Exai Bio announced new data from a 3,300 subject study demonstrating that its novel RNA- and AI-based platform can detect multiple cancer types at the earliest stages using a single, standard blood sample.
Notably, Exai’s platform maintained high sensitivity and specificity in both independent training and testing cohorts. In the independent testing cohort, overall stage I sensitivity across eight cancer types was 87% at 95% specificity. These data will be presented at a poster session at the American Society for Clinical Oncology (ASCO) 2023 annual meeting.
Exai’s multi-cancer early detection (MCED) study analyzed eight cancer types including lung, stomach, pancreas, kidney, colorectal, breast, prostate, and bladder, thus representing the majority of the cancer burden on society. The study demonstrates successful detection of stage I breast and prostate cancers which, to date, have both been extremely challenging to detect at early stages using other blood-based approaches. When a cancer signal was detected in the testing cohort, Exai’s platform could identify the tissue of origin with 88% accuracy using the top predicted cancer type and 95% accuracy considering the top two predictions.