Venetoclax consolidation in high-risk CLL treated with ibrutinib for ≥1 year achieves a high rate of undetectable MRD

Patients receiving ibrutinib for CLL rarely achieve undetectable measurable residual disease (U-MRD), necessitating indefinite therapy, with cumulative risks of treatment discontinuation due to progression or adverse events. This study added venetoclax to ibrutinib for up to 2 years, in patients who had received ibrutinib for ≥12 months (mo) and had ≥1 high risk feature (TP53 mutation and/or deletion, ATM deletion, complex karyotype or persistently elevated β2-microglobulin). The primary endpoint was U-MRD with 10–4 sensitivity (U-MRD4) in bone marrow (BM) at 12mo. Forty-five patients were treated. On intention-to-treat analysis, 23/42 (55%) patients improved their response to CR (2 pts were in MRD + CR at venetoclax initiation). U-MRD4 at 12mo was 57%. 32/45 (71%) had U-MRD at the completion of venetoclax: 22/32 stopped ibrutinib; 10 continued ibrutinib. At a median of 41 months from venetoclax initiation, 5/45 patients have progressed; none have died from CLL or Richter Transformation. In 32 patients with BM U-MRD4, peripheral blood (PB) MRD4 was analyzed every 6 months; 10/32 have had PB MRD re-emergence at a median of 13 months post-venetoclax. In summary, the addition of venetoclax in patients treated with ≥12mo of ibrutinib achieved high rate of BM U-MRD4 and may achieve durable treatment-free remission.

Leukemia (2023) 37:1444-1453; https://doi.org/10.1038/s41375-023-01901-4 BACKGROUND Patients receiving the Bruton tyrosine kinase (BTK) inhibitor ibrutinib for chronic lymphocytic leukemia (CLL) or small lymphocytic lymphoma (SLL) rarely achieve complete remission (CR) with undetectable measurable residual disease at a sensitivity of 10 -4 (U-MRD4). Therefore, continuous and indefinite maintenance therapy is standard of care. Long-term treatment with ibrutinib, even in the first-line setting, results in a cumulative risk of treatment discontinuation due to adverse events or disease progression [1,2], as well as high cost [3]. In one large, real world analysis, the rate of discontinuation due to adverse events was 41% at a median follow-up of 17 months [4]. Adverse effects of particular interest include atrial fibrillation, major hemorrhage, and hypertension. Most recently, concern has emerged regarding an increased risk of ventricular arrhythmia and sudden cardiac death (relative risk 8.13 in a meta-analysis, of which, half resulted in sudden cardiac death) [5].
The risk of progression, most commonly associated with mutation at the cysteine residue in codon 481 of BTK that prevents covalent binding of ibrutinib [6], is highest in patients with complex karyotype and/or TP53 deletion or mutation [7,8]. Median progression-free survival (PFS) is approximately 26 months in patients with relapsed/refractory disease and del[17p]; in first line patients, approximately 60% remain progression-free at 6 years [9,10]. Some series suggest increased risk in patients with del [11q], which may be more important in heavily pre-treated patients [11] or persistently elevated β 2 -microglobulin [12]. There is no plateau on the PFS curve, with a continuous pattern of relapses seen [8]. After failure of ibrutinib, a standard-of-care salvage therapy is treatment with the Bcl-2 inhibitor venetoclax, with or without rituximab. Response rate to venetoclax monotherapy after failure of ibrutinib is approximately 70%, with median PFS of just over 2 years [13].
Venetoclax is synergistic with ibrutinib [14], has a largely nonoverlapping toxicity profile, and is able to achieve U-MRD alone or combined with CD20 monoclonal antibodies [15][16][17]. Combining venetoclax with ibrutinib is attractive for multiple reasons: first, the completely distinct mechanisms of action could reduce the risk of on-treatment emergence of resistance through selective pressure, which has been seen in the majority of high risk patients treated with BTK inhibitor or venetoclax monotherapy; second, synergistic CLL cell killing may achieve deep remissions, allowing treatment discontinuation, reducing both risk of toxicity from continued BTK inhibitor therapy and also cumulative financial burden (provided the duration of off-treatment remission exceeds the duration of combined therapy); third, the largely non-overlapping toxicity profiles allows safe combination of the two agents, as has been shown in other studies [18][19][20][21].
