Long-term follow-up of venetoclax monotherapy in previously treated patients with Waldenström macroglobulinemia Open Access

The B-cell lymphoma 2 (BCL2) antagonist venetoclax is safe and effective for patients with relapsed Waldenström Macroglobulinemia (WM) and is included as a treatment option in various guidelines, including the National Comprehensive Cancer Network. A multicenter, prospective, phase 2 study in previously treated patients with WM who were treated with finite-duration therapy with venetoclax monotherapy (50% had previous exposure to Bruton tyrosine kinase [BTK] inhibitors) reported an overall response rate (ORR) of 84% and a median progression-free survival (PFS) of 30 months.¹ 

However, the long-term durability of venetoclax therapy in terms of PFS, treatment-free survival (TFS), overall survival (OS), the emergence of BCL2 mutations, and the outcomes of venetoclax retreatment in patients with WM have not been described. This report will provide insights into these issues.

The study design, methods, and results have been reported previously.¹ Briefly, the participants were previously treated adults with a diagnosis of WM and an indication to treat based on the Second International Workshop for WM criteria.^2,3 Venetoclax was ramped up to a maximum dose of 800 mg orally once daily for 2 years. Tumor lysis syndrome monitoring was mandatory during the venetoclax ramp-up.

The study’s primary objective was to achieve an ORR ≥70% (vs an alternative ORR ≤40%) with a double-sided α value of 0.05 and a power of 80%. Categorical responses were assessed using the modified Sixth International Workshop for WM criteria⁴; extramedullary disease resolution was not required to attain a very good partial response (VGPR). The median follow-up time was estimated using the reverse Kaplan-Meier method. PFS was defined as the time between treatment initiation and disease progression, last follow-up, or death from any cause. TFS was defined as the time between treatment initiation and the next treatment, the last follow-up, or death from any cause. OS was defined as the time between treatment initiation and the last follow-up or death from any cause.

All bone marrow samples underwent CD19 selection before genetic testing. Polymerase chain reaction (PCR) assays were used to detect MYD88 L265P and BCL2 G101V mutations. The CXCR4 mutational status was assessed using Sanger sequencing and allele-specific PCR.

Institutional review board approval and participants’ written informed consent were obtained before research activities. The study followed the Declaration of Helsinki and the International Conference on Harmonization Guidelines for Good Clinical Practice.

Table 1 shows the baseline characteristics of the 32 patients with WM who received at least 1 month of venetoclax monotherapy. As previously reported, the ORR was 84%, the rate of PR or better was 81%, and the VGPR rate was 19%. No additional long-term toxicities were observed beyond those previously reported with neutropenia (grade ≥2) being the most common, occurring in 53% of patients.

With a median follow-up of 81 months (95% confidence interval [CI], 71-86), 23 patients (72%) had disease progression, 17 (53%) began a new treatment, and 3 (9%) had died. Of the 3 deaths, 2 were because of WM progression at 24 and 48 months, and 1 was of unknown cause at 54 months after treatment completion.

The median PFS was 36 months (95% CI, 27-44) with a 5-year PFS rate of 24% (95% CI, 10-41; Figure 1A). Participants who attained a PR or better had a longer median PFS of 42 months vs 7 months for those who did not achieve PR (P < .001; Figure 1B). The CXCR4 mutational status (Figure 1C), VGPR attainment, time to response, and previous exposure to BTK inhibitors did not impact the PFS (P > .05 in all cases).

The median TFS was 43 months (95% CI, 37-66) with a 5-year TFS rate of 39% (95% CI, 20-57; Figure 1D). The median TFS was numerically longer in participants who attained a PR or better than in the patients who did not achieve PR (54 vs 25 months; P = .07; Figure 1E). The CXCR4 mutational status (Figure 1F), VGPR attainment, time to response, and previous exposure to BTK inhibitors did not impact TFS (P > .05 in all cases).

The median OS was not reached, and the 5-year OS rate was 95% (95% CI, 74-94).

As part of a dedicated subgroup analysis, we evaluated outcomes in patients with previous exposure to covalent BTK inhibitors. As reported in the initial trial publication, response rates tended to be lower in patients previously treated with a BTK inhibitor (ORR, 75% vs 93%; P = .33) with a significantly longer time to response (4.5 vs 1.4 months; P < .001). Previous BTK inhibitor exposure was not associated with significant differences in the PFS (28 vs 36 months; P = .33) or TFS (42 vs 54 months; P = .90).

