Paper Details

ABSTRACT

Introduction: We conducted an RWE study of pirto in R/R CLL to validate results of the BRUIN and BRUIN-CLL-321 trials.

Methods: In this retrospective cohort study, patients with CLL who at any point had received a cBTKi and then received pirto outside of a clinical trial were eligible. Patients with a history of Richter transformation before pirto were excluded. The primary endpoints were event-free survival (EFS) and the best overall response rate (ORR) by investigator assessment. Secondary endpoints were time-to-next treatment (TTNT), overall survival (OS), and grade ≥3 adverse events (AE) of interest. EFS was defined as the time from pirto start to disease progression (PD), start of a new line of therapy (LOT), or death, whichever occurred earlier. The log-rank test was performed to compare survival curves. A multivariate (MV) Cox regression model was used to calculate adjusted hazard ratios (aHR) and 95% confidence intervals (CI).

Results: The study included 266 patients treated at 20 US centers. One hundred and seventy-six (66%) were male, with a median age of 72 years (47–95) at pirto start; 233 (88%) were White non-Hispanic, 19 (7%) Black, 5 (2%) Asian, and 3 (1%) Hispanic. IGHV mutation status was available for 178 (67%) patients, and 145 (81%) had unmutated IGHV. Of the 249 (94%) patients who had TP53 data available, 116 (47%) had either a del(17p) or a TP53 mutation (37, 15%, had both). NGS prior to pirto start was available for 133 (50%) patients: 57 (43%) had ≥1 BTK mutation, including 56 C481 (49 C481S and 7 other C481X) and 6 T474I; 5 patients had a PLCG2 mutation (4/5 known hotspot). Patients received a median of 4 (range 1–15) previous LOTs, and 29 (11%) patients received pirto as the 2nd LOT after a cBTKi (17 [6%] after PD on a cBTKi). All had prior cBTKi, including 204 (77%) ibrutinib, 118 (44%) acalabrutinib, and 37 (14%) zanubrutinib. Other LOT included chemo-immunotherapy in 169 (64%) patients, venetoclax (ven) ± an anti-CD20 monoclonal antibody (mAb) in 172 (65%) patients, and CAR-T in 17 (6%) patients; 139 (52%) patients were refractory to their last LOT.

Median cumulative exposure to prior cBTKi was 51 months (range 0.3 – 122). One hundred ninety-seven (74%) stopped the cBTKi for PD, 51 (19%) for AE, and 18 (7%) for other reasons. Pirto was started a median of 12 months (0–115) after stopping cBTKi. The starting dose was 200 mg daily in 96% of patients. Forty-two (17%) patients received pirto in combination: 24 with ven ± an anti-CD20 mAb, 15 with anti-CD20 mAb, and 3 with other agents. Eighty-seven patients (33%) developed grade (G) ≥3 AE: 43 (16%) infectious, 38 (14%) grade 4 neutropenia, 15 (6%) cardiac (including 9 atrial fibrillation/flutter), and 11 (4%) bleeding AE. Thirty-one patients (12%) required a dose reduction, and 112 (42%) stopped pirto: 51 (19%) due to PD, 22 (8%) due to AE, and 23 (9%) as they completed bridging.

Responses were available for 227 (86%). Best ORR was 64% (clinical complete response rate 10%). At a median follow-up of 15.4 months (95% CI 13 – 18), the 12-month EFS and OS were 66% (60%–72%) and 83% (77%–88), respectively. The 1-year cumulative incidence of initiating next LOT was 21% (26%–16%). In patients who responded to pirto, the 1-year EFS was 80% (72%–86%), while in those who received pirto as 2^nd LOT after cBTKi, the 1-year EFS was 89% (69%–96%). In double-exposed patients, the 1-year EFS was 51% (42%–59%). On univariate analysis, double-exposed disease, PD as best response to cBTKi, shorter cumulative time on cBTKi (continuous), and refractoriness to last LOT predicted worse EFS (p<0.05), while patients who received pirto as 2^nd LOT had improved EFS (p<0.05). Conversely, age (by quartiles), TP53 aberration, IGHV status, and reason to stop cBTKi (PD vs non-PD) were not predictive of EFS (p >0.05). In a MV model (with double-exposed disease, PD as best response to cBTKi, cumulative time on cBTKi, and pirto as 2^nd LOT), PD as best response to cBTKi (aHR 2.52, 1.22-5.18), shorter time on cBTKi (aHR 1.01, 1.01-1.02), and pirto as 2^nd LOT (aHR 0.29, 0.10-0.83) were significant predictors of EFS.

Conclusions: In this RWE study of heavily pretreated patients with R/R CLL, safety and efficacy of pirto were consistent with those of the pivotal trials, with low rates of discontinuation due to AEs. Double-exposed disease and inferior efficacy of cBTKi negatively predicted pirto efficacy, while patients who received pirto as 2^nd LOT after cBTKi had improved outcomes.