We hypothesized that adding venetoclax to ibrutinib in patients at high-risk for CLL progression, who had received ibrutinib therapy for at least 12 months, would achieve U-MRD4 in bone marrow (BM) and be associated with low risk of disease progression. In the early studies of venetoclax therapy, tumor lysis syndrome (TLS) was noted, necessitating gradual dose escalation [15]. Risk for TLS with venetoclax initiation is proportionate to disease bulk at time of therapy initiation. Therefore, we also hypothesized that TLS risk would be low when utilizing venetoclax as an "add on" to ibrutinib after at least 12 months of ibrutinib therapy, given most patients would have low disease burden, unlike the situation potentially faced if venetoclax is initiated at the time of disease progression on ibrutinib.

METHODS
We designed a phase II, investigator-initiated, response-adapted clinical trial with the addition of venetoclax to ibrutinib in patients with one or more high-risk features for disease progression who had received at least 1 year of ibrutinib, regardless of line of treatment. There was no stipulated maximum duration of ibrutinib therapy. Patients had detectable disease (defined as ≥0.01% MRD assessed by flow cytometry in BM; measurable disease by CT was not required) at the time of study entry and did not meet IWCLL criteria for progression. High-risk features were defined as the presence of at least one of: del[17p] or TP53 mutation, complex karyotype (defined as 3 or more unrelated abnormalities on conventional, metaphase karyotyping), del[11q]; or β 2 -microglobulin above the upper limit of normal at the time of study screening. Patients could satisfy genomic eligibility criteria if detected within the 3 months prior to starting ibrutinib or at any time during ibrutinib therapy. This provision was utilized to account for the fact that some patients were expected to have a low frequency of CLL cells present in BM at the time of study screening, which could lead to false negative testing for the relevant genomic aberrations. Ibrutinib was continued at the previously tolerated dose (140-420 mg once daily) and standard TLS risk-adjusted initiation and escalation of venetoclax was performed to the target dose of 400 mg once daily. Treatment with combined ibrutinib and venetoclax continued for a maximum duration of twenty-four 28-day cycles, or approximately 2 years (Fig. 1). Full inclusion and exclusion criteria are outlined in the data supplement.
Patients had BM evaluation for MRD by flow cytometry (sensitivity 10 -4 ) performed prior to treatment and after every 6 cycles of venetoclax. At the initiation of the study, this was performed utilizing the standardized 4-color flow cytometry method of Rawstron et al. [22], which was upgraded to an in-house 10-color method on January 8, 2020. The 10-color flow cytometry assay was extensively cross-validated with the standardized 4-color assay and demonstrated at least equivalent sensitivity of 0.01%. Details of the in-house flow cytometry assay are provided in the supplementary methods. Patients also had a pre-treatment computed tomography (CT) scan of neck, chest, abdomen, and pelvis for evaluation of nodal disease and re-staging CT every 6 cycles of treatment until achievement of CR. All scans were read by a single, expert radiologist (NG) and responses were assigned according to IWCLL 2008 criteria [23]. Patients in CR with BM U-MRD4 on two consecutive evaluations, 6 cycles apart, stopped venetoclax, but could continue ibrutinib at physician discretion. Patients who were BM MRD4-positive post-24 cycles of treatment continued ibrutinib maintenance (Fig. 1).
The primary endpoint for the study was BM U-MRD4 after 12 cycles of therapy. For the purposes of sample size calculation, the assumption was that after 12 cycles of ibrutinib plus venetoclax, at least 25% of patients would achieve BM U-MRD4. To ensure patient safety, the probability of toxicity (defined as grade ≥3 non-hematologic adverse events) was monitored using a Bayesian monitoring rule [24]. The detailed statistical plan is provided in the supplementary appendix.
Adverse events were assessed from time of initiation of venetoclax until completion of venetoclax therapy according to the National Cancer Institute Common Terminology Criteria for Adverse Events (version 4.03). Dose modification of ibrutinib and/or venetoclax were allowed, with guidelines provided in the protocol as per the U.S. Prescribing Information.
