Seasonal influences on the efficacy of anti-programmed cell death (ligand) 1 inhibitors in lung cancer.

BACKGROUND: Seasonal variations in systemic immunity have been reported. This study aimed to evaluate whether seasonality affects the efficacy of anticancer immunotherapy. METHODS: A total of 604 patients with lung cancer receiving single anti-programmed cell death (ligand) 1 (anti-PD-[L]1) inhibitors from two prospective observational cohorts were screened. Primary outcomes were progression-free survival (PFS) and overall survival (OS). Patients were classified into two groups according to the season when the treatment started: winter (November-February) and other seasons (March-October). Kaplan-Meier analysis and Cox proportional hazards models were fitted to evaluate the impact of seasonality on survival. For validation, propensity score matching was performed. RESULTS: A total of 484 patients with advanced non-small cell lung cancer were included. In an unmatched population, multivariable analysis demonstrated that the winter group (n = 173) had a significantly lower risk of progression or death from immunotherapy than the other group (n = 311) (PFS: hazard ratio [HR], 0.77 [95% confidence interval (CI), 0.62-0.96]; p = .018; OS: HR, 0.77 [95% CI, 0.1-0.98]; p = .032). In a propensity score-matched population, the winter group (n = 162) showed significantly longer median PFS (2.8 months [95% CI, 1.9-4.1 months] vs. 2.0 months [95% CI, 1.4-2.7 months]; p = .009) than the other group (n = 162). The winter group's median OS was also significantly longer than that of the other group (13.4 months [95% CI, 10.2-18.0 months] vs. 8.0 months [95% CI, 3.6-8.7 months]; p = .012). The trend toward longer survival in the winter group continued in subgroup analyses. CONCLUSIONS: Starting an anti-PD-(L)1 inhibitor in winter was associated with better treatment outcomes in patients with lung cancer compared to other seasons.

Factors related to tumor response rate from TCGA three omics data-variants, expression, methylation.

The Cancer Genome Atlas (TCGA) and its patient-derived multi-omics datasets have been the backbone of cancer research, and with novel approaches, it continues to shed new insight into the disease. In this study, we delved into a method of multi-omics integration of patient datasets and the association of biological pathways related to the disease. First, across thirty-three types of cancer present in TCGA, we merged genomic mutations and drug response datasets and filtered for the viable variant-drug response combinations available in TCGA, containing more than three samples for each drug response label with RNA sequencing (RNA-seq) and genomic methylation data available for each patient. We identified two distinct combinations in TCGA, one being pancreatic adenocarcinoma patients with/without rs121913529 variant in KRAS gene treated with gemcitabine, and the other low-grade glioma with/without rs121913500 variant in IDH1 gene administered with temozolomide. In these two groups, different patterns of gene expression were observed in the pathways often associated with cancer progression, such as mTOR and PDGF between the patients with complete response and progressive disease. Our result will provide yet another example of the relevance of these biological pathways in cancer drug response and a way for multi-omics integration in cancer datasets.