Artificial Intelligence for Prostate Cancer Treatment Planning

Protocol for the Development of a Machine Learning Model for Prostate Cancer Treatment Planning

This project's goal is to develop and test an application that uses Artificial Intelligence (AI) to improve consistency and quality of Radiation Treatment (RT) plans for prostate cancer. By understanding expert planner preferences in structure contouring and treatment planning, and combining this framework with planning data and outcomes amassed in NRG clinical trials, AI models may be trained to produce contours and treatment plans that are indistinguishable or even potentially deemed superior to those produced by individual experts.

At the conclusion of this contract, the awardees will provide a software product which, when given the input of a description of desired anatomical target volumes and target doses along with a patient's CT scans, will generate target volumes and radiation treatment plans based upon a "gold standard" amalgamated from the input of multiple experts, thereby achieving desired doses to target volumes while meeting or exceeding the dose-volume constraints imposed by adjacent normal tissues.

Study Overview

• Status: Completed
• Conditions: Prostate Cancer; Artificial Intelligence; Radiotherapy
• Intervention / Treatment: Other: AI-assisted RT modelling

Detailed Description

PHASE 1

1. Develop a process and tools (DAST) to capture the rationale, criteria, and logical basis behind the treatment planning process using well understood Human Factors knowledge gathering methodologies and Machine Learning tools.
2. Build the AI technology to learn the process and apply it to generating treatment plans. Images and expert-drawn volumes from Radiation Therapy Oncology Group (RTOG) 0938 will be used for initial training of the AI system. These data are not Dartmouth-Hitchcock Medical Center (DHMC) patients, but rather were consented and acquired through RTOG 0938. These data are housed at NRG/RTOG headquarters in Philadelphia. NRG already has an established, IRB-approved protocol for exploring AI systems for the 0938 data set. A minimum of 30 cases will be used for this initial training work for the AI system to "learn" volume segmentation of important structures and targets. Additional patients from the available 200+ patients on 0938 may be added for additional AI learning of volume segmentation as initial software programming is implemented.

3. Determine the optimal number of historic treatment plans to train the AI technology and test it. 45 cases will be provided by NRG from NRG/RTOG studies 0415, 0126, and 0521 (totaling 135 cases), respectively representing favorable-, intermediate-, and high-risk prostate cancer treatments. These scans and expert-defined volumes are part of NRG datasets housed in Philadelphia. All these patients already signed study-specific consents which included permissions to allow personally specific clinical information to be used for other cancer-related studies. IRB review of the use of these data for this specific study is anticipated to be achieved through NRG mechanisms in the context of these prior consents, to assure that the use of these data for this specific AI study is appropriate and approved. Provision is anticipated of a total of 30 additional patient cases as "gold standards," 10 each from DHMC, University of Massachusetts (UMass), and Oregon Health Sciences University (OHSU), respectively, all initially planned and treated within the 2015-2018 time frame. The first 5 from each institution will be favorable-risk patients, and the next 5 from each institution will be high-risk patients (thereby achieving a wide range of treatment approaches, with more to be added subsequently). As part of this effort, each individual institution will contact its own specific patients (5 favorable-risk, 5 high-risk) to obtain study-specific consent for the use of their data for this protocol. Once anonymized, these scans and plans will be shared across all three institutions. For each of these patients, the other two ("non-host") institutions will create their own volumes and plans. Using a modified Delphi approach, the three teams will then meet to generate an agreed-upon "composite" plan for each patient. Thus, in total there will be four treatment plans for each of these 30 patients, yielding 120 total plans that will serve as the "gold standard" for this AI project, and will be inputted for testing/validation to the AI system.

   PHASE 2

4. Expand the database to include intermediate-risk patients, 5 respectively from each institution, following the above procedures, to yield an additional 60 plans to serve as additional inputs for the AI system.
5. Validate and test the AI technology by inputting patient images and target delineations from historic case data and assessing whether the AI technology-generated plans are "consistent" with the final plans that were created by expert clinicians.
6. Test the technology with new patient case data and validate the plan with a team of expert clinicians. This will involve "modified Turing tests," as developed in NRG-RTOG studies exploring AI applications.

• Study Type: Observational
• Enrollment (Actual): 5

Contacts and Locations

Study Locations

• United States
  • New Hampshire
    • Lebanon, New Hampshire, United States, 03756
      • Dartmouth-Hitchcock Medical Center

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 21 years to 90 years (Adult, Older Adult)
• Accepts Healthy Volunteers: N/A

• Sampling Method: Probability Sample

Study Population

Male patients diagnosed with prostate xancer who have received RT and now have elected to provide data for the development of AI-models.

