Page 128 - Haematologica June

1030
D. Drandi et al.
Introduction
In recent years, the MYD88L265P mutation has been recur- rently identified in Waldenström macroglobulinemia (WM),1,2 an indolent lymphoplasmacytic lymphoma (LPL) characterized by the accumulation in the bone marrow (BM) of monoclonal lymphocytes, lymphoplasmacytic cells and plasma cells, responsible for monoclonal IgM pro- tein secretion.3 Several studies using different techniques such as Sanger sequencing, polymerase chain reaction (PCR) and allele-specific quantitative PCR (ASqPCR), on CD19-sorted BM samples, found that about 90% of WM patients carry the MYD88L265P mutation, while it is present in 14-29% of “activated B-cell” diffuse large B-cell lym- phomas,4,5 6-10% of marginal zone lymphomas, 3-5% of chronic lymphocytic leukemia and is absent in multiple myeloma and in non-IgM monoclonal gammopathies of uncertain significance (MGUS).6–8 Therefore, MYD88L265P is now considered a hallmark of WM and may be helpful in the differential diagnosis from other lymphoproliferative neoplasms with overlapping clinical features, such as mul- tiple myeloma.9,10 Furthermore, it might represent an ideal marker for minimal residual disease (MRD) monitoring in a disease whose therapeutic scenario is rapidly changing, with many newly available and highly effective drugs.11–17 Moreover, MYD88L265P has been demonstrated in CD19- sorted BM samples from 50-80% of patients with IgM- MGUS, an asymptomatic phase which represents a pre- neoplastic condition, suggesting a potential role in disease progression.6,8,18–20 BM biopsy is mandatory for the differen- tial diagnosis between WM and IgM-MGUS, but patients with an asymptomatic M component do not readily agree to undergo such an invasive procedure. The availability of accurate diagnostic tools based on the use of peripheral blood (PB), or even urine samples, would overcome this problem and avoid the risk of misclassification of patients. Additionally, the current MYD88L265P ASqPCR method lacks sensitivity (1.00x10-3) and is not, therefore, suitable for MRD.6,21 Indeed, ASqPCR is suboptimal for testing specimens such as unsorted BM or even PB, which con- tains low concentrations of circulating tumor cells (espe- cially after immunochemotherapy), or for assessing cell- free tumor DNA (ctDNA), usually present in only very small amounts in plasma, including cerebrospinal fluid and pleural effusions.22,23 Recently, digital PCR has been shown to be a powerful technique that provides improved sensi- tivity, precision and reproducibility, overcoming some of the pitfalls of qPCR.24,25 We here describe a newly devel- oped, highly sensitive, droplet digital PCR (ddPCR) assay for the identification of the MYD88L265P mutation in patients affected by WM or LPL, suitable for screening and MRD monitoring on BM and PB cells, as well as on cell- free DNA.
Methods
Patients and samples collection
BM, PB, plasma and urine samples were collected at baseline and during follow up from patients affected by WM, IgM-MGUS or IgG-secreting LPL (Online Supplementary Figure S1). Consecutive patients were included in this study and were classified based on the 2008 World Health Organization classification criteria.3 PB from 40 healthy subjects and BM from 20 patients with multiple myeloma were used as negative controls. Sample collection and
storage as well as nucleic acid extraction procedures are described in the Online Supplementary Appendix.
All patients provided written informed consent to the use of their biological samples for research, in accordance with Institutional Review Board provisions and the Declaration of Helsinki. This study was approved by the local ethics committee.
Droplet digital polymerase chain reaction assays for detection of the MYD88L265P mutation
The mutation detection assay was designed as reported in Online Supplementary Figure S2A. ddPCR was performed using the QX100 Droplet Digital PCR system (Bio-Rad Laboratories) as detailed in the Online Supplementary Appendix. Samples were test- ed in triplicate and results are expressed as merge of wells. The cut-off for defining the presence of the mutation was set based on the highest MYD88L265P level detected within the control group and is indicated in figures as a red dashed line. Samples with a ratio value below the red dashed line are considered MYD88L265P wild- type (WT). Each experiment included a no template control, a known highly mutated positive control sample (mutation rate = 70%, mutated/WT ratio 7x10-1), previously tested by Sanger sequencing, as reported by Treon et al.,1 and a negative control (healthy donor or multiple myeloma gDNA). dMIQE guidelines (Minimum Information for Publication of Quantitative Digital PCR Experiments) for ddPCR experiments are listed in Online Supplementary Table S1.25
Allele-specific quantitative polymerase chain reaction assay for detection of the MYD88L265P mutation
The sensitivity of MYD88L265P ddPCR was compared to that of the ASqPCR assay previously described by Xu et al.6 on a standard curve of 10-fold serial dilutions. In parallel, 100 WM samples from the Torino series and 23 patients from the Salamanca series were analyzed, following the strategy described by Jiménez et al.21 Additionally, MYD88L265P mutation detection by ddPCR was com- pared to that of the qBiomarkerTM Somatic Mutation Assay kit for MYD88_85940 (Qiagen) on 15 samples from the University of Pisa.
Allele-specific quantitative polymerase chain reaction assay for tumor-specific IGH-VDJ rearrangement
In order to perform a comparison to the worldwide standard- ized ASqPCR technique for immunoglobulin-VDJ rearrangement (IGH-VDJ) MRD analysis, patient-specific IGH-VDJ was amplified and directly sequenced26 (Online Supplementary Figure S2B). This IGH-based MRD analysis was performed according to the Euro- MRD guidelines.27
Statistical analysis
Associations between categorical variables were analyzed by the Fisher exact test, while Mann-Whitney and Kruskal-Wallis tests were used for the inference on continuous variables. Results of the analyses of continuous variables are expressed as the median (range); ddPCR and ASqPCR results are expressed as mutated:WT ratio. The interrater agreement on categorical data was estimated by computing the Fleiss kappa (k) index. All reported P-values were estimated by the two-sided exact method with the conventional 5% significance level. Data were analyzed as of July 2017 using R 3.4.1 (R Foundation for Statistical Computing, Vienna, Austria).28
Results
Patients and samples
Overall, 291 samples from 148 patients from the three series were analyzed (133 WM, 1 amyloid-associated
haematologica | 2018; 103(6)