NEW YORK – In a single-cell analysis of myeloid malignancies, a team led by researchers at Memorial Sloan Kettering Cancer Center has found that acute myeloid leukemia (AML) is dominated by a small number of clones, which frequently harbor co-occurring mutations in epigenetic regulators. Conversely, mutations in signaling genes often occur more than once in distinct subclones, which is consistent with increasing clonal diversity.
In a study published on Wednesday in Nature, the researchers noted that AML and other myeloid cancers arise from the expansion of hematopoietic stem and progenitor cells that acquire somatic mutations. Bulk molecular profiling has suggested that mutations are acquired in a stepwise fashion, with mutant genes with high variant allele frequencies appearing early in leukemogenesis, and mutations with lower variant allele frequencies being acquired later.
Although bulk sequencing can provide information about leukemia biology and prognosis, it can't distinguish which mutations occur in the same clones, accurately measure clonal complexity, or definitively elucidate the order of mutations, the researcher said. In order to delineate the clonal framework, they performed single-cell mutational profiling on 146 samples from 123 AML patients, mapping clonal trajectories for each sample and observing mutations that combined to promote clonal expansion and dominance.
The investigators also combined protein expression with mutational analysis to map somatic genotype and clonal architecture with immunophenotype. Overall, they said, these findings provided insights into the pathogenesis of myeloid transformation and how clonal complexity evolves with disease progression.
In a similar study published last week in Nature Communications, researchers from the University of Texas MD Anderson Cancer Center explored the clonal diversity of acute myeloid leukemia using single-cell sequencing, detecting driver mutations that tended to co-occur as well as mutations that were mutually exclusive, and teasing out the evolutionary history of some of those mutations.
For their study, the MSKCC researchers and their colleagues used a custom amplicon panel from Mission Bio, covering 31 frequently mutated genes, to perform single-cell sequencing. They sequenced 740,529 cells from 146 samples from patients at diagnosis and/or relapse. The most common mutations they identified were in DNMT3A, TET2, NPM1, and FLT3.
They next investigated disease subtypes, subdividing the cases into samples with epigenetic mutations, samples with signaling mutations, samples without epigenetic mutations, and samples with epigenetic and co-mutated signaling effectors. The number of mutations per sample was significantly higher in AML than in myeloproliferative neoplasms (MPN), and in MPN than in clonal hematopoiesis (CH), they found. The increase in mutations per sample was more pronounced in cases of AML with signaling effector mutations, specifically those in RAS and FLT3.
When they explored clonal repertoire, with clones defined as cells with identical protein-encoding single-nucleotide variants, the researchers observed a significant increase in clone number in samples from patients with AML compared to MPN or CH, with the highest number of clones in FLT3-mutant AML samples. They also assessed the diversity of clone size on a per-sample basis and observed a significant increase from CH or MPN to AML. Clonal diversity was higher in samples with mutations in RAS and FLT3 than in samples from patients with CH or MPN or in AML samples without mutations in RAS or FLT3.
The investigators were then able to identify gene-specific contributions to clonal expansion, finding that IDH2, NPM1, and JAK2 mutations were nearly always present in the dominant clone, whereas FLT3 and RAS mutations were present only in minor subclones in some patients, and in dominant clones in others. Of the 80 AML samples with epigenetic mutations, nearly 53 percent harbored mutations in more than one epigenetic modifier. In nearly all cases, epigenetic regulator mutations were in the same clone, and in 81 percent of cases the co-occurring mutations were within the dominant clone, suggesting cooperativity between epigenetic mutations.
When the researchers focused on six patients who harbored mutations in both DNMT3A and IDH1/2, and had concurrent signaling effector mutations, they found that a high fraction of cells had concurrent mutations in DNMT3A and IDH1/2 but that few clones possessed more than one mutation in a signaling effector. This suggested that the presence of additional mutations with epigenetic modifiers may influence subsequent mutational trajectories.
The researchers also sought to determine whether clonal architecture was altered during disease transformation and response to therapy. In four out of six patients in whom the disease transformed from MPN to AML, they observed a significant alteration in clonal architecture, with emergence of new dominant clones. Pre- and post-therapy samples from three patients bearing FLT3 mutations who were treated with the FLT3 inhibitor gilteritinib (Astellas Pharma's Xospata) showed that all three patients had a decrease in FLT3-mutant clones in response to gilteritinib, with significant clonal architecture changes. Two of the three patients had an outgrowth of clones with RAS mutations, which has previously been described as a potential resistance mechanism to FLT3 inhibitor therapy, often with RAS mutations acquired in the FLT3 mutant clone. The overall results indicated that transformation and therapeutic perturbations can alter clonal architecture in both a linear and branched manner.
"These data suggested that myeloid malignancies manifest as a complex ecosystem of clones that evolves over time, and that scDNA-seq gives a glimpse into this milieu that is not seen with conventional bulk sequencing," the authors wrote. "Our studies of clonal architecture at a single-cell level give us insights into how clonal complexity contributes to the pathogenesis of myeloid transformation. Similar studies across different pre-malignant and malignant contexts will give new information about how malignancies initiate and progress and will lead to new therapeutic strategies."