The study was approved by the M.D. Anderson Institutional Review Board and conducted according to the Declaration of Helsinki. All patients provided informed consent prior to participating in any study-related procedures. The trial was registered at clinicaltrials.gov (NCT03128879).

RESULTS
Pre-treatment patient characteristics of the 45 enrolled patients are shown in Table 1. All patients had completed venetoclax therapy as of the final data cutoff for this analysis (August 22, 2022). The median number of prior therapies was 1 (range 0-3); 22/45 patients were receiving ibrutinib as initial therapy. Median duration of ibrutinib therapy prior to venetoclax was 32 months (range, 12-73). Sixty-three percent of patients had del[17p] and/or mutated TP53. Thirty-one percent of patients with stimulated karyotype result available had complex karyotype. Median prevenetoclax MRD percentage in BM was 7.3% (range 0.03 to 70.1%). No patients were high-risk for venetoclax-associated TLS, 5/45 were medium-risk and 40/45 were low-risk. Pre-treatment disease response status is detailed in Table 2.
As shown in Fig. 2A, the majority of patients who achieved BM U-MRD4 did so within the first 12 months of therapy. Serial BM MRD analysis predicted likelihood of achieving BM U-MRD4 status. At 6 months, patients with detectable MRD at a level of <1% had an 82% likelihood of achieving BM U-MRD4 (14/17) at the end of therapy, while only 1/9 patients with ≥1% MRD achieved BM U-MRD status at the end of therapy. Among 16 evaluable patients who had detectable MRD after cycle 12, 6 (38%) achieved BM U-MRD4 subsequently (although one, who had U-MRD4 after cycle 18 was MRD-positive at 0.014% at cycle 24). The highest MRD level after cycle 6 in these 6 patients who converted from MRD-positive after cycle 12 to U-MRD4 after cycle 24 was 0.08%.

Nodal responses
Two of 45 patients (5%) were in MRD + CR prior to venetoclax. Of the remaining patients, 23/42 (55%) patients improved their response to CR/CRi during venetoclax treatment; one patient was not evaluable for nodal response. All evaluable patients had reduction in lymph node size with addition of venetoclax. Median change in the sum of the product diameters of 6 measurable lymph nodes was −70% (range 1-100%), Fig. 2B.

Patient disposition, PFS and survival
Patient remission and treatment status are shown in Fig. 3 and in the swimmer plot (Fig. 4). Median follow-up is 41 months from the time of venetoclax initiation. Forty-one patients completed venetoclax treatment per protocol. Thirty-two patients achieved BM U-MRD4 status at the completion of planned study treatment. Of these 32 patients, 1 (with U-MRD4 after 6 cycles) came off study due to metastatic pulmonary adenocarcinoma; 10 stopped venetoclax prior to C24 per protocol (5 after C12 and 5 after C18), after achieving CR with BM U-MRD on two consecutive response assessments; the remaining 21 patients No patient died during venetoclax treatment. Patients continued to be followed for overall survival after completing study therapy. No patient has died from CLL or Richter Transformation. Two patients died after receiving treatment for other primary cancers (metastatic melanoma, AML) and one after coming off protocol to receive acalabrutinib, from unknown causes (the patient was lost to follow-up prior to his death). (Fig. 2C). The patient who  Lymph node enlargement >1.5 cm (+/− splenomegaly) 27 Bone marrow involvement (>30% lymphocytes or B lymphoid nodules) +/− splenomegaly 2 Splenomegaly only 0 a All 14 patients with absolute lymphocyte count >4 × 10 9 /L also had at least one lymph node >1.5 cm in maximum diameter. P.A. Thompson et al. developed MDS and subsequently AML had received chemoimmunotherapy on a clinical trial with combined cyclophosphamide, fludarabine, alemtuzumab, and rituximab (CFAR) prior to treatment with ibrutinib.
Clonal evolution at disease progression and subsequent treatments Additional genomic abnormalities at disease progression are shown in Table 3.
Only one patient developed disease progression during combined therapy (patient 39). This patient's CLL had a BTK C481R mutation detected prior to venetoclax initiation, indicating emergence of an ibrutinib-resistant cell population prior to venetoclax initiation. The BTK C481R mutated cells remained at progression and the CLL developed 11 new karyotypic abnormalities on conventional cytogenetic analysis, without any associated BCL2 mutation identified. Three patients developed disease progression while taking continuous ibrutinib (one MRD-positive,  two with U-MRD4) after completion of venetoclax (patients 5, 14 and 28): Patient 5, with MRD-positive disease, had difficulty tolerating venetoclax due to neutropenia and thrombocytopenia and required dose reduction to 50 mg daily. In addition to CLL, he has clonal cytopenia of uncertain significance, with a U2AF1 mutation present, and this may have contributed to inability to maintain venetoclax dose intensity. The two patients with BM U-MRD4 who progressed on ibrutinib maintenance (14 and 28) had different patterns of resistance: one, who had del[11q] but no TP53 abnormality, had emergence of a CLL clone containing multiple new chromosomal abnormalities, including tetraploidy, as well as a BTK C481S mutation; intriguingly, patient 28, whose CLL had del[17p] and complex karyotype pre-venetoclax, had no identifiable mechanism of resistance to ibrutinib. Her CLL had 2 separate BCL2 mutations at progression (including G101V), despite rapidly achieving BM U-MRD4 (at 6 months) during combination therapy. One patient (patient 39), whose CLL had del[17p] and complex karyotype developed Richter Transformation during observation after achieving BM U-MRD4 status and stopping therapy. He achieved CR, with U-MRD6 by Clonoseq on a clinical trial with venetoclax plus R-CHOP and proceeded with allogeneic stem cell transplant.

MRD-free survival
During follow-up, in patients with BM U-MRD4 at the completion of therapy, PB MRD testing was obtained every 6 months (Fig. 5); 31 of 32 patients who had BM U-MRD4 at the completion of venetoclax have had at least one PB MRD measurement; 10 have had confirmed PB MRD re-emergence, 7/22 patients who were not taking maintenance ibrutinib and 3/10 patients on ibrutinib maintenance, (p = 0.48, Fig. 5B). There was also no difference in MRD-free survival according to remission status (CR/CRi vs no CR/CRi), (p = 0.42, Fig. 5A), previously treatment-naïve vs relapsed/refractory (p = 0.62 Fig. 5C), or TP53 abnormalities (p = 0.66, Fig. 5D). Three patients subsequently developed disease progression, of whom two were receiving ibrutinib maintenance per physician discretion at the time of progression.

Adverse event profile
Treatment was well-tolerated. All adverse events (AEs) occurring in ≥15% of patients, regardless of attribution, are shown in Table 4. Similar to other previously published studies of combined ibrutinib plus venetoclax, the most common AE was diarrhea in 21 patients (47%), grade 1-2 in all but 1 patient. No patient developed clinical or biochemical TLS. The most common grade 3-4 AE was neutropenia, seen in 20% of patients. No patient had febrile neutropenia. Eighteen patients experienced a total of 25 SAEs: nine patients had non-melanoma skin cancers diagnosed during the trial; 2 patients developed myelodysplastic syndrome (both with a history of chemotherapy treatment for CLL); 6 developed other solid tumors (see supplementary table); 3 patients developed grade 3 infections (pneumonia, pyelonephritis, skin abscess). There were no opportunistic infections.  No patient stopped venetoclax treatment due to toxicity. However, 14/45 patients required venetoclax dose reduction (300 mg, n = 5, 200 mg n = 7, 100 mg, n = 1 and 50 mg, n = 1), most commonly due to neutropenia (n = 7, or diarrhea, n = 6). Prior to study entry, 24/45 patients had ibrutinib dose reduced to 280 mg or 140 mg. During the study an additional 6 patients had dose reductions of ibrutinib due to adverse events (30/45 patients total had ibrutinib dose reductions either prior to or during the study) and 5 patients stopped ibrutinib (due to atrial fibrillation, n = 3, non-sustained ventricular tachycardia, n = 1 and unexplained syncope complicated by subdural hematoma, n = 1).

DISCUSSION
Venetoclax added to ibrutinib in patients with high-risk CLL was well-tolerated and achieved a cumulative BM U-MRD4 rate of 73%,   Thompson et al. with no significant differences in rate of achievement of BM U-MRD according to presence or absence of TP53 abnormalities, complex karyotype or prior treatment status. This allowed treatment discontinuation in a significant proportion of patients. The high rate of BM U-MRD4 was similar to results from studies utilizing ibrutinib plus venetoclax as initial therapy [18,20,25] or for treatment of relapsed/refractory disease [19], despite the selection of high genomic risk patients in our cohort. In addition to the high rates of U-MRD4 achieved, 55% of patients improved response to CR, with a median 70% reduction in sum product diameters of target lymph node groups, indicating that the benefit of addition of venetoclax extended beyond the bone marrow compartment.
Given the distinct mechanisms of action of ibrutinib and venetoclax, we hypothesized that the likelihood of simultaneous selection of resistance mutations to both agents would be low. Importantly, in our study, only 1 of 45 patients developed progressive disease during combined therapy (a patient with preexisting BTK mutation indicating emerging ibrutinib resistance prior to venetoclax). In one high risk patient (del[17p] and complex karyotype), BCL2 mutation was detected at progression during ibrutinib maintenance, despite rapid attainment of U-MRD4 at 6 months of combination therapy (an unusual scenario, given only brief exposure to venetoclax); in the other 3 cases of disease progression, no patient had BCL2 mutations detected, while one had emergence of BTK C481S as the mechanism of ibrutinib resistance (in addition to >10 new karyotypic abnormalities). Thus, in high-risk patients, therapy with combined ibrutinib and venetoclax may reduce, but does not eliminate, the risk of selecting treatment-resistant cells. The mechanisms of MRD persistence and resistance to combined BTK and BCL2 inhibition remain to be elucidated.
The optimal duration of combined ibrutinib and venetoclax is unknown. Some studies have used a fixed treatment durations (GLOW [25], CAPTIVATE-FD [26]) while others have utilized an MRD-adapted approach (M.D. Anderson Phase 2 [18], CAPTIVATE-MRD [20], HOVON141 [21], CLARITY [19]). We employed an MRDadapted treatment strategy in this study. The majority of patients who were destined to achieve BM U-MRD did so within 12 cycles; however, of the 16 patients who were MRD-positive after cycle 12, 6/16 achieved U-MRD4 in the 2nd year of therapy. Thus, the MRDadapted treatment approach offered further potential benefit beyond a 12-month, fixed-duration treatment strategy such as employed in the GLOW trial, in allowing some patients to achieve deeper remissions. This deepening of remissions with a 2nd year of combined ibrutinib and venetoclax has also been reported in the CAPTIVATE-MRD and CLARITY trials. The optimal approach to patients who have U-MRD4 after ibrutinib + venetoclax remains unclear; after a median of 3 years of follow-up, the CAPTIVATE study has shown no PFS benefit for ibrutinib maintenance vs placebo in patients with U-MRD, but these are first line patients with a minority having high risk features [27]. The benefit from continued ibrutinib maintenance may be greater in high risk cohorts such as ours. The majority of patients in our trial who had U-MRD stopped ibrutinib (22/32) and their long-term outcomes (including response to covalent BTKi re-treatment) will be important to monitor. We also continued ibrutinib maintenance in patients who were MRD-positive at the end of therapy, of whom 8/10 remain in ongoing remission and are potentially benefiting from maintenance therapy. The optimal approach to patients who remain MRD + after 2 years of combined ibrutinib + venetoclax is unknown. CLARITY and an M.D. Anderson phase 2 study [28] have explored a 3rd year of combination therapy in patients who remain MRD+. However, the additional benefit in terms of deepening of responses in these data sets appears modest and it is unclear if any improvement in PFS from long-term doublet therapy vs monotherapy with ibrutinib would justify the additional cost and toxicity.
