While follicular lymphoma (FL) prognosis is known to be influenced by clinical characteristics and age, investigation of modifiable factors in the modern treatment era with prognostic significance has been limited. Binding of the active vitamin D metabolite to the nuclear vitamin D receptor results in autocrine and paracrine effects possibly relevant to both cancer prevention and prognosis, including regulation of cell proliferation, induction of apoptosis and differentiation, and immune modulation. Recent literature reports a potential association between high vitamin D and improved prognosis in multiple myeloma, breast and colorectal cancer, and most recently, diffuse large B cell lymphoma and chronic lymphocytic leukemia. Using stored serum from SWOG and LYSA (formerly GELA) FL trials, we evaluated the impact of pretreatment vitamin D on outcome.
Subjects included in the SWOG cohort were previously untreated FL patients (pts) enrolled on one of three SWOG clinical trials (S9800, S9911, S0016) involving CHOP chemotherapy plus an anti-CD20 antibody (rituximab or I-131 tositumomab, enrolled 1998–2008); enrolled pts with pre-treatment serum stored and available through the SWOG serum banking protocol (S8947) were eligible for this analysis. Subjects included in our second independent cohort were also previously untreated FL pts enrolled on LYSA's PRIMA trial (rituximab plus chemotherapy, randomized to rituximab maintenance versus observation; enrolled 2004–2007), who had pre-treatment serum samples stored and available for serum 25(OH)D analysis. Baseline samples for both cohorts were sent to the Mayo Clinic Medical Laboratories where 25(OH)D2 and 25(OH)D3was measured directly using the gold standard liquid chromatography-tandem mass spectrometry method. Our primary endpoint was progression free survival (PFS), defined as time from date of enrollment (SWOG) or registration (PRIMA) to date of progression or death from any cause. Overall survival (SWOG enrollment/PRIMA registration to date of death from any cause) was also evaluated. Kaplan-Meier survival curves were estimated, and differences in survival time by vitamin D status were assessed using the log-rank test.
The SWOG cohort included 183 pts enrolled and treated in centers across the US: 55% male, 96% Caucasian, and 30% age 60 or older. Pts largely had grade 1–2 FL (90%) advanced stage (68% stage IV) FL; only 16% were poor prognosis according to IPI. Median serum 25(OH)D was 31 ng/ml. After median follow-up of 5.4 years, pts with serum 25(OH)D < 20 ng/ml (insufficiency threshold per 2010 Institute of Medicine recommendations) had significantly inferior PFS (HR 2.00, p=0.011) and OS (HR 3.57, p=0.003) as compared to those with higher levels (analyses stratified by treatment trial and adjusted for IPI). The PRIMA subset cohort included pts primarily enrolled and treated in France and Belgium: 55% male, and 39% over age 60. The majority of these pts were high risk (91% Stage III/IV, 46% ≥3 by FLIPI, 34% with B symptoms). Surprisingly, the PRIMA serum 25(OH)D distribution was notably shifted in comparison to the SWOG cohort (median 17 ng/ml). As such, we conducted the analysis in this cohort with serum 25(OH)D dichotomized at both the median (17ng/ml) and first quartile (10 ng/ml). After median follow-up of 4.5 years, pts below the median (25(OH)D <17 ng/ml) that received R-CHOP induction (n=237) had significantly inferior OS (HR 3.8, p=0.02) as compared to those with higher levels. Moreover, when dichotomized at the first quartile (10 ng/ml), PFS and OS were significantly inferior for those with lower values (PFS HR 1.73, p=0.0086; OS HR 2.7, p=0.03).
In this international collaborative study of newly diagnosed FL pts uniformly treated with chemotherapy and anti-CD20 therapy, we report a robust association between low vitamin D levels and FL outcomes in two independent cohorts. While the threshold for sufficiency with regard to FL outcomes will need to be further defined, and may vary by population and region, the observed prognostic significance of low vitamin D for FL in this study is surprisingly strong, with OS HRs indicating a magnitude of association at or greater than the individual FLIPI prognostic factors, which we currently rely upon clinically. Moreover, serum vitamin D is the first potentially modifiable factor to be associated with FL survival.
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
Asterisk with author names denotes non-ASH members.