医学
肺癌
肿瘤科
内科学
免疫疗法
PD-L1
阶段(地层学)
免疫组织化学
组织学
生物标志物
癌症
生物
古生物学
生物化学
化学
作者
Noy Meshulami,Sooyun Tavolacci,Diego Pérez,Christian Rolfo,Philip C. Mack,Fred R. Hirsch
标识
DOI:10.1016/j.cllc.2023.03.014
摘要
Lung cancer is responsible for 1.8 million annual deaths. Non-small cell lung cancers (NSCLC) represent 85% of lung cancer tumors. While surgery is an effective early-stage treatment, the majority of newly identified US lung cancer cases are stage III/IV. Immunotherapy, using programmed death-ligand 1 (PD-L1) or programmed death 1 (PD-1) receptor antibody therapeutics, has increased survival for patients with NSCLC. PD-L1 protein expression is widely used as a predictive biomarker informing treatment decisions. However, only a minority of patients (27%-39%) respond to PD-L1/PD-1 treatment. PD-L1 protein expression by immunohistochemistry assay has deficiencies in identifying responding and refractory patients. Given the different characteristics of squamous and nonsquamous NSCLC, the predictability of PD-L1 levels in determining which patients would benefit from immunotherapy could vary between the 2 histologies. We analyzed 17 phase-III clinical studies and a retrospective study to determine if the predictive capability of PD-L1 expression varies between squamous and nonsquamous NSCLC. For patients with NSCLC treated with mono or dual-immune checkpoint inhibitors (ICI), PD-L1 expression was more predictive of benefit for patients with nonsquamous NSCLC than squamous NSCLC. Patients with nonsquamous histology and PD-L1 high tumor proportion scores (TPS) survived 2.0x longer compared to those with low TPS, when treated with monotherapy ICI. Among patients with squamous NSCLC, that difference was 1.2 to 1.3x. For patients treated with ICIs and chemotherapy, there was no clear difference in the predictive value of PD-L1 levels between histologies. We encourage future researchers to analyze the predictability of PD-L1 biomarker expression separately for squamous and nonsquamous NSCLC.
科研通智能强力驱动
Strongly Powered by AbleSci AI