The treating oncologist as a potential barrier to enrollment in molecular targeted trials.

407 Background: It has been hypothesized that the rarity of molecular inclusion criteria hampers trial accrual. In this study, we attempted to address this challenge in the context of a SETD2-mutation targeted trial. Methods: Patients who received molecular sequencing services from Tempus were analyzed for the presence of SETD2 molecular variants that matched trial enrollment criteria, and resided within 50 miles from the treatment center or were receiving treatment at clinical sites with strong referral histories. A written notification was issued to the ordering physician whenever a likely match was identified, and additional phone and email outreach follow up, including reasons for not pursuing the clinical trial, was conducted. Results: Over one year, 38 eligible patients were identified, none of which were enrolled. The most common reason for not being enrolled was cohort closure following initial identification and outreach. More specifically, cohort closure occurred < 2 months, 2-4 months, 4-6 months and > 6 months following identification of the mutation in 5, 8, 2 and 7 patients, respectively. Physicians for additional potential patients cited trial consideration as a future option, presence of prohibitive comorbidities, lack of interest, and patient death in 6, 3, 2 and 1 patients, respectively. Physicians for 4 patients did not respond to multiple follow up attempts. In regards to individual follow up, physicians received 0-1, 2-3, and 4 or more follow up outreach notifications regarding trial availability for 14, 19, and 5 patients respectively. Conclusions: Despite identification of a large number of potential trial candidates for a molecular targeted therapeutic trial through use of matching algorithms and extensive treating physician follow up, no patients were enrolled. This raises the hypothesis that the biggest barrier to enrollment, as in other settings, is treating physician motivation and that AI algorithms for identifying potential subjects are insufficient. Larger target populations than anticipated, better coordination between outreach efforts and enrollment status, and directed patient outreach may be necessary.