Progression Risk Stratification of Asymptomatic Waldenström Macroglobulinemia

Abstract

Background: Waldenström macroglobulinemia (WM) is preceded by asymptomatic WM (AWM), for which the risk of progression to overt disease is not well defined.

Methods: We studied 439 patients with AWM, who were diagnosed and observed at Dana-Farber Cancer Institute between 1992 and 2014.

Results: During the 23-year study period, with a median follow-up of 7.8 years, 317 patients progressed to symptomatic WM (72%). Immunoglobulin M 4,500 mg/dL or greater, bone marrow lymphoplasmacytic infiltration 70% or greater, β2-microglobulin 4.0 mg/dL or greater, and albumin 3.5 g/dL or less were all identified as independent predictors of disease progression. To assess progression risk in patients with AWM, we trained and cross-validated a proportional hazards model using bone marrow infiltration, immunoglobulin M, albumin, and beta-2 microglobulin values as continuous measures. The model divided the cohort into three distinct risk groups: a high-risk group with a median time to progression (TTP) of 1.8 years, an intermediate-risk group with a median TTP of 4.8 years, and a low-risk group with a median TTP of 9.3 years. We validated this model in two external cohorts, demonstrating robustness and generalizability. For clinical applicability, we made the model available as a Web page application ( www.awmrisk.com ). By combining two cohorts, we were powered to identify wild type MYD88 as an independent predictor of progression (hazard ratio, 2.7).

Conclusion: This classification system is positioned to inform patient monitoring and care and, for the first time to our knowledge, to identify patients with high-risk AWM who may need closer follow-up or benefit from early intervention.

FIG 1.

Cumulative probability of progression among patients. The Kaplan-Meier method was used for estimation of cumulative incidence of progression (A) among patients with asymptomatic Waldenström macroglobulinemia and stratified by (B) immunoglobulin M (IgM) levels, (C) bone marrow (BM) involvement, (D) β2-microglobulin, and (E) albumin.

FIG 2.

Cumulative probability of disease progression among patients with different risk scores according to the proportional hazards model and model performance in external validation cohorts. (A) The model was built using four variables: BM involvement, immunoglobulin M (IgM) levels, β2-microglobulin, and albumin. It divided the cohort into three risk groups, corresponding to low-, intermediate-, and high-risk AWM with a median time to progression (TTP) of 9.3, 4.8, and 1.8 years, respectively. Dashed and solid lines represent the results of training set and cross-validation, respectively. (B) Mayo Clinic, Rochester, MN, cohort: median TTP of 10.2, 5.7, and 2.4 years, for the low-, intermediate-, and high-risk groups, respectively. (C) National and Kapodistrian University, Athens, Greece, cohort: median TTP of not reached, 7.3, and 2.9 years, for the low-, intermediate-, and high-risk groups, respectively. BM, bone marrow; AWM, Asymptomatic Waldenström macroglobulinemia.

FIG 3.

MYD88 mutation status is an independent risk factor for progression to symptomatic Waldenström macroglobulinemia. The Kaplan-Meier method was used to compare progression-free survival between patients with MYD88 L265P and wild-type (WT) disease.