Journal for immunotherapy of cancer · 2026
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.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Miller David M, Merkin Ross D, Kaufman Howard L, Gupta Sameer G, Wolkow Natalie, Adelaja Adewunmi, Sullivan Ryan J, Patel Chirayu, Sorrentino Alexandra, Cohen Sonia, Cimoch Christine, Emerick Kevin S
Аннотация
BACKGROUND: Programmed death receptor 1 (PD-1) blockade produces high response rates in resectable and locally advanced cutaneous squamous cell carcinoma (CSCC), but how many doses are needed and whether surgery or radiation after immunotherapy (consolidation) adds benefit in deep responders (patients with substantial clinical responses) remain unclear. Observational inference is challenging because treatment decisions are response-guided, doses accumulate over time, and treatment selection depends on patient characteristics that also affect outcomes.
METHODS: We conducted a retrospective cohort study of 189 patients with resectable, borderline-resectable, locally advanced, or limited metastatic CSCC treated with immune checkpoint inhibitors as part of management (2019-2025). We summarized responses, treatment discontinuation patterns, and time-to-event outcomes. To evaluate dose-response relationships, we fit Bayesian regression models adjusting for prespecified baseline confounders. Directed acyclic graphs were used to formalize causal assumptions and identify minimally sufficient adjustment sets. For event-free survival (EFS), we used a prespecified landmark analysis among patients receiving ≥2 doses to reduce guarantee-time bias.
RESULTS: Objective response occurred in 119/189 (63%), including 52/189 (27.5%) complete responses. Nearly half of patients received ≤2 doses (92/189, 48.7%). Among 50 complete responders managed without surgery, 1 recurrence was observed over a median follow-up of 25.4 months. Dose-response models showed a consistent but modest positive association between additional doses and response: in a linear model, each additional dose was associated with ~1.09 fold higher odds of response (posterior probability >95%). Flexible threshold models suggested early concentration of benefit, strongest at 2 versus 1 dose (93.6% posterior probability of benefit), with persistent uncertainty in effect magnitude (66% probability of ≥6 percentage-point absolute increase). In the EFS landmark modeling cohort (n=177), receipt of ≥3 versus 2 doses showed a 93% posterior probability of reduced event risk, but the 89% credible interval spanned the null (HR 0.61-1.01), indicating substantial uncertainty regarding incremental downstream benefit.
CONCLUSIONS: In real-world CSCC, durable disease control frequently occurred after limited immunotherapy exposure, and clinical responders often did well without routine surgical consolidation. Although additional doses may modestly improve outcomes, observed gains appear concentrated early with uncertain incremental value beyond two doses. These patterns support conceptualizing treatment as frontline immunotherapy with response-guided subsequent treatment rather than fixed-duration neoadjuvant therapy.
METHODS: We conducted a retrospective cohort study of 189 patients with resectable, borderline-resectable, locally advanced, or limited metastatic CSCC treated with immune checkpoint inhibitors as part of management (2019-2025). We summarized responses, treatment discontinuation patterns, and time-to-event outcomes. To evaluate dose-response relationships, we fit Bayesian regression models adjusting for prespecified baseline confounders. Directed acyclic graphs were used to formalize causal assumptions and identify minimally sufficient adjustment sets. For event-free survival (EFS), we used a prespecified landmark analysis among patients receiving ≥2 doses to reduce guarantee-time bias.
RESULTS: Objective response occurred in 119/189 (63%), including 52/189 (27.5%) complete responses. Nearly half of patients received ≤2 doses (92/189, 48.7%). Among 50 complete responders managed without surgery, 1 recurrence was observed over a median follow-up of 25.4 months. Dose-response models showed a consistent but modest positive association between additional doses and response: in a linear model, each additional dose was associated with ~1.09 fold higher odds of response (posterior probability >95%). Flexible threshold models suggested early concentration of benefit, strongest at 2 versus 1 dose (93.6% posterior probability of benefit), with persistent uncertainty in effect magnitude (66% probability of ≥6 percentage-point absolute increase). In the EFS landmark modeling cohort (n=177), receipt of ≥3 versus 2 doses showed a 93% posterior probability of reduced event risk, but the 89% credible interval spanned the null (HR 0.61-1.01), indicating substantial uncertainty regarding incremental downstream benefit.
CONCLUSIONS: In real-world CSCC, durable disease control frequently occurred after limited immunotherapy exposure, and clinical responders often did well without routine surgical consolidation. Although additional doses may modestly improve outcomes, observed gains appear concentrated early with uncertain incremental value beyond two doses. These patterns support conceptualizing treatment as frontline immunotherapy with response-guided subsequent treatment rather than fixed-duration neoadjuvant therapy.