Background: More than two-thirds of cancers have no effective screening options; however, early detection may reduce cancer-related morbidity and mortality.1-4 A circulating cell-free DNA–based multicancer early detection (MCED) test has been shown to detect cancer signals across 50+ cancer types and accurately predict the cancer signal origin (CSO) >90% of the time.5 When a cancer signal is detected, CSO is provided to help guide the next steps in diagnostic evaluation. PATHFINDER (NCT04241796) is a prospective study that returned results from a targeted methylation-based MCED test to physicians and participants and evaluated the extent of ensuing diagnostic workups prompted by a positive screening test.
Objectives: To assess the extent of clinical evaluation required to achieve diagnostic resolution (ie, confirmed cancer or no confirmed cancer) for individuals with a positive MCED test result and to assess MCED test performance.
Methods: An interim analysis of a prospective interventional study conducted at outpatient clinic sites in 7 US health networks is presented. Between December 2019 and December 2020, 6796 adults aged ≥50 years with or without additional risk factors for cancer (including smoking history, genetic predisposition, or prior diagnosis of successfully treated cancer) were recruited for MCED testing. The primary end point was the extent and type of diagnostic testing required to achieve resolution following a “cancer signal detected” MCED test result. Test performance measures (cancer signal detection rate, positive predictive value [PPV], and accuracy of CSO prediction), participant-reported outcomes, and adverse events were also assessed. In a prespecified analysis, an MCED test that was further refined for screening use also was evaluated using the previously collected blood samples.
Results: 6662 participants enrolled, and 6629 had analyzable MCED test results and a median follow-up of 6.1 months. The MCED test detected a cancer signal in 1.4% (92/6629). To date, 70.7% (65/92) have reached diagnostic resolution (29 with cancer and 36 without cancer diagnosis), and 29.3% (27/92) are not yet resolved. Sixty-three participants with evaluation triggered by the MCED test had a median time to diagnostic resolution of 50 days (interquartile range = 28.0, 91.0); 90.5% (57/63) had ≥1 imaging tests, primarily PET/CT or CT scans, and 47.6% (30/63) had ≥1 invasive procedures, including most (77.8%) participants with confirmed cancer. Most participants (86.6%; 26/30) had only minimally invasive procedures (eg, colonoscopy). The PPV was 44.6% (95% CI, 33.2%-56.7%), and accuracy of the first CSO prediction in those with a cancer diagnosis was 85.2% (95% CI, 67.5%-94.1%). The further refined test version demonstrated cancer signal detection with similar PPV and maintained high accuracy of CSO prediction together with fewer hematologic CSO predictions relative to the earlier MCED test.
Conclusions: In this interim analysis, this blood-based MCED test was safely administered and detected signals from a broad range of cancers. Among signal-detected participants with resolution, diagnostic resolution was achieved within a median time of less than 2 months. Follow-up 12 months from MCED testing will allow for a final assessment of the diagnostic workup and MCED test performance for both versions of the test.
- American Cancer Society. Cancer Facts & Figures 2021. Atlanta, GA: American Cancer Society; 2021.
- DaCosta Byfield S, Smyth EN, Mytelka D, et al. Healthcare costs, treatment patterns, and resource utilization among pancreatic cancer patients in a managed care population. J Med Econ. 2013;16:1379-1386.
- Croswell JM, Kramer BS, Kreimer AR, et al. Cumulative incidence of false-positive results in repeated, multimodal cancer screening. Ann Fam Med. 2009;7:212-222.
- Ahlquist DA. Universal cancer screening: revolutionary, rational, and realizable. NPJ Precis Oncol. 2018;2:23.
- Liu MC, Oxnard GR, Klein EA, et al. Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. Ann Oncol. 2020;31:745-759.