Baseline PET Metabolic Tumor Volume Predicts Outcome in Advanced Follicular Lymphoma Patients Who Received First-Line Immunochemotherapy but Not Those Treated with Lenalidomide-Rituximab in the Phase III Relevance Study

Abstract

Introduction The prognostic value of total metabolic tumor volume (TMTV) at diagnosis in follicular lymphoma (FL) remains debated. Its impact was reported for patients treated mostly with Rituximab (R)-CHOP chemotherapy without R maintenance (FOLLCOLL study, Meignan et al, J Clin Oncol. 2016; 34(30):3618-3626) but manual TMTV assessment was not confirmed as prognostic in the GALLIUM trial where most patients had bendamustine and anti-CD20 antibody maintenance (Barrington et al, Blood 2018; 132(Suppl 1):2882). There is no data on the prognostic value of TMTV in chemo-free regimens such as the immunomodulatory combination of lenalidomide and rituximab (R2). Here, we investigated the prognostic value of baseline TMTV for FL patients with previously untreated advanced FL included in in the phase 3 RELEVANCE trial comparing R2vs R-chemo with both regimens followed by rituximab maintenance therapy (NCT01476787 and NCT01650701). Methods TMTV was computed on baseline PET/CT using the 41% SUVmax method. To separate high and low TMTV, we tested the previously published cut-off of 510cm3 (Meignan et al. J Clin Oncol. 2016; 34(30):3618-3626). Survival was estimated using Kaplan Meier curves. Analyses were performed on the total PET-evaluable population (n=406/1032) with 196 pts on the R2 arm and 210 pts on the R chemo arm, because baseline PET was not mandatory in this trial. Results Clinical baseline characteristics of PET evaluable patients were similar to the overall trial population for age, Ann Arbor stage, bone marrow infiltration, LDH, beta 2 microglobulin and FLIPI, and well balanced between the two treatment arms. After a median 6.5 years follow-up, 5y-PFS was 67.6%, 5y-OS was 93.5%, with no significant difference between the two arms. Median baseline TMTV was 284 cm3 in the whole cohort (Q1-Q3; 144-553) : 274 cm3 (129-526) in the R2 arm and 288 cm3 (148-562) in the R chemo arm (p=0.579). Overall, patients with TMTV > 510 cm3 had higher Ann Arbor stage (p<0.001), more extra nodal sites (p<0.001), more frequently elevated beta 2 microglobulin (p<0.001), higher FLIPI (p=0.002) and FLIPI2 scores (p<0.001). High TMTV was significantly associated with an inferior PFS (p=0.0121) and POD 24 (p=0.0014) but not OS (p=0.247). Patients with a high TMTV (n=111, 27%) had a 5-year PFS of 53.9% (CI 95% 43.6-63.1) vs 70.5% (CI 95% 64.7-75.6) for low TMTV patients. By contrast, bulk as measured by CT (either >6cm or >7cm) was not prognostic of PFS (p=0.978 and p=0.661 respectively). When testing FLIPI with TMTV both were independently associated with PFS (p=0.0119 and p=0.0379, respectively). When testing FLIPI2 with TMTV, only TMTV remained significant (p=0.2444 for FLIPI2, p=0.0285 for TMTV). The combination of FLIPI and TMTV stratified the overall population into 3 risk groups : patients with no risk factors had a 5y-PFS of 76.3% (n=162, 40% reference), declining to 63.4% for patients with 1 risk factor (n=180, 44%; HR=1.45 ;[1.01-2.06], p=0.042) and 47.1% (n=64, 16%; HR=2.15 [1.39-3.31], p<0.001) for patients with both. Although the interaction between TMTV and randomization arm was not statistically significant for PFS (p=0.31), the prognostic impact of TMTV appeared different between the two arms: a high TMTV was associated with a lower PFS in the R chemo arm (Figure 1 A: p=0.0093) whereas it was not significant in the R2 arm (Figure 1B: p=0.3369). Conclusion: Baseline TMTV was a strong prognostic factor independent of FLIPI or FLIPI2 status, in advanced FL patients treated with first-line immunochemotherapy followed by antibody maintenance but not R2. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal