Obinutuzumab, lenalidomide, and venetoclax for the initial treatment of patients with advanced‐stage follicular lymphoma: Results of the Phase Ib/II LEVERAGE study Authors: Chan Y. Cheah, Piers Blombery, Amanda Johnston, Joel Wight, Katharine L. Lewis, Rory Bennett, Jing Xie, Stella Vlachos, Sally M. Hunter, Joyce Gonzales, Sally Binks, John F. Seymour Published: 2026 ABSTRACT AbstractFollicular lymphoma (FL) is the most common indolent lymphoma. Although chemoimmunotherapy is effective, toxicity remains problematic, and novel treatments are required. We performed a Phase Ib/II study of obinutuzumab, lenalidomide, and venetoclax in patients with treatment‐naïve, high tumor burden, advanced‐stage FL. During induction (6 × 28‐day cycles), patients received obinutuzumab and venetoclax (in four dose levels from 400 mg day [D]1–10 to 800 mg daily) from Cycles 1 to 6 and lenalidomide (20 mg Days 1–21) from Cycles 2 to 6. Patients with complete response (CR) received 2 years maintenance obinutuzumab, while partial responders received 2 years obinutuzumab plus 6 months of venetoclax and lenalidomide. Genomic characterization and measurable residual disease (MRD) monitoring were performed using targeted next‐generation sequencing (NGS) by circulating tumor DNA (ctDNA). Fifty patients were enrolled with a median age 60. No maximum tolerated dose was defined, and the Phase II dose of venetoclax was 800 mg daily. The CR rate and objective response rate were 86% and 92%, respectively. At median follow‐up of 18.3 months, the 2‐year progression‐free survival and overall survival rates were 92% and 100%, respectively. The most common any‐grade adverse events were neutropenia (72%), diarrhea (52%), and upper respiratory infection (48%). Undetectable MRD after one cycle of therapy had a 95% negative predictive value for relapse, while detectable MRD after three cycles of therapy had a 75% positive predictive value. Obinutuzumab, lenalidomide, and venetoclax are active in patients with treatment‐naïve high tumor burden FL; however, myelosuppression impacted deliverability. ctDNA MRD is a promising biomarker in FL. INTRODUCTION Follicular lymphoma (FL) is the most common indolent lymphoma. Although patients with high tumor burden and advanced‐stage disease typically respond well to chemoimmunotherapy, relapse invariably occurs. Cytotoxic chemotherapy can result in both acute and late toxicity, which varies according to the specific agents employed but can include neuropathy, neutropenic sepsis, hemorrhagic cystitis, cardiotoxicity, steroid‐induced hyperglycemia, osteoporosis, and secondary myeloid neoplasms. There is therefore a need for non‐cytotoxic regimens that remain highly active but have improved safety profiles. The combination of the CD20 monoclonal antibody rituximab and lenalidomide has been established as an effective and tolerable regimen in both first‐line and relapsed FL. 1 , 2 , 3 The more potent CD20 monoclonal antibody obinutuzumab has been shown to improve progression‐free survival (PFS) compared with rituximab when combined with chemotherapy in treatment‐naïve patients with FL in the GALLIUM study. 4 , 5 Obinutuzumab has also been combined with lenalidomide in both first‐line and relapsed/refractory FL with favorable efficacy and safety. 6 , 7 Venetoclax is a BCL2 inhibitor with single‐agent activity in FL and has previously been investigated in combination with CD20 monoclonal antibodies, chemoimmunotherapy, and/or checkpoint inhibitors in Phase II studies. 8 , 9 , 10 , 11 At the time of writing, only one other study evaluating obinutuzumab, venetoclax, and lenalidomide has been conducted, a Phase I study in patients with relapsed/refractory B‐cell lymphomas. 12 Among 22 patients treated (5 with FL) with a median of 2 prior lines of therapy, 4 dose levels were explored with escalating doses of venetoclax combined with the standard dosing of lenalidomide (20 mg day [D]1–21 of a 28‐day cycle) and obinutuzumab (1000 mg Cycle 1, D1, D8, and D15 and Cycles 2–6 D1). The highest dose level of venetoclax 800 mg once daily, D1–28, was declared the recommended Phase II dose (RP2D), but no Phase II data were reported. Responses were observed in 4/5 FL patients (2 complete responses [CRs] and 2 partial responses [PRs]) with the most common Grade ≥3 adverse events: neutropenia (91%), infections (27%), anemia (23%), and thrombocytopenia (14%). No prior study had evaluated this combination in the initial treatment of patients with high tumor burden FL. We hypothesized that venetoclax could be safely added to the combination of obinutuzumab and lenalidomide, and that this triple regimen would deliver effective disease control in patients with treatment‐naïve FL. PATIENTS AND METHODS We performed a Phase Ib/II study of obinutuzumab, lenalidomide, and venetoclax in patients with treatment‐naïve advanced‐stage high tumor burden FL. Key inclusion criteria were the following: histologically confirmed FL (Grades 1–3A); Stage II (non‐contiguous or bulk >7 cm)–IV disease; requiring treatment per Groupe d'Etude des Lymphomes Folliculaires criteria; measurable disease; Eastern Cooperative Oncology Group (ECOG) ≤ 2 and adequate organ function. Key exclusion criteria were the following: Grade 3B/follicular large B‐cell lymphoma; proven/suspected diffuse large B‐cell lymphoma (DLBCL); history of other malignancy unless treated with curative intent >2 years ago; prior systemic therapy for lymphoma (except steroids ≤7 days); and inability to swallow tablets. Full inclusion and exclusion criteria are included in the Supplementary Appendix. This study was carried out in compliance with principles of the International Conference on Harmonization (ICH), Good Clinical Practice (GCP), Declaration of Helsinki, and all applicable national and local regulations governing clinical studies. This study was approved by the Human Research and Ethics Committee of Peter MacCallum Cancer Centre, and all authors had access to primary clinical trial data. Peter MacCallum Cancer Centre was the study sponsor. Treatment was divided into induction (6 × 28‐day cycles) and maintenance (12 × 8‐week cycles) phases. During induction, patients received obinutuzumab in the standard schedule (1000 mg IV D1, 8 and 15 during Cycle 1; D1 Cycles 2–6). Lenalidomide was commenced during Cycle 2 at 20 mg D1–21 from Cycles 2 to 6. Venetoclax was given from Cycles 1 to 6 during induction. During Phase Ib, four dose levels of venetoclax were serially explored: 400 mg D1–10, 800 mg D1–10, 400 mg continuous, and 800 mg continuous. After six cycles, patients who attained CR received obinutuzumab maintenance (1000 mg every 8 weeks for 2 years). Patients who attained PR received a further six cycles of lenalidomide (10 mg D1–28) and venetoclax and the 2 years of obinutuzumab maintenance (as above). Treatment delays of up to 4 weeks were permitted to allow recovery of hematologic toxicities to Grade ≤1, but delays of >4 weeks required permanent treatment discontinuation. The primary endpoint of Phase Ib was to determine the RP2D and maximum tolerated dose (MTD). Dose escalation in Phase I was according to a 3 + 3 design with MTD defined as the highest dose level at which the incidence of dose‐limiting toxicity (DLT) was less than 2/6 patients. DLT definitions and the dose escalation strategy are summarized in the Supplementary Appendix. The DLT window was the first two cycles (as lenalidomide was introduced from Cycle 2). The primary endpoint of Phase II was to determine the CR rate at the end of induction (by Lugano 2014). 13 Key secondary endpoints included safety, objective response rate (ORR), PFS, overall survival (OS), duration of response (DOR), and time to next lymphoma treatment. Positron emission tomography (PET)‐computed tomography (CT) scan was performed at baseline and end of induction, or at any stage if progression was suspected clinically. A third PET‐CT scan was to be performed at month 30 or the end of maintenance for patients with treatment delays. Bone marrow biopsy was not mandated to confirm CR, given that large datasets have demonstrated this adds little value. 14 , 15 For the clinical endpoints, the database was locked on 13 June 2024. Additional progression events after the clinical database lock were included in the correlative analyses given the low event rate. Correlative analyses Samples were prospectively collected for protocol‐specified pre‐planned correlative studies, including baseline genomic analysis. Plasma was collected at baseline and prior to each cycle of therapy, while tissue was collected where possible at suspected disease progression to assess for genomic evolution and potential resistance mechanisms. Sequencing was performed by targeted next‐generation sequencing (NGS) using KAPA HyperCap DS NHL panel (Roche Diagnostics) according to the manufacturer's specifications. ctDNA was extracted from 2 to 4 mL of plasma collected in Streck tubes and sequenced using an Illumina NovaSeq. 6000® or NovaSeq X Plus achieving a median of 2200‐fold coverage (range 481–7630). This panel covers the coding regions of 383 genes recurrently mutated in lymphoid malignancy as well as intergenic and non‐ coding regions of the genome including the immunoglobulin heavy and light chain loci. Baseline genomic characterization was derived from both pretreatment tissue and ctDNA samples. MRD reporter lists were generated by filtering for somatic variants (≥2% variant allele frequency in tissue or ≥1% ctDNA and ≥10 variant reads) using a modified ctDNAtools workflow in combination with manual curation for removal of sequencing artefacts and germline polymorphisms. 16 Alignments were manually reviewed to confirm phasing of variants; phased variant analysis was incorporated as an established strategy to improve the sensitivity of MRD detection in the context of background sequencing errors. 16 The final refined reporter list was queried against ctDNA samples, and a statistical model was used to determine the detected or undetectable status for MRD. Statistical considerations Sample size determination Dose escalation phase A pragmatic sample size was planned for Phase I, with 6–30 evaluable patients (expected n = 21). Patients who did not initiate treatment were replaced. Dose expansion phase Sample size was calculated via Simon's minimax two‐stage design, 17 with the primary endpoint of CR rate by PET‐CT at induction completion. The obinutuzumab + lenalidomide + venetoclax triplet required ≥70% CR for further development (≤50% CR warranted termination). The maximum total sample size was 37 evaluable patients in dose expansion (23 in Stage 1, 14 in Stage 2), with the first 6 from the RP2D determined in dose escalation. Stage 1 accrual would have stopped if ≤12/23 patients achieved CR. As the CR rate after 23 patients were treated was ~90%, Stage 2 proceeded. The triplet would have been deemed unsuitable for further development if ≤23/37 patients achieved CR at the RP2D. Calculations used a one‐sided 5% significance level and 80% power. This primary analysis occurred when all patients had completed six cycles of therapy. All eligible patients who received at least one dose of study medication were considered evaluable. The CR rate and ORR (CR/PR) were estimated using the exact method, and patients who are known to have deceased or withdrawn consent prior to a response assessment were considered non‐responders. The complete and overall response rate using the 2014 Lugano criteria was also estimated using the Koyama and Chen method with corresponding 90% confidence intervals (CIs) to reflect the inclusion of an interim analysis for futility in the trial design. Safety was assessed using CTCAE v5.0, and the maximum toxicity grade per patient of each adverse event was derived and presented in tabular form. Time to event data are presented as Kaplan–Meier plots and summary tables of Kaplan–Meier estimates for criterion rates at fixed timepoints, with 95% CIs. Median follow‐up was estimated by the reverse Kaplan–Meier method. The trial was registered at Clinicaltrials.gov (NCT03980171). RESULTS A total of 50 patients were enrolled between September 2019 and November 2023: 18 patients in dose escalation and 32 in dose expansion. Six patients were treated at dose Level 4 during dose escalation, making a total of 38 patients treated at the RP2D. Key baseline characteristics include median age 60 (range 33–78) years, Stage III or IV in 39 (78%), B symptoms in 8 (16%), and PRIMA‐PI scores of 0, 1, and 2 in 21 (42%), 21 (42%), and 8 (16%) patients, respectively (Table 1). As the total number of nodal sites was not collected for all patients, the FLIPI score could not be calculated. Patients received a median of 6 (range 1–6) cycles of induction therapy, with 34 (68%) experiencing treatment delays during induction, due to neutropenia in 28 (56%), infections in 5 (10%), and thrombocytopenia in 2 (4%). These delays were >7 days in 52% of instances. During maintenance, 53% of patients experienced dose delays: due to infection in 8 (16%), thrombocytopenia and neutropenia in 2 each (4%). Table 1Baseline characteristics. n = 50Median age (range), years60 (33–78)Male25 (50%)ECOG041 (82%)19 (18%)StageII11 (22%)III29 (58%)IV10 (20%)PRIMA‐PI021 (42%)121 (42%)28 (16%)B symptoms8 (16%)GELF criteria presentNodal mass ≥7 cm27 (54%)Involvement of ≥3 sites ≥3 cm24 (48%)Presence of constitutional symptoms8 (16%)Splenomegaly2 (4%)Organ compression8 (16%)Pleural effusion or ascites3 (6%)Leukemic phase disease0 (0%)Cytopenias0 (0%)Elevated beta‐2‐microglobulin or lactate dehydrogenase8 (16%)Abbreviations: BMI, body mass index; ECOG, Eastern Cooperative Oncology Group.John Wiley & Sons, Ltd. Baseline characteristics. Abbreviations: BMI, body mass index; ECOG, Eastern Cooperative Oncology Group. During induction, the median relative dose intensity for obinutuzumab, lenalidomide, and venetoclax was 94.0% (interquartile range [IQR]: 89.0−100.0), 91.2% (IQR: 82.3−99.1), and 92.3% (IQR: 3.6−110.0), respectively. Dose adjustments for venetoclax and lenalidomide occurred in 28 (54%) and 21 (42%) of patients during induction, respectively. The most common reasons for venetoclax dose adjustment were neutropenia (n = 5), nausea (n = 5), and infection (n = 4). The most common reasons for lenalidomide dose adjustment were rash (n = 6) and nausea (n = 3). Obinutuzumab dose modifications were not permitted per protocol. Patient disposition Patient disposition at the time of data cutoff (13 June 2024) is summarized in Figure 1. A single patient had deceased, due to T‐cell prolymphocytic leukemia 30 months following study entry. In retrospect, on baseline peripheral blood flow cytometry, a small population of CD4/CD8 dual‐positive T‐cells had been present, consistent with the subclinical presence of the second malignancy prior to commencement of study medications. Four patients were recorded as having progressive disease (PD), with three of these events as histologic transformation to DLBCL (n = 2) or high‐grade B‐cell lymphoma with rearrangements in MYC and BCL2 (n = 1). The remaining patient with a persistent FDG‐avid mesenteric node at the end of induction did not undergo biopsy to confirm the nature of the apparent refractory disease. However, this patient received only rituximab maintenance off study without further progression, and negative ctDNA MRD in the peripheral blood at this timepoint suggests the PET‐CT was a false positive. Patient disposition. ALT, alanine transaminase; DL, dose level; IRR, infusion related reaction; PD, progressive disease. Safety and tolerability Seven patients (14%) discontinued treatment permanently during the induction phase and a further 4 (8%) during the maintenance phase due to toxicity. One was due to a DLT of Grade 3 alanine transaminase elevation after C1D1 venetoclax, which was asymptomatic and resolved within 72 h without clinical sequelae or treatment. No other DLTs occurred. The other reasons for permanent discontinuation included infection (n = 4), hematologic toxicity (n = 4), T‐PLL (n = 1), and infusion reaction (n = 1). Adverse events (regardless of causality) occurring in ≥10% of patients are summarized in Figure 2A. Neutropenia was the most frequent adverse event, with Grade 3 or 4 events in 70% of patients. Febrile neutropenia was rare, occurring in 2 (4%) patients. No venous thromboembolic events occurred. Biochemical tumor lysis syndrome (TLS) occurred in two patients (both on C1D1), but clinical TLS was not observed. The next most common adverse events were diarrhea (any grade 62%, Grade 3+ 4%), upper respiratory infection (48%, 2%), nausea (38%, 2%), and maculopapular rash (38%, 6%). Second primary malignancies occurred in 4 patients: T‐PLL (n = 1) and 13 episodes of non‐melanoma skin cancer in 3 patients (12 Grade 2, 1 Grade 3). Two of the three had a history of non‐melanoma skin cancers prior to study entry. (A) Adverse events (all causality) occurring in >10% of patients. (B) Progression‐free survival (PFS). (C) Overall survival. GORD, gastroesophageal reflux disease. Efficacy The CR rate at the end of induction among the 32 patients treated at the RP2D (venetoclax 800 mg daily continuous) was 86% (95% CI 71, 95). Among all 49 evaluable patients, the CR rate was 83% (95% CI 70, 93). The ORR was 92% (95% CI 78, 98) at the RP2D and 90% (95% CI 78, 97) among all evaluable patients. After a median follow‐up of 18.3 months, the estimated 2‐year PFS and OS rates were 92% (95% CI 80, 97) and 100%, respectively (Figure 2B,C). ctDNA MRD assessment Molecular characterization of the cohort was performed by NGS from baseline tumor samples for 48 patients with results shown in Figure 3A. Variant reporter lists were generated for each patient from both baseline