P1284: LIQUID BIOPSY BY ULTRA-DEEP SEQUENCING PLUS PET-TC MONITORING IN REAL-LIFE FOLLICULAR LYMPHOMA PATIENTS

Abstract

Background: Follicular lymphoma (FL) is considered an indolent disorder with a relatively favorable course. With modern day treatments, long remissions are often achieved both in front-line and relapsed setting. However, a subset of patients has a more aggressive course and a poorer outcome. Although understanding prognosis is important, ultimately the goal in FL should be to guide therapy. Studies of risk-adapted therapy based on MRD remains investigational. Aims: To prospectively evaluate the predictive value of a liquid Biopsy MRD (LiqBio-MRD) SNV-NGS method in combination with the Deauville criteria using the 5-point scale (D5PS) in the evaluation of response in a cohort of 72 FL patients that require treatment from 2018-2021. Methods: 57 patients received first-line therapy (39 R-CHOP, 10 R-B, 5 Rituximab in monotherapy and 3 radiotherapy) and 15 patients salvage treatment. Genomic DNA from formalin-fixed paraffin embedded (FFPE) lymph node biopsies and/or cell-free DNA (cfDNA) was obtained before start treatment. These samples were sequenced with short length Ampliseq Custom Panel targeting 56 important genes in B-cell dyscrasias. Somatic mutations were selected as disease biomarkers, and cfDNA samples were analysed at mid-induction (n = 26) and at the end of treatment (n= 41). PET/CT examinations were usually performed after 4 cycles (n = 44) and at the end of treatment (n = 63). PET positive was considered when D5PS was 4 or 5. Statistical analysis was performed with SPSS® using Cox regression, Kaplan-Meier survival curves and log-rank test. Results: Median age at diagnosis was 66 years (35 – 90) and 58% were female. At diagnosis only the lymphocyte to monocyte ratio < 2.5 had prognostic value (p=0.04). Median PFS was 13 months for PETEOT (+) cases and 10 months for MRDEOT (+) ones. 71% of patients with PETEOT (+) presented POD24, in contrast to only 9% of patients with PETEOT (-), p < 0.001. Regarding MRD, 73% of patients with MRDEOT (+) presented POD24 vs 12% of patients with MRDEOT (-), (p < 0,001; HR 8.30, 95%CI 1.70–40.38). The sensitivity (SE) and specificity (SP) of both tests to predict POD24 were: PETEOT (SE77%, SP68%) and MRD (SE77%, SP86%). 40 patients had MRDEOT and PETEOT available and results were concordant in 80% of cases. In a stratified analysis combining the information on MRD and PET, the 2-yr PFS were; 96% for MRD/PET -/-, 71% for MRD/PET -/+, and 58% MRD/PET +/-. Strikingly, patients that were positive by both tests presented a 2-yr PFS of only 10%. Using only concordant results, SE91% and SP96% was achieved. Similar results in PFS and in predicting POD24 were obtained when including only patients with first line therapy. Figure 1. When interim results were analysed, also PETINT(+) and MRDINT(+) were associated with higher risk of relapse (p<0,001 and p=0.009, respectively). The negative predictive value of PETINT results was 96%. Concordance between both tests was clearly inferior (47%), although no patient with PETINT(-) and MRDINT(-) results relapsed. Image:Summary/Conclusion: Our results demonstrate for the first time that NGS based MRD quantification is feasible in Liquid Biopsies from FL. The achievement of negative MRD after treatment, but also in an interim analysis, enhances prognostic information on patient outcome. PET/CT and cfDNA are complementary increasing SE and SP to predict progression and POD 24 with a high accuracy. Although, more patients are necessary to draw meaningful conclusion, the combination of these techniques could aid to the develop clinical trials with a response-adapted precision therapy.