Of the 17 participants who began a new treatment, 9 received retreatment with indefinite-duration venetoclax in a subsequent line of therapy. Six received venetoclax monotherapy (4 attained a PR, 1 a VGPR, and 1 showed no response). Two patients received venetoclax with zanubrutinib (1 attained VGPR, and 1 showed no response), and 1 received venetoclax with acalabrutinib (attained VGPR). All the patients underwent venetoclax ramp-up and tumor lysis syndrome monitoring upon venetoclax reinitiation. A swimmer plot depicting the durability of response upon retreatment is shown in Figure 2.

Of the 8 patients who started a new treatment but did not receive venetoclax, 4 received BTK inhibitor monotherapy (3 received ibrutinib and 1 received zanubrutinib; all attained VGPR), and 4 received anti-CD20 monoclonal antibody combination therapy (2 with bendamustine attaining VGPR and PR, and 2 with proteasome inhibitors without a response).

One event of histologic transformation to diffuse large B-cell lymphoma was observed 2 years into venetoclax retreatment in a patient without CXCR4 mutations. The patient was treated with 6 cycles of polatuzumab vedotin, cyclophosphamide, doxorubicin, and prednisone and attained a complete response. Twenty-two months after the transformation event, the patient is alive without evidence of diffuse large B-cell lymphoma.

A total of 52 bone marrow samples from 27 patients were available for BCL2 G101V mutation assessment with more than 1 sample available in 17 patients. Nineteen samples were obtained at cycle 6, 16 at cycle 12, 1 at cycle 15, and 16 at the end of treatment. BCL2 G101V mutations were not detected in any of the samples.

In this study, we provide insights into the long-term efficacy of venetoclax monotherapy in patients with relapsed WM. We observed a median PFS of 36 months and a median TFS of 43 months with superior outcomes in patients who attained at least PR. We also observed that retreatment with venetoclax is reasonable in patients who did not have progressive disease while on venetoclax therapy. Notably, the initial course of fixed-duration venetoclax did not induce BCL2 G101V mutations, permitting retreatment.

This study included only patients with MYD88 L265P mutated disease, confirmed by highly sensitive CD19-selected bone marrow and allele-specific PCR testing,⁵ and therefore supports venetoclax use in this population. Trialing venetoclax or repeating genotyping may still be reasonable in patients deemed to have wild-type MYD88 by less sensitive methods. A recent retrospective study of WM patients treated with venetoclax off-trial reported a median PFS of 28.5 months with a median follow-up of 33 months; notably, 4 of the 76 patients had wild-type MYD88, although their outcomes were not analyzed separately.⁶ 

The median PFS on venetoclax compares favorably with other treatment options in WM previously exposed to BTK inhibitors and chemoimmunotherapy. Specifically, the noncovalent BTK inhibitor pirtobrutinib, administered until disease progression or unacceptable toxicity, was associated with a median PFS of 20 months despite a high response rate.⁷ However, the patient population in the pirtobrutinib study was more heavily pretreated with a median number of therapies of 3 and 100% previous exposure to covalent BTK inhibitors. The single-arm, nonrandomized nature of this trial with a relatively small sample size may have limited the power for subgroup analyses, precluding direct comparisons with alternative agents. Consequently, cross-trial comparisons must be interpreted with caution. Given the evolving landscape of WM treatment with other agents under development, including the radioisotope iopofosine I-131,⁸ the antibody-drug conjugate loncastuximab tesirine,⁹ the bispecific antibody epcoritamab,¹⁰ the second-generation BCL2 antagonist sonrotoclax,¹¹ and BTK degraders,¹² the optimal positioning and sequencing of venetoclax remains unclear, but it should be considered after covalent BTK inhibitor failure because of its comparatively lower efficacy^13-16 or in patients in whom a BTK inhibitor is a less preferred approach, such as in patients with cardiac disease or bleeding diathesis.