tissue and ctDNA samples with unique and overlapping variants being contributed by each compartment. Across the cohort, 2416 variants were shared, 2920 variants were tissue‐restricted, and 798 variants were ctDNA‐restricted. Variant reporter lists were able to be generated for all 48 patients with available baseline samples. The median number of reporters identified per patient was 111 (range 21–366) with 73% of SNV phased across the cohort (49%–88% per patient). The most frequent contributing loci were IGH, IGK, IGL, BCL2, and BCL6 (Figure 4A). (A) Oncoplot of baseline molecular characteristics of the cohort determined by next‐generation sequencing. Individual patients are listed in columns with genes harboring genomic variants listed in rows. The number in each individual cell indicates the number of variants detected in that gene, with empty cells indicating no variants detected. Patients with disease progression for correlative analysis are indicated in the first row in red cells. (B) Histogram of most frequently mutated genes contributing to reporter variants. The most frequently mutated genes were immunoglobulin light and heavy chain loci, BCL2 and BCL6. (C) Circulating tumor DNA (ctDNA) measurable residual disease (MRD) detection from Cycle 1 to Cycle 3 of LEVERAGE therapy. Progressing patients are indicated with colored circles. NA indicated sample not available for testing. Patients with progressive disease at any timepoint are indicated as colored circles within each vertical timepoint bar. (D) Variant allele fraction (VAF) over time for eight patients with progressive disease. Each different colored line represents a different patient. Enlarged dots indicate time of progression after treatment commencement (months). Star indicates transformation to high‐grade disease (three patients not biopsied at progression). (A) Genes contributing reporter variants across the entire cohort. Each bar indicates a different gene listed in increasing genomic coordinate order across the genome. The height of the bar indicates the number of variants contributed to reporter lists (log2 scale). Selected high‐frequency genes are annotated. (B) Genomic characterization in baseline versus progression biopsies. Blue bars indicate individual variants detected in baseline biopsies with pink bars indicating those present in relapse. The height of the bars is the variant allele fraction detected in tissue samples. Prior to treatment, 46 (100%) of patients with available ctDNA samples had detectable MRD. After Cycle 1 (obinutuzumab and venetoclax), 19/46 (41%) of patients had undetectable MRD. The undetectable MRD rate continued to increase with ongoing therapy being 81% (38/46) after Cycle 2 and 91% (42/46) after Cycle 3 (Figure 3B). Of the 19 patients with undetectable MRD after Cycle 1, only one patient has developed PD to date, resulting in a negative predictive value (NPV) for relapse for undetectable MRD after Cycle 1 of 95% (95% CI 75.4%–99.1%). After Cycles 2 and 3, the NPV dropped to 89.5% (95% CI 75.9%–95.8%) and 88% (95% CI 75%–94.8%), respectively. Of the four patients with detectable MRD after Cycle 3, three patients have developed disease relapse resulting in a positive predictive value (PPV) for relapse of detectable MRD after Cycle 3 of 75%. (95% CI 30.1%–95.4%). PPV increased from 26% (95% CI 13.3%–44.7%) after Cycle 1, when seven patients with PD were positive, to 44.4% (95% CI 18.9%–73.3%) after Cycle 2 when four patients with PD were positive. Analysis of the kinetics of the ctDNA response throughout treatment for eight patients with subsequent PD was performed (Figure 3C). One patient with PD had undetectable MRD after the first cycle of therapy; this patient had early progression diagnosed on PET‐CT due to a single FDG‐avid mesenteric lymph node, which was not biopsied. In addition to achieving early ctDNA MRD negativity, the patient had undetectable MRD at the point of metabolic progression. To date, this patient has only received post‐trial single‐agent rituximab suggesting a false positive for progression on imaging. Another patient had PD at 54 months post‐therapy. The patient had undetectable ctDNA MRD at progression using the reporter list generated from baseline tissue. Genomic characterization of the progression biopsy showed significant clonal evolution when compared to baseline (Figure 4B) with comparison of the baseline and progression tissue reporters showed almost