Total Metabolic Tumor Volume Is Confirmed As Independent Prognostic Factor in Treatment Naïve Follicular Lymphoma Patients and Can be Combined with FLIPI2 to Improve Prognostic Accuracy. a FOLL12 Substudy By the Fondazione Italiana Linfomi

Abstract

Aim/Introduction: The prognostic role of FDG PET/CT parameters before first line therapy in patients with Follicular Lymphoma (FL) remains unclear, with conflicting data published in the literature. To this end we evaluated the prognostic value of baseline Total Metabolic Tumor Volume (TMTV) in a large cohort of treatment-naïve FL patients enrolled in the FOLL12 trial (NCT02063685). Materials and Methods: FOLL12 is a multicenter, randomized, phase III trial that compared standard (Arm A) vs response adapted maintenance (Arm B) in patients with grade 1-3a advanced stage FL who achieved complete (CR) or partial response (PR) to induction rituximab based immunochemotherapy (R-ICT). In the current study we included patients for whom baseline FDG PET/CT was available and centrally reviewed. The TMTV was obtained by expert nuclear physicians from baseline scans by summing the metabolic volumes of all individual nodal and extra nodal lesions, using the 41% SUVmax threshold method. Univariate and multivariate analyses were performed using Cox proportional hazard model. Main study endpoint was Progression Free Survival (PFS). Results: starting from the FOLL12 trial population, 692/807 patients were included in this study; 48% were older than 60 years, 89% had stage III-IV disease and 40% had a high-risk FLIPI-2 score(3-5 risk factors). Overall, the 5-year PFS was 79% (95% CI, 76 to 82%) and the median TMTV was 242 mL (IQR=446). The optimal cutoff identified for TMTV was 200ml. TMTV values were higher in association with unfavorable prognostic features (Hb <12 g/dl, LoDLIN > 6cm, elevated beta2 microglobulin and LDH, and number of nodal areas >4) whereas were similar between study arms, and ICT regimens. In univariate analysis, 5-year PFS was significantly lower for patients with high vs low TMTV (60% vs 75%; HR=1.87 [95%CI: 1.41-2.49], p<0.001). The correlation between TMTV and PFS was confirmed both in the standard maintenance arm and in the response adapted arm of the FOLL12 trial and the independent prognostic role of TMTV was confirmed when adjusted by study arm, FLIPI-2, and type of ICT. In multivariate analysis, high TMTV (HR 1.47 (1.09 - 2.00); p=0.012) and high risk FLIPI2 (3-5 RFs: HR2.05 (1.54 - 2.73) p<0.001) retained their independent prognostic role for predicting PFS. Three groups of patients were identified which were associated with different 5-year PFS rates; TMTV < 200ml and intermediate FLIPI2: N=241 (35%), 5yr PFS 78% (95% CI 71-83); TMTV > 200ml or high risk FLIPI2: N=237 (34%), 5yr PFS 68% (60-74), HR 1.53 (1.06 - 2.22); TMTV >200ml and high risk FLIPI2: N=214 (31%) 5-yr PFS 51% (43-58), HR 2.96 (2.10-4.19). The combined TMTV + FLIPI2 model was internally validated with 1000 bootstrap resampling (c-Harrel 0.616) Conclusion: Pre-treatment TMTV is confirmed as a strong independent predictor of PFS in patients with FL receiving frontline R-based ICT The combination of TMTV and FLIPI2 may contribute to improve risk stratification in guiding risk adapted therapy. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal