Whole-Genome Sequencing Identifies 2 Distinct Myeloma Precursor Conditions

Whole-genome sequencing (WGS) has the potential to accurately differentiate between stable and progressive precursor conditions to multiple myeloma in low disease burden clinical states and the use of this technology in the clinic may result in a significant shift in the management of these patients, according to data from a study published in Nature Communications.

The research showed that clinically stable cases of monoclonal gammopathy of undetermined significance (MGUS) and smoldering myeloma (SMM) are characterized by a different genomic landscape and by variations in the temporal acquisition of myeloma defining genomic events vs progressive entities.

“Despite this limited and retrospective series, the distribution of genetic events reveals striking differences and the existence of 2 biologically and clinically distinct entities of asymptomatic monoclonal gammopathies: (i) one entity characterized by a sufficient number of myeloma genomic defining events to confer malignant potential and which is associated with progressive disease, and (ii) another entity with a lower burden of genetic events characterized by high likelihood of a prolonged, indolent, and clinically stable course,” the authors of the study wrote.

MGUS and SMM are found in 2% to 3% of the general population who are older than 40 years of age. Although only a small percentage of MGUS cases will progress to multiple myeloma, approximately 60% of people with SMM will progress within 10 years of their initial diagnosis.

Indirect measures and surrogate markers of disease burden are utilized to differentiate between MGUS and SMM. Although this approach has successfully identified SMM who are high risk for disease progression, it is significantly less effective in predicting risk for those who have a low disease burden. The advent of next-generation sequencing has shed more light on the genomic complexity of multiple myeloma and its precursor conditions. It has been shown that the immune changes linked with progressive and stable myeloma precursor conditions are notably distinct, although whether these patterns impact pre-myeloma clonal selection and progression or are a consequence of pre-myeloma clonal expansion remains unclear.

Because WGS interrogates the full repertoire of disease-defining genomic events like single nucleotide variants (SNVs), mutational signatures, copy number variants, and structural variants, it has proven to be able to adequately characterize multiple myeloma and its precursor conditions. The use of this testing on precursor conditions has been limited by the availability of tumor DNA.

In the study, investigators applied multi-parameter flow cytometry sorting with a low-input WGS approach to address the challenge of low cellularity in MGUS and SMM samples. Using this strategy, they set out to characterize the genomic landscape of normal tissue from a few thousand cells.

Investigators performed WGS on a total of 32 patients with myeloma precursor conditions defined by the International Myeloma Working Group 2014 criteria; of these patients, 18 had MGUS and 14 had SMM. Only 1 of the patients who had SMM were determined to be high risk and no patients exhibited signs of disease progression at the time that the samples were collected.

At a median follow-up of 24 months from sample collection, 53% of patients (n = 17/32) who had a precursor condition went on to develop multiple myeloma and began treatment; this included 13 patients with SMM and 4 patients with MGUS. The median time to disease progression was 14.5 months (range, 2-105). Those who experienced progression were defined as having progressive myeloma precursor condition, while those who had at least 1 year of follow-up without progression were defined to have stable myeloma precursor condition.

Investigators noted that patients who had stable myeloma precursor conditions also had substantially less bone marrow plasma cell infiltration (BMPC) vs those who had progressive conditions (Wilcoxon rank-sum test, P = .0005), which was considered to potentially be reflective of the higher proportion of MGUS. Moreover, stable condition was found to have a median mutational burden of 3406 SNVs, which was significantly lower than the 5518 SNVs noted in the progressive condition (Wilcoxon rank-sum test, P = .005) and the 5482 SNVs with multiple myeloma (Wilcoxon rank-sum test, P = .005). The correlation between mutational burden and BMPC infiltration was determined to be weak (P = .02 and R2 = 0.125).

