NIH/NCI 476 - Digital Twin Software for Optimization of Cancer Radiation Therapy

Fast-Track proposals will be accepted.

Direct-to-Phase II proposals will be accepted.

Number of anticipated awards: 2-4

Budget (total costs, per award):

Phase I: up to $400,000 for up to 12 months

Phase II: up to $2,250,000 for up to 2 years

PROPOSALS THAT EXCEED THE BUDGET OR PROJECT DURATION LISTED ABOVE MAY NOT BE FUNDED.

Summary

Digital Twins (DTs) are virtual multidimensional models of systems designed to allow virtual testing and prediction. DTs are a new and highly promising tool for medicine as noted in a recent report from the National Academies of Science, Engineering and Medicine (NASEM). Recent publications indicated that the application of DT-based disease models could greatly benefit patients under treatments with complex scales, where intervention is expensive, and outcomes if improper processes are employed, can be suboptimal. Research evidence has shown that radiation therapy, which has multidimensional variables, could take advantage of DT’s tools to optimize treatments.

The purpose of this solicitation is to support the development of DT software that can help optimize the use of therapeutic radiation in patient care by utilizing multi-scale data, with examples of dimensions or scales being the following: molecular fingerprinting, the organ at risk location, and treatment modality (brachytherapy, proton, electron, photon, and all subtypes of these options) including potential combination therapy available to the patient that may include other classes of agents such as chemotherapy, biologic therapy, immunotherapy, hormone therapy, and other types of therapies based on geography, standard treatment options and/or real-time trial availability. The outcomes sought via the DT software will be variable based on proposals, but the underlying overall goal is to help patients receive optimal treatment, follow-up, and referral by assisting in decision-making on all levels. It should be noted that DTs are relevant to the NCI Cancer Moonshot Program in terms of the integration of multi-scale data as a priority area.

Project Goals

The short-term goal of this project is to have DTs developed that can help in clinical decision-making before, during, and after radiation therapy. While there are many areas of cancer therapy and medicine in general, that would benefit from DTs, the constraint of radiation therapy is used in this contract topic to keep this initial opportunity set focused and is based on the fact that radiation therapy is an area of medical care that is built around the use of treatment simulation at baseline. The first procedure a patient typically undergoes in their course of radiation treatment after their consultation is called a “simulation.”

Current unmet needs are many, but a subset of which include optimal treatment technique, optimal clinical trial selection if a patient can go on more than one, optimal tumor volume and clinical target volume delineation, and optimal follow-up scheduling and testing. The long-term goal of this project is the deployment of DTs in radiation therapy to optimize patient care decisions. Additionally, we hope these methods and platforms will ultimately go beyond radiation therapy so that other aspects of cancer care and potentially general medical care can benefit from digital twin technology.

The expected activities under this topic should focus on the development of a tool/product that can help decision-making related to any aspect of radiation therapy. For example, DT-based tools could assist clinicians with selection of treatment protocol based on multidimensional data (e.g., lowest compliance risk for the given patient), treatment optimization (e.g., beam angle, patient positioning, fractionation, or beam type), or follow-up timing optimization (how often to see a patient and order tests given some patients have many other doctors and medical issues). DTs can also apply to devices: e.g., how often to do maintenance. As a requirement, DTs should consist of data from at least three different scales, e.g., molecular, cellular, organ, organism, societal, geographic, modalities available, family history, cost, and toxicity. DTs must be able to be updated dynamically – in this way experience can impart improved decision-making, e.g., how to best compensate for missed or delayed treatment. Applicants are encouraged to leverage cloud-based multimodal data from the NCI Cancer Research Data Commons (CRDC) including multi-omics and imaging data.

Phase I Activities and Deliverables:

Offerors are expected to demonstrate data access to three or more domains of data on patients having undergone or about to undergo radiation therapy (within the prior year to the next year), and prior predictive modeling expertise in the form of products, publications, or code demonstrations in the Phase I proposal. The proposal should also clearly articulate the potential clinical use for the DT model and how this could benefit patients. The Phase I proof-of-concept deliverables must include:

Develop a DT model that can accurately model appropriate and relevant parameters on a small training core of patients or normal controls. Offerors should establish key metrics for the DT model proof of concept, and also provide a clear rationale for the proposed metrics. At a minimum, the proof-of-concept study can be based on retrospective datasets, but approaches that propose to use prospective data will be permitted.
Develop a clear framework for implementation of the DT model and how can be used to benefit patients. This can include but is not limited to reduction of toxicity, increasing treatment efficacy, improved patient care, cost savings, etc.
Develop a detailed experimental plan to test the digital twin(s) in a future clinical trial, including the ability to input data into the DT from ongoing clinical sources to enrich the digital twin for a given patient in real or close to real-time so that the very patient or the future patients in the process benefits from the input of these dynamic data.
Discuss the regulatory requirements for the proposed project with the FDA.

Phase II Activities and Deliverables:

Phase II proposals should focus on validation of the DT models in prospective studies.

Validate the performance of the DT model prospectively in relevant patient population. Offerors should establish metrics for expected performance of the model and provide a clear rationale for the proposed metrics.
Demonstrate containerization and ability to be deployed in a clinically relevant capacity (local, smartphone app, interface to the cloud, secure website, too compatible with radiation departmental software packages, etc.).
When appropriate, offerors should benchmark the DT model against competing technologies and demonstrate a clear competitive advantage.
Conversation(s) with the FDA regarding steps needed to produce a software product. If this requires working with other commercial products, if the support of those products’ companies is required, letters of support providing evidence that the support exists must be submitted to NCI.
Development of an appropriate regulatory strategy to file an IND translation plan for clinical adoption.

NOTE: For offerors where the cooperation of other vendors or collaborators is critical for implementing the proposed technology, the offeror should provide a letter of support from the partner(s) in the proposal.

Receipt date: Friday, October 18, 2024, 5:00 p.m. ET

Apply for this topic on the Contract Proposal Submission (eCPS) website.

View the full PHS2025-1 Contract Solicitation.