no overlap of reporters. ctDNA MRD assessment using an updated reporter list including reporters from progression tissue was positive at the time of progression. The baseline tissue sample of the patient was tested using the reporters identified in the tissue at PD and was positive at a low level consistent with the presence of the progressing clone at a low level in the baseline biopsy. Tumor biopsy was available for three other patients at PD. Molecular characterization of these tumor timepoints showed evidence of clonal evolution in all samples; however, in contrast to Patient 1, all three samples had significant overlap with baseline reporters (Figure 4B). DISCUSSION The combination of obinutuzumab, lenalidomide, and venetoclax is highly active in patients with treatment‐naïve FL, with a CR rate of 86% at the RP2D and estimated 1‐year PFS and OS rates of 92% and 100%, respectively. With the caveats of cross‐study comparison, the results from this small study compare favorably to the CR rates observed from chemoimmunotherapy in GALLIUM (72%–79%) and rituximab and lenalidomide in RELEVANCE (55%, but determined by CT rather than PET‐CT). 18 When obinutuzumab has been combined with lenalidomide in treatment‐naïve FL, the CR rate observed was comparable to that observed in this study. Investigators from MD Anderson Cancer Centre treated 90 patients with treatment‐naïve FL (41% high risk FLIPI) with a longer duration of lenalidomide maintenance (24 cycles) with a 24‐week CR rate was 97% and a 3‐year PFS rate 91% (Lugano 2014). 7 LYSA investigators used a similar dosing schedule in a Phase II study, which included 100 patients (43% high risk FLIPI) and reported an end‐of‐induction CR rate of 80% (using 1999 IWG criteria, although use of PET‐CT for response assessment was near universal). 19 After a median follow‐up of 3.7 years, the estimated 3‐year PFS was 82%. The Phase I SAKK 35/15 of obinutuzumab and venetoclax in treatment‐naive patients with FL resulted in a CR rate of 68% in PET‐CT staged patients with a 30‐month PFS of 73%. 9 The only other prior study to our knowledge examining the combination of obinutuzumab, venetoclax, and lenalidomide has not been published as a manuscript, and included only patients with relapsed/refractory B‐NHL subtypes. 12 While the efficacy observed in the present study was encouraging, it was comparable to Phase II studies of obinutuzumab and lenalidomide without venetoclax in a similar population. A key question raised is therefore what additional benefit venetoclax added. This is challenging to address directly given that the data available are from Phase II studies with differing dosing schedules, methods of response assessment, and median follow‐up. These caveats aside, the efficacy appears comparable with the obinutuzumab/lenalidomide doublet studies, and perhaps favorable in comparison to the obinutuzumab/venetoclax doublet, though longer follow‐up will be necessary to make any robust evaluation. What is clearer is that the addition of venetoclax added substantially to hematologic toxicity compared with the doublet studies. The rates of Grade ≥3 neutropenia were 19% in the MD Anderson study, 47% in the LYSA studies, and 28% in the SAKK study in comparison to 70% observed here. While this did not result in significant rates of febrile neutropenia, it did result in substantial treatment delays and dose reductions, which would pose barriers for the future development of the regimen. In aggressive lymphomas (DLBCL and Richter transformation), intermittent dosing of venetoclax in combination with multiagent chemotherapy has been employed, although hematologic toxicity has still been problematic. 20 , 21 , 22 The 24% rate of Grade ≥3 thrombocytopenia was also higher than the other studies: 8% in the SAKK study and not reported in the MD Anderson and LYSA studies. The rate of discontinuation due to toxicity was also higher in this study (14%) compared with the MD Anderson (2%), LYSA (7%), and SAKK trials (4%). We may speculate that if a lower dose level had been selected for the RP2D, the regimen may have been better tolerated, given that much of the hematologic toxicity appeared cumulative and occurred well after the DLT period (2 cycles) had concluded. Molecular analysis of the patients in this study revealed a potential role of ctDNA MRD as a biomarker for relapse. Although follow‐up to date is relatively short, ctDNA MRD status is a potential