Lymphadenopathy is an independent prognostic factor for patients with Waldenström macroglobulinaemia

Waldenström macroglobulinaemia (WM) is a rare B-cell lymphoproliferative disorder characterized by the clonal expansion of mature B cells within the bone marrow and the production of monoclonal immunoglobulin M (IgM). While the bone marrow is the primary site of disease involvement, extramedullary manifestations, such as lymphadenopathy, often occur in WM patients.1 In other indolent B-cell lymphomas, lymph node-associated tumour burden serves as a predictor of adverse outcomes and an indicator for initiating treatment.2 However, the prognostic significance and clinical impact of lymphadenopathy in WM remain underexplored, whereas the presence of lymphadenopathy is not included in the revised international prognostic score system for WM.3, 4 This report presents findings from a retrospective analysis of a prospectively maintained database, aiming to assess the impact of lymphadenopathy findings at diagnosis on the clinical presentation and outcomes of patients with WM, following the pertinent results of the recent publication by Ostergaard et al.5 This multicentre study included consecutive, newly diagnosed and treatment-naive patients diagnosed with symptomatic WM who received treatment between 1977 and 2024. Individual patient data were retrieved from a prospectively maintained, comprehensive database. Lymphadenopathy at diagnosis was assessed through clinical examination and all patients were evaluated by imaging with computed tomography (CT), as a standard practice. Abnormal lymph nodes in WM were defined as nodes ≥1.5 cm (short axis) on CT scans. First-line WM-directed therapies were documented, and treatment responses (based on the modified sixth criteria6) were categorized based on IgM reduction and imaging response was also recorded for patients with lymphadenopathy and/or organomegaly. Clinical characteristics were reported using descriptive statistics and chi-squared, Fisher or Wilcoxon tests were used for comparison between groups. Median follow-up was calculated by the reverse Kaplan–Meier approach. Time-to-event curves were generated using the Kaplan–Meier approach and differences were compared using the log-rank test. Univariate and multivariate analyses were performed using a Cox proportional hazards regression model. Population-wide risk factors to be included in this analysis were prespecified and included the updated risk stratification criteria: serum variables (albumin, haemoglobin [Hb], platelet count, β2 microglobulin, lactate dehydrogenase [LDH] and monoclonal IgM concentrations) and age (treated as a two-level categorical variable, ≤65 years or >65 years).3, 7 Moreover, inherent heterogeneity factors of the population such as diagnosis year, gender and therapy type are included in all models. Finally, in order to enhance interpretability and avoid overfitting, parsimonious multivariate Cox models were constructed by retaining only covariates with influence on the outcome studied, assessed through stepwise selection in both directions, using Akaike information criterion (AIC). Of a total of 707 patients with symptomatic WM, lymphadenopathy at diagnosis was present in 232 (32.8%); their characteristics are shown in Table S1. The median follow-up was 10.39 years (95% confidence interval [CI]: 9.54–11.41 years) for the whole cohort, 10.63 years (95% CI: 9.88–13.40 years) for the subgroup of patients with nodal involvement at baseline and 10.20 years (95% CI: 8.85–11.19 years) for those without. There were no statistically significant differences between patients with and without lymphadenopathy in terms of age, presence of MYD88 L265P and CXCR4 mutations, B symptoms and anaemia (Hb < 10 g/dL) at diagnosis, although the number of females was lower in the patient subgroup with lymphadenopathy (33.2% vs. 44.6%, p = 0.005). The median progression-free survival (PFS) was significantly lower among patients with lymphadenopathy at baseline (3.17 years, 95% CI: 2.57–3.99), compared to those without (5.36 years, 95% CI: 4.45–6.25). The overall survival (OS) was also inferior for patients with lymphadenopathy compared to those without (8.8 years [95% CI: 8.0–10.9] vs. 10.0 years [95% CI: 9.36–11.7], respectively) (Figure 1A,B). As per univariate Cox analysis, patients with lymphadenopathy at baseline demonstrated a 63% increased risk for disease progression or death (Hazard Ratio (HR) = 1.63, 95% CI: 1.36–1.97, p < 0.001) and a 29% increased risk for death (HR = 1.29, 95% CI: 1.03–1.60, p = 0.024), compared to those without lymph node involvement. Further exploration via multivariate Cox regression models underlined lymphadenopathy as a statistically significant predictor of adverse prognosis, as it was associated with a 57% increased risk for disease progression or death (adjusted Hazard Ratio (aHR) = 1.57, 95% CI: 1.25–1.96, p < 0.001) and a 31% increased risk for death (aHR = 1.31, 95% CI: 1.01–1.73, p = 0.049). This established lymphadenopathy at diagnosis as an independent predictor of inferior survival outcomes in our cohort (Table 1). Moreover, while no significant differences were observed with regard to second primary malignancies between groups, there was an increase in disease transformation to high-grade lymphoma in the subgroup of patients with lymphadenopathy present at diagnosis (6.9% vs. 2.1%). However, there were no statistically significant differences regarding the time to transformation between the two groups (aHR = 1.44, 95% CI: 0.62–3.36, p = 0.398). Regarding the first-line treatment for WM, 63 (8.9%) patients were treated with bortezomib, dexamethasone, rituximab±cyclophosphamide (BDR±C); 57 (8.1%) with Bruton's tyrosine kinase inhibitors (BTKis); 261 (36.9%) with dexamethasone, rituximab and cyclophosphamide (DRC); and 326 (46.1%) with other older regimens including conventional chemotherapy. The percentage of patients treated with BTKi was lower in the subgroup of patients with lymphadenopathy (4.3% vs. 9.9%, Table S2). Lymphadenopathy at diagnosis remained an adverse prognosis factor in the subgroup of patients treated with DRC (PFS aHR = 1.99, 95% CI: 1.44–2.74, p < 0.001 and OS aHR = 1.57, 95% CI: 1.05–2.36, p = 0.027). Regarding response to treatment in terms of IgM reduction, there were no significant differences observed between the two groups (39.7% vs. 35.4%, no response; 32.3% vs. 36.2%, minor response; 25.0% vs. 22.9%, major response). Among patients with lymphadenopathy, the depth of response differed significantly depending on treatment (Fisher's, p = 0.030). More specifically, treatment with BTKis was the most effective (70% minor and 20% major responses), followed by DRC (38.3% minor and 27.2% major responses), while BDR/B-DRC demonstrated the highest percentage of major responses (33.3%) (Figure 1C). Furthermore, 120 of the 232 (51.7%) patients with lymphadenopathy at diagnosis exhibited response on CT imaging, with 40 (17.2%) achieving partial response (at least 50% but not 100% reduction of lymph node enlargement) and 80 (34.5%) complete response (disappearance of lymph node enlargement). Finally, median time to next treatment (TTNT) was significantly lower in the lymphadenopathy subgroup (4.75 years vs. 8.03 years) (Figure 1D), and those patients had a 53% increased risk of requiring salvage therapy, compared to those without (aHR 1.44, 95% CI: 1.14–1.82, p < 0.001). In this study, lymphadenopathy was evident in 33% of the patients with symptomatic WM at diagnosis. The presence of lymphadenopathy emerged as an independent adverse prognostic factor, associated with shorter survival, earlier disease progression and a higher risk of disease transformation to high-grade lymphoma. Importantly, our results are in concordance with a recent Danish study that reported nodal involvement in one-third of 469 patients with WM.5 However, we included only patients with symptomatic WM, in comparison to the Danish study that included patients with both asymptomatic and symptomatic disease. Two other studies from Spain and Latin America have previously reported lymphadenopathy in less than 30% of 217 and 159 patients with symptomatic WM respectively.8, 9 In line with our results, the Danish study also found an adverse association between lymphadenopathy and survival outcomes, along with an increased risk of WM transformation.5 The risk of histological transformation of WM to high-grade lymphoma is estimated at about 4%, whereas it may affect both nodal and extranodal sites.10-12 However, new or enlarging lymph nodes in patients with WM, especially in the presence of elevated levels of serum lactate dehydrogenase, stable IgM levels and wild-type MYD88, are highly suspicious of transformation and lymph node biopsy is essential. We may postulate that marked lymphadenopathy associated with high tumour burden may increase the probability of clonal events driving transformation; however, this has to be investigated in future studies. Among the limitations of our retrospective analysis, we should note the heterogeneity in terms of diagnostic and therapeutic standards for WM throughout the 48 years that spanned the study. Furthermore, data on mutational status for MYD88 and CXCR4 were available only for a subset of patients. Although all the patients with lymphadenopathy had been evaluated with CT scans, the available data on positron emission tomography (PET)/CT scans were limited, and thus, they were not included in the present analysis. Fluorodeoxyglucose (FDG)-PET/CT may reveal nodal involvement in WM that may be occult by conventional imaging.13 Another point to consider in future studies would be the differentiation between bulky and non-bulky lymphadenopathy and the evaluation of the prognostic role of the burden of lymph node involvement. In conclusion, our analysis showed that the presence of lymphadenopathy at the time of diagnosis of symptomatic WM was associated with poor patient outcomes. Our results have important implications both for clinical practice regarding the need for long-term close monitoring even in patients who have responded to first-line treatment, as well as for clinical trial design in terms of integrating specific subanalysis for this subgroup of patients in order to optimize their management. Conceptualization: IN-S and MG; methodology: IN-S and CF; software: CF; formal analysis: CF; investigation: IN-S, VL, NG, ES, SD, DF, PM, MMig, MMic, SG, EH, EKat, ET, EKas, MAD and MG; data curation: CF; writing—original draft preparation: IN-S and CF; writing—review and editing: MG and IN-S; supervision: MG, MAD and EK. All authors have read and agreed to the final version of the manuscript. The authors have nothing to report. This research received no external funding. The authors declare no relevant conflicts of interest. Ethical review and approval were waived for this study due to the retrospective nature of the non-experimental, observational study. Patients provided informed consent for data collection. Further data available upon reasonable request from the corresponding author. Tables S1–S2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.