OBJECTIVES: Unanchored ITCs should adjust for effect modifiers and prognostic variables when comparing treatment arms between single-arm trials or observational studies that do not share a common comparator. To inform future analyses, we performed a literature review to identify effect modifiers and/or prognostic factors included in unanchored ITCs for TCE RRMM.

METHODS: Medical Literature Analysis and Retrieval System Online (MEDLINE) and Excerpta Medica database (EMBASE) were queried to identify published unanchored ITCs comparing the efficacy of treatments for TCE RRMM patients (July 2022). Additionally, manual searches of American Society of Clinical Oncology, American Society of Hematology, and European Hematology Association conferences were performed for January 2020-July 2022. Data was extracted for publication type, trial-characteristics (population, sample size, and data type); analysis type (base case/sensitivity); characteristics adjusted for by population sub-setting or as model covariates; covariate selection process; and effective sample size (ESS).

RESULTS: A total of 36 unanchored ITCs in TCE RRMM were identified, representing 21 unique comparisons involving CARTITUDE-1 (N=10), MajesTEC-1 (N=5), KarMMa (N=3), DREAMM-2 (N=2), and STORM-2 (N=1). These analyses included comparisons to 15 different target populations. Only 11 analyses reported information on the covariate selection process. Base case analyses included between 3 to 15 covariates, most commonly adjusting for refractory status (N=20), number of prior lines (N=18), International Staging System (ISS)/Revised ISS disease stage (N=18), and cytogenetics risk profile (N=16). ESS values were only reported for 13 analyses, ranging from 33 to 248 in base case analyses.

CONCLUSIONS: There are important differences in unanchored ITCs evaluating comparative efficacy and safety of treatments for TCE RRMM in terms of the target population of interest, number of covariates, and covariates adjusted for, which limits our ability to compare results across analyses. Improved transparency for unanchored ITCs is required, including evaluation of structural uncertainty in terms of different covariate combinations.