OBJECTIVES

Complete response (CR) is one of the main criteria for judging effectiveness of treatments for indolent non-Hodgkin’s lymphoma (iNHL). Objective response (ORR) has also been used for regulatory evaluation. A third outcome is the percent reduction in sum of the products of the longest perpendicular diameters (SPD), which is one of the criteria to determine the complete and overall response and indicates the collective change in disease status for the population rather than the proportion who reach a specific threshold. Here, we determine whether the choice of which treatment is better differs based on the efficacy measure being compared.

METHODS

Clinical trials for the relapsed/refractory iNHL treatments ibrutinib (ibr), copanlisib (cop), idelalisib (idel) were identified reporting CR, ORR, and percent SPD reduction for each individual patient. The number of patients and number of responders for each treatment were estimated by beta distributions to estimate the probability of each treatment being better than the others using CR and ORR as decision criteria. The probability of each treatment being better using the SPD outcome was estimated from 10,000 bootstrap samples of the digitized patient-level SPD reductions.

RESULTS

Cop was reported to have the highest CR (12%) and ORR (59%). Consequently, it was estimated to have 78% and 97% probability of being better than ibr and idel when comparing CR, whilst 97% and 70% when comparing ORR. Comparing ibr and idel, the probability of the former being better was 84% and 9% for CR and OR. Using the SPD outcome, cop was also estimated to be better than ibr (94%) and idel (84%), and idel was estimated to be better than ibr with 75% probability.

CONCLUSIONS

The discordance in comparisons among iNHL treatments when using 3 common clinical outcome measures demonstrates the need to evaluate all potential outcomes for comprehensive decision making.