Feasibility of Circulating Tumor DNA Analysis in Patients with Follicular Lymphoma

Abstract

Purpose: The feasibility of sequencing circulating tumor DNA (ctDNA) in plasma as a biomarker to predict early relapse or poor prognosis in patients with follicular lymphoma (FL) receiving systemic immunochemotherapy is not clear.

Materials and methods: We sequenced DNA from cell-free plasma that was serially obtained from newly diagnosed FL patients undergoing systemic immunochemotherapy. The mutation profiles of ctDNA at the time of diagnosis and at response evaluation and relapse and/or progression were compared with clinical course and treatment outcomes.

Results: Forty samples from patients receiving rituximab-containing immunochemotherapy were analyzed. Baseline sequencing detected mutations in all cases, with the major detected mutations being KMT2C (50%), CREBBP (45%), and KMT2D (45%). The concentration of ctDNA and tumor mutation burden showed a significant association with survival outcome. In particular, the presence of mutations in CREBBP and TP53 showed poor prognosis compared with patients without them. Longitudinal analysis of ctDNA using serially collected plasma samples showed an association between persistence or reappearance of ctDNA mutations and disease relapse or progression.

Conclusion: Analysis of ctDNA mutations in plasma at diagnosis might help predict outcome of disease, while analysis during follow-up may help to monitor disease status of patients with advanced FL. However, the feasibility of ctDNA measurement must be improved in order for it to become an appropriate and clinically relevant test in FL patients.

Keywords: Circulating tumor DNA; Follicular lymphoma; Mutations; Prognosis.

Fig. 1.

Overview of study design: design for sample collection. Target genes for mutation profiling. BR, bendamustine and rituximab; ctDNA, circulating tumor DNA; FL, follicular lymphoma; MRD, minimal residual disease; RCHOP, rituximab, cyclophosphamide, vincristine, doxorubicin, and prednisolone; RCVP, cyclophosphamide, vincristine, and prednisolone.

Fig. 2.

Comparison of mutations according to sample types and treatments (A). Concordance of somatic mutations between tissue and plasma cell-free DNA (cfDNA) samples (n=12). (B) The landscape of somatic mutation detected in plasma cfDNA samples from 40 patients, with mutational profiles grouped according to their treatments: BR (bendamustine plus rituximab; n=19), R-CHOP (rituximab, cyclophosphamide, vincristine, doxorubicin, and prednisolone; n=19), and R-CVP (cyclophosphamide, vincristine, and prednisolone; n=2).

Fig. 3.

Relationship between circulating tumor DNA (ctDNA) concentration, tumor mutation burden (TMB) (A), and treatment response (B). Comparison of response rates according to TMB high and low group across entire targeted 61 genes (C) and the region covered by only 7-FLIPI (Follicular Lymphoma International Prognostic Index) genes (D). hGE, human genome equivalents; PD, progression of disease.

Fig. 4.

Progression-free survival (left panel) and overall survival (right panel) according to tumor mutation burden (A), total circulating tumor DNA concentration (B), mean ctDNA concentration (C), absence of mutation on 7-FLIPI (Follicular Lymphoma International Prognostic Index) genes (EZH2, ARID1A, MEF2B, EP300, FOXO1, CREBBP, and CARD11) (D) and TP53/CREBBP gene mutation (E-G). DN, double negative; DP, double negative; hGE, human genome equivalents; SP, single positive.

Fig. 5.

Monitoring circulating tumor DNA (ctDNA) concentrations of patients in therapeutic response (A) and disease relapse (B) groups. The chemotherapeutic history is displayed above each monitoring plot. The patients have their positron emission tomography computed tomography (PET-CT) results at nearby time points and the cell-free DNA (cfDNA) sampling and response were classified using Deauville criteria (which is an internationally recommended scale in PET-CT assessment). BR, bendamustine and rituximab; R, rituximab; RB, rituximab, bendamustine; R-CHOP, rituximab, cyclophosphamide, vincristine, doxorubicin, and prednisolone; R-CVP, cyclophosphamide, vincristine, and prednisolone; hGE, human genome equivalent.