The AIDPRO-CRC Trial

ConditionColo-rectal Cancer

Trial statusRecruiting

Trial phaseNot applicable

Trial typeInterventional

Biological sexAll

Age18+

SponsorZealand University Hospital

About this trial

The AIDPRO-CRC trial aims to improve outcomes for patients undergoing surgery for colorectal cancer by using artificial intelligence (AI) to assist surgeons in risk assessment. The trial will evaluate whether AI can help surgeons better predict the risk of complications and death, leading to improved care, fewer complications, and better use of healthcare resources.

In this nationwide, randomized clinical trial, participants will be divided into two groups. One group will have their risk assessed by a surgeon using standard clinical methods, while the other group will have their risk assessed by a surgeon using AI assistance. Based on the risk level, patients will receive varying levels of perioperative care. The AI-assisted risk assessment aims to tailor the treatment more precisely to each patient's individual needs, precisely allocating care to those who need it to more efficiently allocate heath system resources while having no deterioration in patient outcomes.

The primary hypothesis is that AI-assisted risk assessment will lead to more efficient and economic patient care without a deterioration in patient outcomes. The trial also aims to explore clinician satisfaction with the platform and its perceived effect. This is paired with a substudy exploring the variability of suggested treatment plans by clinicians with and without access to the MDT presentation platform.

The trial will include patients at seven hospitals across Denmark, involving patients diagnosed with colorectal cancer who are scheduled for curative surgery. All patients will receive standard treatment according to national guidelines, with the only difference being the modality of risk assessment. For the evaluation of the clinicians satisfactory with the device and the substudy of variability of suggested treatment plans, the trial will enroll clinicians using the device.

This study is a researcher-initiated, nationwide, randomized clinical trial involving patients diagnosed with colorectal cancer across eight hospitals in Denmark. Participants will be randomly assigned to one of two groups: AI-assisted risk assessment or standard surgeon-led assessment. The intervention focuses on optimizing perioperative care based on individual risk levels determined by either AI or the surgeon's clinical judgment.

The study builds on a successful pilot project (AID-SURG) that showed promising results in reducing complications, hospital stays, and readmissions.

Eligibility criteria

This trial does not accept healthy volunteers

Qualifiers

patients

Histologically confirmed diagnosis or strong clinical suspicion of first-time colon or rectal cancer, clinical stage I-IV.

Signed written informed consent obtained prior to any study-specific procedures.

Age ≥18 years at the time of consent.

Disqualifiers

patients

Be licensed medical doctors.

Be either board-certified specialists in surgical oncology or currently in training to become one.

Trial design

Design model

Parallel

Treatments tested in this trial

AI augmented risk-stratification
Device
A state-of-the-art artificial intelligence (AI) model called AIDPRO manual CRC is used as a decision support tool to estimate the 1-year mortality risk for each patient
Expert-based Risk-stratification
Other intervention
Experienced colorectal surgeons assess patient risk based on clinical judgment and national guidelines.

Treatment groups

1,200 Participants

are divided into 2 treatment groups

Group A: AI-augmented risk-stratificationExperimental treatment 1 intervention

Group B: Expert-based Risk-stratificationActive comparator 1 intervention

Trial outcomes

Primary outcomes

Cost Effectiveness

This primary endpoint is the saved marginal cost of perioperative intervention bundles achieved by integrating AI-augmented decision support. This will be assessed by comparing the overall marginal cost per patient between the AI-assisted arm (Intervention-arm) and the standard clinician-based stratification arm (control-arm). This cost-calculation factors in the following: * Distribution of patients across risk strata (A, B, C, D) * Cost per intervention bundle * Total perioperative expenditure per patient pr. bundle

