Reevaluating the IMWG Multiple Myeloma Complete Response Criterion in the Era of Mass Spectrometry: A Critical Analysis

Introduction:

In multiple myeloma (MM), "Complete Response" (CR) is defined by the IMWG as negative immunofixation (IFE) on the serum and urine, disappearance of any soft tissue plasmacytomas, and less than 5% plasma cells (PC) in the bone marrow (BM). However, 24-hour urine collections and BM examinations can be challenging and uncomfortable for patients (pts), leading to noncompliance in routine clinical practice. The Spanish Myeloma Group already showed the limited value of urine IFE for defining CR. In this study, we aimed to evaluate the value of BM examinations in pts with suspected CR based on negative IFE results in serum (sIFE) and to explore the utility of mass spectrometry (MS) as a marrow sparing, single serological marker to determine a potentially mass-spec based CR category.

Methods:

We analyzed a total of 1050 paired serum and BM samples obtained from 380 newly diagnosed transplant eligible MM pts enrolled in the GEM12MENOS65, GEM14MAIN, and GEM-CESAR clinical trials. Samples were obtained at four predefined time points: post-induction, after ASCT, post-consolidation and after 2 years of maintenance. Serum samples were analyzed using Quantitative Immunoprecipitation Mass Spectrometry with anti IgG/A/M, total k and total l beads in the EXENT^® Analyzer (The Binding Site, part of Thermo Fisher Scientific) and SPEP/IFE, carried out as per each center protocol. First-pull bone marrow aspirations were used for morphological assessment with May-Grümwald-Giemsa staining. PC counts were obtained from a 200-cell differential count, using conventional bright-field microscopy. The treatment schemas and dosing of the three trials have been previously described.

Results:

First, we pooled together the results of all samples (including all time points and independently of the conventional response achieved by the pts) and analyzed the individual clinical value of the two main factors defining CR (i.e., PC counting [<5% vs ≥5%] and sIFE [positive vs negative]). Results showed that PC counting and sIFE status were not able differentiate two groups of pts with different median progression-free survival (mPFS). In contrast, MS status effectively separated two groups with different mPFS (MS negative not reached [NR] vs MS positive 5.55 years; p<0.0001; HR. 0.5674 [95% CI: 0.46 – 0.69]).

Focusing only in the 426 samples (from 189 pts) with “suspected CR” based solely on sIFE negative results, we observed a trend towards inferior PFS in the 18 samples (from 15 pts) with ≥5% PC (mPFS: <5%PCs NR vs ≥5%PCs 5.42 years; p=0.1884). However, a highly significant reduction in mPFS was observed among these cases with a sIFE negative but positive for MS (mPFS: MS^- negative NR vs MS positive 4.98 years; p=0.0002; HR. 0.52 [95%CI 0.37 – 0.74]). Further, analysis of the combined results of PC counting and MS (among sIFE negative cases) showed that only 8 out of the 408 cases had ≥5%PC and were MS negative, being all of them also MRD negative by NGF. Therefore, the negative predictive value of MS using the results of PC counting among sIFE negative cases was of 97% (95%-99%, p=0.0184). Among sIFE positive cases, MS discriminated 2 subgroups with different mPFS (NR in negative cases vs 5.96 years in positive cases) although not reaching statistical significance.

We then combined the results of PC counting and sIFE, therefore dividing the global cohort in samples in ≥CR (<5%PC and sIFE negative) vs <CR cases (≥5%PC and/or sIFE positive). No significant differences in PFS between the ≥CR and < CR groups defined as above were observed. In contrast, when we examined the impact of MS in pts in ≥CR, we found out that the MS status effectively segregated two groups with different mPFS. Importantly, MS was able to identify a subgroup of cases with a significantly inferior outcome (mPFS: MS negative NR vs MS positive 4.98 years; p=0.0009; HR 0.55 [95% CI: 0.38 – 0.78]).

Conclusion: Our study suggests that MS surpasses the clinical value of the CR category and the limitations of PC counting and sIFE, and thus supports the consideration of a new, MS-based, category of treatment response in pts with MM. Also, our results show that performing a BM aspiration in cases with “suspected CR” would only be justified if the analysis of MRD is planned, given the limited added value of PC counting.