First-Line Pembrolizumab Demonstrates Promising Activity in Phase 2 of Cutaneous Squamous Cell Carcinoma

Frontline treatment with the immune checkpoint inhibitor pembrolizumab as monotherapy generated promising activity against cutaneous squamous cell carcinoma with durable responses and a manageable safety profile.

Frontline treatment with the immune checkpoint inhibitor pembrolizumab (Keytruda) as monotherapy generated promising activity against cutaneous squamous cell carcinoma (CSCC) with durable responses and a manageable safety profile as treatment of patients with CSCC, according to findings from the phase 2 CARSKIN clinical trial (NCT02883556).

“In conclusion, the results of this prospective trial of first-line pembrolizumab monotherapy for patients with unresectable CSCCs demonstrated its efficacy,” the study authors, led by Eve Maubec, MD, PhD, wrote. “Pembrolizumab had robust antitumor activity, with sustained responses and manageable safety in this setting, confirming the interest of checkpoint inhibitors in treating CSCCs.”

Fifty-seven patients enrolled in the study between March 2017 and September 2018, but 7 were not evaluable for tumor response. Thirty-nine patients were included in the primary cohort, while 18 were placed in the expansion cohort.

The primary cohort received a median of 11 infusions of pembrolizumab (range, 0-33) during the 24-month trial. Two patients were untreated and 10 of the study participants received less than 5 infusions. Fourteen patients (37%) had received 5 to 20 infusions, and 14 patients (37%) had received > 20 infusions. At week 96, 9 patients were still receiving pembrolizumab. Six patients had to discontinue treatment due to treatment-related adverse events (TRAEs).

The primary end point of CARSKIN was response rate, while secondary end points include safety profile, response rates in PD-L1-positive patients, disease control rate, overall survival (OS), progression-free survival, and duration of response, among other key efficacy points.

The overall response rate (ORR) at week 15 was 41% in the primary cohort (95% CI, 26%-58%), which included 3 complete responses (CRs) and 13 partial responses (PRs). For all patients, the ORR was 42% (95% CI, 29%-56%), which included 4 CRs and 20 PRs. The disease control rate at week 15 was 54% in the primary cohort (95% CI, 37%-70%) and 60% in the overall population (95% CI, 46%-72%).

The best ORR in the primary cohort was 41% (95% CI, 26%-58%), and 8 patients (21%) achieved a CR between weeks 9 and 72, 2 of which were sustained 9 to 12 months after stopping treatment. One patient had stable disease and discontinued treatment for 10 months and did not experience progression. Two additional patients underwent conservative surgery for initially unresectable regional disease after pembrolizumab treatment.

The median time to response was 9 weeks (range, 8-15), and the median time to a CR was 24 weeks (range, 8-71).

The ORR in the expansion cohort was 44% (95% CI, 22%-69%), which included 1 CR and 7 PRs. The disease control rate was 72% (95% CI, 47%-90%).

Among patients with PD-L1-positive tumors at baseline, the ORR at week 15 was significantly higher at 55% compared with patients with PD-L1-negative disease (17%; P =.02).

“Baseline PD-L1 expression in at least 1% of tumor cells was associated with better pembrolizumab efficacy. The high ORRs and durations of response warrant additional investigation in these elderly populations for whom chemotherapy is often inappropriate, and poorly tolerated, with limited efficacy and durability,” the study authors wrote.

The median progression-free survival was 6.7 months in the primary cohort, and none of the responders experienced subsequent disease progression. The median duration of response was not reached, but the probability of maintaining response at week 48 was 93% (95% CI, 82%-100%). The median OS was 25.3 months (95% CI, 14.2-not estimable). The estimated 6-month OS rate was 81.3% (95% CI, 69.7%-94.8%), and the 12-month OS rate was 75.5% (95% CI, 62.7%-90.8%).

TRAEs were mostly grades 1 or 2, occurring in 71% of the patients. The most common TRAEs in the intention-to-treat population included fatigue (18%), diarrhea (13%), hypothyroidism (13%), pruritus (11%), and eczematous eruption (11%). Twenty-two patients had died during either treatment or follow-up.

Grade 3 TRAEs included diarrhea (2%), colitis (2%), cholestasis (2%), and cutaneous vasculitis (2%). Grade 4 CSCC hyper progression occurred in 1 patient, and grade 5 second de novo head and neck occurred in 1 patient.

The patients in this study were predominantly men, which made up 79% of the primary cohort and 83% in the expansion cohort, and the median age was 79 years (range, 42-99). The most frequent site of CSCC was HN (63%), and 12% had unresectable local disease, 63% regional lymph node involvement, and 25% distant metastases. The majority of patients (79%) had PD-L1 expression.

While 19% were referred for their initial diagnoses of advanced CSCC, unresectable at presentation, or de novo metastatic disease, 32% had a first relapse and 49% had at least 2 relapses. Thirty-seven percent had previously undergone radiotherapy, and no patients had received prior concurrent systemic therapy or radiosensitizing agents with radiation.


Maubec E, Boubaya M, Petrow P, et al. Phase II study of pembrolizumab as first-line single-drug therapy for patients with unresectable cutaneous squamous cell carcinomas. J Clin Oncol. doi: 10.1200/JCO.19.03357