Vitiligo under Anti-Programmed Cell Death-1 Therapy is Associated with Increased Survival in Melanoma Patients.

To the Editor: Anti–programmed cell death-1 (anti-PD-1) agents have demonstrated their efficacy in the treatment of advanced melanoma with reasonable toxicities.1Weber J.S. D'Angelo S.P. Minor D. et al.Nivolumab versus chemotherapy in patients with advanced melanoma who progressed after anti-CTLA-4 treatment (CheckMate 037): a randomised, controlled, open-label, phase 3 trial.Lancet Oncol. 2015; 16: 375-384Abstract Full Text Full Text PDF PubMed Scopus (2012) Google Scholar Although anti-PD-1–related adverse events (AEs), especially vitiligo, have been reported as being associated with higher response rates and prolonged survival, this association remains controversial because of conflicting results.2Postow M.A. Sidlow R. Hellmann M.D. Immune-related adverse events associated with immune checkpoint blockade.N Engl J Med. 2018 11; 378: 158Crossref PubMed Scopus (2135) Google Scholar Consequently, the impact of AEs under anti-PD-1 agents on survival in patients treated for advanced melanoma in the setting of daily practice was investigated. All patients treated with anti-PD-1 agents for stages III to IV melanoma between 2011 and 2018 were retrospectively reviewed. Type and severity of AEs according to the Common Terminology Criteria for Adverse Events version 5.0 and interval from anti-PD-1 agent initiation to AEs occurrence were collected. Using Cox regression models, the prognostic impact of clinical factors including AEs on progression-free survival and overall survival (OS) was investigated. In addition, a landmark analysis at 3 time points to address lead-time bias inherent to the time dependence of the occurrence of AEs was performed. Overall, 111 patients treated with anti-PD-1 agents were evaluated. Patient characteristics are listed in Table I. Over a median follow-up of 438 days (range, 8-1687), 61 patients experienced AEs (55%) after a median time of 51 days (range, 1-758). Cutaneous AEs (n = 32, 29%), including vitiligo (n = 15, 13.5%), endocrine (n = 35, 32%), and gastrointestinal (n = 17, 15%), were the most frequent. Specifically, vitiligo occurred at a median time of 256 days (range, 30-758). Severe AEs occurred in 12 patients (11%), leading to therapy discontinuation in 8 patients (7%).Table ICharacteristics of patients with or without AEsTotal patients (N = 111, 100%)Patients with AEs (n = 61, 55%)Patients without AEs (n = 50, 45%)UnivariablePMedian age, y [range]67 [25-89]66 [32-89]68 [25-88].585Sex ratio Male58 (52)3325.667 Female53 (48)2825Median body mass index, kg/m2 [range]25.5 [15.7-45.4]27.0 [15.7-45.5]25.4 [18.3-39.5].229History of autoimmune disease Yes9 (8)54.970 No102 (92)5646Use of steroids Yes15 (14)105.327 No96 (86)5145American Joint Committee on Cancer stage III30 (27)1416.285 IV81 (73)4734Brain metastases Yes35 (32)1718.359 No76 (68)4432Median lactate dehydrogenase level, UI/L [range]250 [139-2367]242 [139-2367]250 [156-1876].092BRAFV600 mutation Yes37 (21)1819.345 No74 (79)4331Targeted therapies∗Patients who previously received a targeted therapy were either in treatment failure or experienced toxicity contraindicating targeted therapies. Yes30 (17)1515.523 No81 (73)4635Ipilimumab monotherapy†Patients who previously received ipilimumab. Yes44 (24)2618.478 No67 (76)3532Anti–programmed cell death-1 monotherapy Pembrolizumab monotherapy854738.897 Nivolumab monotherapy261412Values are n (%) unless otherwise defined.AEs, Adverse events.∗ Patients who previously received a targeted therapy were either in treatment failure or experienced toxicity contraindicating targeted therapies.