Screen failure among patients consenting for early phase cancer clinical trials (EPCTs).

Abstract

107 Background: EPCTs represent an important treatment option for patients with cancer. However, trials often have strict eligibility criteria and screen failures (SF) can limit enrollment. We sought to identify reasons for SF among potential EPCT participants, evaluate associations between SF and overall survival (OS), and describe use of palliative care (PC) and subsequent therapies among patients who had a SF. Methods: Using an institutional database, we identified adult patients who signed consent for an EPCT between 1/2021 – 10/2025 but did not receive treatment on the trial. We retrospectively reviewed the electronic health record to obtain patient demographics, clinical characteristics, reasons for SF, receipt of PC, advanced care planning (ACP), future therapies, and OS. Results: The table provides SF reasons. Among 273 identified patients (median age = 64.1 years [range 18.4 – 90.2], 52% female), most common SF reasons were clinical decline (CD; 28%) and lab abnormality (19%); table excludes 6% with unknown reason. Patients with CD had increased risk of death (HR 2.3, p<.001) compared to other SF reasons, while those with SF due to patient preference and lack of measurable disease had longer survival (HR 0.4, p=.001; HR 0.2, p<.001; Table). Among all patients, median OS was 3.9 months (95% CI 3.3-5.1), with median follow up of 26.1 months (95% CI 22.9-29.8); survival rates at 1 and 2 years were 18% and 8%, respectively. Most patients (66%) were referred internally and 48% had received 3+ lines of therapy; 30% had participated in a prior EPCT. Among internal referrals, most did not have prior PC (70%), ACP documentation (70%), or advance directive (92%). Most patients (64%) received more therapy after SF, including 11% who enrolled on another EPCT; 5% had unknown subsequent treatment. Of those with CD, 49% pursued hospice/supportive care, 36% received another line of therapy, 11% died during screening. Among all other patients, 13% pursued hospice, 81% another line, and 1% died. Conclusions: Reasons for SF vary, with CD and lab abnormality as the most common. We found that most patients had no prior PC receipt or ACP documentation, despite having advanced disease. Notably, most patients received further therapies after SF and some experienced extended survival. Further work is needed to better understand and address SF reasons to improve care for this population. Reason for SF N (%) Median OS (months, 95% CI) OS HR (95% CI) P Clinical decline 76 (28) 1.6 (1.2 - 5.6) 2.3 (1.7 - 3.1) <.001 Lab abnormality 52 (19) 3.9 (2.2 - 7.8) 1.1 (0.8 - 1.6) .450 Patient preference 28 (10) 20.2 (3.4 - NR) 0.4 (0.3 - 0.7) .001 Multiple reasons 23 (8) 2.7 (1.3 - 4.0) 1.4 (0.9 - 2.3) .148 New CNS disease 22 (8) 3.9 (2.5 - 6.3) 1.3 (0.8 - 2.1) .248 Comorbidity 19 (7) 6.3 (3.9 - NR) 0.7 (0.4 - 1.2) .221 Trial logistics 14 (5) 6.0 (2.6 - 11.7) 0.9 (0.5 - 1.5) .587 Medication interaction 12 (4) 11.2 (2.1 - NR) 0.6 (0.3 - 1.3) .179 No measurable disease 11 (4) NR (14.0 - NR) 0.2 (0.1 - 0.5) <.001