-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

545 Defining Genomic Probability of Progression to Identify Low-Risk Smoldering Multiple Myeloma

Program: Oral and Poster Abstracts
Type: Oral
Session: 652. Multiple Myeloma and Plasma cell Dyscrasias: Clinical and Epidemiological: Reoptimizing Standards and Redefining Approaches
Hematology Disease Topics & Pathways:
Translational Research, Plasma Cell Disorders, Clinically Relevant, Diseases, Lymphoid Malignancies
Sunday, December 12, 2021: 5:30 PM

Anil Aktas-Samur, PhD1*, Mariateresa Fulciniti, PhD2, Sanika Derebail, MS3*, Raphael Szalat, MD, PhD4, Giovanni Parmigiani, PhD5*, Jill Corre6*, Herve Avet-Loiseau7*, Mehmet K. Samur, PhD8 and Nikhil C. Munshi, MD, PhD9

1Department of Medical Oncology, Dana Farber Cancer Institute, Boston, MA
2Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston
3Medical Oncology, Dana-Farber Cancer Institute, Boston, MA
4Department of Hematology and Medical Oncology, Boston University Medical Center, Boston, MA
5Dana-Farber Cancer Institute, Boston, MA
6Unité de Génomique du Myélome, IUCT- Oncopole, Toulouse, France
7Unite de Genomique du Myelome, IUC-Oncopole, Toulouse, France
8Department of Data Science, Dana-Farber Cancer Institute, Harvard TH Chan School of Public Health, Boston, MA
9Dana-Farber Cancer Institute, Harvard Medical School, Boston

On an average, 1% of monoclonal gammopathy of undermined significance (MGUS) and 10% of smoldering Multiple Myeloma (SMM) progress to symptomatic MM every year within the first five years of diagnosis. The probability of progression significantly decreases for SMM patients after first 5 years. However, a distinct subset of SMM patients progress within 2 years and are re-classified as high-risk patients based on risk markers such as 20/2/20 or certain genomic features. Although recent studies have evaluated the high-risk genomic features for SMM but genomic background of SMM patients who do not progress to MM after long-term follow-up (>= 5 years) has not been described.

Here, we evaluated transcriptomic and genomic changes enriched in non-progressor (NP) (no progression after 5 years of follow-up) precursor conditions (N=31) with those progressed within short period of time (N=71) and compared them with changes observed in newly diagnosed MM (N=192). Additionally, using transcriptome, epigenome and whole genome profiling we also studied additional unique samples from 18 patients at their precursor stage as well as when progressed to MM.

Overall, we have observed significantly lower mutational load for NP SMM from progressor SMM (median SNV 4900 vs. 7881 p < 3e-04) with high sensitivity (0.83) and specificity (0.65) to separate NP from progressors. We have further developed a deep learning model by using more than 4500 genome wide features using ten-fold cross validation. This model indicated that not only the load but also the patterns of mutations (type, location, frequency) are different between two conditions. We also found that NP samples have significantly lower heterogeneity (p < 0.05). However, progressed samples showed similar mutational load and heterogeneity at precursor stage and MM. Among CNA differences, absence of gain or deletion of chr8 (not involving MYC region) were strong predictor of NP (OR=7.2 95% CI 2.2-24). Focal genomic loss was also significantly lower for NP (p=0.004) which was also reflected by low genome scar score (GSS) (p=0.07). Structural variant and copy number signature analysis also showed that NPs were showing significantly low exposure to non-clustered variable size genomic deletions. We observed similar frequency of primary translocations [t(11;14), t(4;14), and t(14;16)] in both progressor and NP samples as well as newly diagnosed MM. MYC translocation with any partner was not observed in NP samples, whereas 37% of progressor samples had a MYC translocations (OR=12.8). Adding all these differences including chr8 CNAs, MYC translocations, mutation burden, GSS, focal deletions, all driver mutations as well as primary translocations into recursive partitioning model to predict non-progressor SMM, we have identified a simple genomic model only involving chr8 CN changes and overall mutational burden to achieve a high sensitivity (0.82) and specificity (74%).

Our transcriptomic analysis measured the distance between progressor and NP SMM as well as MM and found that NP SMM has greater difference with MM which is closer to progressor SMM. We quantified transcriptomic heterogeneity by using molecular degree of perturbation. This analysis showed that consistent with DNA changes, DNA repair pathway and MYC target genes are expressed similarly in NP SMM as in normal plasma cells compared to progressor SMM. Epigenomic analysis yielded 75 SEs regions differentially utilized between precursor and symptomatic MM stage using paired samples. The targeted genes included BMP6, PRDM1, STAT1, SERTAD2 and RAB21 and possibly regulating genes related to oncogenic KRAS activities.

In conclusion, we define genomic characterization of non-progressor SMM and our results now provide the basis to develop molecular definition of SMM as well as risk driving features.

Disclosures: Munshi: Janssen: Consultancy; Pfizer: Consultancy; Legend: Consultancy; Novartis: Consultancy; Adaptive Biotechnology: Consultancy; Oncopep: Consultancy, Current equity holder in publicly-traded company, Other: scientific founder, Patents & Royalties; Takeda: Consultancy; Abbvie: Consultancy; Karyopharm: Consultancy; Amgen: Consultancy; Celgene: Consultancy; Bristol-Myers Squibb: Consultancy.

*signifies non-member of ASH