Treatment for pancreatic cancer? How AI helped target ‘undruggable’ protein

The Mayo Clinic, a prestigious American medical institution, recently unveiled a groundbreaking collaboration with Bengaluru-based AI startup Sravathi AI Technology. This joint effort aims to develop a novel cancer treatment targeting a specific protein known as GIPC1, which has the potential to fuel the growth of various types of cancer, including pancreatic cancer.

The findings of this pioneering research were published in the esteemed scientific journal Cell Reports under the title, "AI-driven discovery and validation of a GIPC1 PDZ domain inhibitor for pancreatic ductal adenocarcinoma".

GIPC1 is not inherently cancerous; rather, its association with cancer arises from aberrant expression or signaling within the body. When this protein becomes overproduced, as observed in pancreatic ductal adenocarcinoma (PDAC), the most aggressive form of exocrine pancreatic cancer, it can stimulate tumour growth and thwart the effectiveness of chemotherapy.

Tragically, PDAC boasts an abysmal five-year survival rate of less than 13.3%, primarily due to late diagnosis, rapid progression, and existing therapeutic resistance.

The PDZ domain, a key component of GIPC1, presents a formidable challenge for conventional drug development. Comprised of amino acids, this domain is found in the signaling proteins of various organisms, including bacteria, yeast, plants, viruses, and animals. Because of its broad and shallow surface, small molecules often struggle to bind effectively to this domain.

Moreover, the PDZ domain functions as a signalling hub, rendering it essential to inhibit multiple pathways simultaneously to achieve successful treatment outcomes.

To overcome these obstacles, the research team at Sravathi AI Technology employed advanced computational modeling, machine learning, and predictive analytics. Starting with a pool of 40,000 molecules, they meticulously narrowed down the candidates to five, eventually synthesizing and testing two. Generative AI played a pivotal role in designing the initial molecules, while predictive AI facilitated the evaluation of crucial properties such as toxicity and absorption.

With Sravathi AI completing their research within five months, Mayo Clinic researchers proceeded to conduct animal trials. If the treatment proves suitable for human use, it could potentially be available within three years. Debabrata Mukhopadhyay, a senior author of the study and a cancer researcher at the Mayo Clinic Florida, expressed optimism regarding the prospects of this AI-assisted discovery, stating that it offers a promising avenue for identifying novel therapeutic opportunities against previously undruggable targets.

Although the findings are yet to be tested in human trials, they lay a solid foundation for future research endeavors.