Identifying drug synergistic with cancer immunotherapy

项目摘要 Avinash D Sahu博士，是一名计算生物学家，其首要职业目标是解决长期存在的问题 癌症免疫学和转化精确肿瘤学使用人工智能（AI），并设计新的治疗方法， 晚期癌症患者的治疗策略。为了确定与癌症免疫治疗协同的药物， 研究将尖端的AI技术与免疫肿瘤学（IO）相结合，以产生一种系统的方法， 确定与免疫疗法协同作用的药物，并将其优先用于晚期黑色素瘤的临床试验， 膀胱癌肾癌和肺癌 职业发展计划：Sahu博士是迈克尔逊奖的获得者，他的研究使命是启动精确 通过将患者从姑息性化疗转移到更个性化的IO治疗来改善免疫肿瘤学。他 之前在人工智能，统计学，方法开发，癌症和翻译生物学方面的培训使他准备好进行 提议的研究。Sahu博士概述了具体的培训活动，以扩大他在四个领域的技能： 免疫学，2）人工智能，3）翻译研究和4）新的免疫测定。这套技能将是必要的， 研究独立性。导师/环境：Sahu博士的指导和咨询团队汇集了世界领先的 计算生物学、翻译和临床研究、人工智能、统计学和免疫学专家。此外，萨胡博士 发展了学术合作和工业合作伙伴，为他的提议提供实验支持。 利用癌症免疫数据共享（CIDC）提供的最先进的软件和谷歌云基础设施; 来自DFCI、哈佛和布罗德研究所的计算资源;以及对最大免疫疗法的独特访问 Sahu博士从合作者的患者数据中识别出最有前途的IO药物组合。 研究：缺乏一种原则性的方法来确定有前途的IO药物组合，这往往导致 任意设计的IO临床试验，没有合理的生物学基础。该提案制定了第一个计算机预测器 评估药物的免疫调节作用和与免疫疗法协同作用的潜力。目的1建立一个小说深 学习预测器-DeepImmune-预测转录组的免疫治疗反应。目标2估计， 使用DeepImmune从药物诱导的转录组学变化中观察药物的免疫调节作用。目标3 优先考虑最佳预测的免疫调节药物，并在临床前模型中验证其效果。 结果/影响：成功完成提案将产生一个强大的预测器，以合理地将联合收割机 癌症治疗与免疫疗法，并为临床试验奠定基础，以测试最有前途的联合治疗。 职业发展奖和指导研究将使萨胡博士成为新研究领域的领导者 精准免疫肿瘤学和人工智能的交叉点

PROJECT SUMMARY Avinash D Sahu, Ph.D., is a computational biologist whose overarching career goal is to solve longstanding problems in cancer immunology and translational precision oncology using artificial intelligence (AI) and to devise new therapeutic strategies for late-stage cancer patients. Entitled Identifying drug synergistic with cancer immunotherapy, the proposed research combines cutting-edge AI technology with Immuno-oncology (IO) to produce a systematic approach to identifying drugs that synergize with immunotherapy, and prioritize them for clinical trials for advanced melanoma, bladder, kidney, and lung cancer. Career development plan: Dr. Sahu is a recipient of the Michelson Prize, and his research mission is to initiate precision immuno-oncology by moving patients away from palliative chemotherapy to more personalized IO treatments. His previous training in AI, statistics, method development, cancer, and translation biology have prepared him to conduct the proposed research. Dr. Sahu has outlined specific training activities to expand his skill set in four areas: 1) cancer immunology, 2) AI, 3) translation research and 4) new immunological assays. This skill set will be necessary to gain research independence. Mentors/Environment: Dr. Sahu mentoring and the advisory team assembles world-leading experts in computational biology, translation and clinical research, AI, statistics, and immunology. Also, Dr. Sahu has developed academic collaborations and industry partners to provide him experimental support for the proposal. Leveraging the state-of-art software and google-cloud infrastructure provided by Cancer Immune Data Commons (CIDC); computational resources from DFCI, Harvard, and Broad Institute; as well as unique access to largest immunotherapy patient data from collaborators, Dr. Sahu is uniquely placed to identify most promising IO drug combinations. Research: There is a lack of a principled approach to identify promising IO drug combinations that has often led to arbitrarily designed IO clinical trials without a sound biological basis. The proposal formulates the first in silico predictor to estimate drug’s immunomodulatory effect and potential to synergize with immunotherapies. Aim 1 builds a novel deep learning predictor —DeepImmune— to predict immunotherapy response from transcriptomes. Aim 2 estimates the immunomodulatory effects of drugs from for its drug-induced transcriptomic changes using DeepImmune. Aim 3 prioritize top predicted immunomodulatory drugs and validate their effect in pre-clinical models. Outcomes/Impact: The successful completion of the proposal will result in a robust predictor to rationally combine cancer therapies with immunotherapy and set the basis for a clinical trial to test the most promising combination therapy. The career development award and mentored research will enable Dr. Sahu to become a leader in the new field of research at the intersection of precision immuno-oncology and AI.