prognostic biomarker for earlier disease progression for patients treated with obinutuzumab, venetoclax, and lenalidomide. While ctDNA can potentially identify false‐positive disease progression on PET, we also identified potential challenges in tracking ctDNA MRD in the context of significant clonal evolution with the possibility of false‐negative MRD results depending on the reporter list used. Newer technologies and analytical approaches will be potentially useful for resolving these issues. Limitations of this study include the modest cohort size, time taken to complete enrolment (close to 4 years), lack of availability of FLIPI scores, and non‐randomized design. Strengths include comprehensive genomic profiling and novelty of ctDNA MRD evaluation using tumor‐informed targeted NGS sequencing. The sample size does mean estimates of NPV and PPV are uncertain. In conclusion, the combination of obinutuzumab, lenalidomide, and venetoclax is highly active in patients with treatment‐naïve FL, but the addition of venetoclax when given continuously at 800 mg/day appears to increase hematologic toxicity and impacts treatment deliverability. Since this study was designed, bispecific T‐cell‐engaging antibodies (mosunetuzumab, epcoritamab, odronextamab, and surovatamig) have emerged as a promising novel therapeutic class in the management of patients with relapsed/refractory FL and are now being evaluated in treatment‐naïve patients. 23 , 24 , 25 , 26 , 27 Given their efficacy, safety, and deliverability, most attention and investment in FL drug development is understandably centered on these agents. As such, it seems unlikely that this regimen will be further evaluated despite the encouraging efficacy signal unless a biomarker for specific venetoclax sensitivity was developed. AUTHOR CONTRIBUTIONS Chan Y. Cheah: Conceptualization; investigation; funding acquisition; writing—original draft; methodology; validation; visualization; writing—review and editing; formal analysis; project administration; data curation; supervision; resources. Piers Blombery: Formal analysis; software; data curation; visualization; writing—review and editing. Amanda Johnston: Writing—review and editing; resources; project administration. Joel Wight: Resources; project administration; writing—review and editing. Katharine L. Lewis: Writing—review and editing; resources. Rory Bennett: Writing—review and editing; resources. Jing Xie: Writing—review and editing; visualization; methodology; formal analysis. Stella Vlachos: Writing—review and editing; visualization; project administration; data curation. Sally M. Hunter: Writing—review and editing; resources. Joyce Gonzales: Resources; writing—review and editing. Sally Binks: Writing—review and editing. John F. Seymour: Conceptualization; investigation; writing—original draft; methodology; validation; visualization; writing—review and editing; formal analysis; project administration; supervision; resources. CONFLICT OF INTEREST STATEMENT C.Y.C.: Consulting/advisory/honoraria—Roche, Janssen, Gilead, AstraZeneca, Lilly, BeOne Medicines, Menarini, Dizal, AbbVie, Genmab, Sobi, CRISPR Therapeutics, BMS, and Regeneron. Speaker's bureau—Janssen, AstraZeneca, BeOne Medicines, Genmab, AbbVie, Roche, and MSD. Research funding—BMS, Roche, AbbVie, MSD, and Lilly. Travel expenses—Lilly, BeOne Medicines. J.F.S.: Consultancy—TG Therapeutics and BMS. Research funding—AbbVie, BMS, Genmab, Lilly, and Roche. Honoraria—AbbVie, AstraZeneca, BeOne Medicines, BMS, Gilead, Jansen, Loxo, and Roche. J.W.: Consulting/advisory/honoraria—Roche, J&J, Gilead, AstraZeneca, BeOne, AbbVie, and Sobi. R.B.: Honoraria—AbbVie. Travel expenses—Roche. ETHICS STATEMENT This study was carried out in compliance with principles of the International Conference on Harmonization (ICH), Good Clinical Practice (GCP), Declaration of Helsinki, and all applicable national and local regulations governing clinical studies. This study was approved by the Human Research and Ethics Committee of Peter MacCallum Cancer Centre at and all authors had access to primary clinical trial data. FUNDING The study was supported by F. Hoffman‐La Roche Ltd, AbbVie, and Bristol Myers Squibb. Open access publishing facilitated by The University of Western Australia, as part of the Wiley ‐ The University of Western Australia agreement via the Council of Australasian University Librarians. Supporting information Supporting Information.