Clinical Evaluation of Cancer Signal Origin Prediction and Diagnostic Resolution Following Multi-Cancer Early Detection Testing in the PATHFINDER Study.

Abstract

Blood-based multi-cancer early detection (MCED) tests represent a new approach for cancer detection. To gain insight into the utility of various approaches of evaluating a cancer signal detected (CSD) test result, we evaluated diagnostic journeys of a subset of 6,662 participants in PATHFINDER who had a CSD on both an initial and refined version of an MCED test that also provided a prediction of cancer signal origin (CSO). We sought to determine whether CSO prediction-guided diagnostic evaluations led to diagnostic resolution; whether participants with known risk factors for cancer beyond age alone and a negative initial diagnostic evaluation had a cancer diagnosis during study follow-up; the utility of whole body imaging (WBI) in reaching diagnostic resolution; and differences in the diagnostic journeys needed to reach diagnostic resolution for both true- and false-positive results. Of the 39 participants in this analysis, 82% (32/39) achieved diagnostic resolution after the initial evaluation, including 78% (25/32) who reached resolution specifically with a CSO prediction-directed workup. 18% (7/39) required additional evaluation for persistent clinical suspicion of cancer, all of whom achieved resolution (3 with and 4 without cancer). WBI contributed to diagnostic resolution in only 49% of cancer signal detected cases. Approximately 90% of true- and false-positive cases had imaging tests; more true- versus false-positives (81.0% vs 38.9%) had non-surgical and/or surgical procedures. In conclusion, CSO prediction-directed evaluations enabled diagnostic resolution for most participants, although some with negative initial evaluations but persistent suspicion of cancer required additional testing.