Objective: We aimed to systematically review the extent of reporting and definition heterogeneity.
Methods: We included randomized controlled trials (RCTs) identified from SRs comparing adjuvant treatments after TURBT or TURBT alone in patients with NMIBC (with or without carcinoma in situ) published between 2000-2020. Abstracts and full texts were screened independently by two reviewers. Data were extracted by one reviewer and checked by another.
Results: We screened 807 abstracts; from 15 SRs, 57 RCTs were included. Verbatim outcome names were coded to standard outcome names and organised using the Williamson and Clarke taxonomy. Recurrence (98%), progression (74%), treatment response (in CIS studies) (40%), and adverse events (77%) were frequently reported across studies. However, overall (33%) and cancer-specific (33%) survival, treatment completion (17%) and treatment change (37%) were less often reported. Quality of Life (3%) and economic outcomes (2%) were rarely reported. Heterogeneity was evident
throughout, particularly in the definitions of progression and recurrence, and how CIS patients were handled in the analysis of studies with predominantly papillary patients, highlighting further issues with the definition of recurrence and progression vs treatment response for CIS patients. Data reporting was also inconsistent, with some trials reporting event rates at various time-points and others reporting time-to-event with or without Hazard Ratios. Adverse events were inconsistently reported. QoL data was absent in most trials.
Conclusions: Heterogenous outcome reporting is evident in NMIBC effectiveness trials. This has profound implications for meta-analyses, SRs and evidence-based treatment decisions. A core outcome set is required to reduce heterogeneity.
|Publication status||Accepted/In press - 23 Feb 2021|