A Case-Control Study Brings to Light the Causes of Screen Failures in Phase 1 Cancer Clinical Trials

Abstract

IntroductionEnrolling cancer patients in phase I clinical trials (P1s) requires that they fulfill specific criteria. Between the time they sign the consent form and the 1st administration of the experimental drug, some patients may be excluded and considered as screen failures (SFs). Our objective was to assess SF patients profiles and the reasons and risk factors for SFs.Materials and MethodsAll patients included in P1s at Gustave Roussy from 2008 to 2013 were reviewed retrospectively. SFs were matched with control P1 patients who were successfully enrolled. Patient and tumor characteristics, P1 types and the reasons for SF were analyzed.ResultsAmong 1,293 patients, 192 (15%) were SF cases; 182 SF cases were matched with 182 controls: median age was 57 (48–64) and 55 (47–63), median home-cancer center distance was 69 vs 55 km, 45% vs 34% had more than 2 metastatic sites, median screening period was 14 vs 11 days, median progression-free survival during the previous line was 12 vs 14 weeks, 37% vs 29% of LDH values were above the upper limit of normal, 42% vs 36% of albumin values were < 35 g/L, respectively. Reasons for SFs were cancer progression (44%), sponsor decision unrelated to a clinical reason (25%), patient retrieval (13.5%), relevant comorbidity (13.5%). Multivariate analysis revealed that a high Royal Marsden Hospital (RMH) prognostic score was potentially associated with higher risk of SFs (OR = 2.3; 95% CI [1.0–5.7], p = 0.06).ConclusionCancer progression led to half of the SFs in P1s. Physicians should pay attention to the RMH score at the time of patient inclusion to avoid further SFs.

Plasma total cell-free DNA is a prognostic biomarker of overall survival in metastatic solid tumour patients

“…An improved risk assessment could support the clinical decision when enrolling patients in phase I studies as early discontinuation of patients in phase 1 trials most frequently are due to progression and clinical deteoriation. 21 The clinical implementation of the proposed model therefore has the potential to improve selection of patients to phase I trials. Furthermore, low plasma concentration of cfDNA at baseline is correlated with a higher chance of receiving further treatment.…”

Natural language processing for patient selection in phase I/II oncology clinical trials

“…We could not directly compare survival scores (RMH or GRIm) to our model prediction which is a binary screening status, independent from the time. However, in a cohort similar to ours, RMH score wrongly discarded half of potential good candidates for inclusion in early-phase clinical trials 6 . This is of major concern, because early-phase clinical trials was shown to improve clinical outcomes for up to 30% of patients 27 The screen failure rate observed in this study (including screening and DLT period) may seem high (about 40%), but was consistent with previous reports and can be explained by the known tendency of physicians to be overoptimistic regarding their patients prognosis 7,28 .…”

“…The rate of patients not eligible to the trial after this period -defining screening failureusually scales between 10% and 30% [5][6][7] . For patients who actually start the study treatment, there is an additional crucial period -defined as the dose-limiting toxicity (DLT) evaluation period -that usually prolongs of another 28 days the length of time during which if a patient is withdrawn from the trial and not evaluable for DLT (e.g.…”

“…Thus, early discontinuation directly extends the duration of the study and majors development-associated costs. Altogether, these early discontinuation events have been described in approximately 40% of phase I patients and are mainly linked to detrimental cancer progression 6 . Those are largely unpredictable by the current selection methods available.…”

Academic Cancer Center Phase I Program Development

“…A large patient base—17,000 new cancer patients in 2015—aided referrals for site #6. Screen failures are common and occur due to cancer progression during the screening period and/or poor performance status . The Royal Marsden Hospital prognostic score has been validated by an MD Anderson Cancer Center phase I clinic .…”