Clonal Tracking By Whole Genome Sequencing Permits Comprehensive Mapping of the Genomic Landscape in Pre- and Post-Gene Therapy Sickle Cell Patients

Recent advances in clonal stem cell tracking strategies have enabled interrogation of unperturbed human hematopoiesis. Whole genome sequencing (WGS) can be used to map the clonal dynamics of hematopoietic stem and progenitor cells (HSPCs) by employing spontaneous somatic mutations as unique clonal tags (Lee-Six et al., Nature, 2018). These tags allow for retrospective analysis of individual stem cell clones and the construction of phylogenetic trees mapping out stem cell relatedness, with mutations being acquired in a near-linear fashion over the course of an individual's life. The unprecedented level of information obtained in these studies is particularly well-suited to understanding genomic changes in gene therapy trials aimed at curing diseases such as sickle cell disease (SCD). In addition to mapping relatedness between stem cells, sequencing data can be used to better define mutational signatures for HSPC clones that have been successfully gene-modified as well as those that lack an integrated copy of the therapeutic vector. Given this method's ability to identify low frequency mutations in individual HSPC clones, mutations with extremely low variant allele frequencies can be detected much more readily than through traditional bulk sequencing approaches, something that is particularly relevant given recent safety concerns in some SCD gene therapy trials. In this study, we have mapped the clonal dynamics of HSPCs obtained from pre- and post-gene therapy samples from 4 SCD patients who have undergone autologous gene therapy performed using a BCL11A shmiR lentivirus vector (NCT 03282656, 12-36 months follow-up). HSPCs from mobilized peripheral blood (pre-gene therapy), bone marrow aspirates (both pre- and post-gene therapy) or unmobilized peripheral blood (post-gene therapy) were expanded as single clones and 1508 individual colonies were then sequenced using WGS to an average sequencing depth of 12.3x. Initial results indicate that the mean mutation burden per cell in a pre-gene therapy sample is elevated for some patients compared to what would be expected based on patient age in similar studies. In pre-gene therapy samples, the structure of the phylogenetic trees appeared to be highly polyclonal, indicating that there were no significant clonal expansion events prior to gene therapy. In one patient where we undertook extensive profiling, approximately 15-20 excess mutations per HSPC were observed across the entire genome 24 months after transplantation, presumably acquired as a consequence of gene therapy and/or reconstitution post-transplantation, which is equivalent to approximately one year of normal ageing without a transplantation intervention. However, no clonal expansions or driver mutations were identified at this 24 month follow-up timepoint, suggesting that no strong selective advantage or pre-leukemic events were present prior to or following the gene therapy protocol. Extending this approach to a wider range and larger number of patients will allow for comprehensive mapping of the genomic landscape and clonal evolution of stem cells in sickle cell patients and will also set the stage for improved assessment of safety and potential leukemia-initiating events in the context of gene therapy.