Gemini is reportedly a highly accurate computer model of disease progression and drug response at the individual patient level. Clinical development applications include discovering markers of response/nonresponse for clinical trial design, predicting optimal combination therapies and running head-to-head in-silico trials. Market access applications include generating evidence for line of therapy switching and optimizing treatment sequencing.

 

GNS built Gemini to model a multiple myeloma patient based on several factors. Colin Hill—chairman, CEO and founder of GNS Healthcare—has a long history with the Multiple Myeloma Research Foundation, which has collated a huge reservoir of multi-modal data. The availability of data has been bolstered by the advent of electronic medical records and those companies that have emerged to process and annotate the huge volume of information available.

 

In addition, there have been many new drugs targeting multiple myeloma released in the last several years, thus broadening the model’s data. Gemini includes the drug mechanisms most commonly used to treat multiple myeloma—such as proteasome inhibitors, IMiDs, corticosteroids, alkylating agents, anti-SLAMF7, anti-CD38 and others—connecting the impact of these drugs to clinical endpoints, including progression-free survival and overall survival.

 

“Over the past decade, there have been a dozen treatments approved for multiple myeloma, but there is still a lack of evidence to ensure patients receive optimal treatments in first-line and subsequent lines of therapy,” said Hill. “Creating Gemini, the in-silico patient, allows us to break the bottleneck of understanding what treatments work for which patients, driving better clinical trial design, generating real-world evidence for market positioning, and ultimately creating better outcomes for patients.”

 

GNS licenses the software to customers who are able to work together with GNS and input their own data to explore outcomes. Gemini has made this work much easier for customers to utilize the features without needing experts in computer science. Gemini allows customers to exploit the ability to unravel which drugs are working, what combinations may increase efficacy—and why. According to Hill, it enables a deeper analysis of causal relationships—not just correlation, but also influence.

 

“[Human disease] is more complex than a SpaceX journey or a driverless car,” Hill explains. “Those things are hard, but disease is harder. The number of components and the astronomical number of ways they can interact is the difference. We don’t have the blueprint. We just got the parts list 20 years ago, and there are more parts lists to discover.”

 

This iteration of Gemini is likely the first of many, with solid tumors a potential next target for Gemini investigation. To support the future formation of in-silico patients, GNS recently convened an in-silico patient advisory board to guide development and commercialization strategy. The in-silico patient for multiple myeloma represents the first of several poised to expand the world’s understanding of causal response to therapeutics across a range of diseases within oncology, immunology and neurology.

 

“The rise and proliferation of probably the most powerful new treatment for cancer in the history of humankind is immuno-oncology drugs, but it is still unclear which cancers they work in and why. Why are they super powerful in renal cell carcinoma and melanoma but not in colorectal cancer? Why in some patients and not others? Why in combination with some drugs but not others? Unraveling the drivers of the safety, efficacy, resistance and safety will be the key thing that truly unlocks the power of these drugs. Therefore, our next efforts will be very much in solid tumors, and there will be more than one,” Hill concludes.