Description

Inclusion Criteria:

• Favorable-risk inclusion criteria (as per RTOG 0415)

  1. Histologically confirmed prostate adenocarcinoma
  2. Gleason Score <= 3+4 = 7 ( with less than 50% of all cores positive, and no more than one core with Gleason 3+4=7)
  3. Clinical stage T1-T2b
  4. Prostate Specific Antigen (PSA) <10 ng/ml within 180 days prior to treatment planning. PSA may not have been acquired within 30 days of stopping finasteride, or within 90 days of stopping dutasteride
  5. RT treatment initiated between 1/1/15 and 12/31/16
  6. Prostate MRI used as part of RT treatment planning
  7. No previous hormonal therapy, such as LHRH agonists, estrogens, anti-androgens, or surgical castration
  8. No previous use of finasteride within 30 days prior to planning
  9. No previous use of dutasteride within 90 days prior to planning

• High-risk inclusion criteria (as per RTOG 0521)

  1. Histologically confirmed prostate adenocarcinoma
  2. PSA < 150

  3. One of the following combinations:

     1. Gleason 7 or 8 and PSA >= 20
     2. Gleason 8 and clinical T-stage > T2a
     3. Gleason 9 or 10
  4. Negative bone scan within 180 days of planning
  5. XRT treatment initiated between 1/1/15 and 12/31/16
  6. Prostate MRI used as part of RT treatment planning
  7. No previous hormonal therapy, such as LHRH agonists, estrogens, anti-androgens, or surgical castration, prior to prostate cancer diagnosis

• Intermediate-risk inclusion criteria

  1. Histologically confirmed prostate adenocarcinoma
  2. PSA < 20
  3. Gleason 7 or 8
  4. Not meeting criteria for favorable- or high-risk disease, as per above
  5. XRT treatment initiated between 1/1/15 and 12/31/16
  6. Prostate MRI used as part of RT treatment planning
  7. No previous hormonal therapy, such as LHRH agonists, estrogens, anti-androgens, or surgical castration, prior to prostate cancer diagnosis

Exclusion Criteria:

1. Prior or concurrent invasive malignancy (except non-melanomatous skin cancer) or lymphomatous/hematogenous malignancy unless continually disease free for a minimum of 5 years
2. Evidence of distant metastases
3. Regional lymph node involvement
4. Previous radical prostate surgery or cryosurgery
5. Previous pelvic irradiation or prostate brachytherapy
6. Previous or concurrent cytotoxic chemotherapy for prostate cancer

7. Severe, active comorbidity, defined as follows:

   1. Unstable angina, congestive heart failure, and/or transmural myocardial infarction requiring hospitalization within the last 6 months
   2. Acute bacterial or fungal infection requiring intravenous antibiotics
   3. Hepatic insufficiency resulting in clinical jaundice or coagulopathy
   4. Acquired immune deficiency syndrome based upon current CDC-defined criteria
8. Zubrod performance status 2 or worse
9. Previous use of finasteride within 60 days of planning
10. Previous use of dutasteride within 180 days of planning

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
All patients for enrollment and analysis; Patients previously treated with RT for prostate cancer who are now being enrolled into this study for data analysis and incorporation into Artificial Intelligence (AI) models	Other: AI-assisted RT modelling; Artificial Intelligence assisted Radiation Treatment

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change (trend) in AI System performance over time	Validate the AI technology by inputting images and targets from historic case data and then assessing whether the AI-generated plans are comparable or even improved when measured against plans created by expert clinicians for the same patient data. To accomplish this, methodology will employ "modified Turing tests," as previously developed in NRG-RTOG studies exploring AI applications in other venues, whereby blinded experts evaluate alternative plans, score them on a variety of criteria, and more generally assess whether these were generated via machine learning or via human intelligence. These quantified comparisons will be performed several times. The first will be at the end of Phase I (spring of 2021). Phase II will introduce further data and plans. Repeat comparisons of AI- versus human-performance will be performed every six months during Phase II. Final comparisons and overall trend analysis will be reported at conclusion of Phase II in 2023.	Baseline measure (Phase 1): June 2021. Repeat assessments every 6 months (Phase 2): June 2021 - June 2023.

Collaborators and Investigators

• Sponsor: Dartmouth-Hitchcock Medical Center
• Collaborators: National Cancer Institute (NCI); Oregon Health and Science University; NRG Oncology; University of Massachusetts, Worcester; Nicolalde R&D
• Investigators: Principal Investigator: Alan C. Hartford, MD PhD FACR, Dartmouth-Hitchcock Medical Center

Study record dates

Study Major Dates

• Study Start (Actual): June 22, 2020
• Primary Completion (Actual): November 30, 2021
• Study Completion (Actual): March 4, 2022

Study Registration Dates

• First Submitted: June 6, 2020
• First Submitted That Met QC Criteria: June 19, 2020
• First Posted (Actual): June 22, 2020

Study Record Updates

• Last Update Posted (Actual): October 13, 2023
• Last Update Submitted That Met QC Criteria: October 12, 2023
• Last Verified: October 1, 2023

More Information

Terms related to this study

• Keywords: Prostate Cancer; Artificial Intelligence; Radiation Therapy; Radiation Planning
• Additional Relevant MeSH Terms: Neoplasms; Urogenital Neoplasms; Neoplasms by Site; Genital Neoplasms, Male; Prostatic Diseases; Urogenital Diseases; Male Urogenital Diseases; Genital Diseases, Male; Genital Diseases; Prostatic Neoplasms
• Other Study ID Numbers: D20003; 75N91019C00054-0-9999-1 (U.S. NIH Grant/Contract)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: There are no plans to share individual participant data at this time.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No