The sarcoma proteomeLeiomyosarcoma proteomics-based analysisThe leiomyosarcoma MS-based proteomeDifferential protein abundanceGastrointestinal stromal tumor proteomics-based analysisThe gastrointestinal stromal tumor MS-based proteomeDifferential protein abundanceAdditional informationRelevant links and publications

Cancer methods

The sarcoma proteome

Sarcomas are a diverse group of malignant tumors that arise from mesenchymal tissues, including connective tissue, muscle, fat, blood vessels, and bone. Unlike carcinomas, which originate from epithelial cells and account for the majority of cancers, sarcomas are relatively rare and comprise numerous histological subtypes with distinct molecular characteristics and clinical behavior. Because sarcomas can develop almost anywhere in the body, symptoms vary, and many present as a painless, growing lump. Treatment is based mainly on surgery, sometimes combined with radiotherapy or chemotherapy.

The data available in the Cancer Resource include two types of mesenchymal tumors: leiomyosarcoma and gastrointestinal stromal tumors (GISTs). Leiomyosarcoma is a soft tissue sarcoma originating from smooth muscle cells that can arise in various anatomical locations, including the uterus, abdomen, and blood vessels. GISTs are mesenchymal tumors of the gastrointestinal tract, most commonly occurring in the stomach and small intestine. Most GISTs are driven by mutations in the KIT or PDGFRA genes, and drugs that block these proteins have greatly improved the treatment of advanced disease. Leiomyosarcomas are often aggressive, and the outlook depends mainly on tumor size, grade and whether the tumor can be completely removed.

Molecular characterization of sarcoma

Here, we explore the molecular characteristics of gastrointestinal stromal tumors and leiomyosarcoma using mass spectrometry (MS)-based proteomics data from ProteinTalks-DB. The dataset enables comparisons of protein abundance across cancer types, as well as between tumor and adjacent non-tumor tissue, providing insights into proteins with elevated expression and differential protein abundance in these two tumor types. Transcriptomics data from TCGA and MS-based proteomics data from CPTAC are not available for these tumor types.

Leiomyosarcoma proteomics-based analysis

MS-based proteomics data from ProteinTalks-DB were used to classify proteins in leiomyosarcoma according to their cancer specificity and to compare paired tumor and adjacent non-tumor tissues.

The leiomyosarcoma MS-based proteome

MS-based proteomics data from ProteinTalks-DB were available from 24 patients with paired tumor and adjacent non-tumor samples of leiomyosarcoma. The proteomics analysis showed that 6920 of all measured proteins (n=10718) were detected in leiomyosarcoma tumors. Proteins were classified into five categories based on their relative abundance in leiomyosarcoma compared to the other analyzed cancer types, identifying proteins with elevated and cancer-specific detection.

Figure 1. The distribution of all proteins across the specificity categories (left) and distribution categories (right), based on protein abundance in leiomyosarcoma and all other analyzed cancer types.

123 proteins show some level of elevated detection in leiomyosarcoma compared to other cancers (Figure 1). The elevated category is further subdivided into three categories as shown in Table 1.

Table 1. The number of proteins in the subdivided categories of elevated detection in leiomyosarcoma.

Differential protein abundance

Differences in protein abundance between tumor and adjacent non-tumor tissue were analyzed using ProteinTalks-DB. Proteins showing significantly increased or decreased abundance in leiomyosarcoma compared to adjacent non-tumor tissue are illustrated in a volcano plot below.

ProteinTalks DB relative protein data

Proteins that are significantly down- or upregulated in leiomyosarcoma compared to adjacent non-tumor tissue are illustrated in a volcano plot using data-independent acquisition mass spectrometry (DIA-MS) data from ProteinTalks-DB, based on the paired analysis of 24 tumor samples and 24 adjacent non-tumor samples from 24 patients. Proteins detected in at least 10% of the samples in both groups were tested. Proteins detected in fewer than 10% of the samples in one group were not tested, but were classified as mainly detected in tumor or mainly detected in non-tumor tissue if their detection rate was at least 50 percentage points higher in the other group.

In leiomyosarcoma, 554 and 207 proteins are down- (blue) and upregulated (red) compared to adjacent non-tumor tissue, respectively. In Table 2, the 20 most significant genes are listed.

Table 2. The 20 genes with the highest significance associated with a downregulated or upregulated protein abundance in leiomyosarcoma compared to adjacent non-tumor tissue.

Selected proteins with altered abundance in tumors

Two examples of proteins with altered abundance in tumor compared to adjacent non-tumor tissue in ProteinTalks-DB are shown below.

SERPINH1 (serpin family H member 1, also known as HSP47), a collagen-specific chaperone required for collagen production by mesenchymal cells, is detected in all tumor and adjacent non-tumor samples, but its protein levels are significantly higher in tumors, together with other proteins involved in collagen synthesis such as P3H1 and CRTAP.

ALDH1A1 (aldehyde dehydrogenase 1 family member A1), an enzyme involved in the synthesis of retinoic acid, shows the opposite pattern. It is detected in all tumor and adjacent non-tumor samples, but its protein levels are significantly lower in tumors.

Gastrointestinal stromal tumor proteomics-based analysis

MS-based proteomics data from ProteinTalks-DB were used to classify proteins in gastrointestinal stromal tumor according to their cancer specificity and to compare paired tumor and adjacent non-tumor tissues.

