Session Item

Generalized pairwise comparisons (GPCs) have been proposed to simultaneously analyze several outcomes of any type (discrete, continuous, possibly censored). GPCs are especially useful when (1) the outcomes of interest can be prioritized (from clinically most important to least important), and (2) clinical thresholds are deemed relevant for some of these outcomes (for instance, survival gains should exceed 6 months to be considered clinically worthwhile). In randomized clinical trials comparing Treatment to Control, GPCs consist of comparing all possible pairs of patients formed by taking one patient from the Treatment group and one patient from the Control group. Each pair is classified as a win, a loss or a tie for the outcome of highest priority. Ties are then classified using the next outcome of lower priority, and the process is repeated until all outcomes have been analyzed. The Net Treatment Benefit (NTB) is defined as the number of wins minus number of losses divided by the number of pairs. NTB is an absolute measure that directly addresses patient-centric questions about the probabilities of benefits and harms from treatment. As such, GPC can potentially be used to individualize treatment choices. This approach may be useful to improve statistical power (because several outcomes are considered in the analysis) as well as to inform decision-making for cancer treatments that may cause serious adverse events.