Real-world assessment of response to anti-programmed cell death 1 therapy in advanced cutaneous squamous cell carcinoma

Abstract

To the Editor: Immunotherapy has revolutionized the treatment of advanced cutaneous squamous cell carcinoma (advCSCC) not amenable to curative surgery and/or radiotherapy. Phase I/II clinical trials1Migden M.R. Rischin D. Schmults C.D. et al.PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma.N Engl J Med. 2018; 379: 341-351Crossref PubMed Scopus (685) Google Scholar,2Grob J.J. Gonzalez R. Basset-Seguin N. et al.Pembrolizumab monotherapy for recurrent or metastatic cutaneous squamous cell carcinoma: a single-arm phase II trial (KEYNOTE-629).J Clin Oncol. 2020; 38: 2916-2925Crossref PubMed Scopus (86) Google Scholar excluded poor performance status (PS) and immunosuppressed patients. Data on the efficacy of anti-programmed cell death 1 (PD-1) therapy in real-world cohorts are lacking, especially in the advCSCC population generally deemed trial ineligible.We performed an Institutional Review Board-approved retrospective study of patients with advCSCC who received immune checkpoint inhibitors from 2016 to 2020 at Massachusetts General Hospital. Response to immune checkpoint inhibitors was evaluated using Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1. The Kaplan-Meier method was used to estimate overall survival (OS), progression-free survival (PFS), and duration of clinical benefit. In a preplanned exploratory analysis, univariable and multivariable Cox proportional hazards regression were used to model associations between clinicopathologic features and PFS or OS (for details see the Supplementary Methods via Mendeley at https://doi.org/10.17632/g769x5dt5r.1).Of the 76 patients that met inclusion criteria (median age, 74 years), 43 patients (57%) had unresectable/locally advCSCC only, and 33 (43%) had distant metastatic disease (Table I). Given standard of care guidelines at the time of treatment, 47 patients (62%) received anti–PD-1 therapy as first-line systemic therapy, 17 patients (22%) as second-line, and 12 patients (16%) had ≥2 lines of prior systemic therapy. Nineteen patients (25%) were immunosuppressed, and 26 patients (34%) had an Eastern Cooperative Oncology Group Performance Status ≥2. Additional clinicopathologic characteristics are summarized in Table I.Table IPatient characteristics at time of immune checkpoint inhibitor (ICI) initiationClinical features∗Data are presented as median (range) or as number (%).Patient data (N = 76)Age, y74 (18-98) <75 y39 (51) ≥75 y37 (49)Race Black/African American1 White75Sex Male51 (67) Female25 (33)ECOG Performance Status score 08 (11) 142 (55) 222 (29) 33 (4) 41 (1)Primary site of CSCC Head/neck52 (68) Extremities19 (25) Trunk5 (7)Anti–PD-1 therapy Cemiplimab38 (50) Pembrolizumab30 (39) Nivolumab8 (11)Previous systemic therapy for CSCC No previous systemic therapy (eg, first-line ICI)47 (62) 1 previous line of systemic therapy17 (22) ≥ 2previous lines of systemic therapy12 (16)Extent of CSCC Locoregional only43 (57) Distant metastasis33 (43)Immunosuppressed†The immunosuppressed cohort included patients with chronic lymphocytic leukemia, non-Hodgkin lymphoma, and multiple myeloma, patients who received a stem cell transplant for acute myeloid leukemia and had a long-standing history of immunosuppressive therapy (prednisone and azathioprine), and solid organ transplant recipients receiving immunosuppressive therapy (sirolimus and prednisone in all 3 cases). No57 (75) Yes19 (25)Chronic lymphocytic leukemia10Multiple myeloma2Non-Hodgkin lymphoma3Solid organ transplant3Allogeneic stem cell transplant1Xeroderma pigmentosum2 (3)CSCC, Cutaneous squamous cell carcinoma; ECOG, Eastern Cooperative Oncology Group; PD-1, programmed cell death 1.∗ Data are presented as median (range) or as number (%).† The immunosuppressed cohort included patients with chronic lymphocytic leukemia, non-Hodgkin lymphoma, and multiple myeloma, patients who received a stem cell transplant for acute myeloid leukemia and had a long-standing history of immunosuppressive therapy (prednisone and azathioprine), and solid organ transplant recipients receiving immunosuppressive therapy (sirolimus and prednisone in all 3 cases). Open table in a new tab The objective response rate was 34% (95% confidence interval [CI], 24%-46%), and the disease control rate was 70% (95% CI, 58%-80%) (Table II). Notably, 72% (95% CI, 62%-84%) of patients remained alive at 12 months, with a median OS of 3.3 years (95% CI, 1.8–not reached