A de novo signature extraction was done across a cohort of plasma cell disorders (n = 112) to determine the key mutational signatures of multiple myeloma. APOBEC was found to be the most differentially active. Notably, only 13% (n = 2/15) of the stable condition cases had significant evidence of APOBEC activity vs 82% (n = 14/17) with progressive condition cases and 85% (n = 68/80; P = .0046) and multiple myeloma (P = .002). The 2 stable condition cases that had a APOBEC signature were characterized by a high APOBEC3A:3B ratio and both had a translocation between IGH and MAFB; this underscored that APOBEC activity needs to be evaluated in light of APOBEC3A:3B ratio.

Investigators also combined indels and SNVs to evaluate the distribution of mutations in 80 known myeloma driver genes within the precursor conditions. Two whole-exome sequencing datasets were utilized. The first dataset consisted of 33 MGUS, 2 of which experienced progression, while the second dataset included 947 patients with multiple myeloma who were enrolled to the CoMMpass trial (NCT01454297).

Here, investigators found that those with stable condition had a significantly lower number of mutations in known driver genes vs those with progressive condition (Wilcoxon rank-sum test, P = .002) and those with multiple myeloma (Wilcoxon rank-sum test, P <.0001). Additionally, when looking at patterns of positive selection across different stages, a significant signal reflective of positive selection in known driver genes was identified in those with progressive condition and multiple myeloma; this was not observed in those with stable condition. Only mutations in HIST1H1E (n = 2) were found to indicate positive selection among stable conditions.

To recognize mutational hotspots within myeloma driver genes, investigators ran sitednds and found that patients with stable condition had a lower number of mutations in known hotspots than those with progressive condition or multiple myeloma. Moreover, stable condition was also found to have a higher proportion of mutations within known AID targets vs multiple myeloma.

“These results suggest that the mutational landscape of stable myeloma precursor condition is significantly different in comparison to progressive myeloma precursor condition and multiple myeloma, in terms of both number of mutations in myeloma driver genes and the mutational processes involved,” the authors wrote.

No substantial differences in recurrent aneuploidies were discovered between those with progressive condition and those with multiple myeloma. Patients with stable condition had a significantly lower prevalence of known recurrent multiple myeloma abnormalities vs those with progressive condition or multiple myeloma.

Investigators used the comprehensive resolution of WGS to evaluate the distribution and prevalence of SVs and complex SV events, as they are key players in multiple myeloma pathogenesis. Overall, stable precursor cases were defined by a lower SV burden; this proved to be true for single SVs (Wilcoxon rank-sum test, P = .0005) and even more notably with regard to complex SVs (Wilcoxon rank-sum test, P <.0001).

“Overall, the progressive myeloma precursor condition SV landscape was similar to that observed in multiple myeloma, itself,” the study authors wrote. “This finding was confirmed by looking at the genomic distribution of SV: in progressive precursors, and to a greater extent in multiple myeloma, the distribution was significantly associated with H3K27a and chromatin accessibility loci.”

Additionally, progressive myeloma precursor condition did not demonstrate any significant differences in SV hotspot prevalence vs either stable condition or multiple myeloma.

When evaluating the time lag between initiation and sample collection, stable condition was found to be a distinct genomic entity vs multiple myeloma, while progressive condition was found to have a genomic profile that was comparable to multiple myeloma. Investigators concluded that stable condition is acquired at a later age, without myeloma-defining genomic events, and has a much lower tendency to progress to multiple myeloma vs progressive condition.

“Despite its limited sample size, this study provides proof of principle that WGS has the potential to accurately differentiate stable and progressive precursor conditions in low disease burden clinical states,” the authors concluded. “…Going forward, improved and biology-oriented strategies to accurately identify patients with progressive myeloma precursor condition before clonal expansion (i) will allow earlier initiation of therapy before onset of end-organ damage to avoid severe clinical complications, (ii) will prevent patients with precursor conditions from being overtreated.”

Reference

Oben B, Froyen G, Maclachlan KH, et al. Whole-genome sequencing reveals progressive versus stable myeloma precursor conditions as two distinct entities. Nat Commun. 2021;12(1):1861. doi:10.1038/s41467-021-22140-0