A new risk stratification model for Waldenström macroglobulinemia (WM), a rare hematologic B-cell malignancy, better differentiates between patients at higher risk and lower risks than the existing staging systems while also using fewer variables, researchers said in a new paper in the Journal of Clinical Oncology.
The model — called the Modified Staging System for WM (MSS-WM) — uses age, serum albumin, and serum lactate dehydrogenase (LDH) to assess patient risk and has been shown to more reliably identify the differences between patient outcomes, said the researchers, led by Prashant Kapoor, MD, of Mayo Clinic in Rochester, Minnesota.
“This led us to conclude that we could further simplify the original IPSS (International Prognostic Scoring System) and utilize very simple parameters that we routinely use in clinical practice to distinctly separate patients into four groups with completely different survival rates,” he said.
Accurately assessing survival risk at the time of starting treatment is an important part of managing patients with WM.
Existing risk stratification models for WM — the IPSSWM and the revised IPSSWM (rIPSSWM) — were either developed before the currently used front-line chemoimmunotherapies or did not look at the effects of molecular parameters, including myeloid differentiation primary response 88 (MYD88), which is now routinely assessed at diagnosis. Most patients fell into one of two groups under the IPSS.
Researchers turned to a cohort of 889 patients diagnosed with symptomatic WM at Mayo Clinic. Using hazard ratios for overall survival (OS) for these factors as a guide, they assigned a score of 1 point each to serum albumin of less than 3.5 g/dL and an age of 66 to 75 years old, and 2 points each to an age of greater than 75 and elevated serum LDH with a value above the upper limit of normal for the laboratory.
Those with a composite score of 0 were assigned to the low-risk group, a score of 1 to the low-intermediate risk group, 2 to the intermediate group, and a score of 3 to 5 to the high-risk group, since those patients had similar outcomes and relatively small numbers of patients had scores of 4 or 5.
Twenty-one percent of patients were scored as low-risk, 32% as low-intermediate, 24% as intermediate, and 23% as high-risk. The estimated median OS for those groups was 14.6 years, 11.2 years, 8.3 years, and 5.5 years, respectively. Their five-year OS rates were 93%, 82%, 69%, and 55%, respectively.
Using an external validation cohort, the groups were similarly differentiated, with five-year OS rates of 93%, 90%, 75%, and 57%, respectively.
When grouped using the rIPSSWM model, patients in the Mayo derivation cohort had overlapping survival curves for some of the groupings, researchers found.
Moreover, among the 220 patients with data available to make estimations using both IPSSWM and MSS-WM models, nearly one in 10 in the high-risk group under IPSSWM was reclassified as low-risk under the MSS-WM, and only about a third who were in the high-risk IPSSWM group remained high-risk under MSS-WM, highlighting the value of the proposed four-tiered staging system, Dr. Kapoor said.
OS of those with the MYD88L256P mutation, at a median of 11.4 years, was similar to OS of those without the mutation, at a median of 10 years, with a p-value of 0.10. That mutation has been found to be predictive mostly for those treated up front with BTK inhibitors (BTKis) and fewer than one in five were treated with BTKis in this cohort, which might explain MYD88’s poor predictive power seen in this analysis, Dr. Kapoor said. The prognostic impact of MYD88 mutational status should be examined further, specifically in larger cohorts treated initially with BTKis, he said.
The impact of other potentially predictive mutations, including CXCR4, also should be examined, he said, although due to the complexity of the assay, capable of detecting the entire breadth of alterations potentially encountered within the C-terminal domain of CXCR4, such tests are typically only performed at academic medical centers currently.
“We are at a good distinguishing capacity,” Dr. Kapoor said. “But there is clearly room for improvement.”
Any conflicts of interest declared by the authors can be found in the original article.
Reference
Zanwar S, Le-Rademacher J, Durot E, et al. Simplified risk stratification model for patients with Waldenström macroglobulinemia. J Clin Oncol. 2024;42(21):2527-2536.
For more on WM, check out the recent Algorithms in Hematology about using genomics and BTKis to treat WM.
