Background BRUIN CLL-314 (NCT05254743) is a global phase 3 open-label randomized controlled trial (RCT) comparing pirtobrutinib with ibrutinib (ibr) in both treatment-naïve (TN) and relapsed/refractory patients (pts) with CLL/SLL and no prior BTKi exposure. This study met its primary endpoint of noninferiority of overall response rate (ORR) versus ibr. As this is the first head-to-head study of BTK inhibitors in the TN setting, a network meta-analysis (NMA) was conducted to contextualize the ORR and PFS of pirtobrutinib vs other therapies. Unlike a matching-adjusted indirect comparison (MAIC), an NMA preserves randomization to enable more robust comparisons across a network of RCTs through common comparators. As a result, NMAs are reproducible and widely accepted.

Methods A systematic literature review (SLR) identified RCTs of pts with BTKi-naive CLL/SLL across multiple databases. All studies in the TN setting included within the network connected to BRUIN CLL-314, provided at least one of the studied therapies were recommended in the US National Comprehensive Cancer Network (NCCN) Guidelines for CLL (version 3.2025). A Bayesian NMA was performed to compare independent review-assessed ORR and PFS between BRUIN CLL-314 and other RCTs identified. ORR was evaluated by calculating odds ratios (ORs) from published rates of partial response or better for each treatment arm, while PFS was evaluated using hazard ratios (HRs) with 95% confidence intervals.

Data extraction for each included RCT was performed by one abstractor and independently validated by two additional abstractors using the most recent publications containing the data required for analysis. The NMA was conducted via Markov Chain Monte Carlo simulations (80k burn-in iterations; 480k total samples; thinning rate of 24). Heterogeneity, consistency, and model fit were assessed using residual deviance and the deviance information criterion. An independent random effects model was applied in base case, and all analyses were performed using the R2JAGS package.

Results The TN network consisted of nine RCTs including a total of 4,072 participants with TN CLL: BRUIN CLL-314, SEQUOIA, ALLIANCE- A041202, RESONATE-2, CLL13, ELEVATE-TN, CLL14, GLOW, and AMPLIFY. This resulted in two disconnected networks of NCCN-recommended therapies: Network1 included pirtobrutinib, ibr, and zanubrutinib (zanu); Network2 included venetoclax (ven) + obinutuzumab (obi), ven + ibr, acala (acala), acala + obi, acala + ven, and acala + obi + ven.

In Network1, pirtobrutinib had an 87.5% probability of being ranked first for ORR (surface under the cumulative ranking curve [SUCRA]=96.8%), followed by zanu (5.7% probability of being first, SUCRA=64.2%). OR point estimates for ORR were >1, favoring pirtobrutinib vs all comparators. OR (95% credible interval [CrI]) results for pirtobrutinib vs covalent (c) BTKi monotherapies were: 2.57 (0.71-9.58) vs zanu and 2.20 (0.91-5.72) vs ibr. For PFS, pirtobrutinib had an 82.3% probability of being ranked first (SUCRA=95.0%), followed by zanu, with a probability of 13.5% (SUCRA=77.1%). PFS results demonstrate favorable HRs (<1.0) for pirtobrutinib vs all comparators. HR (95%CrI) of pirtobrutinib vs cBTKi monotherapies were: 0.62 (0.25-1.53) vs zanu and 0.51 (0.24-1.07) vs ibr.

Network2 may be connected to Network1; however, this would require inclusion of a study conducted >13 years ago and connection through a regimen that is no longer an NCCN-recommended therapy. Given these limitations and uncertainty in potential bias that may be incurred from including an older study in which treatment practices and supportive care may have changed, Network2 was not connected to Network1 and no indirect comparison to pirtobrutinib could be made.

ConclusionThis Bayesian NMA provides preliminary evidence supporting that pirtobrutinib's efficacy in the treatment-naive setting is comparable to available cBTKi monotherapy and is consistently ranked as the optimal treatment with high probability. While the efficacy of pirtobrutinib as measured by ORR and preliminary PFS generate favorable point estimates relative to currently available therapy (i.e., ORs > 1 and HRs < 1), caution is needed when interpreting the data due to the wide credible intervals and the immaturity of PFS data for BRUIN CLL-314 at the time of this analysis. Further work is needed to establish a comprehensive NMA that could incorporate all available non-chemotherapy targeted regimens in the first-line setting.

This content is only available as a PDF.
2025
Sign in via your Institution