A Phase 2 Study of Duvelisib and Venetoclax in Patients with Relapsed or Refractory (R/R) Chronic Lymphocytic Leukemia (CLL) or Richter’s Syndrome (RS)

Background

While the MURANO venetoclax (V) + rituximab regimen is highly effective, it spans 2 years and includes 6 months of infusional therapy. Duvelisib (D) is an oral PI3K-δ/γ inhibitor (PI3Ki) FDA-approved for R/R CLL after 2 prior therapies. Preclinical data support V + D in CLL and RS (Patel 2017; Iannello 2019), and we previously demonstrated that V + D was well-tolerated at the FDA-approved doses of each drug (Crombie, ASH 2020). Here, we report the phase 2 results of our investigator-sponsored trial (NCT03534323) in patients (pts) with R/R CLL and RS.

Methods

Key eligibility criteria for CLL pts: requiring treatment by iwCLL criteria, ≥ 1 prior therapy with no prior V or D, ECOG PS ≤ 2, and for RS pts: no V within 1 year. Treatment began with a 7-day lead-in of D 25mg BID, and V starting on day 8. V was initiated at 10mg (outpt) or 20mg (inpt), with weekly ramp-up to 400mg daily. Pts with RS could accelerate V ramp-up to 400mg over 5 days (inpt). After 12 cycles, pts with bone marrow (BM)- undetectable minimal residual disease (uMRD) could discontinue treatment and reinitiate V at time of progression, and pts with detectable MRD continued V alone, with an option to discontinue later if they achieved BM-uMRD. Assessments of toxicities were by CTCAE v5, efficacy by 2008 iwCLL criteria, and MRD by central 8-color flow (10^-4 sensitivity).

Results

As of May 2, 2024, enrollment was complete for 38 pts (n=29 CLL and n=9 RS). Baseline characteristics included: median age: 69 years (range 50-79); 70% male; 66% unmutated IGHV; 11% del(11q); 26% complex karyotype (≥3 abnormalities); 48% TP53-aberrant; 18% NOTCH1 mutant; median prior treatments: 2 (range 1-5); 45% with prior BTKi.

In the total population of 38 pts, the median number of cycles was 13 (range 1-50), and all-grade heme tox included: neutropenia (86%; 79% Gr3/4) and thrombocytopenia (58%; 21% Gr3/4). Notable non-heme tox ≥ 20% included: diarrhea (63%; 13% Gr 3/4), nausea (55%, 3% Gr3), increased alkaline phosphatase (45%, all Gr1), increased AST (42%; 8% Gr3/4), hypertension (42%, 8% Gr3), and headache (34%, all Gr1). Other notable SAEs included: colitis (11%, 3% Gr3), tumor lysis syndrome (5% Gr3/4, n=1 lab and 1 clinical), febrile neutropenia (5% Gr3), and infection (5% Gr3). One pt developed a gastric perforation requiring surgical repair while on high-dose steroids to manage diarrhea and discontinued treatment. Gr5 events included n=1 hepatic failure attributed to progressive RS and n=1 COVID-19 pneumonia.

In pts with CLL, the best ORR and CR/CRi (primary endpoint) rates were 97% and 62%, respectively. After 12 cycles, the ORR was 72% (44% CR/CRi). CR/CRi was 64% and 46% for TP53-aberrant (n=14) and post-BTKi (n=13) pts, respectively. At this time point, PB and BM-uMRD were 45% and 41%, respectively. 11 pts of 29 achieved BM-uMRD after 12 cycles and electively discontinued therapy, 3 of whom later restarted V once MRD re-detected (median yrs off treatment=1.2 [range: 0.97-3.7]). In the 9 pts with RS, 3 achieved CR, including one pt who proceeded to alloSCT and remains in remission 2.3 yrs later.

Reasons for discontinuation included: achievement of BM-uMRD (n=11 CLL), alloSCT (n=1 RS), PD (n=3 CLL, n=6 RS), unacceptable tox (n=4 CLL, n=1 RS), and fatal COVID-19 pneumonia (n=1 CLL). For pts with CLL, the 3-year PFS and OS were 68% and 89%, respectively. 3-year PFS in TP53-aberrant and post-BTKi pts were 47% and 46%, respectively.

Conclusions

The all-oral, MRD-driven, time-limited regimen V + D is active in R/R CLL and RS, including in high-risk pts with TP53-aberrant disease and post-BTKi. Toxicities were observed, but were manageable for most patients. Our data support exploration of V + D in larger studies of pts with R/R CLL and RS.