Background: Idelalisib (IDELA) is effective in the treatment of patients (pts) with relapsed/refractory (R/R) CLL. Nonetheless, pts may discontinue (D/C) IDELA due to progressive disease (PD) or adverse events (AEs). In this post hoc analysis, we examined the outcomes of pts with R/R CLL who D/C IDELA and the relationship of these outcomes to IDELA exposure. Methods: Data were pooled for pts treated with IDELA + rituximab (NCT01539512, n = 110), IDELA monotherapy (NCT01539291, extension study), and IDELA + ofatumumab (NCT01659021, n = 173). Time-dependent endpoints, including time to next therapy (TTNT), time to PD (TTPD), and time to death (TTD), were calculated from the dates of permanent IDELA D/C. Kaplan-Meier analysis was used to estimate overall survival (OS) from the date of randomization. Subgroup analyses of OS were performed using the integrated safety data set. Results: Of 283 pts in the safety population, 124 (44%) remained on IDELA, 28 (10%) D/C due to PD, and 131 (46%) D/C for non-PD reasons (AEs, n = 87; “other” reasons, n = 44 [death, n = 7; study withdrawal, n = 21; physician’s decision, n = 14; other, n = 2]). Clinical outcomes for these D/C groups (excluding “other”) are presented in the table. Pts who D/C IDELA due to PD also had longer median IDELA exposure (11 vs 8 mo) than those who D/C due to AEs. Conclusions: Pts who were required to discontinue IDELA due to AEs survived no longer than those required to discontinue due to PD. OS was similar among pts who D/C IDELA, regardless of reason for D/C. Analyses of patient characteristics, type of PD and dose intensities associated with IDELA D/C, as well as outcomes based on timing of D/C due to AEs (early vs late), are ongoing. Clinical trial information: NCT01539512, NCT01539291, NCT01659021

NR = not reached.