Low vitamin d is an unmodifiable adverse prognostic factor in older DLBCL patients: Results from the previd randomized trial by fondazione italiana linfomi (FIL).

Abstract Background: R-CHOP remains the standard of care for fit elderly patients with diffuse large B-cell lymphoma (DLBCL), while modified regimens such as R-mini-CHOP or substituting conventional doxorubicin with its liposomal formulation (R-COMP) are preferred for unfit or frail individuals. Vitamin D (VitD) deficiency has been associated with worse outcomes in DLBCL. The Elderly Prognostic Index (EPI), combining simplified Geriatric Assessment (sGA), hemoglobin, and IPI, is the first index specifically designed for this population. The PREVID study aimed to test whether VitD supplementation before and during treatment could improve survival in older patients. Methods: PREVID is a multicenter, randomized trial conducted by the Fondazione Italiana Linfomi (FIL). Patients aged ≥65 years with untreated DLBCL underwent a standardized prephase with oral prednisone before starting chemoimmunotherapy (R-CHOP, R-mini-CHOP, R-COMP, or R-mini-COMP). Patients were randomized 1:1 to a reference arm (prephase only) or an experimental arm (prephase plus VitD). In the experimental arm, cholecalciferol 25,000 IU/day was administered for 7 days (if baseline 25(OH)VitD was 20–40 ng/mL) or 14 days (if <20 ng/mL), followed by weekly maintenance during treatment. If VitD levels were still <30 ng/mL at cycle 2, a second loading phase was given. All patients underwent sGA before treatment. The primary endpoint was progression-free survival (PFS); secondary endpoints included overall survival (OS) and correlation with EPI and baseline VitD levels. Results: Between 2020 and 2024, a total of 343 patients were enrolled; after the exclusion of 29 screening failures, 314 were evaluable (156 reference arm, 158 experimental arm). Median age was 76 years (range 65–91). According to sGA, 56% of patients were classified as fit, 30% as unfit, and 14% as frail, with similar distribution across study arms. Based on EPI scores, 20.4% were low risk, 49.4% intermediate, and 30.2% high risk, again with balanced representation between study arms. Induction treatments (R-CHOP 29%; R-mini-CHOP 10%; R-COMP 34%; R-mini-COMP 27%) were equally distributed between the two arms (Chi² p=0.066). A total of 239 patients completed all 6 cycles (117 ref, 122 exp), and 230 were evaluable for response. VitD levels significantly increased in the experimental arm post-prephase (p<0.001), while they remained unchanged in the reference arm (p=0.242). After a median follow-up of 21 months (range 1–48), the 3-year PFS was 61% (95% CI, 50–71) in the reference arm and 59% (95% CI, 49–67) in the experimental arm (p=0.405). Baseline VitD levels <16 ng/mL were associated with inferior 3-year PFS (53% vs. 66%, p=0.036) in both arms. EPI was prognostic: patients with high-risk EPI had significantly shorter PFS (HR 2.76, 95% CI 1.45–5.26, p=0.002) and OS (HR 4.40, 95% CI 1.70-11.40, p=0.002). sGA stratification (fit vs. unfit and frail) also correlated with outcome, with PFS 69% vs 49% at 3 years (HR 2.01, 95% CI, 1.34-3.01, p=0.001). At data cutoff, 59 deaths were recorded (19% of the entire cohort). The 3-year OS was 77% (95% CI, 66–85) in the reference arm and 72% (95% CI, 62–80) in the experimental arm. There was no statistically significant difference in OS (HR 1.26, 95% CI 0.76–2.11; p=0.373), and this remained true after adjusting for sGA, IPI, and treatment (adjusted HR 1.21, 95% CI 0.71–2.06; p=0.477). When stratified by treatment regimen, no statistically significant difference in PFS was observed between the reference and experimental arms within the R-CHOP or R-COMP groups. In terms of toxicity, higher rates of grade 3–4 neutropenia were seen in the experimental arm (27.6% vs 15.3% p=0.014) without an increase in infection rates. Conclusions: This is the first randomized trial evaluating VitD supplementation in older DLBCL patients. Although supplementation effectively corrected biochemical deficiency, it did not improve PFS or OS, and low baseline VitD remained a negative prognostic factor. EPI was strongly prognostic in stratifying risk for both PFS and OS. In conclusion, while VitD deficiency retains prognostic value, the reason behind this mechanism remains unknown.