Daniela Drandi 1,*, Elisa Genuardi1,
Irene Dogliotti1, Martina Ferrante1, Cristina Jimenez2, Francesca Guerrini3, Mariella Lo Schirico1, Vittorio Muccio1, Barbara Mantoan1, Milena Gilestro4, Paola Omedè4, Sara Galimberti3,
Lorella Orsucci4, Federica Cavallo1, Ramon Garcia Sanz2, Mario Bocca 1,4,Marco Ladetto5, Simone Ferrero1
1 Department of Molecular Biotechnologies and Health Sciences, Hematology Division, University of Torino, Italy
2 Servicio de Hematología, Hospital Universitario de Salamanca, Salamanca, Spain
3 Clinical and Experimental Medicine, University of Pisa, Pisa, Italy
4 Division of Hematology, ASO S.Giovanni Battista, Torino, Italy
5 Division of Hematology, Az Ospedaliera SS Antonio e Biagio e Cesare Arrigo, Alessandria, Italy;
Background: MYD88L265P mutation might represent an ideal marker for minimal residual disease (MRD) monitoring in Walden- ström Macroglobulinemia (WM). However, the conventional allele-specific quantitative PCR (ASqPCR) is not sensitive enough for MRD monitoring on peripheral blood (PB), harboring low concentrations of tumor cells. Besides, cell-free DNA (cfDNA) is increasingly used for mutational studies. We set up a new, highly sensitive, droplet digital PCR (ddPCR) assay for MYD88L265P detec- tion and described: 1) its feasibility for mutation screening and MRD monitoring in bone marrow (BM) and PB; 2) its application for mutational studies on cfDNA.
Methods: BM, PB and plasma from local series of WM, IgG- lymphoplasmacytic lymphoma (LPL) and IgM-MGUS patients (pts) were collected at baseline and during follow-up (FU). 40 healthy subjects were used as negative controls. Genomic DNA (gDNA) and plasmatic cfDNA were extracted by Maxwell RSC system (Promega). MYD88L265P was assessed on gDNA (100 ng) and cfDNA (from 1 ml of plasma) by a custom ddPCR assay on a QX100 System (Bio-Rad). For comparison ASqPCR was assessed on gDNA (100 ng), as described [Xu L, 2013]. MYD88L265P cut-off was settled based on the healthy samples background level. IGH-based MRD analysis was performed as described [Drandi D, 2015].
Results: Sensitivity of ddPCR versus ASqPCR was assessed on a ten-fold serial dilution standard curve. Whereas ASqPCR con- firmed the sensitivity of 1.00E−03, ddPCR reached a sensitivity up to 5.00E−05. Overall, 291 samples from 148 pts, 194 base- line (128 BM, 66 PB) and 97 follow-up (43 BM and 54 PB), were analyzed. 123/128 (96.1%) diagnostic BM and 47/66 (71.2%) PBsamples scored positive for MYD88L265P (BM median 3.6%, range: 0.02-72.6%: PB median 0.3%, range: 0.01-27.8%). 11/46 (24%) pts with both BM/PB collected at diagnosis showed a positive/negative match. Concordances between ddPCR and qPCR methods were investigate on 100 samples (60 BM, 40 PB) and overall a good concordance was observed (p=0.0005). Of note the majority of discordances were observed in the follow-up samples. Moreover, to investigate whether MYD88L265P ddPCR could be used for MRD monitoring we compared it to the gold standard IGH-based MRD assay in baseline and FU samples (23 BM, 15 PB) from 10 pts. The comparison showed highly superimposable results between meth- ods. Finally, ddPCR performed on cfDNA from 60 plasma samples showed 1 log higher median levels of MYD88L265P mutation by mutation in plasma (1.4%, range 0–72.2%) compared to PB (0.1%, range: 0.01–27.8%).
Conclusion: ddPCR is a feasible and highly sensitive assay for mutational screening and MRD monitoring in WM, particularly in samples harboring low concentrations of circulating tumor cells. Moreover, plasmatic cfDNA represents a promising tissue source and might be an attractive, less invasive alternative to PB or BM for MYD88L265P detection.