Abstract
Background: Anti-CD20 maintenance prolongs progression-free survival (PFS) in follicular lymphoma (FL) but sustains B-cell depletion and blunts vaccine responses. Methods: we conducted a retrospective study comparing two Italian centres that, during the COVID-19 pandemic, adopted opposite policies: Centre A continued maintenance; Centre B discouraged it. Results: Among 62 consecutive patients completing first- or second-line induction in 2019–2022, maintenance was started in 47 (95% and 41% by centre; 92% and 27% among patients aged ≥70 years); patients not receiving it were older (79 vs. 66 years, p < 0.001). After a median follow-up of 70 months, crude 5-year PFS favoured maintenance (59% vs. 27%; hazard ratio [HR] 0.39), but the advantage shrank after adjustment for age (HR 0.72, 95% CI 0.30–1.72) or overlap propensity weighting (HR 0.60, 0.27–1.31); overall survival did not differ. The strong association between more than four cycles and survival (HR 0.22) faded at a 12-month landmark (HR 0.61, 0.19–1.95), consistent with immortal-time bias. All five COVID-19 deaths (10.6%) occurred in maintained patients, 2.0–3.7 months after the last dose and mostly without seroconversion, and none among the older patients without maintenance. Censoring these deaths brought the PFS benefit back within the range of randomised trials (weighted HR 0.41, 0.17–0.96). Conclusions: In this small real-world cohort, COVID-19 mortality appears to have offset much of the expected gain from maintenance during the pandemic.