RASping myeloma genomics

In this issue of Blood, Schavgoulidze et al present their analysis of somatic mutations in 10 189 patients with plasma cell neoplasms using targeted sequencing panels in routine clinical practice.¹ Their large study shows that KRAS/NRAS/BRAF^V600E mutations are the most frequent secondary events in multiple myeloma (MM) and that, unfortunately, these mutated genes may have a limited role as therapeutic targets and biomarkers in monitoring treatment efficacy.

Despite the enormous effort and amount of original data, the first reports of the MM genomic landscape were frustrating.² In contrast to other hematological malignancies, such as Waldenström macroglobulinemia or hairy cell leukemia, there was no unifying somatic mutation present in >90% of MM patients. The identification of MYD88^L265P in Waldenström macroglobulinemia and BRAF^V600E in hairy cell leukemia has led to significant advances in diagnostic criteria, drug development, and monitoring of treatment efficacy. In other hematological malignancies, such as acute myeloid leukemia, there are no unifying genetic events, but there are subgroups defined by specific gene mutations (eg, NPM1) that are associated with individualized treatment strategies and methods to assess response. Thus, it was important to clarify the frequency, location, and clonality of somatic mutations in the MAPK pathway to determine whether these could define a unique MM subgroup.

In their study, Schavgoulidze et al make 3 important observations. First, KRAS/NRAS/BRAF^V600E mutations are the most frequent genetic event in MM, present in 60.7% of patients. This rate is higher than that of immunoglobulin translocations, copy-number alterations, and even the more general observation of hyperdiploidy (see figure). In fact, a recent study by Maura et al showed that isolated KRAS/NRAS/BRAF^V600E mutations define a subset of patients without the alterations just listed.³ Although mutations that constitutively activate the MAPK pathway are not specific to MM and are observed in other tumors, the presence of mutations in nearly two-thirds of patients suggests a major role in the genesis of myeloma.

Second, these mutations appear to be associated with the progression of precursor stages into malignant MM, but not with resistance and relapse after treatment, as recently suggested by others.⁴ A comparison between disease stages shows that KRAS/NRAS/BRAF^V600E mutations were significantly less frequent in precursor states, but the frequency was stable between newly diagnosed and relapsed patients. Furthermore, there was a trend toward early progression of smoldering MM patients carrying KRAS mutations, which is in line with previous reports by Boyle et al⁵ and Bolli et al.⁶ Unfortunately, Schavgoulidze et al did not investigate in their cohort if KRAS/NRAS/BRAF^V600E mutations have null prognostic significance in newly diagnosed MM patients, as shown by others.^3,7,8 

Third, the codons that are mutated and the subclonal nature of these mutations may limit their role as therapeutic targets and biomarkers of disease monitoring. It should be noted that Giesen et al recently reported that combined BRAF/MEK inhibition showed high response rates in relapsed refractory MM having BRAF^V600E mutations.⁹ Interestingly, all patients sequenced in the study carried clonal BRAF^V600E mutations, and resistance to BRAF/MEK inhibition appeared to be mediated by RAS mutations and structural variants involving the BRAF locus.⁹ By contrast, in the study by Schjesvold et al, who investigated an MEK inhibitor alone or in combination with other drugs, no responses were observed in the monotherapy arm regardless of KRAS/NRAS/BRAF^V600E mutation status.¹⁰ Collectively, these results suggest that, with the possible exception of a few patients carrying clonal BRAF^V600E mutations,⁹ the role of BRAF/MEK inhibition may be limited in MM. Unfortunately, the same conclusion probably applies to patients with light-chain amyloidosis and primary plasma cell leukemia, as Schavgoulidze et al provide unique genomic data on the 2 rare diseases, both in need of new treatment options.

Arguably, the main contribution by Schavgoulidze and colleagues is the demonstration that targeted sequencing can be performed during routine diagnostic evaluation. Their analysis of more than 10 000 patients is a testimony to this, along with the fact that they included patients with monoclonal gammopathy of undetermined significance (n = 1168) and light-chain amyloidosis (n = 489), who often present with a low tumor burden in bone marrow. Sequencing was performed with a median depth of 200×, which enabled sensitive detection of mutations with a variant allele frequency of <0.1. Interestingly, these subclonal mutations were also observed using single-cell DNA sequencing, either in the same or different clones. Thus, the same French investigators who led the clinical application of cytogenetics in MM over the past 20 years are now pioneering the implementation of next-generation sequencing in the routine clinical care setting. The authors should be commended for their herculean effort, and both the authors and the editors should be praised for the publication of results that did not identify the hoped-for targetable mutations. These results are equally important to define future treatment and monitoring strategies.

Conflict-of-interest disclosure: B.P. reports honoraria for lectures from and membership on advisory boards with Adaptive, Amgen, Bristol Myers Squibb/Celgene, Gilead, GSK, Janssen, Oncopeptides, Roche, Sanofi, and Takeda; unrestricted grants from Bristol Myers Squibb/Celgene, EngMab, GSK, Roche, Sanofi, and Takeda; and consultancy for Bristol Myers Squibb/Celgene, Janssen, and Sanofi. M.-J.C. reports honoraria received from lectures and participation on advisory boards with Amgen, Bristol Myers Squibb/Celgene, Pfizer, Jazz Pharmaceuticals, Novartis, GSK, Janssen, Astellas Pharma, Sanofi, and Takeda.