† Patients who previously received ipilimumab. Open table in a new tab Values are n (%) unless otherwise defined. AEs, Adverse events. Median OS was 22.5 months (95% confidence interval [CI], 14.9-30.1). Rates of median OS in patients experiencing vitiligo, other AEs, and no AEs were not reached, 28 months (95% CI, 11.1-46.9) and 12 months (95% CI, 4.7-20.3; P < .001), respectively. On multivariable analysis, experiencing vitiligo (hazard ratio [HR], 0.099; 95% CI, 0.013-0.737; P = .024) and serum lactate dehydrogenase level (HR, 3.514; 95% CI, 1.787-6.962; P < .001) were independently associated with OS. Median progression-free survival was 8.9 months (95% CI, 9.2-11.6). On multivariable analysis, experiencing vitiligo (HR, 0.109; 95% CI, 0.025-0.471; P = .003) and serum lactate dehydrogenase level (HR, 3.077; 95% CI, 1.702-5.561; P < .001) were independently associated with progression-free survival. Using a landmark analysis we confirmed that vitiligo was associated with prolonged OS as compared with patients experiencing other AEs or no AEs (Fig 1). In daily practice AEs under anti-PD-1 therapy are frequent, with a predominance of endocrine, cutaneous, and gastrointestinal AEs as previously described.1Weber J.S. D'Angelo S.P. Minor D. et al.Nivolumab versus chemotherapy in patients with advanced melanoma who progressed after anti-CTLA-4 treatment (CheckMate 037): a randomised, controlled, open-label, phase 3 trial.Lancet Oncol. 2015; 16: 375-384Abstract Full Text Full Text PDF PubMed Scopus (2012) Google Scholar Although vitiligo occurrence varies among series, it has been correlated with disease response,1Weber J.S. D'Angelo S.P. Minor D. et al.Nivolumab versus chemotherapy in patients with advanced melanoma who progressed after anti-CTLA-4 treatment (CheckMate 037): a randomised, controlled, open-label, phase 3 trial.Lancet Oncol. 2015; 16: 375-384Abstract Full Text Full Text PDF PubMed Scopus (2012) Google Scholar,3Nardin C. Pelletier F. Puzenat E. Aubin F. Vitiligo repigmentation with melanoma progression during pembrolizumab treatment.Acta Derm Venereol. 2019; 99: 913-914Crossref PubMed Scopus (7) Google Scholar,4Hua C. Boussemart L. Mateus C. et al.Association of vitiligo with tumor response in patients with metastatic melanoma treated with pembrolizumab.JAMA Dermatol. 2016; 152: 45-51Crossref PubMed Scopus (455) Google Scholar yet its association with survival remains disputed.4Hua C. Boussemart L. Mateus C. et al.Association of vitiligo with tumor response in patients with metastatic melanoma treated with pembrolizumab.JAMA Dermatol. 2016; 152: 45-51Crossref PubMed Scopus (455) Google Scholar,5Teulings H.-E. Limpens J. Jansen S.N. et al.Vitiligo-like depigmentation in patients with stage III-IV melanoma receiving immunotherapy and its association with survival: a systematic review and meta-analysis.J Clin Oncol. 2015; 33: 773-781Crossref PubMed Scopus (405) Google Scholar In the current study patients who developed vitiligo under anti-PD-1 therapy experienced longer survival. However, analyzing the correlation between AEs such as vitiligo should take into account lead-time bias because patients dying of disease, especially early after anti-PD-1 therapy initiation, do not develop AEs. This issue has been addressed using a landmark analysis. Further, the occurrence of any AEs other than vitiligo was not associated with any survival benefit as compared with patients experiencing no AEs. This observation highlighted the relevance of monitoring vitiligo under anti-PD-1 therapy. Although inherently limited by its retrospective nature and single-center design, the current study showed that the occurrence of vitiligo under anti-PD-1 therapy seems to be relevant as a surrogate of treatment efficacy and prolonged survival in daily practice. We thank Alexandre Doussot, Elisabeth Homassel and the Association A Fleur de Peau (Besançon, France) for their technical assistance.