IIT Guwahati Develops Adaptive Multi-Stage Clinical Trial Method for Precision Medicine

IIT Guwahati has introduced an AI-driven adaptive clinical trial model designed to enhance personalized patient care by optimizing treatment strategies in real-time.

Researchers at the Indian Institute of Technology (IIT) Guwahati, in collaboration with global institutions, have developed a multi-stage clinical trial method to transform personalised medical care. This innovative approach dynamically adjusts treatment plans in real time based on individual patient responses, ensuring more precise, effective, and tailored healthcare solutions.

Conducted in collaboration with Duke-NUS Medical School, the National University of Singapore, and the University of Michigan, USA, the research focuses on Dynamic Treatment Regimes (DTRs) developed through Sequential Multiple Assignment Randomised Trials (SMARTs). These advanced frameworks address the crucial challenge of optimizing treatment strategies by tailoring sequences of therapies to patients’ evolving responses over time.

Sequential Multiple Assignment Randomised Trials

Multi-stage clinical trials play a crucial role in developing effective Dynamic Treatment Regimes (DTRs), with the SMART methodology allowing researchers to evaluate different treatment sequences to identify the most suitable option for each patient. Unlike traditional trials, SMART incorporates multiple treatment stages, reassigning patients based on their responses to previous interventions.

However, conventional SMART trials allocate patients equally across treatment arms, even when interim data suggests some treatments are less effective, often leading to unnecessary treatment failures. To address this, Dr. Palash Ghosh and his team have developed an adaptive randomization method that dynamically adjusts patient allocation in real-time, favouring better-performing treatment sequences and optimizing outcomes throughout the trial. By focusing on both short-term and long-term outcomes, the method improves the entire treatment process, reducing failures and enhancing patient care.

Speaking about the research Dr. Palash Ghosh, Assistant Professor, Dept. Of Mathematics, IIT Guwahati, said, “Adaptive designs like this would encourage more patient participation in clinical trials like SMART. When patients see they are receiving treatments tailored to their needs, they are more likely to stay engaged. This approach also has vast potential for public health interventions, such as tailoring substance abuse recovery plans to individual needs as well as in other chronic diseases.”

The research findings have been published in the esteemed journal Biometrics in a paper co-authored by Dr. Palash Ghosh and his research scholar Rik Ghosh from IIT Guwahati, alongside Dr. Bibhas Chakraborty from Duke-NUS Medical School, National University of Singapore, and Dr. Inbal Nahum-Shani and Dr. Megan E. Patrick from the University of Michigan.

Read More

• IIT Guwahati Partners with Miraclus Orthotech to Revolutionise Orthopedic Technology
• IIT Guwahati Pioneers Nanomaterial for Mercury Detection in Cells and Environment