Dermatology evaluation for cutaneous immune related adverse events is associated with improved survival in cancer patients treated with checkpoint inhibition.

To the Editor: Cutaneous immune-related adverse events (cirAEs), an adverse effect from immune checkpoint inhibitors (ICIs), can be associated with significant morbidity, at times necessitating termination of their ICI regimen,1Darnell E.P. Mooradian M.J. Baruch E.N. Yilmaz M. Reynolds K.L. Immune-related adverse events (irAEs): diagnosis, management, and clinical pearls.Curr Oncol Rep. 2020; 22: 39https://doi.org/10.1007/s11912-020-0897-9Crossref PubMed Scopus (117) Google Scholar despite improving overall survival (OS) in many patients.2Reynolds K.L. Arora S. Elayavilli R.K. et al.Immune-related adverse events associated with immune checkpoint inhibitors: a call to action for collecting and sharing clinical trial and real-world data.J Immunother Cancer. 2021; 9e002896https://doi.org/10.1136/jitc-2021-002896Crossref Scopus (17) Google Scholar,3Cho Y.T. Lin Y.T. Yang C.W. Chu C.Y. Cutaneous immune-related adverse events among Taiwanese cancer patients receiving immune checkpoint inhibitors link to a survival benefit.Sci Rep. 2022; 12: 7021https://doi.org/10.1038/s41598-022-11128-5Crossref PubMed Scopus (5) Google Scholar Dermatologists play an increasingly critical role in the management of cirAEs;4Nadelmann E.R. Yeh J.E. Chen S.T. Management of cutaneous immune-related adverse events in patients with cancer treated with immune checkpoint inhibitors: a systematic review.JAMA Oncol. 2022; 8: 130-138https://doi.org/10.1001/jamaoncol.2021.4318Crossref PubMed Scopus (12) Google Scholar however, there is a paucity of literature evaluating whether dermatology evaluation may improve clinical outcomes. Our study aims to evaluate the role of dermatology evaluation in the management of cirAEs in patients receiving ICIs with regard to multiple treatment outcomes, in particular ICI retrial and survival. In this retrospective cohort study at Massachusetts General Hospital between January 1, 2016 and June 29, 2021, we used billing data to identify cancer patients who initiated antiprogrammed death-1, antiprogrammed death-ligand 1, and/or anticytotoxic T-lymphocyte antigen-4 therapy and identified patients' clinic visits appended with an International Classification of Diseases-10 code consistent with a possible cirAE (n = 2902). CirAE status was confirmed through review of the medical record (n = 628). Demographic and clinical information was collected among these patients. Associations between dermatology evaluation and rates of ICI retrial following prior ICI discontinuation and survival outcomes (progression-free survival [PFS] and OS) were evaluated. Additional methods are outlined in the supplement. Among the 628 patients (median age, 67; 40.9% female) who developed a cirAE, most had high-stage malignancy (median stage, 4 [IQR, 3-4]). The most prevalent cancer was melanoma (40.9%). CirAEs occurred at a median of 43 days (IQR: 18.5-128.5) after ICI initiation, with a median duration of 71 days (IQR: 27-238). The median Common Terminology Criteria for Adverse Events Version 5.0 severity grade was 1 (IQR: 1-2). CirAE management was evaluated by a dermatologist in 229 patients (36.5%). CirAEs were categorized by their effect on the current ICI regimen including ICI discontinuation due to cirAE without retrial (5.9%) and with retrial (10.4%) (Table I).Table IDemographic and oncologic features of cirAE patients evaluated by dermatologyCharacteristicOverallNo dermatology evaluationN = 399Dermatology evaluationN = 229P value∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment.Age, years—median (IQR)67 (57-75)65 (56-74)69 (59-78).002∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment.Female sex, no. (%)257 (40.9)168 (42.1)89 (38.9).611Cancer type, no. (%).228 Melanoma230 (36.6)155 (38.9)75 (32.8) NSCLC106 (16.9)69 (17.3)37 (16.2) Head or neck53 (8.4)31 (7.8)22 (9.6) Urothelial31 (5.0)17 (4.3)14 (6.1) RCC28 (4.5)18 (4.5)10 (4.4) Breast22 (3.5)10 (2.5)12 (5.2) Other†Other cancer types included genitourinary (renal, uroethelial), gynecologic (cervical, ovarian), breast, hematologic (Hodgkin lymphoma, non-Hodgkin lymphoma), cutaneous (non-melanoma skin cancers, Merkel cell carcinoma), endocrine (neuroendocrine, adrenocortical, thyroid), musculoskeletal/soft tissue (sarcoma), and neurologic (meningioma, glioblastoma) cancers.158 (25.2)69 (17.3)59 (25.8) Cancer stage—median (IQR)‡Cancer stage information was not available for 8 patients.4 (3-4)4 (3-4)4 (3-4).785 Brain metastases, no. (%)311 (49.5)188 (47.1)123 (53.7).274ECOG score, no. (%)§The ECOG performance status score was not available at baseline for 10 patients..750 0278 (45.0)179 (45.4)99 (44.2) 1-2326 (52.8)205 (52.0)121 (54.0) 3+14 (2.3)10 (2.5)4 (1.80)Pre-ICI treatments, no. (%)‖Reflects the median number of pre-ICI treatments when pre-ICI antineoplastic therapies considered by class (ie, radiation, traditional chemotherapy, targeted agent, other immunotherapy). Radiation therapy166 (26.4)105 (26.0)61 (26.3).0765 Traditional chemotherapy162 (25.8)95 (23.8)67 (29.3).233 Targeted therapy52 (8.3)32 (8.0)20 (8.7).374ICI regimen<.035∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment. Anti-PD-1451 (71.8)278 (69.7)173 (75.6) Anti-PD-L154 (8.6)33 (8.3)21 (9.2) Anti-CTLA-416 (2.6)14 (3.5)2 (0.9) Anti-PD-1/PDL-1 +CTLA-4107 (17.0)74 (18.6)33 (14.4)cirAE features Time to cirAE, days—median (IQR)43 (18.5-128.5)41 (17-104)63 (21-181)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment. ICI cycles prior to cirAE—median (IQR)2 (1-5)2 (1-4)3 (1-7)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment. Duration of cirAE, days—median (IQR)71 (27-238)54 (21-178)147 (46-333)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment. CTCAE v5.0 severity grade—median (IQR)1 (1-2)1 (1-1)2 (1-2)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment. Biopsy taken no. (%)79 (12.6)2 (0.3)78 (34.1)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment.ICI changes due to cirAE, no. (%) ICI discontinued without retrial due to cirAE37 (5.9%)6 (1.5%)31 (13.5%)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment. ICI discontinuation with retrial due to cirAE65 (10.4)10 (2.5)55 (24.0)<.001∗P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment.cirAE, Cutaneous immune-related adverse event; CTCAE v5.0, Common Terminology Criteria for Adverse Events Version 5.0; CTLA-4, cytotoxic T-lymphocyte antigen-4; ECOG PS, Eastern Cooperative Oncology Group performance scale; ICI, immune checkpoint inhibitor; IV, intravenous; no., number; PD-1, programmed death-1; PD-L1, programmed death-ligand 1.∗ P values reflect Fischer exact or Mann-Whitney tests for categorical and continuous variables, as appropriate. P values with an asterisk remained statistically significant with the 2-sided P value beneath the critical value during false discovery rate assessment.† Other cancer types included genitourinary (renal, uroethelial), gynecologic (cervical, ovarian), breast, hematologic (Hodgkin lymphoma, non-Hodgkin lymphoma), cutaneous (non-melanoma skin cancers, Merkel cell carcinoma), endocrine (neuroendocrine, adrenocortical, thyroid), musculoskeletal/soft tissue (sarcoma), and neurologic (meningioma, glioblastoma) cancers.‡ Cancer stage information was not available for 8 patients.§ The ECOG performance status score was not available at baseline for 10 patients.‖ Reflects the median number of pre-ICI treatments when pre-ICI antineoplastic therapies considered by class (ie, radiation, traditional chemotherapy, targeted agent, other immunotherapy). Open table in a new tab cirAE, Cutaneous immune-related adverse event; CTCAE v5.0, Common Terminology Criteria for Adverse Events Version 5.0; CTLA-4, cytotoxic T-lymphocyte antigen-4; ECOG PS, Eastern Cooperative Oncology Group performance scale; ICI, immune checkpoint inhibitor; IV, intravenous; no., number; PD-1, programmed death-1; PD-L1, programmed death-ligand 1. Impact of dermatology evaluation was assessed by evaluating the odds of ICI retrial following