Waldenstrom's Macroglobulinemia and Overall Survival in Relation to Prognostic Factors and First Line Treatment: An Observational Study from Swedish Lymphoma Registry

Introduction: Waldenstrom's macroglobulinemia (WM) is a rare disease with a worldwide incidence of approximately 3-4 per million persons per year. The definition of WM includes presence of lymphoplasmacytic lymphoma (LPL) in bone marrow and IgM monoclonal immunoglobulin at any concentration (MI) in serum. LPL with nodal disease is mostly included in most epidemiological studies. In Sweden the age adjusted incidence for WM and LPL is high; 10.6 per million persons per year. The aim of this study is to describe the WM population in Sweden from a prospective nationwide population-based registry and to compare outcome in relation to prognostic factors and first line treatment.

Methods and study population: Between January 1, 2000 and December 31, 2014, in total 1511 patients with WM/LPL were registered in the Swedish Lymphoma Registry (SLR). The diagnosis was verified from medical records in 1135 patients and were classified as following; WM 978 (86.2 %), LPL 124 (10.9 %), other diagnosis 15 (1.3 %), and unclassified 18 (1.6 %).

Results: For the WM population (n = 978) the median age at diagnosis was 73 years (range: 29 to 94) with a male/female ratio of 1.5 and with no difference in age between males and females. Only 26 patients (2.7 %) were diagnosed below 50 years of age and 718 (73.4 %) patients were older than 65 years. The median size of the IgM monoclonal immunoglobulin (MI) in 960/978 patients at diagnosis was 19 g/l, and only 33 (3.4 %) patients had an IgM MI >70 g/l. The most common IgM MI was of kappa type; 642/794 (81 %). Twenty-six (3 %) patients had biclonal immunoglobulins. The median overall survival (OS) for all WM patients was 96 months with a 3-years and 5-year OS of 78 % and 66 %, respectively. The OS has improved over the time. Significant prognostic factors for OS in newly diagnosed patients (symptomatic and asymptomatic) in univariate analysis were; age (p < 0.001), WHO performance status (PS) 1-4 (p < 0.001), B-symptoms (p < 0.001), elevated LDH (p < 0.001), haemoglobin ≤ 115 (p < 0.001), albumin ≤ 35 (p < 0.001), Beta-2-microglobuline (B2M) > 3 (p < 0.001), platelets ≤ 100 (p < 0.001), and lymphocytosis ≥ 5x109/l (p < 0.001), but not the size of the IgM MI (p = 0.36), IgM MI > 70 (p = 0.66), the light chain (kappa/lamda) (p = 0.32) or hypogammaglobulinemia. In multivariate analysis age (p > 0.001), elevated LDH (p = 0.01), PS 1-4 (p < 0.001), and haemoglobin ≤115 (p < 0.001) remained significant (B2M, platelets and lymphocytosis were excluded due to many missing values). Female sex emerged as a positive prognostic factor (p = 0.003).

In Sweden is "watch and wait" (w&w) the recommended strategy in asymptomatic patients and w&w patients had a superior median OS compared to symptomatic patients, 101 and 77 months respectively. Approximately one fourth of the patients were symptomatic at diagnosis and data on first line treatment 0 - 3 months after diagnosis are available for 203 patients. Univariate and multivariate analyses on prognostic factors were performed separately for the patients with systemic therapies initiated 0 - 3 months after diagnosis (n = 203). Age (p < 0.001), PS 1-4 (p < 0.001), and haemoglobin ≤ 115 (p > 0.001) were significant prognostic factors in multivariate analysis. The size of the IgM MI (p = 0.87) or IgM MI > 70 (p = 0.58) was not significant prognostic factors.

Rituximab was introduced in WM therapy in the years 2007/2008 and when including treatment with rituximab (single or in combination with chemotherapies) and treatment with chemotherapies in the multivariate analysis emerged rituximab (p = 0.005), female sex (p = 0.002) as positive prognostic factors and LDH (p = 0.002), albumin ≤ 35 (p = 0.045) as negative prognostic factors.

Conclusion: The incidence of WM is higher in Sweden compared to the worldwide incidence. In this large unselected nation-wide WM population with trough medical records confirmed diagnosis, earlier known prognostic risk factors were confirmed, except the size of the IgM MI. Treatment with rituximab was independently associated with improved OS in multivariate analysis. The OS had improved over the time, the reason for that is unclear but one explanation could be the introduction of rituximab to the treatment.