At a median of 13.3 months of post-venetoclax follow up, most patients who attained BM U-MRD have ongoing PB U-MRD. The protocol did allow continuation of ibrutinib according to physician discretion in patients with BM U-MRD at the end of venetoclax therapy. Assignment to ibrutinib maintenance or not in patients with BM U-MRD was non-randomized (unlike in the CAPTIVATE study) [20] and the study was not powered to evaluate differences in PFS according to receipt of ibrutinib maintenance vs no maintenance. Within the limitations of these parameters, there has so far been no significant difference noted in time-to-MRD reemergence in the group of patients who received ibrutinib maintenance vs no maintenance. No other pre-treatment patient characteristics (notably, number of prior therapies, TP53 abnormality vs not) or IWCLL response category (CR/CRi vs not) were associated with risk of re-emergence of MRD post-venetoclax. There have been too few events to evaluate risk of progression according to pre-treatment patient characteristics.
Our study was not intended to directly answer the question of whether combination therapy is superior to the current paradigm of sequential monotherapy, which requires a randomized study to answer. While we await results from studies such as CLL17, a critical question which must be answered from this and other studies which utilize time limited combined BTK inhibitor plus venetoclax, will be the degree to which the disease remains sensitive to one or both agents at progression. Analysis of data from MURANO [29] and a multi-center retrospective analysis [30], demonstrated that patients with disease progression after timelimited venetoclax treatment have a 72.2-79.5% overall response rate to re-treatment, but with unknown durability. Additionally, emerging data from the fixed-duration cohort of CAPTIVATE show 7/9 patients who progressed within 2 years of completing combined treatment responded to ibrutinib re-treatment, with 2/9 having stable disease [31]. This is in contrast to those with disease progression occurring during BTK inhibitor therapy, in which case the disease is resistant by definition. It will be important, as data mature, when comparing time-limited venetoclax-based combination approaches to standard continuous BTK inhibitor-based therapy, that time-to-therapeutic failure (i.e., resistance to re-treatment), rather than just time-to-next therapy, is considered for the time-limited approach. No data yet exist for BTK inhibitor plus venetoclax re-treatment after prior treatment with limited duration BTK inhibitor plus venetoclax, but we hope that this will emerge over the coming years.
Our study has several limitations. Our cohort is somewhat heterogenous, including both frontline and relapsed/refractory patients and patients who had received ibrutinib for various periods of time. The latter feature was by design, as we wanted to capture all patients with high risk CLL who could benefit from therapy, especially since the risk of progression during ibrutinib monotherapy persists indefinitely, with no plateau. The similarity of the U-MRD rates in our study to studies that have started venetoclax after shorter periods of ibrutinib (2-3 months) is encouraging. Once regulatory approval is obtained for ibrutinib + venetoclax combination, our data could provide justification for addition of venetoclax in high risk patients already taking a BTK inhibitor who desire time-limited therapy. We would not advocate prolonged BTK inhibitor use prior to initiation of venetoclax in BTK inhibitor naïve patients who require therapy, as the safety and efficacy of commencing venetoclax after a 3 month ibrutinib monotherapy phase has been repeatedly demonstrated. We included patients with del[11q] as a high-risk group (who represented approximately 1/3 of the patient). At the time of study, the impact of del[11q] on PFS was unclear, with the longest follow-up, from the PCYC-1103 study suggesting inferior outcomes [11]. However, larger cohorts have subsequently demonstrated that del(11q) is not a high-risk feature in ibrutinib-treated patients. Given the favorable data in our study for patients with TP53 mutation or deletion and the unmet need that these patients represent, the study has recently been amended to treat an additional 45 patients who have received either ibrutinib or acalabrutinib for at least 12 months and have either TP53 abnormal disease or complex karyotype.
In summary, our data show a high rate of achievement of BM U-MRD4, very low risk of on-treatment disease progression and durable off-treatment remissions in high-risk patients with the addition of venetoclax to ibrutinib. An MRD-directed treatment approach allowed early treatment discontinuation in patients with rapid achievement of BM U-MRD4 status and continuation of BTK inhibitor maintenance in those with persistently detectable MRD. This approach may benefit patients by limiting duration of exposure to drug and risk of adverse effects in rapid responders and improving remission durability in sub-optimal responders who remain MRD-positive by utilizing maintenance ibrutinib.