years). The median PFS was 1.2 years (95%, CI 0.5–not reached years), with a median duration of clinical benefit of 2.9 years (95% CI 1.5–not reached years). Data are limited concerning duration of clinical benefit monitoring cessation of therapy. Treatment was discontinued by 60% (32 of 53) of the disease control cohort; interestingly, 78% (95% CI, 63%-97%) remained progression free and alive at 12 months and 59% (95% CI, 37%-92%) remained progression free and alive at 3.5 years after the final dose.Table IIResponse assessment in locally advanced and metastatic cutaneous squamous cell carcinoma (CSCC)VariableFull cohort (N = 76)Immunosuppressed (n = 19)Immunocompetent (n = 57)ECOG ≤ 1 (n = 50)ECOG ≥ 2 (n = 26)Best overall response, No. (%) Complete response11 (14)4 (21)7 (12)9 (18)2 (8) Partial response15 (20)3 (16)12 (21)11 (22)4 (15) Progressive disease23 (30)8 (42)15 (26)11 (22)12 (46) Stable disease15 (20)1 (5)14 (25)11 (22)4 (15) Noncomplete response/nonprogressive disease12 (16)3 (16)9 (16)8 (16)4 (15)Objective response rate, % (95% CI)34 (24-46)37 (16-62)33 (21-47)40 (26-55)23 (9-44)Disease control rate, % (95% CI)∗Disease control rate was defined as the proportion of patients with complete response, partial response, stable disease, or noncomplete response/nonprogressive disease as best overall response; 56.6% of the full cohort never experienced progressive disease on study.70 (58-80)58 (33-80)74 (60-84)78 (64-88)54 (33-73)Durable disease control rate, % (95% CI)†Durable disease control rate was defined as the proportion of patients without progression for 12 months.68 (54-85)79 (56-100)65 (50-85)69 (53-91)62 (40-95)Observed time to progression, median (range), mo‡Observed time to progression is defined as the time (months) from initiating immune checkpoint inhibitors to progressive disease in the subset of patients who experienced progressive disease.2.2 (0.2-24)2.0 (0.4-5.5)2.3 (0.2-24)2.1 (0.4-24)2.4 (0.2-5.7)Observed time to response, median (range), mo§Observed time to response is defined as the time (months) from initial immune checkpoint inhibitors administration until first partial response or complete response in the subset of patients who achieved a partial response or complete response.2.6 (0.7-15)3.3 (1.4-9)2.5 (0.7-15)2.7 (0.7-15)2.4 (1.1-9.4)CI, Confidence interval; ECOG, Eastern Cooperative Oncology Group; No., number.∗ Disease control rate was defined as the proportion of patients with complete response, partial response, stable disease, or noncomplete response/nonprogressive disease as best overall response; 56.6% of the full cohort never experienced progressive disease on study.† Durable disease control rate was defined as the proportion of patients without progression for 12 months.‡ Observed time to progression is defined as the time (months) from initiating immune checkpoint inhibitors to progressive disease in the subset of patients who experienced progressive disease.§ Observed time to response is defined as the time (months) from initial immune checkpoint inhibitors administration until first partial response or complete response in the subset of patients who achieved a partial response or complete response. Open table in a new tab A poor baseline PS was associated with shorter OS (hazard ratio [HR], 2.4; P = .0006) and PFS (HR, 1.99; P = .0035) on univariable (Supplemental Table I) and multivariable (HR, 2; P = .005) analysis (Supplemental Fig 1). However, PFS appeared independent of immunosuppression (HR, 1.3; P = .486) (Supplemental Table I). No associations were found when assessing other clinicopathologic features, including age, sex, primary site of CSCC, and line of anti–PD-1 therapy (Supplemental Table I, Supplemental Fig 1). Furthermore, PFS and OS were independent of the type of anti–PD-1 therapy administered (Supplemental Table I, Supplemental Fig 1).Limitations include the retrospective design, limited sample size, and lack of safety/discontinuation assessment. However, our single-institution analysis demonstrates responses to immune checkpoint inhibitors in patients with advCSCC comparable to pivotal trial results (34%-50%).1Migden M.R. Rischin D. Schmults C.D. et al.PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma.N Engl J Med. 2018; 379: 341-351Crossref PubMed Scopus (685) Google Scholar,2Grob J.J. Gonzalez R. Basset-Seguin N. et al.Pembrolizumab monotherapy for recurrent or metastatic cutaneous squamous cell carcinoma: a single-arm phase II trial (KEYNOTE-629).J Clin Oncol. 