Official websites use .gov

A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS

A lock ( 🔒 ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Need Help

AI-Driven Curation and Analysis of Immune-Related Adverse Events in Cancer Immunotherapy Using Secondary Immune-Oncology Databases

Award Number: R21CA309407
ORGANIZATION: NATIONAL CANCER INSTITUTE
OPDIV: NIH
AWARD CLASS: DISCRETIONARY
AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)
PERIOD OF PERFORMANCE START DATE: 07/08/2026
PERIOD OF PERFORMANCE END DATE: 06/30/2028

Group Awards By:

View Award Description

AI-Driven Curation and Analysis of Immune-Related Adverse Events in Cancer Immunotherapy Using Secondary Immune-Oncology Databases - ABSTRACT Background and Significance Immune checkpoint inhibitors (ICIs) have revolutionized cancer therapy by improving outcomes for patients with various malignancies. However, ICIs can induce immune-related adverse events (irAEs) that result in treatment interruptions or discontinuation, thereby diminishing therapeutic effectiveness and compromising patient outcomes. Meanwhile, there remains an unmet need for a comprehensive data registry to support research on irAEs. The curation of such data often requires extensive manual labeling across multiple sources including clinical notes. Therefore, developing an AI-assisted curation tool to automate irAE information extraction and a predictive model to identify patients at risk early in treatment is critical to advancing precision immunotherapy. Innovation This project leverages the Georgetown Lombardi Comprehensive Cancer Center’s Immune-Oncology (IO) registry, encompassing over 2,500 immunotherapy-treated patients from ten MedStar hospitals, and harmonized data from Hackensack Meridian (HMH) Health’s John Theurer Cancer Center with 1,300 patients. Guided by oncologists’ clinical expertise, this proposal presents a framework that integrates natural language processing (NLP) and hybrid machine learning to automate the extraction, prediction, and interpretation of irAEs using a comprehensive spectrum of clinical data. The resulting tools and models will be shared to advance data curation and treatment management within the cancer research community, supporting the broader goal of precision oncology. Approach and Expected Outcomes We will first develop a large language model (LLM)–based NLP pipeline to extract irAE evidence, type, and grade from de-identified clinical notes, creating a reusable, annotated corpus (Aim 1). Next, we will integrate structured features including demographics, treatment regimens, tumor characteristics, and laboratory trends, into hybrid predictive models that combine static and longitudinal information (Aim 2). The model will be interpreted using explainable AI methods, enabling clinical validation of key predictive features in collaboration with oncologists. External validation using HMH data will test model generalizability across institutions (Aim 3). This project will generate an accurate and interpretable predictive model for irAE prediction and an NLP tool for automated curation. Both resources will be made publicly available to advance cancer information retrieval and enhance clinical decision-making in cancer treatment management.


Issue Date FY	Funding FY	Legal Entity Name	Legal Entity Address	Legal Entity City	Legal Entity State	Legal Entity Zip Code	Legal Entity COUNTY	Legal Entity COUNTRY	Assistance Listing	Award Code	Budget Year	Action Date	Action Type	Action Amount

Issue Date FY: 2026 ( Subtotal = $412,167 )
2026	2026	GEORGETOWN UNIVERSITY	MAIN CAMPUS	WASHINGTON	DC	20057	DISTRICT OF COLUMBIA	USA	Cancer Detection and Diagnosis Research	000	1	7/9/2026	NEW	$412,167
														Subtotal = $412,167

Grand Total All Awards = $412,167

Top

All Categories

About

Search

Reports

Data Submission

Award Information

AI-Driven Curation and Analysis of Immune-Related Adverse Events in Cancer Immunotherapy Using Secondary Immune-Oncology Databases

Award Number: R21CA309407

ORGANIZATION: NATIONAL CANCER INSTITUTE

OPDIV: NIH

AWARD CLASS: DISCRETIONARY

AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)

PERIOD OF PERFORMANCE START DATE: 07/08/2026

PERIOD OF PERFORMANCE END DATE: 06/30/2028

Federal Websites

Department of Health & Human Services

HHS Operating Divisions

HHS Staff Divisions

Download A Document Viewer