Characterization of the tumor microenvironment and identification of spatially predictive biomarkers associated with beneficial neoadjuvant chemoradiotherapy in locally advanced rectal cancer.

Neoadjuvant chemoradiotherapy (nCRT) has become the standard treatment for patients with locally advanced rectal cancer (LARC). However, 20-40% of patients with LARC show little to no response to nCRT. Thus, comprehensively understanding the tumor microenvironment (TME), which might influence therapeutic efficacy, and identifying robust predictive biomarkers is urgently needed. Pre-treatment tumor biopsy specimens from patients with LARC were evaluated in detail through digital spatial profiling (DSP), public RNA sequencing datasets, and multiplex immunofluorescence (mIF). DSP analysis revealed distinct characteristics of the tumor stroma compared to the normal stroma and tumor compartments. We identified high levels of human leukocyte antigen-DR/major histocompatibility complex class II (HLA-DR/MHC-II) in the tumor compartment and B cells in the stroma as potential spatial predictors of nCRT efficacy in the Discovery cohort. Public datasets validated their predictive capacity for clinical outcomes. Using mIF in an independent nCRT cohort and/or the total cohort, we validated that a high density of HLA-DR/MHC-II+ cells in the tumor and CD20 + B cells in the stroma was associated with nCRT efficacy (all p ≤ 0.021). Spatial profiling successfully characterized the LARC TME and identified robust biomarkers with the potential to accurately predict nCRT response. These findings have important implications for individualized therapy.

Development and external evaluation of predictions models for mortality of COVID-19 patients using machine learning method.

To predict the mortality of patients with coronavirus disease 2019 (COVID-19). We collected clinical data of COVID-19 patients between January 18 and March 29 2020 in Wuhan, China . Gradient boosting decision tree (GBDT), logistic regression (LR) model, and simplified LR were built to predict the mortality of COVID-19. We also evaluated different models by computing area under curve (AUC), accuracy, positive predictive value (PPV), and negative predictive value (NPV) under fivefold cross-validation. A total of 2924 patients were included in our evaluation, with 257 (8.8%) died and 2667 (91.2%) survived during hospitalization. Upon admission, there were 21 (0.7%) mild cases, 2051 (70.1%) moderate case, 779 (26.6%) severe cases, and 73 (2.5%) critically severe cases. The GBDT model exhibited the highest fivefold AUC, which was 0.941, followed by LR (0.928) and LR-5 (0.913). The diagnostic accuracies of GBDT, LR, and LR-5 were 0.889, 0.868, and 0.887, respectively. In particular, the GBDT model demonstrated the highest sensitivity (0.899) and specificity (0.889). The NPV of all three models exceeded 97%, while their PPV values were relatively low, resulting in 0.381 for LR, 0.402 for LR-5, and 0.432 for GBDT. Regarding severe and critically severe cases, the GBDT model also performed the best with a fivefold AUC of 0.918. In the external validation test of the LR-5 model using 72 cases of COVID-19 from Brunei, leukomonocyte (%) turned to show the highest fivefold AUC (0.917), followed by urea (0.867), age (0.826), and SPO2 (0.704). The findings confirm that the mortality prediction performance of the GBDT is better than the LR models in confirmed cases of COVID-19. The performance comparison seems independent of disease severity. Supplementary Information: The online version contains supplementary material available at(10.1007/s00521-020-05592-1).

Deep learning with whole slide images can improve the prognostic risk stratification with stage III colorectal cancer.

BACKGROUND AND OBJECTIVE: Adjuvant chemotherapy is recommended as standard treatment for colorectal cancer (CRC) with stage III according to TNM stage. However, outcomes are varied even among patients receiving similar treatments. We aimed to develop a prognostic signature to stratify outcomes and benefit from different chemotherapy regimens by analyzing whole slide images (WSI) using deep learning. METHODS: We proposed an unsupervised deep learning network (variational autoencoder and generative adversarial network) in 180,819 image tiles from the training set (147 patients) to develop a WSI signature for predicting the disease-free survival (DFS) and overall survival (OS) of patients, and tested in validation set of 63 patients. An integrated nomogram was constructed to investigate the incremental value of deep learning signature (DLS) to TNM stage for individualized outcomes prediction. RESULTS: The DLS was associated with DFS and OS in both training and validation sets and proved to be an independent prognostic factor. Integrating the DLS and clinicopathologic factors showed better performance (C-index: DFS, 0.748; OS, 0.794; in the validation set) than TNM stage. In patients whose DLS and clinical risk levels were inconsistent, their risk of relapse was reclassified. In the subgroup of patients treated with 3 months, high-DLS was associated with worse DFS (hazard ratio: 3.622-7.728). CONCLUSIONS: The proposed based-WSI DLS improved risk stratification and could help identify patients with stage III CRC who may benefit from the prolonged duration of chemotherapy.

CMTM6 expression in M2 macrophages is a potential predictor of PD-1/PD-L1 inhibitor response in colorectal cancer.

BACKGROUND: CMTM6 is a novel key regulator of PD-L1. High expression of both CMTM6 and PD-L1 may predict the benefit of PD-1 axis blockade in lung cancer. We aimed to investigate the expression pattern of CMTM6 between mismatch repair-defective (dMMR) and mismatch repair-proficient (pMMR) colorectal cancer (CRC) tissues and assess its correlation with the response to PD-1/PD-L1 pathway blockade. METHODS: Immunohistochemistry (IHC) was used to analyze CMTM6 and PD-L1 expression and immune cell density in dMMR/pMMR CRC. Quantitative multiplex immunofluorescence (IF) was performed to detect CMTM6, PD-L1, CD4, CD8, CD68 and CD163 expression in CRC patients treated with PD-1/PD-L1 inhibitors. RESULT: IHC analysis showed that CMTM6 and PD-L1 were both expressed in tumor cells (TCs) and invasion front immune cells (ICs). CMTM6 and PD-L1 expression and CD4+, CD8+, CD68+ or CD163+ cell density were significantly higher in dMMR CRC patients than in pMMR CRC patients. CMTM6 expression was positively correlated with PD-L1 expression and CD163+ M2 macrophage density in dMMR CRC. IF analysis showed that the coexpression rate of CMTM6/PD-L1 and the expression rate of CMTM6 in CD8+ T cells and CD163+ M2 macrophages were significantly increased in the group that exhibited clinical benefit. CMTM6 expression in M2 macrophages was identified as the best biomarker for predicting the responsiveness to PD-1/PD-L1 inhibitors. CONCLUSIONS: CMTM6 expression in M2 macrophages may predict the PD-1/PD-L1 inhibitor response rate in CRC patients more accurately than dMMR/microsatellite instability-high (MSI-H) status. It can also identify pMMR CRC patients who could benefit from PD-1/PD-L1 inhibitors.