Treatment exposure anddiscontinuation in the PALLAS trial: PALbociclibCoLlaborative Adjuvant Study of palbociclib with adjuvantendocrine therapy for HR+/HER2- early breast cancer

Background :Adherence to and tolerability of oral agents for the treatment of hormone receptor positive (HR+)breast cancer in the adjuvant setting may be challenging and can limit drug exposure. Palbociclib (P) is an oralCDK4/6 inhibitor; P in combination with endocrine therapy (ET) has demonstrated efficacy in HR+/HER-2 negative(HER2-) metastatic breast cancer (MBC). The global PALLAS study ( NCT02513394 ) was designed to determine ifthe addition of P to adjuvant ET improves outcomes over ET alone in early breast cancer. The goal of this analysisis to describe P exposure and discontinuation in PALLAS, and explore any impact on study endpoints. Methods :Pts with stage II-III HR+/HER2- disease were randomized to receive 2 years of P (starting dose 125 mg daily, 3 weekson, 1 week off) in combination with adjuvant ET (Arm A) or ET alone (Arm B). The primary objective was to compareinvasive disease-free survival (iDFS) between arms; secondary endpoints included safety, quality of life, adherence,and translational science. Dose adjustments were defined per protocol in the setting of emergent toxicity.Continuous monitoring of toxicity, dose modifications, and early discontinuations was performed throughout the trial.Statistical analysis included tabulation of dose levels/reductions and reasons for treatment discontinuation, as wellas landmark analysis comparing iDFS between those receiving > 12 months (mo) vs < 12 mo of P. Ongoinganalyses include exploration of relative dose intensity and modeling of P dose intensity/duration and impact on.iDFS. Results:A total of 5760 pts were randomized; 83% had received prior chemotherapy; 68% initiated aromataseinhibitor and 33% tamoxifen, with or without ovarian suppression. Grade 3/4 neutropenia was more common in ArmA vs B (62% vs 0.4%); febrile neutropenia was uncommon (1%). Other all-grade adverse events (AEs) morecommon in Arm A included other hematologic toxicity, fatigue, upper respiratory infection, nausea, diarrhea, andalopecia. A total of 55% of pts required P dose reduction to 100mg, and 34% to 75 mg, at some point duringtreatment. At a median follow-up of 23.7 mo at the second interim analysis, no significant difference in iDFS was observed between the arms (3-year iDFS of 87.9% vs 88.4%, HR 0.93, 95% CI 0.76, 1.15); consequently, pts inArm A stopped P, and all pts moved to long-term follow-up. At time of data cut-off, 32% had completed the planned2 years of P, 26% were still receiving P, and 42% had discontinued P prematurely. P discontinuation was 18% at 6mo, 30% at 12 mo, 38% at 18 mo, and 45% projected at 24 mo. A total of 27% of Arm A pts (770 of 2840)discontinued therapy due to AEs, primarily neutropenia (460, 16%) and fatigue (71, 3%). Other reasons for Pdiscontinuation included non-compliance (128, 5%), recurrent disease or second malignancy (104, 4%), or withdrawal of consent (100, 3%). Last observed dose level was 125mg, 100mg, and 75mg for 45%, 22%, and 33%pts, respectively. Among those discontinuing due to AEs, dose level at time of discontinuation was 75mg for 62%,suggesting some discontinuations occurred without maximum dose reduction. Pts who received > 12 mo of P had a2-year iDFS of 96.6%; those receiving <12 mo had 2-year iDFS of 95.7% (HR 0.86, 95% CI 0.55-1.34).Conclusions:An early analysis of PALLAS has demonstrated that 2 years of adjuvant P in combination withadjuvant ET did not improve iDFS when compared with ET alone in pts with HR+/HER2- breast cancer. While 2years of adjuvant P was feasible for the majority of pts, 42% discontinued P early, primarily for AEs. Further dataexamining drug discontinuations and relationships between drug exposure and clinical outcomes will be presented,expanding upon initial point estimates suggestive of a possible explanation of trial results.