Dataset Information

Machine learning links T cell function and spatial localization to neoadjuvant immunotherapy and clinical outcome in pancreatic cancer.

ABSTRACT: Tumor molecular datasets are becoming increasingly complex, making it nearly impossible for humans alone to effectively analyze them. Here, we demonstrate the power of using machine learning to analyze a single-cell, spatial, and highly multiplexed proteomic dataset from human pancreatic cancer and reveal underlying biological mechanisms that may contribute to clinical outcome. A novel multiplex immunohistochemistry antibody panel was used to audit T cell functionality and spatial localization in resected tumors from treatment-naive patients with localized pancreatic ductal adenocarcinoma (PDAC) compared to a second cohort of patients treated with neoadjuvant agonistic CD40 (αCD40) monoclonal antibody therapy. In total, nearly 2.5 million cells from 306 tissue regions collected from 29 patients across both treatment cohorts were assayed, and more than 1,000 tumor microenvironment (TME) features were quantified. We then trained machine learning models to accurately predict αCD40 treatment status and disease-free survival (DFS) following αCD40 therapy based upon TME features. Through downstream interpretation of the machine learning models' predictions, we found αCD40 therapy to reduce canonical aspects of T cell exhaustion within the TME, as compared to treatment-naive TMEs. Using automated clustering approaches, we found improved DFS following αCD40 therapy to correlate with the increased presence of CD44⁺ CD4⁺ Th1 cells located specifically within cellular spatial neighborhoods characterized by increased T cell proliferation, antigen-experience, and cytotoxicity in immune aggregates. Overall, our results demonstrate the utility of machine learning in molecular cancer immunology applications, highlight the impact of αCD40 therapy on T cells within the TME, and identify potential candidate biomarkers of DFS for αCD40-treated patients with PDAC.

SUBMITTER: Blise KE

PROVIDER: S-EPMC10634700 | biostudies-literature | 2023 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning links T cell function and spatial localization to neoadjuvant immunotherapy and clinical outcome in pancreatic cancer.

Blise Katie E KE Sivagnanam Shamilene S Betts Courtney B CB Betre Konjit K Kirchberger Nell N Tate Benjamin B Furth Emma E EE Dias Costa Andressa A Nowak Jonathan A JA Wolpin Brian M BM Vonderheide Robert H RH Goecks Jeremy J Coussens Lisa M LM Byrne Katelyn T KT

bioRxiv : the preprint server for biology 20231023

Tumor molecular datasets are becoming increasingly complex, making it nearly impossible for humans alone to effectively analyze them. Here, we demonstrate the power of using machine learning to analyze a single-cell, spatial, and highly multiplexed proteomic dataset from human pancreatic cancer and reveal underlying biological mechanisms that may contribute to clinical outcome. A novel multiplex immunohistochemistry antibody panel was used to audit T cell functionality and spatial localization i ...[more]

PMID: 37961410

Similar Datasets

Project description:IntroductionPancreatic ductal adenocarcinoma (PDAC) is an aggressive malignancy with a poor prognosis. Surgical resection remains the only potential curative treatment option for early-stage resectable PDAC. Patients with locally advanced or micrometastatic disease should ideally undergo neoadjuvant therapy prior to surgical resection for an optimal treatment outcome. Computerized tomography (CT) scan is the most common imaging modality obtained prior to surgery. However, the ability of CT scans to assess the nodal status and resectability remains suboptimal and depends heavily on physician experience. Improved preoperative radiographic tumor staging with the prediction of postoperative margin and the lymph node status could have important implications in treatment sequencing. This paper proposes a novel machine learning predictive model, utilizing a three-dimensional convoluted neural network (3D-CNN), to reliably predict the presence of lymph node metastasis and the postoperative positive margin status based on preoperative CT scans.MethodsA total of 881 CT scans were obtained from 110 patients with PDAC. Patients and images were separated into training and validation groups for both lymph node and margin prediction studies. Per-scan analysis and per-patient analysis (utilizing majority voting method) were performed.ResultsFor a lymph node prediction 3D-CNN model, accuracy was 90% for per-patient analysis and 75% for per-scan analysis. For a postoperative margin prediction 3D-CNN model, accuracy was 81% for per-patient analysis and 76% for per-scan analysis.DiscussionThis paper provides a proof of concept that utilizing radiomics and the 3D-CNN deep learning framework may be used preoperatively to improve the prediction of positive resection margins as well as the presence of lymph node metastatic disease. Further investigations should be performed with larger cohorts to increase the generalizability of this model; however, there is a great promise in the use of convoluted neural networks to assist clinicians with treatment selection for patients with PDAC.

Project description:BackgroundBladder cancer is a common malignancy with high invasion and poor clinical outcome. Intratumor heterogeneity (ITH) is linked to cancer progression and metastasis and high ITH can accelerate tumor evolution. Our objective is to develop an ITH-related signature (IRS) for predicting clinical outcome and immunotherapy benefit in bladder cancer.MethodsIntegrative procedure containing ten machine learning methods was applied to develop an IRS with The Cancer Genome Atlas (TCGA), gene series expression (GSE)13507, GSE31684, GSE32984 and GSE48276 datasets. To evaluate the performance of IRS in predicting the immunotherapy benefit, we also used several predicting scores and three immunotherapy datasets, including GSE91061, GSE78220 and IMvigor210.ResultsThe predicting model constructed with Enet (alpha =0.2) algorithm had a highest average C-index of 0.69, which was suggested as the optimal IRS. As an independent risk factor for bladder cancer, IRS had a powerful performance in predicting the overall survival (OS) rate of patients, with an area under curve of 1-, 3- and 5-year receiver operating characteristic (ROC) curve being 0.744, 0.791 and 0.816 in TCGA dataset. Bladder cancer patients with low IRS score presented with a higher level of immune-activated cells, cytolytic function and T cell co-stimulation. We also found a lower tumor immune dysfunction and exclusion (TIDE) score, lower immune escape score, higher programmed cell death protein 1 (PD-1) & cytotoxic T-lymphocyte associated protein 4 immunophenoscore, higher tumor mutation burden (TMB) score, higher response rate and better prognosis in bladder cancer with low IRS score. Bladder cancer cases with high IRS score had a higher half maximal inhibitory concentration value of common chemotherapy and targeted therapy regimens.ConclusionsThe current study developed an optimal IRS for bladder cancer patients, which acted as an indicator for predicting prognosis, stratifying risk and guiding treatment for bladder cancer patients. Further analysis should be focused on the exploration the differentially expressed genes (DEGs) and related underlying mechanism mediating the development of bladder cancer in different IRS score group.

Dataset Information

Machine learning links T cell function and spatial localization to neoadjuvant immunotherapy and clinical outcome in pancreatic cancer.

Publications

Machine learning links T cell function and spatial localization to neoadjuvant immunotherapy and clinical outcome in pancreatic cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets