Spanish real-life practice confirms zanubrutinib as a safe and effective option in relapsed or refractory marginal zone lymphoma: Results of a cohort of 118 patients on behalf of GELTAMO.

Abstract

Abstract Introduction: There is no established standard of care for relapsed or refractory (R/R) marginal zone lymphoma (MZL). Zanubrutinib, a second generation covalent BTK inhibitor, was approved for R/R-MZL based on the results of the phase 2 MAGNOLIA trial, which included 68 patients (pts). We aimed to evaluate the characteristics and outcomes of 118 pts with R/R MZL treated with zanubrutinib in routine clinical practice across Spain. Methods: This retrospective study included adult pts with R/R MZL previously treated with at least one anti-CD20-based regimen and requiring further therapy. Pts received zanubrutinib at 39 Spanish centers (GELTAMO). All MZL subtypes were included: mucosa-associated lymphoid tissue (MALT), splenic (SMZL), and nodal (NMZL). Baseline characteristics, prior lines of treatment (LoT), response rates, survival, adverse events (AEs), and reasons for discontinuation were analyzed. Results: A total of 118 pts were analyzed. Median age at zanubrutinib initiation was 75 years (range, 36-96) and 56% were females. According to MZL type: 55.3% were SMZL, 18.9% MALT, 19.8% NMZL, and 6% no-classifiable. Median number of previous LoT was 2 (range 1-7): 59 pts 1L, 30 pts 2L and 29 pts ≥3L. Thirty pts were refractory to the prior therapy immediately preceding zanubrutinib and 59% (69/116) were POD24 (R/R within the 24 months after first line treatment initiation). Median time from diagnosis to zanubrutinib was 56.7 months (IQR 25.2-93). At zanubrutinib start, 84% had stage III/IV disease, 32% had B-symptoms, and 15% had ECOG ≥2. MYD88 L265P status was available in 35 pts; 7 cases (20%) mutated. Thirty-one pts (26.3%), including 8 with ≥1 MAGNOLIA exclusion criteria, would not have qualified for the trial. The most common reasons were thrombocytopenia <75 × 10⁹/L (n=8), neutropenia <1 × 10⁹/L (n=6), elevated bilirubin (n=6), poor performance status (n=4), creatinine clearance <30 mL/min (n = 4), AST/ALT >2.5× ULN (n = 3), prior treatment with BTKi (n = 2), and CNS involvement (n = 1). No pts had prior histological transformation. Comorbidities were frequent: 68% had ≥1 cardiovascular risk factor (hypertension 48%, diabetes 29%, cardiopathy 22%, including 10 with AF), and fifty-one pts (43%) presented with ≥2 comorbidities. Anticoagulant or antiplatelet therapy was used in 33 pts (28%), with 2 pts receiving both therapies. Zanubrutinib was started at standard dose in 111 patients (160 mg BID or 320 mg QD), and at reduced dose in 7 (80 mg BID or 240 mg QD). Overall response rate (ORR) was 76.0%, with complete response (CR) in 22.2%. Median duration of response was 6.8 months (IQR 4.5 -9.6), and median time to best response was 5.8 months (IQR 3.7-9.9). ORR/CR was not significantly affected by MZL subtype, stage, MYD88 status, POD24, MAGNOLIA eligibility, age, or refractoriness to immediately prior line. However, number of prior LoT did influence outcomes (ORR/CR: 84.7%/30.5% for 1 LoT, 73.5%/10% for 2 LoT, 55.6%/17.2% for >2 LoT; p<0.05). With a median follow-up of 14 months, one-year progression-free survival (PFS) was 79.4 (95% CI 72.1-87.6), median PFS not reached. One-year PFS was 77.8% (SMZL), 77% (MALT), 88.6% (NMZL), and 77.8% (unclassifiable)(p=0.475). At last follow-up, 86 pts (72.8%) remained on treatment. One-year OS was 86.8% (95% CI 80.5-93.5). Sixteen pts had died (8 from lymphoma). Response to zanubrutinib significantly impacted both PFS and OS (p<0.001), while number of prior LoT impacted PFS only (p<0.001). No significant differences in survival were seen by MZL subtype, POD24, or MAGNOLIA eligibility. Regarding toxicity, 49 AEs were reported in 40 pts (34.1%), of which 17 (34.6%) were grade ≥3. Most common AEs included bleeding (n=13), infections (n=10), cytopenias (n=8), and diarrhea (n=5). Bleeding was higher among those on anticoagulants (p=0.039). No cases of atrial fibrillation or grade 5 AEs occurred. Nine pts discontinued zanubrutinib due to toxicity (median time to discontinuation: 5.4 months (IQR 2.6-11.1)). MAGNOLIA ineligible pts had higher AE incidence (p=0.017) and discontinuation rates (p=0.009). Conclusions: This is the largest real-world series of zanubrutinib in R/R MZL to date. Zanubrutinib demonstrated favorable efficacy and safety across all MZL subtypes, independently of MAGNOLIA trial eligibility. Outcomes were significantly better in pts treated earlier in their disease course, supporting the use of zanubrutinib as a second-line option in R/R MZL. (*AMP & FMM contributed equally)