AUTHORS
Anath Christopher Lionel, MD, PhD, Lorenzo Gensini, MD2, Sherif Seif, MD3, Lei Feng, MS4, Wenli Dong, MS5, Dingani Nkosi, MD6, Hima Bansal, MBBS7, Mahmoud R. Gaballa, MD8, Oren Pasvolsky, MD9, Krina Patel Jr, MD, MSc10, Jing Christine Ye, M.D., MSc11, Sarvari V. Yellapragada, MD12, Keyur P. Patel, MD13, Donna Marie Weber, MD14, Robert Orlowski, MD, PhD15, Sheeba Koshy Thomas, M.D.16
AFFILIATIONS
1 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
2 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
3 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
4 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
5 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
6 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
7 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
9 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
10 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
11 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
12 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
13 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
14 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
15 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
16 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
ABSTRACT
INTRODUCTION: Somatic mutations affecting CXCR4 occur in around 40% of patients with Waldenström macroglobulinemia (WM) and have been linked to reduced sensitivity to BTK inhibitor (BTKi)–based therapy and inferior outcome after first-line treatment with Bendamustine-Rituximab chemoimmunotherapy (CIT). However, the prognostic impact of CXCR4 mutation subtypes (nonsense (NS) and frameshift (FS)) on outcomes after first-line treatment for symptomatic disease remains incompletely characterized.
METHODS: We retrospectively analyzed the association between CXCR4 mutation status and type with treatment response and survival outcomes. All patients had next-generation sequencing (NGS) of bone marrow specimens prior to first-line therapy using an assay with > 250x coverage of the CXCR4 coding region; this assay also provided mutation statuses of MYD88 and TP53. Overall survival (OS), cancer-specific survival (CSS), progression-free survival (PFS), and time to next therapy (TTNT) were estimated using the Kaplan-Meier method, with group comparisons by log-rank test. A multivariable Cox proportional hazards model evaluated the independent association between CXCR4 and TTNT, adjusting for other significant covariates.
RESULTS: The study cohort included 116 patients with symptomatic WM, of whom 63 (54%) were male and 107 (93%) self-reported as White/Caucasian. Median age at NGS testing was 68 years. Patients were stratified by CXCR4 mutation status (CXCR4Wild-type(WT), n=68 and CXCR4mutated(MUT), n=48) and by mutation subtype (CXCR4FS, n=16 and CXCR4NS, n=31). Significant baseline differences at diagnosis of symptomatic WM between CXCR4MUT and CXCR4WT patients included lower platelet counts (median 185 vs. 234×10⁹/L, p=0.0002), lower β2-microglobulin (median 3.4 vs. 3.8 mg/L, p=0.017), and higher co-occurrence of MYD88 mutations (100% vs. 73.5%, p<0.0001); IPSSWM and MSS-WM risk score categories did not significantly differ between the 2 groups.
Median follow-up after NGS test date was 47 months (95% CI: 42~99 months). CXCR4 mutation status was not significantly associated with major or overall response rates, OS, CSS or PFS after first-line therapy, regardless of whether CIT or BTKi-based therapy was used. However, CXCR4MUT was associated with significantly shorter TTNT compared to CXCR4WT (median TTNT 60 months vs. not reached; p=0.006), an effect driven predominantly by CXCR4FS (median TTNT 11 vs. CXCR4NS (69 months)). Subgroup analyses revealed that these TTNT differences were observed with first-line CIT but not BTKi-based therapy. On multivariable analysis with TP53 mutation status and DLBCL transformation, CXCR4 mutation status remained independently associated with shorter TTNT (HR 2.56, 95% CI 1.25–5.27, p=0.011).
CONCLUSIONS: CXCR4 mutations in WM are associated with a distinct baseline biological profile including lower platelet counts and lower β2-microglobulin but do not appear to compromise depth of response or overall/cancer-specific survival after first-line therapy. However, CXCR4 mutations, particularly frameshift subtype, are independently associated with earlier need for next-line therapy, suggesting a shorter remission after first-line CIT but not BTKi-based regimens. These findings support incorporating CXCR4 mutation status into risk stratification for treatment sequencing and surveillance planning and warrant prospective validation in larger multi-center cohorts.
DISCLOSURES
Anath Christopher Lionel, MD, PhD: No relevant conflicts of interest to declare.