toxic effects. The frequencies of ceived afatinib treatment for non–small cell lung cancer be- dermatologic visits decreased after the first 6 months, dem- tween November 1, 2007, and April 30, 2013. Most of the pa- onstrating that skin toxic effects can be managed effectively tients had received gefitinib or erlotinib hydrochloride similarly, regardless of the causative agent. Therefore, treatment before commencing afatinib therapy, although some aggressive dermatologic care for patients receiving EGFR- patients had been prescribed afatinib as their first EGFR-TKI. TKIs should be mandatory. The dates of our study analysis were November 1, 2013, to De- cember 30, 2014. The inclusion criteria for the present analy- Kai-Lung Chen, MD sis were (1) EGFR-TKI exposure for at least 30 consecutive days, Chia-Chi Lin, MD, PhD (2) a minimum 30-day washout interval between each drug ex- Yung-Tsu Cho, MD posure if 2 or more EGFR-TKIs were used, and (3) clinical fol- Che-Wen Yang, MD low-up for at least 6 months. Any skin toxic effects occurring Yi-Shuan Sheen, MD during each of the EGFR-TKI exposure periods that fulfilled the Hsiao-En Tsai, MD inclusion criteria were reviewed retrospectively. A χ2 test was Chia-Yu Chu, MD, PhD used to compare the incidences of skin toxic effects for the dif- Author Affiliations: Department of Dermatology, National Taiwan University ferent EGFR-TKIs. An unpaired 2-tailed t test was used to com- Hospital, National Taiwan University College of Medicine, Taipei (Chen, Cho, pare the number of dermatologic visits during the first 180 days Yang, Sheen, Chu); Department of Oncology, National Taiwan University of each drug exposure. P ≤ .05 was considered statistically sig- Hospital, Taipei (Lin); Department of Urology, National Taiwan University nificant. College of Medicine, Taipei (Lin); Department of Surgery, National Taiwan University Hospital Hsin-Chu Branch, Hsin-Chu City (Tsai).

Results | Among 146 patients fulfilling the inclusion criteria, Accepted for Publication: September 22, 2015. 61 patients, 117 patients, and 93 patients had ever received Corresponding Author: Chia-Yu Chu, MD, PhD, Department of Dermatology, National Taiwan University Hospital, National Taiwan University College of gefitinib, erlotinib, and afatinib, respectively (Figure, A and Medicine, 7 Chung-Shan S Rd, Taipei, Taiwan 10002 ([email protected]). B). In general, the incidence of acneiform eruption was the Published Online: December 9, 2015. doi:10.1001/jamadermatol.2015.4448. highest (67.2% [41 of 61] to 76.3% [71 of 93]), followed by Author Contributions: Drs Chen and Chu had full access to all the data in the pruritus and xerosis (47.5% [29 of 61] to 63.4% [59 of 93]). study and take responsibility for the integrity of the data and the accuracy of The incidence of paronychia was the lowest but differed sig- the data analysis. nificantly among the 3 EGFR-TKIs (9.8% [6 of 61] for gefi- Study concept and design: All authors. Acquisition, analysis, and interpretation of data: All authors. tinib, 12.8% [15 of 117] for erlotinib, and 39.8% [37 of 93] for Drafting of the manuscript: Chen, Cho, Yang, Sheen, Tsai, Chu. afatinib; P < .001) (Figure, C). Similar findings remained sig- Critical revision of the manuscript for important intellectual content: Chen, Cho, nificant among 21 patients receiving sequential gefitinib- Yang, Sheen, Chu. Statistical analysis: Chen, Cho, Yang, Chu. erlotinib-afatinib treatment courses (Figure, A and D) and Obtained funding: Tsai, Chu. among patients whose first EGFR-TKI treatment course ful- Administrative, technical, or material support: Chen, Lin, Cho, Yang, filled the inclusion criteria (Figure, A and E). Among Sheen, Chu.

340 JAMA Dermatology March 2016 Volume 152, Number 3 (Reprinted) jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Letters

Figure. Study Design and Comparison of Skin Toxic Effects Associated With Gefitinib, Erlotinib Hydrochloride, or Afatinib Treatment