The gastrointestinal stromal tumor MS-based proteome

MS-based proteomics data from ProteinTalks-DB were available from 25 patients with paired tumor and adjacent non-tumor samples of gastrointestinal stromal tumor. The proteomics analysis showed that 7580 of all measured proteins (n=10718) were detected in gastrointestinal stromal tumors. Proteins were classified into five categories based on their relative abundance in gastrointestinal stromal tumor compared to the other analyzed cancer types, identifying proteins with elevated and cancer-specific detection.

Figure 3. The distribution of all proteins across the specificity categories (left) and distribution categories (right), based on protein abundance in gastrointestinal stromal tumor and all other analyzed cancer types.

460 proteins show some level of elevated detection in gastrointestinal stromal tumor compared to other cancers (Figure 3). The elevated category is further subdivided into three categories as shown in Table 3.

Table 3. The number of proteins in the subdivided categories of elevated detection in gastrointestinal stromal tumor.

Differential protein abundance

Differences in protein abundance between tumor and adjacent non-tumor tissue were analyzed using ProteinTalks-DB. Proteins showing significantly increased or decreased abundance in gastrointestinal stromal tumor compared to adjacent non-tumor tissue are illustrated in a volcano plot below.

ProteinTalks DB relative protein data

Proteins that are significantly down- or upregulated in gastrointestinal stromal tumor compared to adjacent non-tumor tissue are illustrated in a volcano plot using data-independent acquisition mass spectrometry (DIA-MS) data from ProteinTalks-DB, based on the paired analysis of 25 tumor samples and 25 adjacent non-tumor samples from 25 patients. Proteins detected in at least 10% of the samples in both groups were tested. Proteins detected in fewer than 10% of the samples in one group were not tested, but were classified as mainly detected in tumor or mainly detected in non-tumor tissue if their detection rate was at least 50 percentage points higher in the other group.

In gastrointestinal stromal tumor, 1991 and 668 proteins are down- (blue) and upregulated (red) compared to adjacent non-tumor tissue, respectively. In Table 4, the 20 most significant genes are listed.

Table 4. The 20 genes with the highest significance associated with a downregulated or upregulated protein abundance in gastrointestinal stromal tumor compared to adjacent non-tumor tissue.

Selected proteins with altered abundance in tumors

Two examples of proteins with altered abundance in tumor compared to adjacent non-tumor tissue in ProteinTalks-DB are shown below.

KIT (KIT proto-oncogene, receptor tyrosine kinase), the receptor tyrosine kinase that is activated by mutation in most gastrointestinal stromal tumors and the main target of their treatment with tyrosine kinase inhibitors, is detected in all tumor samples and in most adjacent non-tumor samples, but its protein levels are markedly higher in tumors. ANO1 (anoctamin 1, also known as DOG1), another diagnostic marker of these tumors, shows a similar pattern.

SORBS1 (sorbin and SH3 domain containing 1), an adaptor protein abundant in smooth muscle and adipose tissue, shows the opposite pattern. It is detected in nearly all tumor and adjacent non-tumor samples, but its protein levels are significantly lower in tumors.

Additional information

Histologically, leiomyosarcomas consist of intersecting bundles of spindle-shaped tumor cells with eosinophilic cytoplasm and elongated, blunt-ended nuclei, with variable nuclear atypia, mitotic activity and necrosis. Immunohistochemically, the tumor cells are typically positive for smooth muscle markers such as smooth muscle actin, desmin and h-caldesmon. Uterine leiomyosarcomas are distinguished from the much more common benign leiomyomas mainly by marked nuclear atypia, high mitotic activity and tumor cell necrosis.

Gastrointestinal stromal tumors are composed of spindle cells, epithelioid cells or a mixture of both. Almost all GISTs are positive for KIT (CD117) and DOG1 (ANO1) by immunohistochemistry, which distinguishes them from other mesenchymal tumors of the gastrointestinal tract, such as leiomyomas and schwannomas. The risk of recurrence is assessed based on tumor size, mitotic count and location, and tumors in the stomach generally behave less aggressively than those in the small intestine. Localized GISTs are treated with surgery, while patients with a high risk of recurrence or with advanced disease are treated with tyrosine kinase inhibitors such as imatinib, whose effect depends on the type of KIT or PDGFRA mutation.

Sarcomas are classified according to the WHO classification of soft tissue and bone tumors and are graded based on differentiation, mitotic count and necrosis, for example using the FNCLCC (French Federation of Cancer Centers) grading system.

Relevant links and publications

Yue L et al., Spatial distribution of the proteome in the human body and in cancers. Nature. (2026)
PubMed: 42310461 DOI: 10.1038/s41586-026-10660-y

Uhlén M et al., Tissue-based map of the human proteome. Science (2015)
PubMed: 25613900 DOI: 10.1126/science.1260419

Sbaraglia M et al., The 2020 WHO Classification of Soft Tissue Tumours: news and perspectives. Pathologica. (2021)
PubMed: 33179614 DOI: 10.32074/1591-951X-213

Blay JY et al., Gastrointestinal stromal tumours. Nat Rev Dis Primers. (2021)
PubMed: 33737510 DOI: 10.1038/s41572-021-00254-5