Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Timing the initiation of multiple myeloma

Maura, Francesco, Rustad, Even H., Bolli, Niccolo, Yellapantula, Venkata, Leongamornlert, Daniel A., Nadeu, Ferran, Angelopoulos, Nicos ORCID: https://orcid.org/0000-0002-7507-9177, Dawson, Kevin J., Mitchell, Thomas J., Osborne, Rob, Ziccheddu, Bachisio, Carniti, Cristiana, Montefusco, Vittorio, Corradini, Paolo, Anderson, Kenneth C., Moreau, Philippe, Papaemmanuil, Elli, Alexandrov, Ludmil, Puente, Xose S., Campo, Elias, Siebert, Reiner, Avet-Loiseau, Herve, Munshi, Nikhil C., Campbell, Peter J. and Landgren, Ola 2019. Timing the initiation of multiple myeloma. Presented at: 61st American Society of Hematology Annual Meeting, Orlando, FL, USA, 7-10 December 2019. Blood. , vol.134 (Supple) American Society of Hematology, 10.1182/blood-2019-124357

Full text not available from this repository.

Abstract

INTRODUCTION: Cancer pathogenesis is usually characterized by a long evolutionary process where genomic driver events accumulate over time, conferring advantage to distinct subclones, allowing their expansion and progression. METHODS: To investigate the multiple myeloma (MM) evolutionary history, we characterized the mutational processes' landscape and activity over time utilizing a large cohort of 89 whole genomes and 973 exomes. To improve the accuracy of mutational signatures analysis, we analyzed both the 3' and 5' nucleotide context of each mutation and we developed the novel fitting algorithm mmSig, which fits the entire mutational catalogue of each patient with the mutational signatures involved in MM pathogenesis. The contribution of each mutational signature was then corrected based on the cosine similarity between the original 96-mutational profile and the reconstructed profile generated without that signature. To reconstruct the genetic evolutionary history of each patient's cancer, we integrated two approaches. First dividing all mutations into clonal (early) or subclonal (late), then subdivided the clonal mutations into duplicated mutations (present on two alleles and therefore acquired before the duplication) or non-duplicated mutations (detected on a single allele), reflecting either pre-gain and post-gain mutations on the minor allele, or post-gain mutations acquired on one of the duplicated alleles. RESULTS Eight mutational signatures were identified, seven of which showed significant similarity with the most recent mutational signature catalogue (i.e SBS1, SBS2, SBS5, SBS8, SBS9, SBS13 and SBS18). The new mutational signature (named SBS-MM1) was observed only among relapsed patients exposed to alkylating agents (i.e melphalan). The etiology of this specific signature was further confirmed by analyzing recent whole genomes public data from human-induced pluripotent stem cells exposed to melphalan (Kucab et al, Cell 2019). Reconstructing the chronological activity of each mutational signature, we identified four different routes to acquire the full mutational spectrum in MM based on the differential temporal activity of AID (SBS9) and APOBEC (SBS2 and SBS13). Our data indicate that AID activity is not limited to the first contact with the GC, but persists in the majority of patients, behaving similarly to a B-memory cells, capable of re-entering the germinal center upon antigen stimulation to undergo clonal expansion several times before MM diagnosis. Next, we confirmed the clock-like nature (i.e constant mutation rate) of SBS5 in MM and other post-germinal center disorders such as chronic lymphocytic leukemia and B-cell lymphomas. Based on the SBS5 mutation rates and the corrected ratio between duplicated and non-duplicated mutations within large chromosomal gains, we could time the acquisition of the first copy number gain during the life history of each MM patient. Intriguingly, the first MM chromosomal duplication was acquired on average 38 years (ranges 11-64) before sample collection. In 23/27 (85%) cases the first multi gain event occurred before 30 years of age, and in 13/27 (48%) before 20 years reflecting a long and slow process potentially influenced and accelerated by extrinsic and intrinsic factors. DISCUSSION Our analysis provides a glimpse into the early stages of myelomagenesis, where acquisition of the first key drivers precedes cancer diagnosis by decades. Defining the time window when transformation occurs opens up for new avenues of research: to identify causal mechanisms of disease initiation and evolution, to better define the optimal time to start therapy, and ultimately develop early prevention strategies.

Item Type: Conference or Workshop Item (Poster)
Date Type: Published Online
Status: Published
Schools: Medicine
Publisher: American Society of Hematology
ISSN: 0006-4971
Date of Acceptance: 2 October 2019
Last Modified: 04 Jan 2023 02:19
URI: https://orca.cardiff.ac.uk/id/eprint/133797

Actions (repository staff only)

Edit Item Edit Item