Recurrent Crebbp Mutations in Follicular Lymphoma Appear Localized to the Committed B-Cell Lineage

Background: Follicular lymphoma (FL) is genetically characterized by translocations involving the BCL2 locus on chromosome 18q21. However, up to 70% of healthy individuals also carry detectable t(14;18)-positive cells, suggesting BCL2 translocation is critical but not sufficient for FL development.

Chromatin modifying genes (CMGs) including KMT2D, CREBBP, EZH2, and EP300 are almost ubiquitously mutated in FL. We previously reported the direct characterization by ultra-deep sequencing of pre-diagnostic blood and tissue specimens from 19 subjects who ultimately developed FL. CREBBP lysine acetyltransferase (KAT) domain mutations were the most commonly observed precursor lesions, detected in blood a median of 7.5 years before diagnosis in patients developing FL (8/19, 42%) but not in healthy adults with or without detected BCL2 translocations (0/13, p=0.01 and 0/20, p<0.001, respectively) (Schroers-Martin et al, ASH Annual Meeting 2017).

While the BCL2 translocation is thought to occur in pre-B-cell precursors in the bone marrow, the high prevalence of CMG mutations raises the possibility that, analogous to myeloid age-related clonal hematopoiesis, such early lesions could occur in hematopoietic stem cells (HSCs) rather than in the committed B-cell lineage.

Methods: To address this question of mutational hierarchy, we studied blood, bone marrow aspirate, and lymph node samples from 6 patients with CREBBP or KMT2D mutations in their FL tumor. We analyzed flow-sorted purified hematopoietic cell populations by deep sequencing, including isotype-specific mature B-cell, mature T-cell, and precursor populations encompassing HSCs and common lymphoid progenitors (CLPs). Patients additionally underwent routine clinical sequencing of tumor biopsy and marrow specimens.

Results: Bulk sequencing of a diagnostic bone marrow aspirate from patient FL002 revealed a CREBBP mutation concordant with FL tumor biopsy. To ascertain the population bearing this mutation, we sorted to high purity viable marrow aspirate cells (Fig. A). The CREBBP mutation was confirmed in the mature B-cell compartment at an AF of 40.3% but was not detected in other cell populations.

To validate this finding, a similar sorting strategy was employed on viable bone marrow aspirate or peripheral blood samples from another 4 FL patients bearing CREBBP mutations (Fig B). In each case, CREBBP was absent from the CD34+ precursor population. In patient CIML004 a characteristic KMT2D stop mutation was likewise absent in precursors.

An atypical case sheds additional light on the localization of early FL mutations. Patient LYM267 was diagnosed with Grade 1-2 FL bearing CREBBP and NRAS mutations. Eight years into a prolonged remission after chemoimmunotherapy, he developed cutaneous and gingival myeloid sarcoma without radiographic or histopathological evidence of FL recurrence. While the CREBBP mutation was not detected in myeloid sarcoma or bone marrow biopsies, concordant NRAS mutations and clonal VDJ rearrangements were seen in all 3 compartments (Fig. C). This unusual clonal lineage favors the occurrence of the CREBBP mutation later than the branch point between morphologically distinct lymphoid and myeloid tumors, likely in the committed B-cell lineage after pre-BCR rearrangements (Fig D).

Conclusions: HSCs are believed to be the cell of origin in several lymphoid leukemias, and mouse models have demonstrated lymphoma development with induced CREBBP lesions in HSCs (Horton et al Nat Cell Bio 2017). However, in sorted hematopoietic cell populations from marrow and peripheral blood, we observed CREBBP mutations in B-cell lineages but never in CD34+/CD20- precursor populations or paired lymphoid/myeloid disease. Our data therefore are not in support of HSCs as a precursor reservoir in FL. Given that cells harboring the t(14;18) translocation in healthy individuals appear derived from the germinal center, recurrent mutations in CREBBP are likely to occur after the pre-B-cell stage.

Figure

View largeDownload slide

Figure

View largeDownload slide

Close modal