Predicting benefit from immune checkpoint inhibitors in patients with non-small-cell lung cancer by CT-based ensemble deep learning: a retrospective study.

BACKGROUND: Only around 20-30% of patients with non-small-cell lung cancer (NCSLC) have durable benefit from immune-checkpoint inhibitors. Although tissue-based biomarkers (eg, PD-L1) are limited by suboptimal performance, tissue availability, and tumour heterogeneity, radiographic images might holistically capture the underlying cancer biology. We aimed to investigate the application of deep learning on chest CT scans to derive an imaging signature of response to immune checkpoint inhibitors and evaluate its added value in the clinical context. METHODS: In this retrospective modelling study, 976 patients with metastatic, EGFR/ALK negative NSCLC treated with immune checkpoint inhibitors at MD Anderson and Stanford were enrolled from Jan 1, 2014, to Feb 29, 2020. We built and tested an ensemble deep learning model on pretreatment CTs (Deep-CT) to predict overall survival and progression-free survival after treatment with immune checkpoint inhibitors. We also evaluated the added predictive value of the Deep-CT model in the context of existing clinicopathological and radiological metrics. FINDINGS: Our Deep-CT model demonstrated robust stratification of patient survival of the MD Anderson testing set, which was validated in the external Stanford set. The performance of the Deep-CT model remained significant on subgroup analyses stratified by PD-L1, histology, age, sex, and race. In univariate analysis, Deep-CT outperformed the conventional risk factors, including histology, smoking status, and PD-L1 expression, and remained an independent predictor after multivariate adjustment. Integrating the Deep-CT model with conventional risk factors demonstrated significantly improved prediction performance, with overall survival C-index increases from 0·70 (clinical model) to 0·75 (composite model) during testing. On the other hand, the deep learning risk scores correlated with some radiomics features, but radiomics alone could not reach the performance level of deep learning, indicating that the deep learning model effectively captured additional imaging patterns beyond known radiomics features. INTERPRETATION: This proof-of-concept study shows that automated profiling of radiographic scans through deep learning can provide orthogonal information independent of existing clinicopathological biomarkers, bringing the goal of precision immunotherapy for patients with NSCLC closer. FUNDING: National Institutes of Health, Mark Foundation Damon Runyon Foundation Physician Scientist Award, MD Anderson Strategic Initiative Development Program, MD Anderson Lung Moon Shot Program, Andrea Mugnaini, and Edward L C Smith.

The Genomic Landscape and Pharmacogenomic Interactions of Clock Genes in Cancer Chronotherapy.

Cancer chronotherapy, treatment at specific times during circadian rhythms, endeavors to optimize anti-tumor effects and to lower toxicity. However, comprehensive characterization of clock genes and their clinical relevance in cancer is lacking. We systematically characterized the alterations of clock genes across 32 cancer types by analyzing data from The Cancer Genome Atlas, Cancer Therapeutics Response Portal, and The Genomics of Drug Sensitivity in Cancer databases. Expression alterations of clock genes are associated with key oncogenic pathways, patient survival, tumor stage, and subtype in multiple cancer types. Correlations between expression of clock genes and of other genes in the genome were altered in cancerous versus normal tissues. We identified interactions between clock genes and clinically actionable genes by analyzing co-expression, protein-protein interaction, and chromatin immunoprecipitation sequencing data and also found that clock gene expression is correlated to anti-cancer drug sensitivity in cancer cell lines. Our study provides a comprehensive analysis of the circadian clock across different cancer types and highlights potential clinical utility of cancer chronotherapy.

Inhibition of tumor growth with a vaccine based on xenogeneic homologous fibroblast growth factor receptor-1 in mice.

Angiogenesis is important for the growth of solid tumors. The breaking of the immune tolerance against the molecule associated with angiogenesis should be a useful approach for cancer therapy. However, the immunity to self-molecules is difficult to elicit by a vaccine based on autologous or syngeneic molecules due to immune tolerance. Basic fibroblast growth factor (bFGF) is a specific and potent angiogenic factor implicated in tumor growth. The biological activity of bFGF is mediated through interaction with its high-affinity receptor, fibroblast growth factor receptor-1 (FGFR-1). In this study, we selected Xenopus FGFR-1 as a model antigen by the breaking of immune tolerance to explore the feasibility of cancer therapy in murine tumor models. We show here that vaccination with Xenopus FGFR-1 (pxFR1) is effective at antitumor immunity in three murine models. FGFR-1-specific autoantibodies in sera of pxFR1-immunized mice could be found in Western blotting analysis. The purified immunoglobulins were effective at the inhibition of endothelial cell proliferation in vitro and at the antitumor activity in vivo. The antitumor activity and production of FGFR-1-specific autoantibodies could be abrogated by depletion of CD4+ T lymphocytes. Histological examination revealed that the autoantibody was deposited on the endothelial cells within tumor tissues from pxFR1-immunized mice, and intratumoral angiogenesis was significantly suppressed. Furthermore, the inhibition of angiogenesis could also be found in alginate-encapsulate tumor cell assay. These observations may provide a new vaccine strategy for cancer therapy through the induction of autoimmunity against FGFR-1 associated with angiogenesis in a cross-reaction.