SciClone: inferring clonal architecture and tracking the spatial and temporal patterns of tumor evolution

Abstract

The sensitivity of massively-parallel sequencing has confirmed that most cancers are oligoclonal, with subpopulations of neoplastic cells harboring distinct mutations. A fine resolution view of this clonal architecture provides insight into tumor heterogeneity, evolution, and treatment response, all of which may have clinical implications. Single tumor analysis already contributes to understanding these phenomena. However, cryptic subclones are frequently revealed by additional patient samples (e.g., collected at relapse or following treatment), indicating that accurately characterizing a tumor requires analyzing multiple samples from the same patient. To address this need, we present SciClone, a computational method that identifies the number and genetic composition of subclones by analyzing the variant allele frequencies of somatic mutations. We use it to detect subclones in acute myeloid leukemia and breast cancer samples that, though present at disease onset, are not evident from a single primary tumor sample. By doing so, we can track tumor evolution and identify the spatial origins of cells resisting therapy.

Figure 1. Inferring subclonal architecture objectively in multiple myeloma.

(a) Kernel density plots of VAFs across regions with copy number one, two, or three, posterior predictive densities summed over all clusters for copy number neutral variants, and posterior predictive densities for each cluster/component. (b-d) VAFs plotted versus read depth for each of the three copy number regions. (c) Three mutation clusters (green, dark orange, and blue) were detected using variants from copy number neutral segments. (d) Two clusters centered at VAF 31% and 62% were detected from variants in copy number three segments; they likely result from single-copy amplification of the wild-type or the mutant allele of mutations in the founding clone.

Figure 2. Overcoming uncertainty in sparse exome-sequencing data to determine clonal structure and mutation clonality.

(a) Breast cancer sample with well-defined clones. (b) Endometrial cancer sample with overlapping clusters. PIK3CA mutations are strongly associated with the dominant clone (posterior probabilities 93%), whereas the clonal context of an ATM mutation is more ambiguous (57.8%).

Figure 3. Refining subclonal architecture from longitudinal analysis of tumor/relapse pair in acute myeloid leukemia (AML).

Two-dimensional analysis of tumor/relapse sample (a) dissects clusters one and four, which overlap in the relapse sample (b), and one, two, and three, which overlap in the tumor sample (c). Single-sample analyses (b and c) show histogram (rectangles) with posterior predictive densities. Several genes recurrently mutated in AML are highlighted. (d) Inferred schematic of clonal evolution from a single hematopoietic stem cell, showing percentage of cells belonging to each clone (i.e., twice VAF for this nearly pure sample). Broken vertical white lines correspond to primary tumor sample (before chemotherapy) and relapse subsequent to treatment.

Figure 4. Determining stability of inferred subclones as a function (a) of number of variants, (b) of inter-cluster separation, and (c) of clustering method from AML sample.

(a) A fraction of the ∼800 variants from Fig. 3 were randomly sampled and the resulting number of clusters was inferred using beta mixture modeling. Error bars represent standard deviation (). (b) Mutations from clusters one and two from the AML relapse sample were used to assess the limits of cluster separability. As the distance between the two mutation groups was varied, the resulting clusters were assessed for overlap (the fraction of the data within a single standard deviation of both clusters) and accuracy (the fraction of items that were correctly assigned to a second cluster). (c) Consensus clustering of the AML data set (Fig. 3) for number of initial clusters varied from six to 15 and clustering method varied across beta, Gaussian, and binomial mixture models for a total of 30 runs. consensus matrix holds all variants across both rows and columns and has been reordered so that variants belonging to the same cluster are adjacent to one another. Matrix entry , is the fraction of runs in which variant and were co-clustered; entry corresponds to the top-left of the matrix heat map. The narrowest neutral-colored band corresponds to a single variant alternatively classified by Gaussian mixture modeling (Fig. S2a). The larger neutral-colored band corresponds to variants alternatively classified as a sixth cluster by binomial mixture modeling (Fig. S2b).

Figure 5. Assessing intratumor spatial heterogeneity and treatment response with multiple biopsies.

Three breast tumor samples from a single individual were simultaneously analyzed: two spatially distinct samples from a primary tumor and one sample taken after aromatase-inhibitor treatment. (a–c) Two-dimensional slices and (d) still frame of the full three-dimensional interactive plot.