The National University of Singapore (NUS) has reported good results in employing CURATE.AI, an artificial intelligence (AI) application that helps clinicians identify appropriate and tailored chemotherapy doses for patients. The research is in partnership with clinicians from the National University Cancer Institute, Singapore (NCIS) which is part of the National University Health System (NUHS).
The goal of CURATE.AI is to potentially uncover more treatment responders, as well as to find the best doses for patients to maximise efficacy and extend the time a patient is receptive to treatment. These doses may be lower for some patients than the high amounts that have been used in the past.
“Using CURATE.AI, which is efficacy-driven, we hope to help doctors quickly identify the optimal doses that are customised for each patient at different stages of the treatment cycle,” explained Dr Dean Ho, who heads the NUS Department of Biomedical Engineering and co-corresponding author of the study.
Dr Ho added that the goal is to improve both patient and treatment outcomes. Chemotherapy is frequently administered in fixed doses based on patient parameters. These toxicity-guided doses, however, may not result in an optimal response to treatment.
Dr Ho created the CURATE.AI together with his colleagues. He explained that the platform is an optimisation programme that uses a patient’s clinical data, including medication kind, dose, and cancer biomarkers, to create a customised digital profile that can be used to tailor the best dose during chemotherapy treatment.
The direct involvement of clinicians in building individualised datasets is an important aspect of applying AI in medicine. The pilot trial is an encouraging first step toward incorporating CURATE.AI into the clinical workflow of dynamic dose selection in the treatment of solid tumours.
The primary goal of CURATE.AI will enable truly personalised dosing for patients while also empowering clinicians to determine the best dose for each patient without increasing their workload. Clinicians can thus devote more time to the patient and caregiver.
Moreover, CURATE.AI uses each patient’s clinical data to calibrate medicine dosage. Each patient is given different pharmacological doses and their response is measured. This data is combined with clinical data to create a digital patient profile. It optimises treatment outcomes for each digital profile by linking medicine dosing to efficacy and safety. The dose may change during treatment.
Clinicians were allowed to accept or reject CURATE.AI dose recommendations based on the clinical judgement during the pilot trial, which ran from August 2020 to April 2022 at the National University Hospital.
The results of the pilot trial were presented at the 2022 American Society of Clinical Oncology (ASCO) Annual Meeting as a prospective and interventional study that harnesses an AI-based approach to human treatment. ASCO is a leading professional organisation for cancer caregivers, and the meeting will feature presentations on the most recent advances in cancer research.
Other study findings included patient adherence to suggested doses of 80 per cent and 100 per cent compliance in delivering dosing recommendations within the appropriate timeframe. These preliminary findings are encouraging in terms of CURATE’s downstream deployment. Putting artificial intelligence into clinical practice.
Following this initial step toward incorporating CURATE.AI into clinical workflows for dose selection in solid tumour treatment, the NUS team will move forward with a larger, randomised trial to further validate the platform’s performance.
The research team will also conduct clinical trials involving patients with other types of cancer, such as multiple myeloma, as well as disorders such as hypertension. Notably, the team will soon begin a trial to optimise personalised immunotherapy dosing for solid cancers.