Artificial intelligence (AI) solutions provider Phesi has developed a digital twin for Cytokine Release Syndrome (CRS) patients following CAR-T therapy.
Phesi has developed the digital twin for CRS in order to support the development of new CAR-T therapies for the treatment of different types of cancer.
CAR-T immunotherapies have become widely used in treatment of lymphomas, leukaemia, and multiple myeloma and CRS is one the most frequent and serious side-effects of the treatment.
CRS can lead to widespread organ dysfunction and is common in patients treated by infused CAR-T cells.
Phesi created the digital twin for the CRS patient population based on an analysis of 5,473 CAR-T treated patients.
For its digital twin, Phesi analysed data from 5,473 CAR-T treated patients: 5,335 in a cohort setting and 138 with individual anonymised records all receiving the standard of care (SOC). The patients were identified into eight cohorts and were given grades (1-4) of their CRS. The CRS by grade was consistent across the eight cohorts; a more effective treatment of CAR-T related CRS would reduce high and median-grade CRS. Similar to this digital twin, a synthetic clinical trial arm should show similar CRS by grade distribution pattern.
“Because of the relative scarcity of CRS patients, the search for new therapies is effectively akin to evaluating a rare disease. A decade on from the first immunotherapies, it is essential new treatments are found, given that CAR-T is being applied for more indications – particularly for blood cancer,” commented Dr Gen Li, Founder and CEO, Phesi. “The rapid onset and often life-threatening nature of CRS means the traditional development approach of double-blind trials is not feasible. To overcome this barrier, the pharmaceutical industry must look to artificial intelligence and big data analysis to improve the benefit-risk ratio for CAR-T treatment. Our goal in creating this digital twin dataset that can function as a synthetic control arm is to accelerate the evaluation of treatments for CRS.”
“Today, the industry typically gathers data from individual patients, usually in a randomised, double blind, placebo-controlled setting. But in some indications, developing digital twins for a control experience is a new and more effective approach to drug development enabled by using a combination of cohort data and data from individual patients,” continued Dr Li. “By modernising the industry’s current approach to clinical studies, Phesi can help to reduce the time it takes to find treatments for CRS and ensure CAR-T delivers the best outcomes. Importantly, the digital twin approach also makes clinical development far more patient-centric, reducing both patient and investigator site burden by eliminating the need for placebo treatments in trials.”