Evolution of Management and Outcomes in Waldenström Macroglobulinemia: A Population-Based Analysis

Abstract

Introduction: Waldenström macroglobulinemia/lymphoplasmacytic lymphoma (WM) is a rare lymphoma affecting older patients. Its management largely relies on small phase II trials and it is unclear how their results translate into clinical practice in the community.

Method: We evaluated changes in the presentation, management, and survival among 2,666 Medicare beneficiaries diagnosed with WM between 1994 and 2011, using Medicare claims linked to Surveillance, Epidemiology and End Results data.

Results: Prevalence of transfusions, anemia, thrombocytopenia, and neuropathy at diagnosis significantly increased over time, whereas the use of plasmapheresis was low (2.5%) and stable. The proportion of patients starting chemotherapy within 1 year of WM diagnosis increased from 39% in 1994 to 62% in 2011 (p < .0001). Treatments based on classic alkylators and purine analogs predominated in the 1990s, but were quickly replaced by rituximab-containing regimens after 2000. Rituximab monotherapy has been prescribed for >50% of patients since 2004, and combination chemoimmunotherapy for a further 30%. Most patients initiating multiagent regimens in 2012-2013 received rituximab with bortezomib or bendamustine. These changes were accompanied by significant improvements in overall and WM-related survival, but also by a significant increase in cost of chemotherapy. Mean Medicare payments for chemotherapy drugs accrued in the first year of treatment rose from $9,464 in 1994-2000 to $29,490 after 2008.

Conclusion: Hematologists have rapidly adopted innovative, expensive therapies for WM before completion of randomized trials. This underscores the need to assess the comparative value of such therapies in rare malignancies through a combination of clinical and observational data.

Implications for practice: Most older patients with Waldenström macroglobulinemia currently treated in the U.S. receive rituximab as monotherapy or in combination with bortezomib or bendamustine. Newly designed trials should consider control arms aligned with this prevalent real-life standard. Compared with the 1990s, patients diagnosed according to current criteria are more likely to have anemia or neuropathy, or to receive early chemotherapy, but only 2.5% require plasmapheresis at diagnosis. The incremental clinical value of newly introduced agents needs to be assessed through a combination of clinical and health services research, taking into consideration their associated survival benefits, toxicities, and associated costs of care.

Keywords: Bendamustine; Bortezomib; Cost of therapy; Health services research; Rituximab; Waldenström macroglobulinemia.

Figure 1.

Selection of the analytic cohort from the SEER-Medicare data. Abbreviations: COD, certificate of death; LPL, lymphoplasmacytic lymphoma; SEER, Surveillance, Epidemiology, and End Results; WM, Waldenström macroglobulinemia.

Figure 2.

Administration of first-line chemotherapy in Waldenström macroglobulinemia. (A): Cumulative incidence of starting chemotherapy or death without treatment. (B): Trend in the proportion of patients starting chemotherapy within 1 year of diagnosis.

Figure 3.

Upfront chemotherapy regimens administered to patients with Waldenström macroglobulinemia. Data in bar graphs are stratified by year of treatment (A), mean cumulative Medicare payments for care accrued within 1 year of diagnosis (B), or within 1 year of starting chemotherapy (C), in the context of increasing rituximab use. Range bars indicate SE; p values are for linearized yearly trends in log-γ models for costs.

Figure 4.

Plots of overall survival, period survival, and cumulative incidence of death. (A): Overall survival. (B): Stacked cumulative incidence of death from Waldenström macroglobulinemia or from competing causes. (C): Period survival, which illustrates the probability of 3-year survival using time-at-risk confined to a specific calendar period.