Real-world experiences with a multi-cancer detection (MCD) blood test.

Abstract

10565 Background: Multi-cancer detection (MCD) blood tests hold promise to screen for multiple cancers simultaneously. MCD tests have been assessed in case-control and prospective studies but have not yet been shown to improve cancer outcomes. Mass General Brigham (MGB) offers a commercially available, non-FDA-approved methylation-based MCD test (Galleri; GRAIL) through our Early Detection & Diagnostics Program. Here, we report our initial experience. Methods: In this retrospective, observational study, data were collected from all patients (pts) who underwent MCD testing or were evaluated for a prior positive MCD test at our clinic between May 2023 and December 2024. Pts obtained the test through: 1) self-payment after shared decision making, or 2) an early access demonstration project with GRAIL, Point32Health (insurance company), and MGB whereby pts ≥50y with specific insurance products and an MGB primary care physician were invited to receive a cost-free test if they met ≥1 of 4 criteria: personal history of cancer, 1st-degree family history of cancer, elevated BMI, or any smoking history in the past 10 years. All pts tested at MGB underwent pre-test consultation with a medical oncologist or nurse practitioner to review the benefits and limitations of testing. Result management was coordinated by our team. Longitudinal outcomes were abstracted from electronic health records. Results: 742 pts had MCD testing (125 [17%] self-pay, 617 [83%] demonstration project). Median (med) age was 59 (IQR 55-63); most were White (92%) and up to date on USPSTF-recommended cancer screening (82%). Med follow up (FU) was 469 days (IQR 462-478); 580 (78%) had > 12 months FU. Four pts received a “cancer signal detected” result (+MCD): 2/742 tested at MGB (0.3%) and 2 who presented to workup +MCD tested elsewhere; 2 were true positives (TP), 2 false positives. In total, 16 pts (2.2%) were diagnosed (dx) with cancer during FU; 2/16 (13%) after +MCD, and 14/16 after negative MCD, with med time from MCD to diagnosis 199 days (IQR 92-464). The two TP tests corresponded to stage I breast cancer and stage III follicular lymphoma. The most common cancers dx were prostate (n = 5, with two stage III-IV dx at 128 and 142 days) and breast (n = 4, all mammogram-detected stage I). Other advanced cancers included stage III cholangiocarcinoma and stage IV adenocarcinoma unknown primary (dx at 515 and 475 days, respectively). Conclusions: In this real-world cohort, MCD test positivity was rare, and multiple cancers were dx within 18 months of a negative MCD, including some advanced cases. In particular, prostate cancer was dx in multiple pts after negative MCD, consistent with prior reports of limited MCD sensitivity for prostate cancer. Our cohort may be biased by the requirement to either self-pay or carry a specific insurance product. Larger studies with longer FU are needed to better define the added utility of MCD tests to standard cancer screening.