Refining prognostic tools in Waldenström macroglobulinemia

In this issue of Blood Advances, Zanwar et al¹ report results of an international multicenter investigation evaluating the prognostic value of progression of disease within 24 months (POD24) and mutations affecting MYD88 or CXCR4 as early surrogate end points for survival, in a large cohort of patients with Waldenström macroglobulinemia (WM) treated with first-line bendamustine and rituximab (BR). Significantly worse progression-free survival (PFS) and overall survival (OS) outcomes were observed in the 11.5% of patients with POD24, as well as in those with CXCR4 mutations (CXCR4^MUT); in contrast, patients with MYD88 mutations (MYD88^MUT) did not have inferior survival outcomes.

This study provides important evidence for the prognostic utility of POD24 in WM, a metric that has previously been established as a robust early clinical predictor of poor survival in other indolent B-cell lymphomas, such as follicular lymphoma and marginal zone lymphoma.² In addition to expanding the list of prognostic metrics for survival in WM (see table), this finding has clinical implications for the counseling of patients, because those who did not progress within 24 months of starting BR had a similar median OS to that of a cohort of healthy individuals matched on factors including age, sex, and country of origin. Whether POD24 has similar prognostic utility after other first-line treatments for WM, such as Bruton tyrosine kinase inhibitors (BTKi), requires further investigation. Equally essential is the need to characterize the underlying molecular determinants of the high-risk group of patients who experience POD24 after BR.

Prognostic models and factors associated with OS in WM

Prognostic model or factor	Components	References
Prognostic models
IPSSWM	Age, β2M, hemoglobin, platelet count, serum monoclonal protein	3
Revised IPSSWM	Age, β2M, LDH, serum albumin	4
MSS-WM	Age, LDH, serum albumin	5
Prognostic factors
MYD88 mutation status	Reduced OS with MYD88 wild type after treatment with ibrutinib	6
CXCR4 mutation status	Reduced OS with CXCR4 mutations after first-line therapy with BR	1
TP53 mutation status	Reduced OS with TP53 point mutations or 17p deletions	7
POD24	Reduced OS with POD24 after first-line therapy with BR	1

LDH, lactate dehydrogenase; IPSSWM, International Prognostic Scoring System for WM; MSS-WM, Modified Staging System for WM; β2M, beta-2 microglobulin.

This study also provides clinical evidence of the negative prognostic impact of CXCR4^MUT, present in 30% to 40% of WM cases,⁸ on response rates, PFS, and OS after first-line BR in patients with WM. These findings are consistent with previously reported associations between CXCR4^MUTand inferior response with BTKi.^8,9 Further study with a larger number of patients harboring CXCR4^MUT is required to determine the relative impact of different CXCR4 mutation subtypes on survival outcomes and response to BR in WM. This is of relevance given data showing that among patients treated with ibrutinib, those with CXCR4 nonsense mutations have worse clinical outcomes than those with CXCR4 frameshift mutations,^8,9 whereas outcomes with zanubrutinib were not similarly affected.⁹ Future research is also needed to investigate potential mechanisms explaining the interesting findings in this study of similar response rates and survival outcomes after BR in those with MYD88^MUT compared to those without such mutations, as this is in contrast to reports of inferior responses and survival after ibrutinib among those with MYD88^{wild-type(WT)} WM.⁶

Although the majority (70%) of patients in the POD24 cohort of the present study had CXCR4^MUT, POD24 remained prognostic for reduced OS after adjusting for CXCR4 mutation status. This raises the possibility that mutations in other genes play a role in the type of high-risk disease captured by the POD24 clinical metric. CXCR4^MUT has been associated with the concurrent presence of TP53 mutations (TP53^MUT) in patients with WM.⁹ Of note, the current study by Zanwar et al did not have information available on the TP53 mutation status of patients in the POD24 cohort. Alteration of TP53 by either somatic point mutations or 17p deletions has been observed to have a negative prognostic impact on survival in WM.⁷ There is also emerging evidence for the impact of TP53^MUT on response to therapy in WM, including worse cancer-specific survival in those with such mutations who received chemoimmunotherapy such as BR for first-line treatment when compared to those who received BTKi-based regimens.¹⁰ Furthermore, biomarker analysis of the ASPEN trial found patients with TP53^MUT to have worse prognosis after treatment with BTKi compared to those with TP53^WT; interestingly, those with TP53^MUT who received zanubrutinib had better major response rates and PFS compared to those who received ibrutinib.⁹

In summary, this study by Zanwar et al provides evidence for the prognostic utility of POD24 and CXCR4 mutation status as surrogate end points for OS and PFS after first-line treatment with BR in patients with WM. These findings have clinical relevance for both patient counseling and risk stratification, and will also help facilitate future research to identify molecular determinants of high-risk disease and optimize risk-adapted therapeutic approaches for patients with WM.

Conflict-of-interest disclosure: S.K.T. reports research funding from Bristol Myers Squibb, Janssen Biotech, Cellectar Biosciences, Acerta Pharma, AbbVie Pharma, and Sanofi-Aventis; and has served on the advisory boards of Cellectar Biosciences and Mustang Bio. A.C.L. declares no competing financial interests.