Transformed Waldenström macroglobulinemia is associated with poor outcomes and chemorefractory disease

Abstract

Abstract Background: Histological transformation (HT) of Waldenstrom Macroglobulinemia (tWM) is associated with an overall survival of less than 3 years. Risk factors for HT include wildtype MYD88, CD79B mutation, elevated LDH and time from initial diagnosis to HT that is greater than 5 years. There is a paucity of research focusing on tWM, and given its poor prognosis, dedicated efforts are required to improve outcomes. In an attempt to better understand tWM, we describe the clinicopathologic characteristics of a cohort of tWM patients with tumorigenic profiling using MSK-IMPACT Heme. Methods: We retrospectively identified pathologically confirmed tWM patients from 2005 and 2025 at MSKCC. Demographics, treatment history, clinical response, PFS and OS were extracted. PFS and OS were estimated using the Kaplan-Meier method, and comparisons were made using log-rank test. Associations between PFS/OS, clinical, and treatment characteristics were evaluated using Fischer exact test. Tumors were sequenced by MSK-IMPACT Heme. Cell of origin (COO) was determined using Han's Algorithm. Results: Twenty-three patients with tWM were identified. Median time from initial diagnosis of WM to HT was 25 months (0-205 months). Median IgM level at diagnosis and at HT were 1926 and 445 mg/dL, respectively. Twelve of 23 patients (52%) had extranodal involvement. Eight patients (34%) had documented CNS involvement at baseline/relapse. The median lines of therapy prior to transformation was 1 (0-11). Therapy prior to HT included rituximab (R) (n=4), Ofatumumab (Ofa) alone (n=1), R/Ofa-Benda based (n=12), BTKi (n=10), proteasome inhibitor-based (n=7), FCR (n=2), R-CHOP (n=1), R-Cytoxan (n=1), idealisib (n=1), and venetoclax (n=1). Two patients required plasmapheresis. Treatment for tWM included R-CHOP/Pola-R-CHP (70%), dose intense therapies [R-EPOCH, R-CODOX, HyperCVAD, RICE] (17%), and 14% received other regimens [R-MPV, R-Cytoxan, R-CEPP]. The ORR was 70% (16/23), with a CR rate of 65% (15/23). No patient underwent upfront consolidation with ASCT. With a median follow-up of 24 months after HT, 55% of patients are deceased due to progression. The 2-year PFS and OS are 34% (95% CI 18-62) and 64% (95% CI 46-88%), respectively. The median PFS and OS was 12.3 months, and 29.6 months, respectively. Median time to relapse after chemoimmunotherapy was 8.0 months (1-59 months). Upon relapse, 6 patients received stem cell transplant (n=3; 2 autologous, 1 allogeneic) or CAR T-cell therapy (n=3). All patients who received CAR-T cell therapy are alive without evidence of disease, while none are alive after transplantation. When evaluating for potential variables that may influence outcome, de novo tWM, prior BTKi exposure, prior chemoimmunotherapy, and time to transformation (greater than 5 years) did not impact PFS or OS. The presence of extranodal disease, specifically CNS involvement, was associated with inferior PFS (p=0.002). At initial diagnosis, seventeen patients had documented MYD88 mutational testing via PCR or NGS. Twelve of those patients were MYD88 mutated. Three patients harbored CXCR4 mutation. Non-germinal centerm COO was predominant (18/23); 5 samples were unclassifiable. NGS at transformation (n=8) revealed that 62.5% of patients harbored mutations in MYD88 and TP53; only one patient had wildtype MYD88. TP53 was the most genetically altered gene, with several patients characterized by concomitant copy number alterations and mutations. Other notable genetic alterations include CD79B (62%), CXCR4 (38%), PRMD1 (38%), CDKN2A (38%), ARID1A (25%). Conclusions: Our single-center experience recapitulates the poor OS of tWM patients and reinforces the need for further advances for this vulnerable population. The median time to relapse was less than 12 months, suggestive of chemorefractory disease. The presence of CNS involvement significantly influenced PFS. Although limited by numbers, patients who received CAR T-cell therapy upon relapse of tWM demonstrated promising results, likely due to chemorefactory disease. To our knowledge, this is one of the first studies to describe the genetic pattern of tWM. TP53 alterations were enriched in tWM samples, similar to other transformed indolent lymphomas, and may translate to poor sensitivity to chemoimmunotherapy. It is unclear if this is an acquired mutation or clonal evolution. Ultimately, additional studies focusing on the genetic drivers of tWM is warranted to help guide therapeutic advances.