Genetic subdivisions of follicular lymphoma defined by distinct coding and noncoding mutation patterns

Abstract

Follicular lymphoma (FL) accounts for ∼20% of all new lymphoma cases. Increases in cytological grade are a feature of the clinical progression of this malignancy, and eventual histologic transformation (HT) to the aggressive diffuse large B-cell lymphoma (DLBCL) occurs in up to 15% of patients. Clinical or genetic features to predict the risk and timing of HT have not been described comprehensively. In this study, we analyzed whole-genome sequencing data from 423 patients to compare the protein coding and noncoding mutation landscapes of untransformed FL, transformed FL, and de novo DLBCL. This revealed 2 genetically distinct subgroups of FL, which we have named DLBCL-like (dFL) and constrained FL (cFL). Each subgroup has distinguishing mutational patterns, aberrant somatic hypermutation rates, and biological and clinical characteristics. We implemented a machine learning-derived classification approach to stratify patients with FL into cFL and dFL subgroups based on their genomic features. Using separate validation cohorts, we demonstrate that cFL status, whether assigned with this full classifier or a single-gene approximation, is associated with a reduced rate of HT. This implies distinct biological features of cFL that constrain its evolution, and we highlight the potential for this classification to predict HT from genetic features present at diagnosis.

Graphical abstract

Figure 1.

Oncogene translocations in FL. (A) Translocations involving oncogenes in FL and DLBCL genomes. The frequency is shown relative to the total number of tumors within each group. (B-C) Proportion of tumors with BCL2 (B) and MYC (C) translocations in the paired pre- and post-HT genomes. (D) Circos plots showing the MYC translocation partners in the paired pre- and post-HT genomes. The color of the breakpoint represents an individual patient.

Figure 2.

Coding mutations across FL and DLBCL. (A) Oncoplot depicting coding mutations at the SMGs in FL. (B) Mutation diagrams showing genetic variations of CREBBP in FL and DLBCL tumors. Mutations are colored based on their type. Each mutation is annotated with amino acid substitution. Patients with FL are shown as the top track of variants for each gene, and patients with DLBCL are shown below.

Figure 3.

Resolution of distinct subgroups with FL. (A) Definition of cFL and dFLs. The y-axis depicts the probability of a tumor to be classified as cFL based on the RF model. (B-C) show distribution of tumors between cFL and dFL for every lymphoma type. (D) Overview of coding mutations in newly discovered FL subgroups. (E) Forest plot of mutations with differential frequency between cFL and dFL. Displayed are only genes or aSHM sites significantly differentially enriched between cFL and dFL (q < 0.1; Fisher test with multiple test correction using Benjamini-Hochberg method). (F) Mutation counts at some aSHM target sites across DLBCL and genetic subgroups of FL. The star indicates significant differences compared with dFL. (G) Schematic representation of mutations across CREBBP in cFL (top) and dFL (bottom).

Figure 3.

Resolution of distinct subgroups with FL. (A) Definition of cFL and dFLs. The y-axis depicts the probability of a tumor to be classified as cFL based on the RF model. (B-C) show distribution of tumors between cFL and dFL for every lymphoma type. (D) Overview of coding mutations in newly discovered FL subgroups. (E) Forest plot of mutations with differential frequency between cFL and dFL. Displayed are only genes or aSHM sites significantly differentially enriched between cFL and dFL (q < 0.1; Fisher test with multiple test correction using Benjamini-Hochberg method). (F) Mutation counts at some aSHM target sites across DLBCL and genetic subgroups of FL. The star indicates significant differences compared with dFL. (G) Schematic representation of mutations across CREBBP in cFL (top) and dFL (bottom).

Figure 4.

Novel genetic subgroups of FL are characterized by distinct biological features. (A) Expression of CREBBP, FOXP1, and MYC across tumors in de novo DLBCL and within FL cases in each subgroup. (B) Mutational signatures SBS1, SBS5, and SBS9, but not SBS8, are differentially enriched between cFL and dFL tumors.

Figure 5.

Validation of discovered FL subgroups in a separate cohort. (A-B) Distribution of samples between cFL and dFL for every sample. (C) Forest plot of mutations with differential frequency between cFL and dFL in the validation cohort. (D) Mutation burden at aSHM target sites across tumors in the validation cohort. The star indicates significant differences compared with dFL. (E) Immunohostochemistry results for FOXP1 and KI67 staining in the validation cohort. The star indicates significant differences compared with dFL. (F) The Kaplan-Meier curve showing the TTT (years) between cFL and dFL tumors in the validation cohort.