TP53 mutations in paired bone marrow and cell-free DNA samples of patients with Waldenström macroglobulinemia or IgM monoclonal gammopathy of undetermined significance prospectively enrolled in the FIL_biowm study of fondazione italiana linfomi

Abstract

Abstract Background. TP53 mutations are emerging as a relevant prognostic and predictive biomarker in Waldenström Macroglobulinemia (WM). In two retrospective studies, TP53 mutations were associated with shorter progression-free survival (PFS) and overall survival (OS) with chemoimmunotherapy (CIT) (Poulain S et al, Clin Cancer Res 2017; Gustine JN et al, Br J Haematol 2019), while in a post-hoc analysis of the ASPEN study, TP53 mutations correlated with shorter PFS in patients treated with BTK inhibitors (Tam CS et al, Blood Adv 2024). Aim of the study. In the multicenter, observational prospective BIOWM trial (NCT03521596) sponsored by Fondazione Italiana Linfomi (FIL),we assessed TP53 mutations in paired bone marrow (BM) and peripheral blood (PB) samples of patients with untreated WM or IgM-Monoclonal Gammopathy of Undetermined Significance (IgM-MGUS), in order to assess the rate and prognostic impact of TP53 mutations and to evaluate whether these mutations can be detected with adequate accuracy in cell-free DNA (cfDNA) extracted from plasma. Methods. We analyzed paired BM and PB samples collected at the time of diagnosis (T0) or before treatment (T1). In patients with stable disease not requiring treatment, an additional PB sample was collected during follow-up at 12 months (FT1) and at 24 months (FT2). Targeted next-generation sequencing (NGS) of a 15-gene panel including MYD88, CXCR4, TP53, ARID-1A, NOTCH2, KMT2D, CD79b, CARD11, KLF2, TRAF3, HIST1H1E, PRDM1, MYBBP1A, BTK and PLCγ2 was performed on DNA extracted from BM and from plasma. OS and PFS were analyzed using the Kaplan-Meier method. The prognostic impact of TP53 mutations - either at diagnosis or acquired later - was estimated using a Cox regression model with TP53 mutation as time-varying covariate (TVC). Results. TP53 mutation status was assessed in 279 patients, including 210 with WM and 69 with IgM-MGUS. The median follow-up of the study population was 66 months. Overall, 31/279 patients (11%), including 21/210 patients with WM (10%) and 10/69 with IgM-MGUS (14%), harbored a TP53 mutation. The median variant allele frequency (VAF) was 30% (Inter Quartile Range, IQR 12-40) in WM and 6.8% (IQR 2.2-21) in IgM-MGUS patients (P=0.036). Of 31TP53 mutated patients, 20 were mutated at T0 or T1, while 11 acquired the mutation later during the watch-and-wait phase (FT1 or FT2). The total number of oncogenic mutations detected was 35, including 28 missense mutations and 7 nonsense mutations, and the majority of them (32/35, 91%) occurred in the central DNA binding domain encompassing exons 4-8. The rate of TP53 mutations was higher in cfDNA as compared with BM samples (10% versus 4%, P=0.001). Twenty-one of 31 TP53mutated patients (68%) harbored also the MYD88 (L265P) mutation by NGS, with a median VAF of 30% (IQR: 19-40%). By droplet digital polymerase chain reaction (ddPCR) the MYD88 (L265P) mutation was detected in 30/31TP53 mutated patients. As compared with TP53 wild-type patients, the subgroup of TP53 mutated patients was borderline enriched in CXCR4 mutations (35% versus 22%, P=0.113), and significantly enriched in KMT2D mutations (26% versus 9.7%, P=0.015) and NOTCH2 mutations (23% versus 6.5%, P=0.007). Among TP53 mutated patients, 10 have been treated as they met IWWM treatment criteria, 9 with chemo-immunotherapy (7 with Rituximab+Bendamustine, 2 with Dexamethasone+Rituximab+Cyclophosphamide) and 1 with single agent Cyclophosphamide. WM patients with a TP53 mutation had a trend towards a shorter time to first treatment (TTFT) (hazard ratio, HR 2.24, 95% CI 0.67-7.57, P=0.192) and towards an inferior OS as compared with wild-type ones (HR 1.99, 95% CI 0.97-4.09, P=0.055). In WM patients who were treated so far, we did not find a significant difference in PFS between TP53-mutated and wild-type patients (HR 0.65, 95% CI 0.28-1.49, P=0.310). Conclusions. To the best of our knowledge, this is the first prospective study demonstrating an impact of TP53 mutation status on OS in WM patients. Due to the low number of patients treated so far, an impact of TP53 mutations on PFS could not be confirmed at this time. Cell-free DNA extracted from plasma demonstrated to be at least accurate as BM DNA for the assessment of TP53 mutation status, offering an alternative non-invasive, patient-friendly approach for the detection and monitoring of these mutations. The impact of low VAF (<10%) TP53 mutations in patients with IgM-MGUS deserves further investigation.