Adenoma detection by Endocuff-assisted versus standard colonoscopy in routine practice: a cluster-randomised crossover trial.

OBJECTIVE: Endocuff Vision (ECV) is the second generation of a device designed to improve polyp detection. The aim of this study was to evaluate its impact on adenoma detection rate (ADR) in routine colonoscopy. DESIGN: This cluster-randomised crossover trial compared Endocuff-assisted (ECV+) with standard (ECV-) colonoscopy. Two teams of 11 endoscopists each with prior ECV experience, balanced in terms of basal ADR, gender and case volume were compared. In randomised fashion, the teams started with ECV+ or ECV- and switched group after inclusion of half of the cases. The main outcome criterion was ADR difference between ECV+ and ECV-. Subgroup analysis was done for physicians with low and high ADR (< or ≥ 25%). RESULTS: During two periods of 20 and 21 weeks, respectively, the 22 endoscopists included 2058 patients (1032 ECV- vs 1026 ECV+, both groups being comparable). Overall ADR for both groups taken together was higher with ECV (39.2%) than without (29.4%; p<0.001) irrespective of the sequence of use (ECV+ or ECV- first), but mostly in adenomas <1 cm. In the physician subgroup analysis, only high detectors showed a significant ADR increase (from 31% to 41%, p<0.001), while the increase in the low detectors was not significant (from 24% to 30%, p=0.11). ECV had a positive impact in all colonic locations, except for the rectum. No ECV- related complication was reported. CONCLUSION: We observed a significant ADR difference of approximately 10% by the use of ECV. By subgroup analysis, this increase was significant only in physicians classified as high detectors. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov (NCT03344055).

[Decision making process in oncology: rationale and delineation of a composite index of relative antitumoral efficacy (In-RATE)]. / Critères de choix thérapeutique en oncologie : définition et intérêt d'un nouvel index d'efficacité relative des agents antitumoraux (In-RATE).

Over the last decade, the development of new therapeutic options has made more patients benefit from antitumoral strategies including several lines of chemotherapy, the aim of which is a long-term control of the disease progression. In such a context of "chronic" management, the choice of tumor response as a single parameter appears restrictive to assess those new therapeutic options. For that reason, we have recently proposed a composite index of relative efficacy including response rate as well as parameters related to tumour stabilization and duration of the response. The objective of this index, published as the In-RATE is to allow the comparison of two treatments a and b as follows: In-RATE a/b = (response rate a/response rate b) x (time to progression a/time to progression b) x (progression rate b/progression rate a). Values significantly higher or less than 1 suggest the superiority, in terms of efficacy, of treatments a or b, respectively. When retrospectively applied to randomised studies, the In-RATE showed that some results and conclusions based on the response rate as a unique endpoint might be reconsidered, and that a significant difference between protocols could be detected in published reports having concluded to statistical equivalence. This paper reviews the rationale and principle of this work, and discusses the potential clinical applications of the In-RATE.

Rationale and delineation of a composite index of relative antitumoural efficacy (In-RATE).

Over the last decades, the development of new drugs has allowed cancer patients to experience several lines of chemotherapy, the objective of which is a long term stabilization of the tumour. The objectives of this work was to delineate a composite index of relative antitumoural efficacy (In-RATE) of a regimen over another, including response rate (RR), median time to progression (TTP) and progression rate (PR). When considering two treatments a and b, the In-RATE was defined as RRa/RRb x TTPa/TTPb x PRb/PRa. Values significantly superior or inferior to 1 reveal an advantage for treatment a or b, respectively. The applicability of the In-RATE to published randomized trials in four frequent tumour types (colorectal, non-small cell lung, advanced ovarian and metastatic breast cancers) was suggested to more precisely distinguish the effects of different drugs, and sometimes to detect a significant difference when the published data did not conclude to statistical difference.