Microsatellite instability (MSI) has been approved as a pan-cancer biomarker for immune checkpoint blockade (ICB) therapy. However, current MSI identification methods are not available for all patients. We proposed an ensemble multiple instance deep learning model to predict microsatellite status based on histopathology images, and interpreted the pathomics-based model with multi-omics correlation. Methods: Two cohorts of patients were collected, including 429 from The Cancer Genome Atlas (TCGA-COAD) and 785 from an Asian colorectal cancer (CRC) cohort (Asian-CRC). We established the pathomics model, named Ensembled Patch Likelihood Aggregation (EPLA), based on two consecutive stages: patch-level prediction and WSI-level prediction. The initial model was developed and validated in TCGA-COAD, and then generalized in Asian-CRC through transfer learning. The pathological signatures extracted from the model were analyzed with genomic and transcriptomic profiles for model interpretation. Results: The EPLA model achieved an area-under-the-curve (AUC) of 0.8848 (95% CI: 0.8185-0.9512) in the TCGA-COAD test set and an AUC of 0.8504 (95% CI: 0.7591-0.9323) in the external validation set Asian-CRC after transfer learning. Notably, EPLA captured the relationship between pathological phenotype of poor differentiation and MSI ( P < 0.001). Furthermore, the five pathological imaging signatures identified from the EPLA model were associated with mutation burden and DNA damage repair related genotype in the genomic profiles, and antitumor immunity activated pathway in the transcriptomic profiles. Conclusions: Our pathomics-based deep learning model can effectively predict MSI from histopathology images and is transferable to a new patient cohort. The interpretability of our model by association with pathological, genomic and transcriptomic phenotypes lays the foundation for prospective clinical trials of the application of this artificial intelligence (AI) platform in ICB therapy.
Objectives: Coronavirus disease 2019 (COVID-19) is sweeping the globe and has resulted in infections in millions of people. Patients with COVID-19 face a high fatality risk once symptoms worsen; therefore, early identification of severely ill patients can enable early intervention, prevent disease progression, and help reduce mortality. This study aims to develop an artificial intelligence-assisted tool using computed tomography (CT) imaging to predict disease severity and further estimate the risk of developing severe disease in patients suffering from COVID-19. Materials and Methods: Initial CT images of 408 confirmed COVID-19 patients were retrospectively collected between January 1, 2020 and March 18, 2020 from hospitals in Honghu and Nanchang. The data of 303 patients in the People's Hospital of Honghu were assigned as the training data, and those of 105 patients in The First Affiliated Hospital of Nanchang University were assigned as the test dataset. A deep learning based-model using multiple instance learning and residual convolutional neural network (ResNet34) was developed and validated. The discrimination ability and prediction accuracy of the model were evaluated using the receiver operating characteristic curve and confusion matrix, respectively. Results: The deep learning-based model had an area under the curve (AUC) of 0.987 (95% confidence interval [CI]: 0.968-1.00) and an accuracy of 97.4% in the training set, whereas it had an AUC of 0.892 (0.828-0.955) and an accuracy of 81.9% in the test set. In the subgroup analysis of patients who had non-severe COVID-19 on admission, the
BackgroundGenetic variations of some driver genes in non-small cell lung cancer (NSCLC) had shown potential impact on immune microenvironment and associated with response or resistance to programmed cell death protein 1 (PD-1) blockade immunotherapy. We therefore undertook an exploratory analysis to develop a genomic mutation signature (GMS) and predict the response to anti-PD-(L)1 therapy.MethodsIn this multicohort analysis, 316 patients with non-squamous NSCLC treated with anti-PD-(L)1 from three independent cohorts were included in our study. Tumor samples from the patients were molecularly profiled by MSK-IMPACT or whole exome sequencing. We developed a risk model named GMS based on the MSK training cohort (n=123). The predictive model was first validated in the separate internal MSK cohort (n=82) and then validated in an external cohort containing 111 patients from previously published clinical trials.ResultsA GMS risk model consisting of eight genes (TP53, KRAS, STK11, EGFR, PTPRD, KMT2C, SMAD4, and HGF) was generated to classify patients into high and low GMS groups in the training cohort. Patients with high GMS in the training cohort had longer progression-free survival (hazard ratio (HR) 0.41, 0.28–0.61, p<0.0001) and overall survival (HR 0.53, 0.32–0.89, p=0.0275) compared with low GMS. We noted equivalent findings in the internal validation cohort and in the external validation cohort. The GMS was demonstrated as an independent predictive factor for anti-PD-(L)1 therapy comparing with tumor mutational burden. Meanwhile, GMS showed undifferentiated predictive value in patients with different clinicopathological features. Notably, both GMS and PD-L1 were independent predictors and demonstrated poorly correlated; inclusion of PD-L1 with GMS further improved the predictive capacity for PD-1 blockade immunotherapy.ConclusionsOur study highlights the potential predictive value of GMS for immunotherapeutic benefit in non-squamous NSCLC. Besides, the combination of GMS and PD-L1 may serve as an optimal partner in guiding treatment decisions for anti-PD-(L)1 based therapy.
Liver kinase B1 (LKB1) mutation is prevalent and a driver of resistance to immune checkpoint blockade (ICB) therapy for lung adenocarcinoma. Here leveraging single cell RNA sequencing data, we demonstrate that trafficking and adhesion process of activated T cells are defected in genetically engineered Kras-driven mouse model with Lkb1 conditional knockout. LKB1 mutant cancer cells result in marked suppression of intercellular adhesion molecule-1 (ICAM1). Ectopic expression of Icam1 in Lkb1-deficient tumor increases homing and activation of adoptively transferred SIINFEKL-specific CD8+ T cells, reactivates tumor-effector cell interactions and re-sensitises tumors to ICB. Further discovery proves that CDK4/6 inhibitors upregulate ICAM1 transcription by inhibiting phosphorylation of retinoblastoma protein RB in LKB1 deficient cancer cells. Finally, a tailored combination strategy using CDK4/6 inhibitors and anti-PD-1 antibodies promotes ICAM1-triggered immune response in multiple Lkb1-deficient murine models. Our findings renovate that ICAM1 on tumor cells orchestrates anti-tumor immune response, especially for adaptive immunity.
