Passenger mutations due to activation-induced deaminase as biomarkers in lymphoma

University Of Rochester, Rochester NY

Investigators

Linked publications & trials

Abstract

? DESCRIPTION (provided by applicant): Follicular Lymphoma (FL), the most prevalent lymphoma in the US, is generally considered indolent and patients often do not require therapy for years after diagnosis. However, ~30% of patients will have progressive, life-threatening disease. This high variability requires that trials be long with many enrolled subjects and may still be insufficient to detect meaningful responses. For example, SWOG 0016, the largest phase 3 FL trial ever in the US with 554 subjects and lasting more than 12 years from inception to publication, compared a novel targeted therapy to standard of care. Despite its size and duration, the data were not sufficiently compelling to suggest a modification of therapy. This disappointing result indicates the critical need for new approaches to trial design. A robust prognostic marker is needed to enable these new designs. Progression and aggressive transformation of FL are associated with the acquisition of additional non-coding (passenger) mutations attributable to an endogenous enzyme, Activation-induced (Cytidine) Deaminase (AID). We have developed an ultra-deep, next generation sequencing assay (>20,000x coverage) of two mutational hotspots, 5-UTR of BCL2 and IGHV, to quantify AID-mediated mutagenesis. We have shown that genome-wide damage due to AID is estimated by the number AID-mediated passenger mutations in the 5'-UTR of BCL2, while the number of mutations in IGHV reflects a far more active but independently regulated AID-mediated mutagenic process. Clonal mutations (present in all tumor cells) occur earlier in oncogenesis than subclonal mutations which reflect either clonal evolution or mutagenesis occurring in FL cells at another body site that trafficked into the sampled site. Our central hypothesis is that th ongoing AID activity causes clinically relevant DNA-damage, creating the additional mutations within the initially indolent FL population which cause sub-clonal diversification, thereby leading to progression and aggressive transformation. This model predicts that the greater the cumulative activity of AID in FL, the more likely that an aggressive form of the disease will emerge. Proof-of-principle for clinical trial application has also been established; using FL specimens from 12 patients, we have shown that the extents of AID-induced genome-wide and IGH-directed damage are highly variable between FL specimens. The variability is sufficient that a study of these associations with progression can be appropriately powered with samples collected from a large clinical trial. To validate that the extent of AID-mediated mutagenesis is prognostic, we will apply this NGS assay to a completed, multi-site phase III trial, SWOG 0016, the largest trial- related collection of FL specimens ever assembled in the US. Because FL is so prevalent and the course is so variable, this biomarker, in addition to enhancing identification of novel FL therapies through stream-lined clinical trials, would enable highly relevant prognostication for more than half of the patients living with a diagnosis of any non-Hodgkin lymphoma in the US.