Given the absence of comparative data, the decision to initiate venetoclax in the relapsed setting should be individualized with consideration for patient characteristics and preferences, including the desire for time-limited therapy with venetoclax vs continuous treatment with agents such as pirtobrutinib, or interest in participating in clinical trials. Patients with CXCR4 mutations may benefit from venetoclax, because its activity seems to be unaffected by CXCR4 status, in contrast with a numerically reduced response rate observed with pirtobrutinib in this subgroup.¹⁷ Conversely, the slower response to venetoclax, particularly after previous covalent BTK inhibitor exposure, may favor pirtobrutinib in patients with high immunoglobulin M transitioning from a BTKi given its faster onset, thereby lowering the risk for rebound.¹⁸ Although specific data on the efficacy of venetoclax following pirtobrutinib or BTK degraders, or vice versa, are lacking, there is no biologic rationale to suggest that previous therapy would preclude effectiveness given the distinct mechanisms of action.

Early reports in patients with chronic lymphocytic leukemia suggested venetoclax retreatment is reasonable for patients who progressed after completing fixed-duration venetoclax and did not develop mutations in BCL2.^19,20 We report, to our knowledge, the first experience with venetoclax retreatment in patients with WM. Like patients with chronic lymphocytic leukemia, venetoclax retreatment is not recommended in patients who have progressed on venetoclax therapy. Importantly, no BCL2 G101V mutations were detected in bone marrow samples from patients up to the end of treatment, further supporting venetoclax retreatment. Still, the emergence of toxicities or other less common BCL2 resistance mechanisms (eg, D103Y), which were not assessed in this study, should be considered when considering venetoclax retreatment.²¹ 

The development of venetoclax in WM continues. We recently reported on a phase 2 study that combined ibrutinib and venetoclax in previously untreated patients with WM, which was stopped prematurely because of 2 grade 5 ventricular arrhythmias of unclear etiology.²² Other studies that evaluated venetoclax in the frontline setting include the combination of venetoclax plus rituximab vs ibrutinib plus rituximab (ClinicalTrials.gov identifier: NCT04840602) and the combination of venetoclax and rituximab vs standard chemoimmunotherapy (NCT05099471). In the relapsed setting, a study combining pirtobrutinib and venetoclax is ongoing with no arrhythmia events reported to date. Preliminary results show encouraging activity with a VGPR rate of 56%, exceeding what has been reported with either agent alone. However, the long-term efficacy of this combination and the patient subgroups most likely to benefit remain to be determined.²³ 

In conclusion, despite the limitations associated with the small sample size of this prospective study, our study adds to the current body of evidence supporting the use of venetoclax in WM.

AbbVie provided the drug and funding for this study.

Contribution: J.J.C. and S.P.T. designed the study; J.J.C., J.N.A., T.S., R.H.A., C.A.F., and S.S. provided the patients; K.M., N.B., J.N., A.R.-G., A.G., and C.J.P. provided administrative support; N.T., C.J.P., and S.P.T. performed the laboratory studies; J.J.C., A.G., S.P.T., and S.S. performed the statistical analysis; J.J.C. drafted the initial version of the manuscript; and all the authors critically reviewed and approved the final manuscript.

Conflict-of-interest disclosure: J.J.C. reports honoraria from AbbVie, AstraZeneca, BeiGene, Cellectar, Johnson & Johnson, Kite, Loxo, and Pharmacyclics; and research funds from AbbVie, AstraZeneca, BeiGene, Cellectar, Loxo, and Pharmacyclics. J.N.A. reports consulting fees from AbbVie, Adaptive Biotechnologies, ADC Therapeutics, AstraZeneca, BeiGene, Genentech, Janssen, Eli Lilly Pharmaceuticals, Merck, NeoGenomics, and Pharmacyclics; and research funding from BeiGene, Celgene/Bristol Myers Squibb (BMS), and Genentech. T.S. reports research funding from BMS; and consulting fees from AbbVie, AstraZeneca, BeiGene, BMS, and Gilead. R.H.A. reports research funding from BeiGene. S.P.T. reports research funding and/or consulting fees from AbbVie/Pharmacyclics, Janssen, BeiGene, Eli Lilly Pharmaceuticals, BMS, and Ono Pharmaceuticals. S.S. reports research funds or honoraria from ADC Therapeutics, BeiGene, and Cellectar. The remaining authors declare no competing financial interests.

Correspondence: Jorge J. Castillo, Dana-Farber Cancer Institute, 450 Brookline Ave, Mayer 221, Boston, MA 02215; email: jorgej_castillo@dfci.harvard.edu.