Longitudinal biomarker dynamics for accurate risk prediction in 464 patients with asymptomatic IgM gammopathy: A spanish multicenter study

Abstract

Abstract INTRODUCTION Asymptomatic IgM gammopathies (IgM monoclonal gammopathy of undetermined significance [MGUS] and asymptomatic Waldenström macroglobulinemia [AWM]) are characterized by the presence of monoclonal IgM protein without symptoms. While risk stratification at diagnosis is crucial to predict which patients (pts) should be carefully followed, a dynamic assessment of biomarkers offers a more precise risk prediction in IgG or IgA precursor conditions (Cowan, Lancet Hematology 2023). Understanding the temporal dynamics of laboratory variables in IgM isotype is also essential for identifying early markers of progression to symptomatic disease during follow-up. METHODS In this retrospective multicenter study, 2081 asymptomatic IgM gammopathy pts were identified, of whom 464 (117 IgM MGUS and 347 AWM) had longitudinal data available (baseline and at 6, 12, 24, and 36 months [m]). The objective was to evaluate the prognostic value of longitudinal changes in IgM levels and other biomarkers. Baseline risk of progression was classified according to two previous risk models by our group (Moreno, Hemasphere 2024), based on serum IgM ≥10 g/L, β2-microglobulin (β2M) ≥3 mg/L, and albumin <4 g/dL ± bone marrow (BM) infiltration ≥20% (BM-inclusive/free, respectively). Longitudinal IgM measurements were used to calculate individual IgM slopes using a linear mixed-effects model. Log2 transformation was used for free-light chains (FLC). We tested the association between IgM trends and progression-free survival (PFS) through a Cox proportional hazards model. RESULTS At diagnosis, the median age was 69 years (45.9% female). Baseline mean IgM, albumin, β2M and BM infiltration were 13.62 g/L, 4.18 g/dL, 2.7 mg/L and 20%, respectively. MYD88 L265P mutation occurred in 276/347 (79.5%), while CXCR4 mutations in 27/164 (16.5%). According to the BM-inclusive model, 55.5% and 44.5% were low-risk and intermediate/high-risk, respectively. Of 127 (27.4%) pts who progressed, 70% were intermediate/high-risk (p<0.001). Among 258 pts with complete IgM data at each time point, IgM levels showed an increasing trend over time, with the most pronounced rise occurring between +12 and +24m (p < 0.001). During the first two years, patients who progressed exhibited a 30% mean increase in IgM from diagnosis. This IgM increase was noted even after the second visit (+6m). In 424 pts with at least two consecutive IgM measurements, a higher IgM increase rate predicted shorter PFS (p=0.023). A cutoff of 20% of IgM increase in a subsequent visit distinguished a group of patients at higher risk of progression (N=56), with a 3.0-fold increased hazard of progression (p<0.001). By contrast, non-progressor IgM MGUS pts consistently had mean serum IgM levels under 5 g/L during the whole follow-up (+36m 95% CI, 3.9–6.1). An abnormal baseline FLC ratio, not previously systematically analyzed in IgM gammopathy, was an important risk factor for progression (hazard ratio [HR] 2.2, p=0.006). As a continuous value, FLC ratio remained stable over time in non-progressors (fold-change close to 0), whereas patients who progressed exhibited a 40% increase only beyond 36m (N=158). A similar trend was observed regarding hemoglobin values: non-progressors showed almost no significant differences over time, while pts who progressed had a continuous decrease from baseline, being only significant after +24m of diagnosis (mean decrease of 0.5 - 0.6 g/dL, N=299). These data highlight that the serum IgM increase is the earliest biomarker of progression. Longitudinal data on albumin and β2M allowed the assessment of risk groups during follow-up using the BM-free risk model. The Harrell C-index increased in each visit (baseline 0.60, +6m 0.64, +12m 0.65, +24m 0.70, +36m 0.69). During follow-up, 38 (13.7%) low-risk pts changed to high-risk, especially after +12m from diagnosis, underscoring the importance of continuous evaluation for progression prediction. CONCLUSION This study provides the first multi-timepoint analysis of biomarkers in a large cohort of asymptomatic IgM gammopathy pts. Longitudinal increases in serum IgM levels strongly predict progression in asymptomatic IgM gammopathy. Baseline and rising FLC ratios are also associated with progression. Continuous biomarker monitoring including hemoglobin, albumin, and β2M, improves dynamic risk stratification and early identification of patients requiring a closer follow-up. Pts with IgM less than 5 g/L rarely progress and stay stable over time.