Clinical trials (TEMPO 3:4 [NCT00428948]; REPRISE [NCT02160145]) demonstrated that TLV significantly slowed the decline in estimated glomerular filtration rate (eGFR) in ADPKD. Given the slight difference in TLV treatment effects between studies, a post hoc analysis evaluated whether the effects of TLV on eGFR were attenuated by the higher rate of TLV discontinuation in TEMPO 3:4 (23.0%) compared to REPRISE (15.0%). In TEMPO 3:4, eligible subjects (aged 18-50 years) had earlier-stage chronic kidney disease (CKD) associated with ADPKD (>80% had stage 1 or 2 CKD) with high risk of disease progression and were randomized to TLV or placebo (PBO) for 36 months. REPRISE randomized an older ADPKD population (aged 18-65 years) with more advanced CKD (75% had stage 3 and 20% had stage 4 CKD) to 12 months of TLV or PBO. For this analysis, a subgroup of treatment completers was identified, defined as participants who had 1) taken ≥70% of prescribed doses, 2) not discontinued study drug for ≥30 days, and 3) remained on study drug to end of study. The treatment effect of TLV vs PBO on kidney function (for TEMPO 3:4, slope of change in eGFR; for REPRISE, annualized change in eGFR) was compared between the subgroup of TLV completers and PBO completers within each study. Further, the treatment effect of TLV vs PBO on eGFR slope was compared between subjects from TEMPO 3:4 and REPRISE who were matched by propensity score for gender, age, and baseline eGFR. The number of subjects from the total population of each study who met completer criteria was 713/961 (74.2%) in the TLV arm and 410/484 (84.7%) in the PBO arm of TEMPO 3:4 and 576/683 (84.3%) in the TLV arm and 634/687 (92.3%) in the PBO arm of REPRISE. Rates of eGFR decline (in ml/min/1.73 m2) in TLV-treated subjects were similar between the total study population and the completer subgroup in both TEMPO 3:4 (total population, -2.72; completers -2.74) and REPRISE (total population, -2.34; completers -2.38), yielding similar treatment differences between completers and the total population within each study (Table). Propensity score matching generated an analysis set of 108 subjects for each study (54 in each of the TLV and PBO arms). The mean baseline eGFR for the matched subjects from each study (~53 ml/min/1.73 m2) fell within CKD stage 3a. In the matched population, the treatment difference (TLV vs PBO) was 1.41 ml/min/1.73 m2 for TEMPO 3:4, and 1.89 ml/min/1.73 m2 for REPRISE (P=NS for the comparison between studies). TLV effect size on eGFR was similar between study completers and the overall study populations within each of TEMPO 3:4 and REPRISE. Propensity-matched subjects from both studies with similar baseline CKD (i.e., within the area of overlap in the study populations between subjects with early to moderate stage CKD [TEMPO 3:4] and moderate to late stage CKD [REPRISE]) exhibited similar treatment effects. These results suggest that differences in TLV treatment effects between TEMPO 3:4 and REPRISE were due less to study drug adherence level than to variances in subject CKD stage distribution between the studies. Consistent with earlier data (Torres VE et al Clin J Am Soc Nephrol 2016;11:803-11), these results indicate that the effects of TLV in slowing eGFR decline are most easily discernible in later-stage patients, given that kidney function decline accelerates with ADPKD progression.