PO.BCS02.04 · 生物信息与计算

利用人工智能（AI）对接受一线（1L）系统治疗的IV期结直肠癌（CRC）患者进行基于影像的预后评估（IPRO）

Evaluation of imaging-based prognostication (IPRO) using artificial intelligence (AI) in stage IV colorectal cancer (CRC) patients treated with first-line (1L) systemic therapy

编号 2787 展板 18 时间 4/20 02:00–05:00 区域 Section 4 主讲 Omar Khan, MBA;MD

分会场 Radiomics and AI in Medical Imaging

查看 PDF 下载 PDF 🔒 查看 / 下载完整 PDF 需登录并开通下载套餐 · 查看套餐 / 开通 AACR 官方页面

作者与单位 Authors & Affiliations

Mohammed A. Alvi¹, Ronald Bridges², Marina Salluzzi², Felipe Soares Torres³, Kartik Jhaveri³, Natasha B. Leighl⁴, John Riskas¹, Shahid Haider¹, Vignesh Sivan¹, Oleksandra Samodorova¹, Jay Hennesy¹, Duoaud Shah¹, FELIX BALDAUF-LENSCHEN¹, Omar F. Khan⁵

¹Altis Labs, Inc., Toronto, ON, Canada,²Cumming School of Medicine, University of Calgary, Calgary, AB, Canada,³Department of Medical Imaging, University of Toronto Temerty School of Medicine, Toronto, ON, Canada,⁴University Health Network, Toronto, ON, Canada,⁵Department of Oncology, Univ. of Calgary Faculty of Medicine, Calgary, AB, Canada

摘要 Abstract

中文摘要

引言：IV期CRC的准确预后评估有助于治疗决策并在临床试验中对患者进行分层。源自治疗前计算机断层扫描（CT）影像的肿瘤、淋巴结、转移（TNM）分期可对疾病范围进行分类，但可能无法充分表征预后。IPRO-alpha是一种源自治疗前CT影像的AI生成预后评分，评分越高代表生存越佳。IPRO-alpha经过训练和验证，用于预测晚期非小细胞肺癌的生存。本研究评估IPRO-alpha在接受一线系统治疗的真实世界IV期CRC患者中的可推广性和预后效用，并与TNM亚分期进行比较。方法：我们在一个真实世界数据集中回顾性评估了IPRO-alpha和TNM亚分期，该数据集包含2010-2018年间在17个癌症中心接受一线系统治疗的IV期CRC患者。我们采用Cox比例风险模型评估中位总生存期（mOS）和风险比（HR），涵盖IV期各亚分期（A、B、C），并对IPRO-alpha匹配了相对分布。结果：372例患者具备可用的治疗前CT和已知的TNM亚分期（IVA=141，IVB=162，IVC=69）。中位年龄为61岁（IQR 52-69），女性占32.8%（n=122）。TNM亚分期IVA的mOS显著优于IVB，而IVB期与IVC期之间无统计学显著差异（表1）。分布匹配的IPRO-alpha分组在高、中、低评分之间显示出显著的生存差异。结论：在转移性CRC患者中，IPRO-alpha相比TNM亚分期可能以更强的预后判别力对生存进行分层。IPRO-alpha虽在肺癌数据上训练，却学习到了共通的预后特征，使其生存预测能够推广至完全不同的肿瘤部位。未来的工作将评估IPRO-alpha在各类CRC治疗亚组中对生存进行分层的能力。按TNM亚分期和IPRO-alpha生存评分分层的IV期CRC患者中位OS。N 中位OS（月）（95% CI）HR（95% CI）p值 IPRO-alpha高 141 24.0（19.1-29.6）0.74（0.58-0.93）0.010 IPRO-alpha中 162 17.9（16.2-20.9）参照 - IPRO-alpha低 69 10.0（7.4-13.0）2.28（1.70-3.06）< 0.001 IVA期 141 20.8（17.9-27.5）0.77（0.61-0.98）0.032 IVB期 162 17.2（14.4-20.1）参照 - IVC期 69 13.0（10.9 - 16.0）1.14（0.85-1.53）0.363

查看英文原文 English abstract

INTRODUCTION: Accurate prognostication in stage IV CRC informs treatment decisions and stratifies patients in clinical trials. Tumor, node, metastasis (TNM) staging derived from pre-treatment computed tomography (CT) imaging classifies extent of disease but may not fully characterize prognosis. IPRO-alpha is an AI-generated prognostic score derived from pre-treatment CT imaging, with higher scores representing improved survival. IPRO-alpha was trained and validated to predict survival in advanced non-small-cell lung cancer. This study evaluates generalizability and prognostic utility of IPRO-alpha in real-world stage IV CRC patients receiving 1L systemic therapy, compared to TNM substage. METHODS: We retrospectively evaluated IPRO-alpha and TNM substage in a real-world dataset of stage IV CRC patients treated with 1L systemic therapy between 2010-2018 at 17 cancer centers. We evaluated median overall survival (mOS) and hazard ratios (HR) using Cox proportional hazards models across stage IV substages (A, B, C) and matched relative distributions for IPRO-alpha. RESULTS: 372 patients had available pre-treatment CT and known TNM substage (IVA=141, IVB=162, IVC=69). The median age was 61 years (IQR 52-69), with 32.8% (n=122) females. TNM substage IVA mOS was significantly better than IVB, with no statistically significant difference between stage IVB and IVC (Table 1). Distribution-matched IPRO-alpha groups showed significant survival differences across high, intermediate and low scores. CONCLUSIONS: IPRO-alpha may stratify survival with greater prognostic discrimination than TNM substage in metastatic CRC patients. IPRO-alpha, trained on lung cancer data, learned shared prognostic features allowing generalizability of survival predictions to entirely different tumour sites. Future work will evaluate IPRO-alpha's ability to stratify survival in various CRC treatment subsets. Median OS for stage IV CRC patients stratified by TNM substage and IPRO-alpha survival scores. N mOS in months (95% CI) HR (95% CI) p-value IPRO-alpha High 141 24.0 (19.1-29.6) 0.74 (0.58-0.93) 0.010 IPRO-alpha Intermediate 162 17.9 (16.2-20.9) reference - IPRO-alpha Low 69 10.0 (7.4-13.0) 2.28 (1.70-3.06) < 0.001 Stage IVA 141 20.8 (17.9-27.5) 0.77 (0.61-0.98) 0.032 Stage IVB 162 17.2 (14.4-20.1) reference - Stage IVC 69 13.0 (10.9 - 16.0) 1.14 (0.85-1.53) 0.363

利益披露 Disclosure

F. Soares Torres, None.. K. Jhaveri, None. O. F. Khan, Cogent Biosciences (Institutional Funding) ). Altis Labs, Inc. (Institutional Funding) ). Pfizer Other, Honoraria/Speaking Fees. AstraZeneca Other, Honoraria/Speaking Fees. Novartis Other, Honoraria/Speaking Fees. Gilead Other, Honoraria/Speaking Fees. Merck Other, Honoraria/Speaking Fees. Knight Therapeutics Other, Honoraria/Speaking Fees. Breast Cancer Canada (Institutional Funding) ).

← 返回 AACR 2026 检索