As part of its bid to transform the future of cancer screening and seize control of a potentially vast and increasingly competitive market, Grail announced fresh data on Friday from a large U.S. study of its flagship blood-based test for detecting dozens of tumor types. The results reinforce some of the company’s arguments for a new approach to screening; experts said the firm seems to have improved the test’s accuracy but noted that major questions about the real-world impacts of such tests remain.

The Pathfinder 2 study enrolled nearly 36,000 adults over the age of 50 to evaluate the biotech’s screening test, Galleri. When researchers measured the test’s performance in participants who’d been followed for over a year, they found that it caught 40.4% of cancer cases, a test feature known as sensitivity. A little more than half of these cancers were found early, in stage 1 or 2, and approximately three-quarters of them aren’t part of current screening regimens, such as pancreatic, liver, and head and neck cancers.

Among patients who had a positive test, nearly 62% were found to in fact have cancer, while 38% of positive results were false alarms. In earlier studies, Galleri’s positive predictive value was a bit lower, ranging from 43% to 50%. Conversely, nearly all participants who got a negative result did not have cancer, with a negative predictive value of 99.1%.

Continue to STAT+ to read the full story…