A self-assessment survey to identify the risk of screen failure in phase I clinical trials (SCITEP score).

Abstract

e18725 Background: Screen failures (SF) in early phase clinical trial are deleterious for patients, physicians and drug companies. It is therefore crucial to identify risk factors of SF, in order to optimize drug development and maximize patient’s benefit. Self- patient reporting is a new method of clinical evaluation based on the patient's experience that can meaningfully evaluate patient’s outcome. Methods: We created a self-assessment survey comprising 15 questions extracted from the ESAS, EQ-5D and PALLIA 10 scores, which evaluate patient’s (pts) symptoms and quality of life, each item being scored from 0 to 10, for a total of 0 to 150. Patients entering a phase I clinical trial at Gustave Roussy cancer center between January 10th and October 26th 2022 were surveyed at screening (T0), and subsequently every month (T1, T2,..) until end of treatment. Results: 227 of 275 pts entering a Ph1T were assessed between January 10th and October 26th 2022. Most frequent tumor types were lung (18%), GI (16%), hematological (10%) and gynecological cancer (9.3%); 56% were female. At data cut off (December 19th 2022), 25 pts were considered SF. There was no difference observed between SF pts and pts who went on with the trial in terms of ECOG, RMH score, albumin and LDH levels, number of previous lines of treatments and of metastatic sites at baseline (Table 1). At T0, there was a significant difference in the SCITEP score between SF pts and the others (median score 41 [IQR 26-63] vs 20 [IQR 10-38], p < 0.001). Internal validity of the score was good (cronbach alpha > 0.7). With a cutoff of ≥35, sensitivity of the survey to identify pts at high risk of SF was of 66.6% and specificity of 70.5%. Regarding overall survival, pts with a score ≥35 (n = 68) at T0 were at higher risk of death (survival rate at 3 months of 53% vs 65%, p < 0.01). Conclusions: Self-assessment of risk of SF by pts entering a phase I clinical trial is feasible with the SCITEP score. A score ≥35 seemed associated with a higher risk of SF and of death at 3 months. Independent cohorts are needed to revalidate this score as a tool to better detect pts at risk of SF and thereby optimize drug development. [Table: see text]