A Flowchart of study enrollment B No. of patients receiving each drug

225 Patients Erlotinib 1 Exposure ≥30 d (n = 117) Inclusion criteria 2 Washout ≥30 d 3 Follow-up ≥6 mo 146 Patients 25

G E A GE GA EA GEA 21 50 (n = 7) (n = 25) (n = 10) (n = 21) (n = 12) (n = 50) (n = 21) 21 (i) Pooling (ii) Same patient (iv) Exposure ≥180 d 40 61 Gefitinib (G + GE + GA + GEA) 33 Gefitinib and afatinib (GA + GEA) Gefitinib 10 66 7 117 Erlotinib (E + GE + EA + GEA) 71 Erlotinib and afatinib (EA + GEA) Erlotinib 12 49 93 Afatinib (A + GA + EA + GEA) 21 Gefitinib, erlotinib, and afatinib Afatinib Gefitinib Afatinib (n = 61) (n = 93) (iii) First EGFR-TKI treatment 53 Gefitinib 73 Erlotinib 15 Afatinib

C Among patients receiving the 3 EGFR-TKIs D Among 21 patients receiving gefitinib-erlotinib-afatinib sequential treatment

Gefitinib (n = 61) Erlotinib (n = 117) Afatinib (n = 93) Gefitinib Erlotinib Afatinib

80 80

60 60 a

a a

40 40 a

20 20 Incidence of Skin Toxicities, % Incidence of Skin Toxicities, % Incidence of Skin Toxicities,

0 0 Acneiform Paronychia Pruritus Xerosis Acneiform Paronychia Pruritus Xerosis Eruption Eruption Skin Toxicity Skin Toxicity

E Among patients receiving first-line EGFR-TKIs F Comparison of total dermatologic visit

Gefitinib (n = 53) Erlotinib (n = 73) Afatinib (n = 15) All patients Patients with skin toxicities

80 8

60 6 a

a 40 4

20 2 No. of Dermatologic Visits No. Incidence of Skin Toxicities, % Incidence of Skin Toxicities,

0 0 Acneiform Paronychia Pruritus Xerosis GefitinibErlotinib Afatinib Eruption Skin Toxicity Drug Exposure

A and B, Study design and patient enrollment. C-E, Comparison of the incidence receptor tyrosine kinase inhibitors; G, gefitinib; GA, gefitinib-afatinib; of 4 major skin toxic effects. F, Comparison of the total dermatologic visits for GE, gefitinib-erlotinib, GEA, gefitinib-erlotinib-afatinib. any kind of skin toxicity within the first 180 days of drug exposure. A indicates a P <.05. afatinib; E, erlotinib; EA, erlotinib-afatinib; EGFR-TKIs, epidermal growth factor

Conflict of Interest Disclosures: Dr Chu reported receiving consulting fees, International GmbH and reported receiving honoraria from AstraZeneca and travel support, and payment for lectures from Boehringer Ingelheim Roche. No other disclosures were reported.

jamadermatology.com (Reprinted) JAMA Dermatology March 2016 Volume 152, Number 3 341

Downloaded From: https://jamanetwork.com/ on 09/29/2021 Letters

Table. Comparison of the Incidence and Time to Onset of Skin Toxic Effects for EGFR-TKIs Within the First 180 Days of Drug Exposurea

Gefitinib Erlotinib Hydrochloride Afatinib (n = 40) (n = 66) (n = 49) Time to Onset, Time to Onset, Time to Onset, Skin Toxicity Incidence, % Median (Range), d Incidence, % Median (Range), d Incidence, % Median (Range), d Acneiform eruption 42.5 (17 of 40) 37 (13-153) 54.5 (36 of 66) 22 (1-164) 77.6 (38 of 49) 42 (7-168) Pruritus 52.5 (21 of 40) 31 (7-145) 54.5 (36 of 66) 22 (2-167) 63.3 (31 of 49) 55 (7-133) Xerosis 42.5 (17 of 40) 43 (13-194) 53.0 (35 of 66) 36 (11-168) 69.4 (34 of 49) 49 (7-168) Paronychia 7.5 (3 of 40) 101 (70-157) 7.6 (5 of 66) 78 (63-136) 46.9 (23 of 49) 56 (28-168) Abbreviation: EGFR-TKIs, epidermal growth factor receptor tyrosine kinase treatment courses exceeding 180 days, 31 patients with 2 different EGFR-TKI inhibitors. treatment courses exceeding 180 days, and 84 patients with 1 EGFR-TKI a Treatment courses that fulfilled the inclusion criteria and lasted longer than treatment course exceeding 180 days. In total, 119 patients with 155 treatment 180 days were selected. There were 3 patients with 3 different EGFR-TKI courses were analyzed.

