Response-Guided PD-1 Therapy May Support Treatment De-Escalation in Advanced Cutaneous Squamous Cell Carcinoma

Clinical Summary:  

• Design/Population: This retrospective cohort study evaluated 189 patients with resectable, borderline-resectable, locally advanced, or limited metastatic cutaneous squamous cell carcinoma treated with immune checkpoint inhibitors as part of initial management between 2019 and 2025.
• Key Outcomes: Objective response was achieved in 63% of patients, including complete responses in 27.5%, despite nearly half receiving 2 or fewer immunotherapy doses. Among 50 complete responders managed without surgery, only 1 recurrence occurred over a median follow-up of 25.4 months. Dose-response modeling suggested that any incremental benefit from additional treatment was concentrated early, with uncertain and likely modest benefit beyond 2 doses.
• Clinical Relevance: These real-world findings support early response assessment to guide treatment intensity, suggesting that abbreviated PD-1 therapy and omission of routine surgical consolidation may provide durable disease control in selected deep responders, although prospective studies are needed to define optimal treatment duration and identify patients who can safely avoid local therapy.

Results from a real-world study found that durable disease control was frequently achieved after limited PD-1 inhibitor exposure in patients with locally advanced or resectable cutaneous squamous cell carcinoma (cSCC), while the incremental benefit of continuing immunotherapy beyond the early treatment period remained uncertain. 

 “PD-1 blockade produces high response rates in resectable and locally advanced cSCC, but how many doses are needed and whether surgery or radiation after immunotherapy adds benefit in deep responders remain unclear,” stated David Miller, MD, Mass General Brigham Cancer Institute, Boston, Massachusetts, and coauthors. 

This retrospective cohort included 189 patients with resectable, borderline-resectable, locally advanced, or limited metastatic cSCC treated with immune checkpoint inhibition as part of initial management between 2019 and 2025. Patients who previously received systemic antineoplastic therapy for cSCC or widely metastatic disease were excluded.

Most patients received cemiplimab (n = 129) or pembrolizumab (n = 59), and 1 patient received ipilimumab plus nivolumab. Immunotherapy exposure varied substantially, ranging from limited treatment to as many as 46 doses. Nearly half the cohort (48.7%) received 2 or fewer doses. 

Despite the relatively limited exposure in many patients, the overall objective clinical response rate was 63%, with 119 patients responding. Complete responses occurred in 52 patients, corresponding to a complete response rate of 27.5%, while 67 patients achieved partial responses. Median time to first documented response was 42 days. 

Durable disease control was particularly notable among patients achieving complete clinical responses without subsequent surgical consolidation. Among 50 complete responders managed without surgery, only 1 recurrence was observed over a median follow-up of 25.4 months.

Among 141 patients who ultimately achieved no evidence of disease, estimated recurrence-free survival was 95.1% at 1 year, 91.2% at 2 years, and 88.1% at 3 years. Ten recurrences occurred among 86 surgically managed patients compared with 1 among 55 patients managed without surgery who achieved no evidence of disease. 

Investigators used Bayesian regression modeling to examine whether additional immunotherapy exposure was associated with improved response. In a linear model, each additional dose was associated with approximately 1.09-fold higher odds of objective response, with a 99% posterior probability that the dose effect was positive. However, the estimated incremental improvement associated with each additional dose was modest.

Models allowing for diminishing treatment effects suggested that the greatest potential benefit occurred during the earliest doses. In a categorical analysis, the posterior probability that receiving 2 rather than 1 dose improved response was approximately 94%. Evidence for incremental benefit at subsequent dose thresholds was less consistent.

The investigators subsequently examined whether treatment beyond 2 doses translated into improved of event-free survival. In a prespecified landmark cohort of 177 patients who remained event free long enough to receive at least 2 doses, 81 stopped after 2 doses and 96 received at least 3. 

Receiving at least 3 doses was associated with a 93.4% posterior probability of lower event risk compared with stopping after 2 doses. However, the median hazard ratio was 0.79 with an 89% credible interval of 0.61 to 1.01, reflecting substantial uncertainty regarding the magnitude of benefit. The probability of a large effect was low, supporting the interpretation that any incremental benefit beyond 2 doses was more likely modest than substantial.

Local treatment patterns also evolved during the study period. Overall, 92 patients underwent surgical resection, most commonly after a partial response. Among 48 surgical patients with a clinical partial response who subsequently underwent surgery, 48% had a pathologic complete response, while another 24 showed pathologic evidence of treatment response. 

The discrepancy between clinical and pathologic response suggested that residual abnormalities following immunotherapy did not necessarily represent viable disease. Over time, investigators increasingly reserved surgery for patients with suboptimal responses or persistent disease while managing selected deep responders without surgical consolidation. 

“Taken together, these findings suggest that minimal effective dosing may be sufficient for many patients, while the added benefit of extended immunotherapy or routine surigical consolidation in deep responders remains unclear,” concluded Dr Miller et al. 

Source: 

Miller DM, Merkin R D, Kaufman HL, et al. Frontline immunotherapy with response-guided subsequent treatment (FIRST) in cutaneous squamous cell carcinoma: A Bayesian causal analysis of dose intensity, early benefit, and treatment de-escalation. J Immunother Cancer. Published online: July 20, 2026. doi: 10.1136/jitc-2026-015029