Intratumoural heterogeneity (ITH) underlies many of the challenges we face in cancer medicine, including therapy-resistance, disease progression/evolution and relapse after seemingly effective therapy. Distinct tumour cell subpopulations selectively evade therapy and drive disease-progression and technologies that reveal key aspects of ITH are therefore critical for the application of precision cancer medicine. Although bulk genomic analysis has without question provided many insights into genetic ITH, this approach faces a number of fundamental limitations: ITH in cancer occurs at many levels, not restricted to genetics (mutations) but also other factors, such as presence of cancer stem cells in some tumours. Furthermore, bulk genomic analysis reveals patterns of somatic mutations, but not their molecular consequences within distinct (and therapy-resistant) cancer subclones.
Whilst many of the scientific questions relating to ITH have remained the same over many decades, our ability to address these questions has advanced dramatically not least because of advances in technology. Ultimately, as the unit of evolution and clonal selection by therapy in cancer is the cell, techniques that resolve heterogeneity at the single-cell level are ideally placed to unravel ITH and provide entirely new insights into cancer biology, with enormous potential to accelerate the development of new approaches to improve outcomes for patients. However, the lack of coverage across key mutation hotspots when studying cancers using single-cell RNA-sequencing techniques has precluded the correlation of genetic and transcriptional readouts from the same single cell, limiting their application to the study of tumors.
To overcome such limitation, we developed TARGET-seq, a single cell multi-omic method for the high-sensitivity detection of mutations within single cells in parallel with whole transcriptome analysis. TARGET-seq achieved extremely low allelic dropout rates, allowing resolution of clonal hierarchies with over 98% accuracy, while obtaining unbiased high quality transcriptomes from the same single cell.
We have applied TARGET-seq to the study of over ten thousand haematopoietic stem and progenitor cells (HSPCs) from JAK2-mutant myeloproliferative neoplasms. This analysis revealed a high degree of genetic heterogeneity, identifying both linear and branching patterns of clonal evolution. At the transcriptome level different genetic subclones showed distinct transcriptional signatures, indicating that each of them was molecularly distinct. Wild-type cells from MPN patients also showed disrupted gene expression as compared to cells from normal donors, upregulating molecular pathways associated with inflammation (TNFα, TGFβ and IFN signalling). This suggests cell-extrinsic effects disrupting gene expression in non-mutant cells, which has been shown to have prognostic significance and might underlie therapy response. Moreover, TARGET-seq analysis allowed us to identify putative biomarkers of JAK2V617F mutant cells, including novel therapeutic targets to selectively eradicate JAK2-mutant cells and importantly, potential candidates for antibody-based immunotherapy.
Analysis of samples from MPN patients undergoing disease transformation to Acute Myeloid Leukemia (sAML) revealed striking patterns of clonal evolution in different immunophenotypically-defined cell types. We identified pre-leukemic and leukemic subclones emerging from hematopoietic stem cells rather than more mature progenitors, in contrast to evolution patterns in de novo AML, which might indicate different cancer stem cell reservoirs.
In summary, TARGET-seq allowed us to identify distinct and biologically relevant molecular signatures of different genetic subclones of HSPCs in myeloproliferative neoplasms. TARGET-seq could also be broadly applied to the study of other types of tumours, providing a powerful tool for biomarker and therapeutic target discovery for precision medicine.
Disclosures
Mead: Bristol Myers-Squibb: Consultancy; Pfizer: Consultancy; Novartis: Consultancy, Honoraria, Other: Travel/accommodation expenses, Research Funding, Speakers Bureau; CTI: Honoraria, Research Funding; Celgene: Consultancy, Research Funding.
Bacterial single-cell genomics enables phylogenetic analysis and reveals population structures from in vitro evolutionary studies