Autologous Stem Cell Transplantation Improves Survival for Patients with Follicular Lymphoma in First or Second Relapse: Results of a Comparative Effectiveness Instrumental Variable Analysis

Abstract

IntroductionThe role of high dose therapy with autologous stem cell transplantation (ASCT) in relapsed follicular lymphoma (FL) remains controversial, with lack of comparative studies in the Rituximab (R) era. We conducted a comparative effectiveness analysis to evaluate the use of ASCT at first or second (1st/2nd) relapse on survival outcome in two cancer centers in Alberta. Both centers treat a similar catchment population within the same government-funded health care system, providing simultaneous access to new cancer treatments. Treatment center A however, routinely offers ASCT at 1st/2nd FL relapse whereas treatment center B tends to offer ASCT later or never. Both centers occasionally offer allogeneic transplantation later in the disease course.MethodsThe Alberta Cancer Registry was interrogated to identify FL patients diagnosed between 2001 and 2010. Patients who were aged 18-60 years at initial diagnosis and experienced a relapse post initial chemotherapy treatment regardless of first-line management (watch and wait, chemotherapy, chemoimmunotherapy or radiation) were included in this analysis. Patients who had grade 3B or areas of concurrent diffuse large B-cell lymphoma at initial diagnosis were excluded.ResultsA total of 568 patients aged 18-60 years were diagnosed with FL during the 10 year study period in Alberta, and 180 patients meeting inclusion criteria experienced relapse from chemotherapy and received treatment at center A (n=96) or center B (n=84). There were no statistically significant differences in patient and disease characteristics at diagnosis or at relapse between the two centers. The median follow-up was 104.4 months (28.1-156.1) from diagnosis and 65.3 months (0.3-151.8) from first relapse post chemotherapy.The proportion of patients ever receiving ASCT was higher at center A compared to B (61.46% vs.16.67%, p<0.001) especially those receiving at 1st/2nd relapse (58.3% vs. 7.1%, p<0.001). Center A also utilized allogeneic transplantation more frequently (16.7% vs. 3.6%, p=0.004). The use of R-chemotherapy (89.6% vs. 82.1%, p=0.46), R-maintenance therapy (34.4% vs. 39.3%, p=0.50), Doxorubicin-based chemotherapy (60.4% vs. 61.9%, p=0.84), Fludarabine (19.8% vs. 29.8%, p=0.12) were similar for center A vs. B, respectively, although center A less commonly enrolled patients on clinical trials (12.5% vs. 39.3%, p<0.01).The projected 5-year overall survival (OS) from first relapse was significantly higher in center A vs. B (88.9% vs. 59.5%, HR 3.24, p<0.001) (Figure 1). For these 180 patients, the projected 10-year OS from initial diagnosis was 85.4% in center A vs. 60.5% in center B (HR 3.02, p=0.003).The factors associated with improved OS from relapse at univariate analysis were female gender (p=0.01), no co-morbidities (p=0.01), FLIPI score 0-2 at diagnosis (p<0.001) and TTP >1 year (p=0.002). Receiving ASCT was highly associated with improved OS (p=0.006) as well as timing of the ASCT (p<0.001). Patients receiving ASCT at 1st/2nd relapse had projected 5-year OS of 92.4% vs. 66.5% for no ASCT vs. 62.5% for ASCT beyond second relapse (p<0.001) (Figure 2). In contrast, allogeneic transplantation did not affect OS (p=0.62).In multivariate analysis, factors that were independently associated with improved OS were treatment center A (HR 3.16, p=0.001), FLIPI score 0-2 at diagnosis, no transformation, ever use of Rituximab with chemotherapy, and ever use of R-maintenance. Another multivariate model examining the OS of all patients without considering treatment center found that ASCT at 1st/2nd relapse was associated with improved OS (HR 4.55, p=0.002) independent of FLIPI score 0-2 at diagnosis, no transformation, ever use of Rituximab with chemotherapy, and ever use of R-maintenance. Because treatment center and use of ASCT at 1st/2nd relapse were highly correlated, they could not be combined within the same multivariate model.ConclusionsThis is the first study assessing the comparative effectiveness of different treatment approaches in FL patients in their first relapse post chemotherapy, in particular the use of ASCT. Our results illustrate significantly improved survival in patients treated with high dose therapy with ASCT as well as highlighting the importance of timing of ASCT. We therefore support the use of ASCT at 1st/2nd relapse in relapsed FL patients. [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.