Introduction: Lenalidomide in combination with anti-CD20 monoclonal antibody is a standard second line treatment for relapsed/refractory follicular lymphoma (R/R FL). However, in the pivotal phase 2 trial of mosunetuzumab for third line and later treatment of R/R FL, only 13% of patients had previously received lenalidomide, raising concern about its applicability to real world practice. Due to lenalidomide's immunomodulatory effects, mechanisms of resistance may overlap, and the efficacy of mosunetuzumab could be impaired in patients previously exposed to lenalidomide.

Methods: To assess potential differences in treatment efficacy and tolerance according to previous exposure to lenalidomide, we conducted a multi-center retrospective analysis of adult patients with R/R FL treated with standard of care mosunetuzumab between June 2023 and February 2025 across 9 centers participating in the Collaborative US Bispecifics Consortium (CUBIC). Cytokine release syndrome (CRS) and Immune Cell Associated Neurotoxicity Syndrome (ICANS) were graded by ASTCT criteria. Response was assessed by local investigators per Lugano 2014 criteria. Differences in categorical or continuous variables between patient groups were evaluated by Chi-square test/Fisher's exact test or Wilcoxon rank sum test, as appropriate. Log-rank test was used to compare the difference in progression-free survival (PFS) between patient groups.

Results: Seventy-seven patients were included in the analysis, and 30 (38.9%) had prior lenalidomide exposure. At the time of mosunetuzumab initiation, median age was 68 years (range, 33-92), 36 (46.8%) were female and 67 (89.3%) were Caucasian. Thirty-three (42.9%) patients had progressed within 24 months of frontline chemoimmunotherapy and 69/75 (92.0%) had CD20 expression by either immunohistochemistry or flow cytometry. The median number of prior lines of systemic therapies was 3 (1-14), the median largest lymph node size was 4.5 cm (0.6-19.7 cm), 13/64 (20.3%) patients had disease greater than 7 cm, and 29/72 (40.3%) had more than 4 nodal disease sites involved. Lenalidomide-exposed patients had received a higher median number of prior lines of systemic therapy compared to the lenalidomide unexposed group (4 vs 3, p=0.0169). No other significant differences in baseline characteristics were observed.

Twenty-six (35.1%) patients developed CRS (grade 1 in 22, grade 2 in 4); no ICANS events were observed. Twenty (26.7%) patients developed infectious complications during treatment. No differences in the frequency of CRS of any grade (27.6% vs 40%, p=0.2748) nor infectious complications (27.6% vs 26.1%, p=0.8863) were observed when comparing lenalidomide-exposed patients to those who were unexposed. Overall response rate (ORR) to mosunetuzumab was 87.7% and complete response (CR) rate was 69.9%. No significant differences in ORR (86.2% vs 88.7%, p=0.8495) or CR rate (72.4% vs 68.2%, p=0.6998) were observed when comparing lenalidomide-exposed patients to those who were unexposed.

After a median follow up of 13.1 months (95% confidence interval [CI], 12.1-14.9), median PFS was 17.7 months (95%CI: 12.5 - not reached) with a 1-year PFS rate of 64.1%. No significant differences in median PFS were observed when comparing lenalidomide-exposed patients to those who were unexposed (20 months vs 17.2 months, p=0.96). No significant differences in median PFS were observed when comparing 8 patients with recent lenalidomide exposure (≤ 24 months) to 69 with either distant lenalidomide (> 24 months) or no exposure (not reached vs 17.7 months, p=0.56). When comparing patients who responded (n=15) to those who were refractory to lenalidomide (n=12), there was also no PFS difference with mosunetuzumab (20.0 months vs 12.5 months, p = 0.79).

Conclusions: In this first real-world multi-center study, prior lenalidomide exposure does not seem to impact safety and efficacy of mosunetuzumab in patients with FL. Longer follow up and cohorts including a larger fraction of patients with recent (< 24 months) lenalidomide exposure are needed to confirm these hypothesis-generating findings.

This content is only available as a PDF.
2025
Sign in via your Institution