Long Term Follow-up of the PRIMA Study: Half of Patients Receiving Rituximab Maintenance Remain Progression Free at 10 Years

In patients with follicular lymphoma and a high tumor burden, the intergroup PRIMA phase III study was designed to evaluate the potential benefit of 2 years of rituximab (R) maintenance after response to first line R-chemotherapy induction regimens (patients registered from 12/2004 until 4/2007). At a median follow up of 3 years (Lancet 2011) and 6 years (Blood 2013 122:509), a significant sustained improvement in progression free survival (PFS) was demonstrated in patients receiving R-maintenance; hazard ratios (HR) of 0.55 and 0.58, respectively. We report here the long term results of this study with 4 additional years of follow-up.

Patients initially randomized between observation (n=513) and R-maintenance (n=505) were all followed for 7 years after randomization and thereafter until the 31/12/16 after having given an informed consent for this extended follow-up period consisting of a yearly physician visit (imaging procedures were according to local center practices after year 7). At data-base lock, the median follow-up of the entire patient cohort was 9 years (9.7 years for the 607 patients agreeing to extended follow-up).

Median PFS for patients in the observation arm was reached at 4.06 years as compared to 10.49 years in the R-maintenance arm (Log-Rank, P<.0001; HR=0.61, 95% confidence interval (95%CI) 0.52-0.73). At 10 years, 51% of the patients in the R-maintenance arm (versus 35% in the observation arm) were estimated to be free of disease progression. The benefit of R-maintenance for PFS was significant in all predefined patient strata: age (cutpoint=60 years), gender, the 3 FLIPI categories, quality of response to induction immunochemotherapy (CR/CRu and PR) and in patients having received R-CHOP (HR=0.57; 95%CI 0.47-0.70), while only a trend was observed for those having received R-CVP (HR=0.75; 95%CI 0.53-1.07). In a Cox regression model, R-maintenance remained an independent covariate from the above characteristics (HR=0.60; 95%CI 0.50-0.71). The median time to new anti-lymphoma treatment was 6.1 years in the observation arm but has not yet been reached in the R-maintenance arm (P<.0001; HR=0.66; 95%CI 0.55-0.78), with a 10-year estimate of 53% of the patients in this arm not having received a new treatment (versus 41% in the observation arm). At data base lock, 84 and 88 patients have died in the observation and R-maintenance arm, respectively. Main causes of death in the observation and R-maintenance arms were respectively: lymphoma progression in 38 and 39 patients, malignancies for 24 and 6 patients (with respectively 7 and 2 myeloid disorders; no other specific distribution pattern) infections in 6 and 11 patients (1 PML in each arm), and cardiovascular disorders in 4 and 8 patients. No new safety signals were identified with the additional 4 years of follow-up. Overall survival estimates (Kaplan-Meier) at 10 years were identical (80%) in each arm. Age >= 60 years, male sex and high FLIPI category, but not randomization arm, were all associated with a higher risk of death in a Cox regression model.

In summary, PRIMA study long term follow-up demonstrates that R-maintenance after induction immunochemotherapy provides a significant long term PFS benefit over observation. Despite the lack of OS benefit, it is noteworthy that more than half of the patients in the R arm remain free of disease progression and have not required new anti-lymphoma treatment beyond 10 years. With the prolonged life-expectancy of patients with follicular lymphoma, it is important to consider long term treatment-related toxicities and the risk of secondary malignancies related to repeated therapeutic interventions. Obtaining truly durable response with 1^st line induction followed by R-maintenance remains an appealing treatment strategy for these patients.