APPLICATION OF A NEW MOLECULAR APPROACH BASED ON NEXT GENERATION SEQUENCING TO MONITOR THE CHIMERISM POST ALLOGENEIC HSCT

After the hematopoietic stem cell transplant, the chimerism analysis is routinely evaluated to monitor donor immune system engraftment. Today, the most common genetic diagnostic method for chimerism is the low sensitivity STR analysis by capillary electrophoresis. Recent studies have shown the application of the chimerism analysis in the risk of relapse prediction when high sensitivity technique is used and in the HLA-loss relapse recognition when genetic markers outside and inside the HLA locus are evaluated. As first objective of the PhD, a new Next Generation Sequencing (NGS) approach to detect the chimerism was developed to meet high informativity and sensitivity. Based on the technology of Ion AmpliSeq, a 44-amplicon custom chimerism SNP panel was designed, and a custom bioinformatics pipeline dedicated to the genotyping and quantification of NGS data was coded. The custom SNP panel informativity was high: 16 informative recipient alleles in unrelated pairs and 9 in sibling ones. The method sensitivity was fixed at 1% due to the NGS background (<1%). The protocol followed the standard Ion AmpliSeq library preparation and Ion Torrent Personal Genome Machine guidelines. The NGS method sensitivity tested on several artificial chimeras and on 20 clinical samples, resulted higher than the STRs one. The NGS method passed the UKNEQAS external quality program validation, resulting in suitable for diagnostic use. As second PhD objective, an upgrade of the NGS method, to decrease the NGS background with a view to reach the 0,1% of sensitivity, was proposed. Thirty-two haplotype blocks were selected, 20 blocks were located inside and 12 outside the HLA locus. The previous custom bioinformatics pipeline was implemented to recognize the NGS background. A cheaper custom NGS library protocol based on 100 ng of input DNA, was developed and successfully tested on four DNA samples. The panel informativity was higher for both outside and inside the HLA blocks. The NGS background reached was ten times lower than (0,01%) the desired sensitivity (0,1%). However, a full work on artificial chimera will be done to experimentally measure the sensitivity. Since alignment errors in some HLA blocks were found, the bioinformatics tools will be improved. Moreover, the method performance will be validated on clinical patient samples. After further investigations, our NGS haplotype block method could promptly recognize the graft rejection, the risk of relapse and the HLA-loss relapse together.