Clinical sequencing of soft tissue and bone sarcomas delineates diverse genomic landscapes and potential therapeutic targets

Abstract

The genetic, biologic, and clinical heterogeneity of sarcomas poses a challenge for the identification of therapeutic targets, clinical research, and advancing patient care. Because there are > 100 sarcoma subtypes, in-depth genetic studies have focused on one or a few subtypes. Herein, we report a comparative genetic analysis of 2,138 sarcomas representing 45 pathological entities. This cohort is prospectively analyzed using targeted sequencing to characterize subtype-specific somatic alterations in targetable pathways, rates of whole genome doubling, mutational signatures, and subtype-agnostic genomic clusters. The most common alterations are in cell cycle control and TP53, receptor tyrosine kinases/PI3K/RAS, and epigenetic regulators. Subtype-specific associations include TERT amplification in intimal sarcoma and SWI/SNF alterations in uterine adenosarcoma. Tumor mutational burden, while low compared to other cancers, varies between and within subtypes. This resource will improve sarcoma models, motivate studies of subtype-specific alterations, and inform investigations of genetic factors and their correlations with treatment response.

Fig. 1. Analysis of 2138 sarcoma samples reveals variation in patient characteristics among subtypes.

This analysis includes 2138 bone and soft tissue sarcoma samples, each from distinct patients. Subtypes with ≥20 samples in the dataset are displayed. A Distribution of number of samples, survival from time of sequencing, sample type (primary or metastatic site), tumor site, sample purity, age, sex, and self-reported race in each subtype. Retro/IA, retroperitoneal or intrabdominal. NA, not applicable. In the age plot, box boundaries indicate 25th and 75th percentiles, interior lines medians, and whiskers 1.5 times the interquartile range. B Overall distribution of sample number for the entire cohort. C 1, 3, and 5-year survival from the time of sequencing. *, 5-year survival = 0. Vertical lines indicate 95% confidence intervals. DES desmoid tumor, ESS endometrial stromal sarcoma, INT intimal sarcoma, LGFMS low-grade fibromyxoid sarcoma, EMCHS extraskeletal myxoid chondrosarcoma, HGESS high-grade endometrial stromal sarcoma, LGESS low-grade endometrial stromal sarcoma, SEF sclerosing epithelioid fibrosarcoma, ASPS alveolar soft part sarcoma, DDCHS dedifferentiated chondrosarcoma, UAS uterine adenosarcoma, SCSRMS spindle cell/sclerosing rhabdomyosarcoma, CCS clear cell sarcoma, EHAE epithelioid hemangioendothelioma.

Fig. 2. Mutation analysis by subtype.

A Alteration type and frequency, fraction of genome altered (FGA) and tumor mutation burden (TMB) by subtype. Oncogenic fusions detected by MSK-IMPACT are classified as drivers. In the FGA and TMB plots, box boundaries indicate 25th and 75th percentiles, interior lines medians, and whiskers 1.5 times the interquartile range. VUS variant of unknown significance. B Significant mutations were identified in all subtypes with n ≥ 20 in our dataset using both MutSig and MuSiC analysis. Percentages indicate the percentage of samples with an oncogenic mutation in the corresponding gene. C FLT4 mutation type, frequency, and location in ANGS vs. other subtypes. D Cancer cell fraction (CCF) and number of mutations for driver mutations and VUS by subtype. Circles indicate medians and vertical lines interquartile ranges.

Fig. 3. Copy number changes by subtype.

Copy number alteration (CNA) and whole-genome doubling events (WGD) compared across subtypes. A Individual sample CNA across the genome for each subtype. WGD, fraction genome altered (FGA), and purity are shown at right. B Aggregate arm-level (left) and gene-level events (right) grouped by subtype. *significant change based on Bonferroni corrected p-values. Significance was evaluated by random permutations testing. Oncogenic (bold) vs. non-oncogenic CNA classifications according to OncoKB. C Frequency of WGD by subtype (green) compared to other cancers with ≥200 samples available for comparison (gray). NSCLC, non-small cell lung cancer; ST, soft tissue. D, Overall survival based on WGD status within primary and metastatic samples. p.adj, adjusted p-value.

Fig. 4. Integrated pathway analysis.

A Oncogenic alterations within each of 12 pathways with relevance to cancer biology in each subtype. Numbers in each cell indicate percentage of samples harboring alterations. Stacked bar graphs indicate the distribution of the type of oncogenic alteration per gene or pathway (top) or subtype (right). CC, cell cycle; DDR, DNA damage repair; EPI, epigenetic. B PI3K pathway alterations in specific subtypes. The percentage of samples with an alteration in a specific gene in each subtype is indicated in each box. C Oncogenic TERT alterations in each of the 9 most altered subtypes. D Oncogenic epigenetic pathway alterations by subtype, grouped by complex and/or biochemical function of the encoded protein. Totals include all alterations in genes that belong to a parent category, not only those affecting specific complexes listed.

Fig. 5. Mutual exclusivity, co-occurrence, ATRX alterations, and unsupervised clustering based on genetic signatures.

A Co-occurrence and mutual exclusivity of gene- (top) and pathway-level (bottom) alterations in each subtype with significant findings shown. Significance was evaluated by two-sided Fisher’s exact test. EPI, epigenetic; DDR, DNA damage repair. B Frequency and types of oncogenic ATRX alterations in each of the 14 most altered subtypes. C Unsupervised clustering of all samples based on oncogenic alteration patterns. D Subtype-specific cluster associations and entropy scores. For clarity, subtypes with n > 5 are displayed.

Fig. 6. Actionability of mutations by subtype and gene.

For each of the 22 most common subtypes: A Frequency of actionable alterations by level of evidence. B Actionable alterations in individual genes, grouped by pathway. Numbers in each cell represent the percentage of samples with actionable alterations in that gene. C Number of actionable alterations per sample. D Frequency of actionable alterations classified by alteration type.