Funding/Support: This work was supported by grant NTUH103-S2366 from formation (Figure, B). These changes were accompanied by in- National Taiwan University Hospital and by grant HCH103-054 from National creased erythrocyte sedimentation rate (47 mm/h), elevated Taiwan University Hospital Hsin-Chu Branch. C-reactive protein levels (25.8 mg/L), leukocytosis (white blood Role of the Funder/Sponsor: The funding sources had no role in the design and cell count, 15 390/uL), eosinophilia (eosinophils, 11%), and high conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to IgE level (7200 μg/L). (To convert C-reactive protein to nano- submit the manuscript for publication. moles per liter, multiply by 9.524; white blood cells to ×109/L, 1. Lacouture ME, Schadendorf D, Chu CY, et al. Dermatologic adverse events multiply by 0.001; and IgE to milligrams per liter, multiply by associated with afatinib: an oral ErbB family blocker. Expert Rev Anticancer Ther. 0.001.) Therapy with terbinafine was discontinued for 3 days 2013;13(6):721-728. and oral prednisone was administered (10 mg/d). Terbinafine 2. Agero AL, Dusza SW, Benvenuto-Andrade C, Busam KJ, Myskowski P, therapy was reintroduced at the same dose (125 mg/d) for Halpern AC. Dermatologic side effects associated with the epidermal growth factor receptor inhibitors. J Am Acad Dermatol. 2006;55(4):657-670. 6 weeks resulting in a complete resolution of skin lesions. 3. Fox LP. Nail toxicity associated with epidermal growth factor receptor inhibitor therapy. J Am Acad Dermatol. 2007;56(3):460-465. Discussion | Among the very few reports of paradoxical or in- 4. Liao BC, Lin CC, Yang JC. Second and third–generation epidermal growth flammatory drug-related reactions during treatment of fun- factor receptor tyrosine kinase inhibitors in advanced nonsmall cell lung cancer. gal infections are those described for Cryptococcus neofor- Curr Opin Oncol. 2015;27(2):94-101. mans meningitis2 or disseminated paracoccidioidomycosis.3 5. Adachi S, Yoshimura T, Matsuoka T, Okada K, Yasuda T, Kamei K. Appearance Only 1 report to our knowledge has been published on para- of skin and nail toxicity in patients with breast cancer who underwent trastuzumab-containing chemotherapy [in Japanese]. Gan To Kagaku Ryoho. doxical reactions during treatment of dermatophytosis. Nikkels 2011;38(9):1453-1456. et al4 described inflammatory flare-up reactions in 5 patients with dermatophyte infection undergoing antifungal therapy. OBSERVATION In the present case, we observed high fever, chills, and malaise within 12 hours of the first dose of antifungal treatment. Paradoxical Reaction During a Course of Terbinafine Elevated temperature was accompanied by intensified inflam- Treatment of Trichophyton interdigitale Infection mation and pustule formation within the area of primary le- in a Child sions, leukocytosis, eosinophilia, increased C-reactive pro- We describe for the first time, to our knowledge, a paradoxi- tein levels, and elevated total IgE levels. The reaction symptoms cal reaction to terbinafine treatment in the context of derma- prompted us to categorize it as the Jarisch-Herxheimer reac- tophyte infection due to Trichophyton interdigitale. tion (JHR), a well-recognized paradoxical complication of the treatment of syphilis and other infectious diseases, including Report of a Case | A child weighing 30 kg and without any leptospirosis, anthrax, meningococcal meningitis, Pneumo- comorbidities and allergies and not taking any long-term cystis carinii pneumonia, and African trypanosomiasis.5 medications was seen for erythematous, infiltrating, well- The present case, while sharing some features with the demarcated skin lesions located on the face, chest, abdomen, flare-up reaction described by Nikkels et al,4 is closer to JHR and limbs (Figure, A). Mycological culture yielded growth of than the earlier report, and this is supported by 2 facts. First, Trichophyton mentagrophytes. The fungus was identified as Nikkels et al observed that the inflammatory exacerbation T interdigitale by sequence analysis of the internal tran- started 12 to 24 hours after drug intake, while a typical JHR oc- scribed spacer and the D1/D2 domains of the large-subunit (26S) curs at half this time. Second, while JHR is always character- rRNA gene within the rDNA cluster using the polymerase chain ized by systemic signs (fever up to 39°C, malaise, and chills), reaction sequencing described elsewhere.1 The patient these were present only in 2 patients with flare-up reactions reported frequent contact with stray cats. (both developed the symptoms 24 hours after drug Initial treatment included 125 mg of terbinafine once daily. administration).4 Twelve hours after the first dose, we observed temperature el- Although eosinophilia and elevated total IgE level, as evi- evation (39°C), chills, malaise, aggravation of inflammatory denced in our case, may suggest an allergy to terbinafine, suc- symptoms including increased erythema, edema, and pustule cessful reintroduction of the drug would negate this. The

342 JAMA Dermatology March 2016 Volume 152, Number 3 (Reprinted) jamadermatology.com

Downloaded From: https://jamanetwork.com/ on 09/29/2021