Mammography utilization among women with a negative circulating tumor DNA-based early cancer detection test.

Abstract

1563 Background: Blood-based tests may enable minimally invasive detection of multiple cancer types. One such test, CancerSEEK, employs ctDNA and protein biomarkers for this purpose. Test performance has been evaluated in women without a history of cancer in an ongoing prospective study called DETECT-A. The introduction of such blood tests holds promise, and their future utility lies in augmenting, not displacing, standard-of-care (SOC) cancer screening. One important safety concern is that a negative test result could provide false reassurance that discourages adherence to SOC cancer screening. To investigate this possibility, we studied delivery of mammography to DETECT-A participants before and after receipt of a negative CancerSEEK result. Methods: DETECT-A screened 10,000 women aged 65-75 using CancerSEEK. Participants completed a survey about cancer screening at enrollment and at one-year post-enrollment. We analyzed only those participants who had received a negative CancerSEEK result, were insured by Geisinger Health Plan (GHP), and had completed both surveys. GHP claims data were used to identify mammograms performed within one year prior-to and post-enrollment. Overall utilization was determined by combining claims and survey data at enrollment and one-year post-enrollment. In addition to comparing SOC screening rates pre- versus post-testing, we evaluated the impact of primary care physician (PCP) type (Geisinger versus any other institution), as screening reminder mechanisms differ between institutions. Results: Of the 2,241 participants who met analysis criteria, 73.6% (n = 1,650) had a mammogram in the year before enrollment while a significantly great number (79.3%, n = 1,777) did so during the one-year follow-up (χ2(1) = 59.05, p < 0.001). At enrollment, there were 591 participants who had not had a mammogram completed in the previous year, but 404 (68.4%) of them did have a mammogram during the one-year follow-up. The rate of change in mammography utilization did not differ between those who had a Geisinger versus a non-Geisinger PCP (χ2(2) = 1.83, p = 0.40). Conclusions: Participants in a study using a novel blood test for earlier cancer detection had a significantly higher rate of annual mammography after study enrollment and testing. These results indicate that introduction of a minimally invasive ctDNA and protein biomarker-based cancer screening test may engender greater, not lesser, utilization of SOC cancer screening. Further study is required to understand the root causes of increased utilization in this context.