interruption due to cirAE (odds ratio: 10.52; 95% CI: 5.15-21.48; P < .001). Patients who received evaluation by dermatology for their first episode of cirAE had improved PFS (hazard ratio: 0.69; 95% CI [0.54-0.87]; P < .002) and OS (hazard ratio: 0.62; 95% CI [0.45-0.84]; P < .002) (Table II).Table IISurvival outcomes among cases who were evaluated by dermatologyAssociation with dermatology evaluationOR/HR∗Treatment outcomes were assessed using logistic regression models and are reported as odds ratios. Survival outcomes were assessed using Cox proportional hazards models and are reported as hazard ratios.P value95% CIICI retrial†The logistic regression model was adjusted for age, sex, cancer type, cirAE morphology∗, and cirAE severity∗ as defined by the CTCAE severity score. Covariates with an asterisk were statistically significant to P < .05.10.52<.0015.15-21.48PFS‡The Cox proportional hazards model examining associations between dermatology referral and progression-free survival/overall survival was adjusted for age, sex, cancer type∗, cancer stage∗, precipitating ICI regimen∗, cirAE severity∗, ICI retrial∗, and dermatology evaluation. Covariates with an asterisk were statistically significant to P < .05.0.69.0020.54-0.87OS‡The Cox proportional hazards model examining associations between dermatology referral and progression-free survival/overall survival was adjusted for age, sex, cancer type∗, cancer stage∗, precipitating ICI regimen∗, cirAE severity∗, ICI retrial∗, and dermatology evaluation. Covariates with an asterisk were statistically significant to P < .05.0.62.0020.45-0.84HR, Hazard ratio; ICI, immune checkpoint inhibitor; OS, overall survival; PFS, progression-free survival.∗ Treatment outcomes were assessed using logistic regression models and are reported as odds ratios. Survival outcomes were assessed using Cox proportional hazards models and are reported as hazard ratios.† The logistic regression model was adjusted for age, sex, cancer type, cirAE morphology∗, and cirAE severity∗ as defined by the CTCAE severity score. Covariates with an asterisk were statistically significant to P < .05.‡ The Cox proportional hazards model examining associations between dermatology referral and progression-free survival/overall survival was adjusted for age, sex, cancer type∗, cancer stage∗, precipitating ICI regimen∗, cirAE severity∗, ICI retrial∗, and dermatology evaluation. Covariates with an asterisk were statistically significant to P < .05. Open table in a new tab HR, Hazard ratio; ICI, immune checkpoint inhibitor; OS, overall survival; PFS, progression-free survival. We evaluated the association between dermatology evaluation in the management of cirAE, ICI use patterns, and survival outcomes among 628 cirAE patients. The patients with cirAEs who received dermatology evaluation were more likely to have their ICI regimen retrialed following interruption and had improved PFS and OS, when controlling for cirAE severity, age, sex, cancer type, and ICI subtype. Our findings support the hypothesis that cirAE evaluation by a dermatologist improves survival outcomes, and we propose ICI retrial as a potential contributor for this finding. Limitations include the retrospective design, reliance on electronic medical record data, single institutional study, and potential differences in care delivery among oncologists. Additionally, the treating provider who originally observed the first incidence of cirAE may have suspended ICI therapy in anticipation of dermatology evaluation and would have continued this regimen had a referral not been placed. Multi-institutional studies are needed to further elucidate the effects of dermatology evaluation on adherence to ICI treatment regimens and other pertinent cancer outcomes. STC has received honoraria from Pfizer and Novartis for serving on an advisory board for digital media. Nicole R. LeBoeuf is a consultant and has received honoraria from Bayer, Seattle Genetics, Sanofi, Fortress Biotechnologies, Silverback, and Synox Therapeutics outside the submitted work.