An adjusted indirect comparison of everolimus and sorafenib therapy in sunitinib-refractory metastatic renal cell carcinoma patients using repeated matched samples

Abstract

Results suggest that sunitinib-refractory metastatic renal cell carcinoma patients treated with everolimus may experience significantly improved OS outcomes compared to those treated with sorafenib. However, because this is not a randomized controlled trial, the results should be interpreted as those from an observational study.

Treatment patterns in metastatic renal cell carcinoma: a retrospective review of medical records from US community oncology practices

“…These results are consistent with other studies of real-world data 15,[23][24][25] and also indicate that sequential use of VEGF inhibitors results in cumulative toxicities, a situation which may pose management challenges for patients and physicians in the community setting. The impact of cumulative toxicity warrants further investigation in future research on treatment sequencing.…”

Comparative Efficacy of Ceritinib and Crizotinib as Initial ALK –Targeted Therapies in Previously Treated Advanced NSCLC: An Adjusted Comparison with External Controls

“…Thus, while propensity score matching of external controls cannot replace a direct head-to-head randomized comparison, it can provide some preliminary insights to comparative efficacy. 11,[21][22][23] In the present study, the comparison of ceritinib patients to external controls who received crizotinib, in which all patients were followed in clinical trial settings and adjustments were made for multiple patient characteristics, represents the best available comparative evidence for these treatments as initial ALK-targeted therapies. …”

“…13,14,[18][19][20] In the absence of head-to-head RCTs, comparative efficacy in advanced oncology settings, including ALK+ NSCLC, is often based on comparisons to external controls. 11,[21][22][23] One technique that can reduce bias in such comparisons is the use of propensity score matching to achieve baseline similarity across treatment groups. These steps include a comparison of trial designs and study settings, selection of patients based on shared inclusion/exclusion criteria, and matching of multiple known baseline prognostic factors based on a propensity score model.…”

A review of methods for comparing treatments evaluated in studies that form disconnected networks of evidence

“…Five articles were excluded because they were one of the 11 used in the citation searching process. Of the remaining 36 articles, full text articles were reviewed during which a further 8 articles were excluded because they discussed the general issue of comparative effectiveness or were about specific clinical aspects, leaving 28 articles …”