Background
The identification of lung squamous cell carcinoma (LUSC) patients who may benefit from first-line chemo-immunotherapy (CIT) remains a challenge. This study aimed to develop a pathomics model to predict the T cell-inflamed gene-expression profile (GEP) status and validate its utility in identifying patients who derive survival benefit from CIT.
Methods
The pathomics model was developed using whole-slide images and RNA-sequencing data from The Cancer Genome Atlas (TCGA) LUSC cohort (n = 334) to predict the GEP status and generate a pathomics score (PS). The predictive value of PS was validated in a prospective, multicenter trial (AK105-302, n = 267) by assessing its interaction with treatment (CIT vs. chemotherapy) for progression-free survival (PFS) and overall survival (OS). Two additional independent cohorts (n = 82 and n = 50) were used for external validation.
Results
The pathomics model accurately predicted GEP status, achieving area under the curve (AUC) values of 0.80 and 0.71 in the TCGA training and validation sets, respectively. In the AK105-302 cohort, significant interactions were identified between PS and treatment modalities for both PFS (interaction p = 0.011) and OS (interaction p < 0.001). Patients with high PS who received CIT exhibited significantly prolonged PFS (hazard ratio [HR]: 0.31, 95 % confidence interval [CI]: 0.21–0.48, p < 0.001) and OS (HR: 0.30, 95 % CI: 0.18–0.50, p < 0.001) compared to high PS patients receiving chemotherapy. However, this survival benefit was not observed in the low-PS patients. These findings were corroborated in two independent clinical cohorts. Furthermore, biological assessments revealed a significant association between high PS and an immune-hot tumor microenvironment.
Conclusion
We developed a GEP-based pathomics model that provides a practical, cost-effective strategy to identify patients most likely to derive superior survival benefit from first-line CIT over chemotherapy alone.
扫码关注我们
求助内容:
应助结果提醒方式:
