Search

PURPOSE Pirtobrutinib is a highly selective, non-covalent (reversible) Bruton tyrosine kinase inhibitor (BTKi). We report the safety and efficacy of pirtobrutinib in patients with covalent (c) BTKi pre-treated mantle cell lymphoma (MCL), a population with poor prognosis. PATIENTS AND METHODS Patients with cBTKi pre-treated relapsed/refractory MCL received pirtobrutinib monotherapy in a multicenter phase 1/2 trial (BRUIN, NCT03740529). Efficacy was assessed in the first 90 consecutively enrolled patients who met criteria for inclusion in the primary efficacy cohort. The primary endpoint was overall response rate (ORR). Secondary endpoints included duration of response (DOR) and safety. RESULTS Median patient age was 70 years (range, 46-87), median prior lines of therapy 3 (range, 1-8), 82.2% had discontinued a prior cBTKi due to disease progression, and 77.8% had intermediate or high risk sMIPI score. The ORR was 57.8% (95% CI, 46.9-68.1), including 20.0% complete responses (n=18). At a median follow-up of 12 months, the median DOR was 21.6 months (95% CI, 7.5-not reached). The 6- and 12-month estimated DOR rates were 73.6% and 57.1%, respectively. In the MCL safety cohort (n=164), the most common treatment-emergent adverse events (TEAE) were fatigue (29.9%), diarrhea (21.3%), and dyspnea (16.5%). Grade ≥3 TEAE of hemorrhage (3.7%) and atrial fibrillation/flutter (1.2%) were less common. Only 3% of patients discontinued pirtobrutinib due to a treatment-related AE. CONCLUSION Pirtobrutinib is a first-in-class novel non-covalent (reversible) BTKi, and the first BTKi of any kind to demonstrate durable efficacy following prior cBTKi therapy in heavily pre-treated relapsed/refractory MCL. Pirtobrutinib was well tolerated with low rates of treatment discontinuation due to toxicity.

Supplemental Table 1 and Supplemental Figure Legends

Purpose:Clinical trials of venetoclax reported negligible rates of clinical tumor lysis syndrome (TLS) in patients with chronic lymphocytic leukemia (CLL) when using an extended dose escalation schedule. We aimed to understand TLS prophylaxis, rates of select adverse events (AE), and impact of dosing modifications in routine clinical practice.Experimental Design:This retrospective cohort study included 297 CLL venetoclax-treated patients outside of clinical trials in academic and community centers. Demographics, baseline disease characteristics, venetoclax dosing, TLS risk and prophylaxis, and AEs were collected.Results:The group was 69% male, 96% had relapsed/refractory CLL, 45% had deletion chromosome 17p, 84% had unmutated IGHV, 80% received venetoclax monotherapy, and median age was 67. TLS risk was categorized as low (40%), intermediate (32%), or high (28%), and 62% had imaging prior to venetoclax initiation. Clinical TLS occurred in 2.7% of patients and laboratory TLS occurred in 5.7%. Pre-venetoclax TLS risk group and creatinine clearance independently predict TLS development in multivariable analysis. Grade 3/4 AEs included neutropenia (39.6%), thrombocytopenia (29.2%), infection (25%), neutropenic fever (7.9%), and diarrhea (6.9%). Twenty-two patients (7.4%) discontinued venetoclax due to an AE. Progression-free survival was similar regardless of number of dose interruptions, length of dose interruption, and stable venetoclax dose.Conclusions:These data provide insights into current use of venetoclax in clinical practice, including TLS rates observed in clinical practice. We identified opportunities for improved adherence to TLS risk stratification and prophylaxis, which may improve safety.

Purpose:Venetoclax-based therapy is a standard-of-care option in first-line and relapsed/refractory chronic lymphocytic leukemia (CLL). Patient management following venetoclax discontinuation remains nonstandard and poorly understood.Experimental Design:To address this, we conducted a large international study to identify a cohort of 326 patients who discontinued venetoclax and have been subsequently treated. Coprimary endpoints were overall response rate (ORR) and progression-free survival for the post-venetoclax treatments stratified by treatment type [Bruton's tyrosine kinase inhibitor (BTKi), PI3K inhibitor (PI3Ki), and cellular therapies].Results:We identified patients with CLL who discontinued venetoclax in the first-line (4%) and relapsed/refractory settings (96%). Patients received a median of three therapies prior to venetoclax; 40% were BTKi naïve (n = 130), and 81% were idelalisib naïve (n = 263). ORR to BTKi was 84% (n = 44) in BTKi-naïve patients versus 54% (n = 30) in BTKi-exposed patients. We demonstrate therapy selection following venetoclax requires prior novel agent exposure consideration and discontinuation reasons.Conclusions:For BTKi-naïve patients, selection of covalently binding BTKis results in high ORR and durable remissions. For BTKi-exposed patients, covalent BTK inhibition is not effective in the setting of BTKi resistance. PI3Kis following venetoclax do not appear to result in durable remissions. We conclude that BTKi in naïve or previously responsive patients and cellular therapies following venetoclax may be the most effective strategies.See related commentary by Rogers, p. 3501

Supplemental Figure 2: Progression free survival of patients without prior BTKi exposure who had progressed on venetoclax and were then treated with BTKi.

Supplemental Figure 1: Overall survival of entire cohort from initiation of venetoclax.

Background: Richter transformation (RT) is an aggressive diffuse large B-cell lymphoma that can occur in 10% of patients with chronic lymphocytic leukemia (CLL). RT has no approved therapies, poor prognosis, and an estimated median overall survival (OS) of 3-11 months. Pirtobrutinib is a well-tolerated highly selective, non-covalent (reversible) Bruton tyrosine kinase inhibitor (BTKi) with favorable oral pharmacology and promising efficacy in patients with poor-prognosis B-cell malignancies following prior covalent BTKi therapy. We now report on RT patients from BRUIN. Methods: Patients previously treated for B-cell malignancies were eligible for pirtobrutinib monotherapy in dose escalation/expansion within the phase 1/2 BRUIN study. Key endpoints: investigator-assessed overall response rate (ORR) and duration of response (DoR) per Lugano 2014 criteria, and safety. The response-evaluable cohort consisted of patients with RT who completed either first response assessment or discontinued therapy. The safety cohort consisted of patients who received ≥1 dose of pirtobrutinib monotherapy (n=725). Data cut was 31 January 2022. Results: Of 57 RT patients (median age: 67 years), 91% (n=52) received at least 1 prior RT-directed therapy. Median lines of prior RT therapy were 2: (anti-CD20 antibody [86%, n=49], chemotherapy [79%, n=45], BCL-2 inhibitor [35%, n=20], BTKi [28%, n=16], PI3K inhibitor [12%, n=7], CAR-T [9%, n=5], immunomodulator [5%, n=3], stem cell transplant [4%, n=2], other systemic therapy [33%, n=19]) and median lines of prior CLL were 2: (BTKi [60%, n=34], anti-CD20 antibody [60%, n=34], chemotherapy [47%, n=27], BCL-2 inhibitor [42%, n=24], immunomodulator [9%, n=5], PI3K inhibitor [7%, n=4], stem cell transplant [7%, n=4], CAR-T [2%, n=1], other systemic therapy [11%, n=6]). ORR for patients within phase 2 (n=56, 200mg once/day starting dose) was 54% (95% CI, 39-68): 5 complete responses, 22 partial responses. Median DoR was 8.6 months (95% CI, 1.9-NE, 63% censored) and median OS was 13.1 months (95% CI, 7.1-NE) at a median follow up time of 5.5 months and 9.7 months respectively. Six patients electively discontinued pirtobrutinib to pursue curative-intent therapy (allogeneic transplant, n=6, 3.8 months median time-on-therapy). The most frequent TEAEs were fatigue (26%, n=191), diarrhea (22%, n=160), and contusion (19%, n=138). Of Grade≥3 TEAEs, neutropenia (20%, n=143) was the most frequent and hypertension (3%, n=20), hemorrhage (2%, n=16), and atrial fibrillation/flutter (1%, n=7) were of low grade. Fifteen (2%) patients discontinued due to a treatment-related AE. Conclusions: Pirtobrutinib demonstrated promising preliminary efficacy and was well tolerated for patients who received prior RT-directed chemoimmunotherapy and covalent BTKi. Presented:ASH2022. Citation Format: Joanna Rhodes, David Lewis, Paolo Ghia, Nirav N. Shah, Catherine C. Coombs, Chan Y. Cheah, Jennifer Woyach, Nicole Lamanna, Marc S. Hoffmann, Shuo Ma, Toby A. Eyre, Talha Munir, Manish R. Patel, Alvaro J. Alencar, Constantine S. Tam, John F. Seymour, Wojciech Jurczak, Ewa Lech-Maranda, Lindsey E. Roeker, Philip A. Thompson, Paolo B. Abada, Chunxiao Wang, Binoj Nair, Hui Liu, Donald E. Tsai, Anthony R. Mato, William G. Wierda. Efficacy of pirtobrutinib, a highly selective, non-covalent (reversible) BTK inhibitor in Richter transformation: Results from the phase 1/2 BRUIN study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr CT167.

Neuroscience models commonly have a high number of degrees of freedom and only specific regions within the parameter space are able to produce dynamics of interest. This makes the development of tools and strategies to efficiently find these regions of high importance to advance brain research. Exploring the high dimensional parameter space using numerical simulations has been a frequently used technique in the last years in many areas of computational neuroscience. High performance computing (HPC) can provide today a powerful infrastructure to speed up explorations and increase our general understanding of the model's behavior in reasonable times.

Dissertation, RWTH Aachen University, 2023; Aachen : RWTH Aachen University 1 Online-Ressource : Diagramme (2023). = Dissertation, RWTH Aachen University, 2023, This dissertation aims to improve the process of translating (mapping) expressions containing matrices and tensors to high-performance code and thus contributes to the linear and multi-linear algebra domains, respectively. The two domains are examined separately. On the one hand, in the linear algebra domain, standardized building-block libraries (BLAS/LAPACK) have long existed, offering a set of highly optimized functions called kernels. The existence of those libraries has made the task of evaluating mathematical expressions containing matrices equivalent to mapping those expressions to a sequence of kernel invocations. However, end-users are progressively less likely to go through the time-consuming process of directly using said kernels; instead, languages and libraries, which offer a higher level of abstraction, are becoming increasingly popular. These languages and libraries offer mechanisms that internally map an input program to lower level kernels. Unfortunately, our experience suggests that, in terms of performance, this mapping is typically suboptimal. In this dissertation, we formally define the problem of mapping a linear algebra expression to a set of available building blocks as the "Linear Algebra Mapping Problem" (LAMP). Then, we investigate how effectively a benchmark of test problems is solved by popular high-level programming languages and libraries. Specifically, we consider Matlab, Octave, Julia, R, Armadillo (C++), Eigen (C++), and NumPy (Python); the benchmark tests both compiler optimizations, as well as linear algebra specific optimizations, such as the optimal parenthesization of matrix products. Finally, we provide guidelines to language developers and end-users for how to prioritize different optimizations when solving a LAMP. On the other hand, in the multi-linear algebra domain, the absence of standardized, building-block libraries has led to significant replication of effort when it comes to software development. While we point out the benefit that such libraries would provide to a number of scientific fields, at the same time we identify and focus on cases where such a library-based approach would not suffice. Specifically, we investigate two such cases in one of the most prominent fields in multi-linear algebra, chemometrics: the Canonical Polyadic (CP) tensor decomposition and jackknife resampling, a statistical method used to determine uncertainties of CP decompositions.For the first case, the CP decomposition, a broadly used method for computing it relies on the Alternating Least Squares (ALS) algorithm. When the number of components is small, regardless of its implementation, ALS exhibits low arithmetic intensity, which severely hinders its performance and makes GPU offloading ineffective. We observe that, in practice, experts often have to compute multiple decompositions of the same tensor, each with a small number of components (typically fewer than 20). Leveraging this observation, we illustrate how multiple decompositions of the same tensor can be fused together at the algorithmic level to increase the arithmetic intensity. Therefore, both CPUs and GPUs can be utilized more efficiently for further speedups; at the same time the technique is compatible with many enhancements typically used in ALS, such as line search, and non-negativity constraints. We introduce the Concurrent ALS (CALS) algorithm and library, which offers an interface to Matlab, and a mechanism to effectively deal with the issue that decompositions complete at different times. Experimental results on artificial and real datasets demonstrate that the increased arithmetic intensity results in a shorter time to completion.For the second case, we examine jackknife resampling. Upon observation that the jackknife resampled tensors, though different, are nearly identical, we show that it is possible to extend the CALS technique to a jackknife resampling scenario. This extension gives access to the computational efficiency advantage of CALS for the price of a modest increase (typically a few percent) in the number of floating point operations. Numerical experiments on both synthetic and real-world datasets demonstrate that the proposed workflow can be several times faster than a straightforward workflow where the jackknife submodels are processed individually., Published by RWTH Aachen University, Aachen