Time frame

Baseline

Perceived Effect of Clinical Support Tool & User Feedback

This primary endpoint domain evaluates the user-perceived satisfaction with and clinical relevance of the AI-driven MDTPlatform medical device which contains the risk prediction algorithm. * Perceived relevance of the MDTPlatform provided information (i.e., the risk stratification and other displayed data) in support of making clinical decisions regarding perioperative treatment for a patient? * Perceived relevance of the displayed information provided by the MDTPlatform to clinical decision-making? * Comparison of the ability of MDTPlatform to provide a better overview of the current patient's treatment course in the context of multidisciplinary decision-making compared to usual practice? Each measured using a 7 point Likert scale where responses of 5, 6 or 7 are considered relevant All perceived clinician satisfaction with the use of the MDTPlatform will be assessed by questionnaire sent to users

Time frame

After 8 weeks of use again at 24-52 weeks of use and at after inclusion of last patient

Variability of Suggested Treatment With and Without MDTPlatform

This primary endpoint relates to the simulation substudy which will be carried out by letting users a subset of a set of patients. Over 1-2 sessions clinicians will evaluate a subset of a predetermined set of realistic patient cases. The total set will include 75 patients with and without MDTPlatform, yielding a total of 150 (2x75) cases. Clinicians will be asked to evaluate a minimum of 20 cases. Clinicians will score risk class based on the given data and will suggest a treatment plan, which will be recorded. The data given will be the same for cases with and without MDTPlatform, except for the risk stratification which will only be included in the cases presented via the MDTPlatform. The cases that are not presented via the MDTPlatform will be presented in the standard manner of the site where the clinician works. These data will show whether treatment recommendations vary less when MDTPlatform is used for risk prediction and case presentation versus standard care.

Time frame

Baseline

Secondary outcomes

The rate of complicated postoperative course 90 days after surgery

Defined by a Clavien-Dindo (CD) score \>2 or a Comprehensive Complication Index (CCI) \>20 within 90 days postoperatively

Time frame

90 days post-operative

Postoperative Complications

Total number of surgical complications (30, 90 days postoperatively) Both total and stratified by ≥ Clavien-Dindo grade 2. Medical complications at 90 days post-surgery Both total and stratified by ≥ Clavien-Dindo grade 2

Time frame

90 days Post-Operative

Length of Hospital stay (LOS) > 4 days

Defined by total postoperative duration of initial hospital stays measured in days. Analysis of LOS within each study arm and subsequent intervention group. Proportion of patients with LOS \>4 days

Time frame

90 days post-operative

Readmission Rates

Hospital readmissions within 30 and 90 days related to postoperative complications

Time frame

90 days postoperatively

Other outcomes

Surgical & Medical Complications

Total number of surgical complications (30, 90 days postoperatively) Complication severity assessed by Comprehensive Complication Index (CCI)

Time frame

90 Days Postoperatively

30, 90 and 365 day mortality

Is the patient is alive up to one year post-surgery?

Time frame

Up to 1 year post-operatively

Return to intended oncological treatment (RIOT) for patients requiring further treatment in addition to surgery, including chemotherapy, radiotherapy, or a combination hereof.

The proportion of patients who are able to initiate their planned adjuvant oncological therapy (chemotherapy, radiotherapy, or both) within the expected timeframe following surgery. This measure reflects the patient's postoperative recovery and readiness for further cancer treatment.

Time frame

30 days post-operative

Patient-reported Postoperative Recovery

Measured by The Quality of Recovery-15 (QoR-15) repeatedly at baseline, 1-14 days after surgery and 30 days after surgery

Time frame

30 days post-operative

Locations8

Select a site below to view it on the map.

Sponsors and contacts

Click on the lead sponsor to view all of their trials.

Zealand University Hospital

Lead sponsor

Aalborg University Hospital

Collaborator

Randers Regional Hospital

Collaborator

Gødstrup Hospital

Collaborator

Odense University Hospital

Collaborator

Hvidovre University Hospital

Collaborator

Viborg Regional Hospital

Collaborator

Hillerod Hospital, Denmark

Collaborator

Bispebjerg Hospital

Collaborator

Source ClinicalTrials.gov

Apply now