Highly Sensitive Droplet Digital PCR for MYD88^L265P Mutation Detection and Minimal Residual Disease Monitoring in Waldenström Macroglobulinemia

Background. Recently, the somatic MYD88^L265P mutation has been found as the hallmark of Waldenström Macroglobulinemia (WM), being detectable in nearly 90% of cases, as well as in up to 50% of IgM MGUS, rarely in other non-Hodgkin lymphomas and never in multiple myeloma (MM). Beyond its potential diagnostic role, this mutation has been associated with tumor growth and therapy resistance. Moreover, MYD88^L265P might represent an ideal marker for minimal residual disease (MRD) monitoring in a disease whose therapeutic scenario has been rapidly changing, with many new available and highly effective drugs (nucleoside analogues, proteasome and BTK-inhibitors). However, the current MYD88^L265P allele-specific quantitative PCR (ASqPCR) diagnostic tool lacks sensitivity (1.00E-03) and thus is not suitable for MRD. Moreover, is not useful to test peripheral blood (PB), that harbors low concentrations of circulating tumor cells (especially after immunochemotherapy), neither to assess cell-free DNA (cfDNA), usually present at very low amount in plasma. Therefore, our study aims: 1) to assess whether a highly sensitive tool as droplet digital PCR (ddPCR) might be helpful in MYD88^L265P screening; 2) to evaluate whether MYD88^L265P might be a suitable marker for MRD monitoring in WM.

Methods. Bone marrow (BM) and PB samples were collected at diagnosis and during follow-up from a local series of patients affected by WM, IgM MGUS and IgG-secreting lymphoplasmacytic lymphoma (LPL), as well as samples from healthy subjects and MM were used as negative controls. Genomic (gDNA) and cell-free DNA (cfDNA) were extracted as recommended (Qiagen). MYD88^L265P was assessed on 100 ng of gDNA by ASqPCR as previously described [Xu 2013] and by ddPCR, using a custom dual labelled probe assay (Bio-Rad). When available, 50 ng of cfDNA were tested for MYD88^L265P, only by ddPCR. ddPCR was performed on 20 µl of reaction at 55°C for 40 cycles, run on QX100 droplet reader and analyzed by QuantaSoft v1.6.6 (Bio-Rad). MYD88^L265P ASqPCR level was estimated as described [Treon 2012]. ΔCT<8.4 identified a MYD88^L265P positive sample. Similarly, MYD88^L265P ddPCR cut-off was settled on the highest healthy samples level. IGH rearrangements identification and IGH-based MRD analysis were performed as previously described [van der Velden 2007].

Results. Once the ddPCR assay was optimized, the sensitivity of MYD88^L265P ddPCR was compared to ASqPCR on a ten-fold serial dilution standard curves built with a 70% MYD88^L265P mutated WM sample, previously identified by Sanger sequencing [Treon 2012]. Whereas ASqPCR confirmed the reported sensitivity of 1.00E−03, ddPCR reached a sensitivity of 5.00E−05. Thereafter, overall 105 samples (48 BM, 57 PB, 52 diagnosis and 53 follow up) from 58 patients (49 WM, 5 IgM MGUS and 4 LPL) as well as 20 controls (15 healthy subjects and 5 MM) were tested by both methods. 32/33 (97%) diagnostic BM scored positive for MYD88^L265P by both ddPCR and ASqPCR (being the only one negative a WM), while ddPCR, was able to detect more mutated cases, than ASqPCR, among diagnostic PB samples: 15/19 (79%) vs 9/19 (47%) (Table1). Moreover, to investigate whether the MYD88^L265P ddPCR tool could be used for MRD detection we compared it to the standardized IGH-based MRD. An IGH-based MRD marker was found in 40/53 (75%) patients (37 WM and 3 LPL). Five Patients, so far analyzed, with baseline and follow up samples (18 BM, 5 PB) showed highly superimposable results between the two methods. Finally, pivotal results on cfDNA from 10 patients showed higher median levels of MYD88^L265P mutation in plasma if compared to PB.

Conclusions. We developed a new tool for diagnosis and MRD monitoring in WM, showing that: 1) ddPCR is a highly sensitive tool for MYD88^L265P detection, especially useful in low infiltrated samples, like PB; 2) MYD88^L265P can be effectively and easily used for MRD monitoring in WM, achieving similar results to standardized IGH-based MRD; 3) cfDNA recovered from plasma might be an attractive alternative for MYD88^L265P detection, deserving further investigation. Methodological validation against IgH-based MRD detection and Flow cytometry and correlations with clinical impact are currently ongoing on external samples series.

TABLE 1. MYD88^L265P mutation detection in diagnostic samples: ddPCR vs ASqPCR