Contradictory characteristics of elevated mutational burden and a “cold” tumor microenvironment (TME) coexist in LKB1-mutant non-small cell lung cancers (NSCLC). The molecular basis underlying this paradox and strategies tailored to these historically difficult-to-treat cancers are lacking. Here, by mapping the single-cell transcriptomic landscape of genetically engineered mouse models with Kras versus Kras/Lkb1 driven lung tumors, we detected impaired tumor-intrinsic IFNγ signaling in Kras/Lkb1 driven tumors that explains the inert immune context. Mechanistic analysis showed that mutant LKB1 led to deficiency in the DNA damage repair process and abnormally activated PARP1. Hyperactivated PARP1 attenuated the IFNγ pathway by physically interacting with and enhancing the poly(ADP-ribosyl)ation of STAT1, compromising its phosphorylation and activation. Abrogation of the PARP1-driven program triggered synthetic lethality in NSCLC on the basis of the LKB1 mutation-mediated DNA repair defect, while also restoring phosphorylated STAT1 to favor an immunologically “hot” TME. Accordingly, PARP1 inhibition restored the disrupted IFN-γ signaling and thus mounted an adaptive immune response to synergize with PD-1 blockade in multiple LKB1-deficient murine tumor models. Overall, this study reveals an unexplored interplay between the DNA repair process and adaptive immune response, providing a molecular basis for dual PARP1 and PD-1 inhibition in treating LKB1-mutant NSCLC.
PurposeDespite the success of targeted therapy in c-ros oncogene 1 (ROS1)-rearranged cancers, especially non-small cell lung cancer (NSCLC), the clinical significance of ROS1 de novo mutation has not yet been understood. We sought to elucidate the predictive effect of ROS1 mutation for immune checkpoint inhibitor (ICI) therapy in melanoma.MethodsThe Cancer Genome Atlas [TCGA (n = 10967)] and Memorial Sloan Kettering Cancer Center [MSK (n = 10,945)] datasets, as well as two clinical cohorts of melanoma received ICI [CA209-038 (n = 73) and MEL-IPI (n = 110)], were included to explore the prevalence, prognostic effect, and immunotherapeutic predictive effect of ROS1 mutation in melanoma. Overall survival (OS) was defined as the primary outcome.ResultsOverall, melanoma accounted for the highest proportion of ROS1 mutation (~20%) which made up the majority (~95%) of the ROS1-alterated cases. Remarkably, ROS1 mutation yielded longer OS from ICI than the wild-type counterpart in the MSK melanoma population [hazard ratio (HR) 0.47, 95% confidence interval (CI) 0.30–0.74], and two external melanoma cohorts (CA209-038: HR 0.42, 95% CI 0.20–0.89; MEL-IPI: HR 0.55, 95% CI 0.34–0.91), without affecting the prognosis of patients. Elevated tumor mutational burden and enrichment of DNA damage repair was observed in ROS1 mutated patients, providing an explanation for the favorable responses to ICI therapy. Precisely, ROS1 mutation in non-protein tyrosine kinase (PTK) domain but not PTK mutation was responsible for the immunotherapy-specific responses of the ROS1 mutated patients in melanoma.ConclusionsCollectively, ROS1 mutation, specifically the non-PTK mutation, is a potential predictor of ICI therapy in melanoma, which is distinct from the well-established role of ROS1 rearrangement for targeted therapy in NSCLC.
BackgroundHomologous recombination deficiency (HRD) is characterized by overall genomic instability and has emerged as an indispensable therapeutic target across various tumor types, particularly in ovarian cancer (OV). Unfortunately, current detection assays are far from perfect for identifying every HRD patient. The purpose of this study was to infer HRD from the landscape of copy number variation (CNV).MethodsGenome-wide CNV landscape was measured in OV patients from the Australian Ovarian Cancer Study (AOCS) clinical cohort and >10,000 patients across 33 tumor types from The Cancer Genome Atlas (TCGA). HRD-predictive CNVs at subchromosomal resolution were identified through exploratory analysis depicting the CNV landscape of HRD versus non-HRD OV patients and independently validated using TCGA and AOCS cohorts. Gene-level CNVs were further analyzed to explore their potential predictive significance for HRD across tumor types at genetic resolution.ResultsAt subchromosomal resolution, 8q24.2 amplification and 5q13.2 deletion were predominantly witnessed in HRD patients (both p < 0.0001), whereas 19q12 amplification occurred mainly in non-HRD patients (p < 0.0001), compared with their corresponding counterparts within TCGA-OV. The predictive significance of 8q24.2 amplification (p < 0.0001), 5q13.2 deletion (p = 0.0056), and 19q12 amplification (p = 0.0034) was externally validated within AOCS. Remarkably, pan-cancer analysis confirmed a cross-tumor predictive role of 8q24.2 amplification for HRD (p < 0.0001). Further analysis of CNV in 8q24.2 at genetic resolution revealed that amplifications of the oncogenes, MYC (p = 0.0001) and NDRG1 (p = 0.0004), located on this fragment were also associated with HRD in a pan-cancer manner.ConclusionsThe CNV landscape serves as a generalized predictor of HRD in cancer patients not limited to OV. The detection of CNV at subchromosomal or genetic resolution could aid in the personalized treatment of HRD patients.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.