2020; 38: 2916-2925Crossref PubMed Scopus (86) Google Scholar Interestingly, the response appears independent of immunosuppression. A caveat to this analysis is the heterogeneous definition of immunosuppression. However, our results are similar to previous studies assessing immunotherapy in several cancers in varying immunosuppressed cohorts.3Danesh M.J. Mulvaney P.M. Murakami N. et al.Impact of corticosteroids on allograft protection in renal transplant patients receiving anti-PD-1 immunotherapy.Cancer Immunol Immunother. 2020; 69: 1937-1941Crossref PubMed Scopus (9) Google Scholar, 4Hanna G.J. Ruiz E.S. LeBoeuf N.R. et al.Real-world outcomes treating patients with advanced cutaneous squamous cell carcinoma with immune checkpoint inhibitors (CPI).Br J Cancer. 2020; 123: 1535-1542Crossref PubMed Scopus (19) Google Scholar, 5Abdel-Wahab N. Safa H. Abudayyeh A. et al.Checkpoint inhibitor therapy for cancer in solid organ transplantation recipients: an institutional experience and a systematic review of the literature.J Immunother Cancer. 2019; 7: 106Crossref PubMed Scopus (133) Google Scholar Furthermore, risk-to-benefit assessment is warranted in solid organ transplant recipients. One patient who underwent renal transplantation 19 years before therapy developed acute interstitial nephritis and acute kidney rejection secondary to the checkpoint inhibitor. Durable clinical benefit for advCSCC on immunotherapy was achieved.Immunotherapy has transformed the treatment of advCSCC, but studies assessing response among the immunosuppressed, advanced age, and patients with other significant medical comorbidities appear warranted. Importantly, our data suggests clinical PS is a predictor of progression. Thus, patients may benefit from PS assessment at baseline to set expectations for immunotherapy response, especially for those with poor PS who may benefit from timely end-of-life discussions. To the Editor: Immunotherapy has revolutionized the treatment of advanced cutaneous squamous cell carcinoma (advCSCC) not amenable to curative surgery and/or radiotherapy. Phase I/II clinical trials1Migden M.R. Rischin D. Schmults C.D. et al.PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma.N Engl J Med. 2018; 379: 341-351Crossref PubMed Scopus (685) Google Scholar,2Grob J.J. Gonzalez R. Basset-Seguin N. et al.Pembrolizumab monotherapy for recurrent or metastatic cutaneous squamous cell carcinoma: a single-arm phase II trial (KEYNOTE-629).J Clin Oncol. 2020; 38: 2916-2925Crossref PubMed Scopus (86) Google Scholar excluded poor performance status (PS) and immunosuppressed patients. Data on the efficacy of anti-programmed cell death 1 (PD-1) therapy in real-world cohorts are lacking, especially in the advCSCC population generally deemed trial ineligible. We performed an Institutional Review Board-approved retrospective study of patients with advCSCC who received immune checkpoint inhibitors from 2016 to 2020 at Massachusetts General Hospital. Response to immune checkpoint inhibitors was evaluated using Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1. The Kaplan-Meier method was used to estimate overall survival (OS), progression-free survival (PFS), and duration of clinical benefit. In a preplanned exploratory analysis, univariable and multivariable Cox proportional hazards regression were used to model associations between clinicopathologic features and PFS or OS (for details see the Supplementary Methods via Mendeley at https://doi.org/10.17632/g769x5dt5r.1). Of the 76 patients that met inclusion criteria (median age, 74 years), 43 patients (57%) had unresectable/locally advCSCC only, and 33 (43%) had distant metastatic disease (Table I). Given standard of care guidelines at the time of treatment, 47 patients (62%) received anti–PD-1 therapy as first-line systemic therapy, 17 patients (22%) as second-line, and 12 patients (16%) had ≥2 lines of prior systemic therapy. Nineteen patients (25%) were immunosuppressed, and 26 patients (34%) had an Eastern Cooperative Oncology Group Performance Status ≥2. Additional clinicopathologic characteristics are summarized in Table I. CSCC, Cutaneous squamous cell carcinoma; ECOG, Eastern Cooperative Oncology Group; PD-1, programmed cell death 1. The objective response rate was 34% (95% confidence interval [CI], 24%-46%), and the disease control rate was 70% (95% CI, 58%-80%) (Table II). Notably, 72% (95% CI, 62%-84%) of patients remained alive at 12 months, with a median OS of 3.3 years (95% CI, 1.8–not reached years). The median PFS was 1.2 years (95%, CI 0.5–not reached