Inferring Clonal Composition from Multiple Sections of a Breast Cancer

Abstract

Cancers arise from successive rounds of mutation and selection, generating clonal populations that vary in size, mutational content and drug responsiveness. Ascertaining the clonal composition of a tumor is therefore important both for prognosis and therapy. Mutation counts and frequencies resulting from next-generation sequencing (NGS) potentially reflect a tumor's clonal composition; however, deconvolving NGS data to infer a tumor's clonal structure presents a major challenge. We propose a generative model for NGS data derived from multiple subsections of a single tumor, and we describe an expectation-maximization procedure for estimating the clonal genotypes and relative frequencies using this model. We demonstrate, via simulation, the validity of the approach, and then use our algorithm to assess the clonal composition of a primary breast cancer and associated metastatic lymph node. After dividing the tumor into subsections, we perform exome sequencing for each subsection to assess mutational content, followed by deep sequencing to precisely count normal and variant alleles within each subsection. By quantifying the frequencies of 17 somatic variants, we demonstrate that our algorithm predicts clonal relationships that are both phylogenetically and spatially plausible. Applying this method to larger numbers of tumors should cast light on the clonal evolution of cancers in space and time.

Assessing intratumor heterogeneity and tracking longitudinal and spatial clonal evolutionary history by next-generation sequencing

“…LICHeE (23) and SCHISM (24) take VAFs of SNAs as input and construct a phylogenetic tree via an acyclic directed graph. Clomial (25), another program designed exclusively for SNAs, performs mixture deconvolution assuming that all mutational loci are heterozygous from copy number-neutral regions. Clomial decomposes the VAF matrix into a product of sample proportions and population genotypes, and uses expectation maximization (EM) to estimate both matrices.…”

“…If all SNAs are heterozygous from copy number neutral regions, as assumed by SciClone (16) and Clomial (25), then VAF = 1=2 × MCF = 1=2 × CCF × ϕ C , where ϕ C is the cancer cell purity, MCF is the fraction of cells that have the SNA, and CCF is the fraction of cancer cell that have the mutation. Pyclone (15), PhyloSub (33), and EXPANDS (14) account for CNAs but make the assumption that was first introduced by ABSOLUTE (13), namely, that there are no subclonal CNA events.…”

“…S6). As a comparison, Clomial (25), which ignores the existence of CNAs, fails to correctly estimate the clonal frequencies and infers incorrect tumor purities (SI Appendix, Figs. S5B and S6B).…”

“…LICHeE (23) and SCHISM (24) infer phylogeny from VAFs as an acyclic directed graph network. Another recent work, Clomial (25), showed that it is possible to obtain precise and informative estimates of the underlying subpopulations through a matrix deconvolution framework. One practical drawback of Clomial (25) is that it takes only SNA input and assumes that all mutational loci are heterozygous from copy numberneutral regions.…”

“…Another recent work, Clomial (25), showed that it is possible to obtain precise and informative estimates of the underlying subpopulations through a matrix deconvolution framework. One practical drawback of Clomial (25) is that it takes only SNA input and assumes that all mutational loci are heterozygous from copy numberneutral regions. SCHISM (24), BitPhylogeny (26), PhyloWGS (20), and SPRUCE (27) adjust for CNAs in their model in different ways, but these methods still require limiting assumptions and do not make full use of the data, as we discuss in detail a bit later.…”