The eREACH study: A randomized study of an eHEALTH delivery alternative for cancer genetic testing for hereditary predisposition in patients with metastatic cancers.

10502 Background: With FDA approval of targeted therapies in patients with germline BRCA1/2 -related advanced cancers there is a need to evaluate efficient and effective delivery models for germline cancer genetic testing. Methods: eREACH is a 4 arm non-inferiority study where traditional standard-of-care pre-test (visit 1) and post-test (visit 2) counseling delivered by a genetic counselor (GC) are replaced with a patient-centered eHealth (digital) intervention in patients with advanced or metastatic cancer. GC visits were offered by telehealth in the home. Arms include: A (GC/GC), B (GC/digital), C (digital/GC) and D (digital/digital). Those assigned to a digital visit could request a visit with a GC if preferred. Surveys were completed at baseline (T0), after visit 1 (T1), visit 2 (T2) and 6 months (T3). Primary non-inferiority outcomes are change in knowledge and anxiety (T0-T1, T0-T2). Secondary outcomes include uptake of testing, depression, cancer specific distress, responses to testing and satisfaction. We used an intention-to-treat approach (ITT) and as-treated approach (secondary). We used ANOVAs and chi-squared tests for hypothesis testing. For non-inferiority testing, we had additional rules based on the magnitude and sign of effects. Results: 229 participants were recruited from 14 states through Penn Medicine, community sites and social media, with 56-60 per arm.Participants were 35-91 YO (mean 67 YO),and37% were male, 17% were non-white and 43% had less than a college education. 70% were from academic sites, 21% from community sites and 9% from social media. Cancer types were: 52% breast, 26% prostate, 18% pancreatic, and 5% ovary. 173 (76%) of patients completed testing (12% had a positive result, 12% had a VUS). In our primary ITT analyses, we met the non-inferiority threshold for all primary and secondary outcomes except for knowledge (T0-T2) and uptake of testing. Increases in knowledge (T0-T2) were greater in Arms A-C as compared to Arm D, although differences were small (averages +2.03-2.54 v +1.23). Uptake of visit 1 was lower in the digital arms (A: 94.7%, B: 92.7% v. C: 76.7%, D: 82.5%), although uptake of testing after visit 1 met non-inferiority. 20% assigned to Arm C and 11% assigned to Arm D requested a GC. As-treated non-inferiority analyses were similar to the ITT results. Conclusions: Offering patient-centered digital delivery models with one digital visit and one visit with a genetic counselor is non-inferior to two visits with a genetic counselor for patients with metastatic cancer. Reminders, outreach or completion of digital platforms in clinic could address differences in uptake. The fully digital model may be associated with small differences in knowledge gain, although the clinical significance may be small., and longitudinal data could help inform the appropriateness of the fully digital model.