years), with a median duration of clinical benefit of 2.9 years (95% CI 1.5–not reached years). Data are limited concerning duration of clinical benefit monitoring cessation of therapy. Treatment was discontinued by 60% (32 of 53) of the disease control cohort; interestingly, 78% (95% CI, 63%-97%) remained progression free and alive at 12 months and 59% (95% CI, 37%-92%) remained progression free and alive at 3.5 years after the final dose. CI, Confidence interval; ECOG, Eastern Cooperative Oncology Group; No., number. A poor baseline PS was associated with shorter OS (hazard ratio [HR], 2.4; P = .0006) and PFS (HR, 1.99; P = .0035) on univariable (Supplemental Table I) and multivariable (HR, 2; P = .005) analysis (Supplemental Fig 1). However, PFS appeared independent of immunosuppression (HR, 1.3; P = .486) (Supplemental Table I). No associations were found when assessing other clinicopathologic features, including age, sex, primary site of CSCC, and line of anti–PD-1 therapy (Supplemental Table I, Supplemental Fig 1). Furthermore, PFS and OS were independent of the type of anti–PD-1 therapy administered (Supplemental Table I, Supplemental Fig 1). Limitations include the retrospective design, limited sample size, and lack of safety/discontinuation assessment. However, our single-institution analysis demonstrates responses to immune checkpoint inhibitors in patients with advCSCC comparable to pivotal trial results (34%-50%).1Migden M.R. Rischin D. Schmults C.D. et al.PD-1 blockade with cemiplimab in advanced cutaneous squamous-cell carcinoma.N Engl J Med. 2018; 379: 341-351Crossref PubMed Scopus (685) Google Scholar,2Grob J.J. Gonzalez R. Basset-Seguin N. et al.Pembrolizumab monotherapy for recurrent or metastatic cutaneous squamous cell carcinoma: a single-arm phase II trial (KEYNOTE-629).J Clin Oncol. 2020; 38: 2916-2925Crossref PubMed Scopus (86) Google Scholar Interestingly, the response appears independent of immunosuppression. A caveat to this analysis is the heterogeneous definition of immunosuppression. However, our results are similar to previous studies assessing immunotherapy in several cancers in varying immunosuppressed cohorts.3Danesh M.J. Mulvaney P.M. Murakami N. et al.Impact of corticosteroids on allograft protection in renal transplant patients receiving anti-PD-1 immunotherapy.Cancer Immunol Immunother. 2020; 69: 1937-1941Crossref PubMed Scopus (9) Google Scholar, 4Hanna G.J. Ruiz E.S. LeBoeuf N.R. et al.Real-world outcomes treating patients with advanced cutaneous squamous cell carcinoma with immune checkpoint inhibitors (CPI).Br J Cancer. 2020; 123: 1535-1542Crossref PubMed Scopus (19) Google Scholar, 5Abdel-Wahab N. Safa H. Abudayyeh A. et al.Checkpoint inhibitor therapy for cancer in solid organ transplantation recipients: an institutional experience and a systematic review of the literature.J Immunother Cancer. 2019; 7: 106Crossref PubMed Scopus (133) Google Scholar Furthermore, risk-to-benefit assessment is warranted in solid organ transplant recipients. One patient who underwent renal transplantation 19 years before therapy developed acute interstitial nephritis and acute kidney rejection secondary to the checkpoint inhibitor. Durable clinical benefit for advCSCC on immunotherapy was achieved. Immunotherapy has transformed the treatment of advCSCC, but studies assessing response among the immunosuppressed, advanced age, and patients with other significant medical comorbidities appear warranted. Importantly, our data suggests clinical PS is a predictor of progression. Thus, patients may benefit from PS assessment at baseline to set expectations for immunotherapy response, especially for those with poor PS who may benefit from timely end-of-life discussions. Dr Miller has received honoraria for participating on advisory boards for Checkpoint Therapeutics, EMD Serono, Pfizer, Merck, Regeneron, and Sanofi Genzyme. Dr Emerick has received honoraria for participating on advisory boards for Regeneron and Sanofi Genzyme. Dr Sullivan has received research funding from Amgen and Merck and has served as a consultant and/or on advisory boards for Asana Biosciences, AstraZeneca, Bristol-Myers Squibb, Eisai, Merck, Novartis, OncoSec, Pfizer, and Replimune. Dr Kaufman is an employee of Immuneering Corporation. Drs Shalhout and Park have no conflicts of interest to declare. We would like to thank Hang Lee, PhD, of the Massachusetts General Hospital-Harvard Catalyst Biostatistics Program for his assistance and guidance with the statistical analysis plan.