Predictive multigenic scale. Analysis of own results in metastatic breast cancer

Background. Breast cancer is one of the most common female malignancies. Molecular diagnostic methods of tumor profiling allow us to analyze individual tumor characteristics, identify new prognostic and predictive markers. Aim. To increase the efficacy of systemic therapy for breast cancer and reduce inappropriate prescriptions using the data on individual molecular tumor characteristics; to develop a polygenic panel to ensure a tailored approach to systemic therapy for breast cancer. Materials and methods. We analyzed 84 tumor tissue samples from pre- and postmenopausal women with metastatic breast cancer who were treated and followed-up in 6 healthcare institutions. We assessed expression of genes involved in breast cancer. In a pilot study, we analyzed archived paraffin-embedded tumor specimens form 12 out of 1,216 patients with T1–2N0M0 breast cancer included into retrospective analysis. Gene expression was assessed using the nCounter technology based on direct digital detection of targets using fluorescent barcodes (nCounter Analysis System; NanoString Technologies, USA). Tumor tissue (biopsy and surgical specimens) was analyzed. The choice of genes was based on the literature data and experience in the development of other polygenic panels, as well as clinical significance of markers of prognostic scales. Gene mutations were confirmed by next generation sequencing and reverse transcription-polymerase chain reaction. Results. We analyzed the expression of 28 genes with a high predictive value that have been substantially studied (including ESR1, PGR, PIK3CA, BCAR4, BCAS2, CCND1, CCND2, CCND3, FOXA1, Erb2, EGFR, CDH3, FOXC1, KRT14, KRT5, CD274, CDK4, CDK6, P53, PTEN, BRCA1, BRCA2, CHEK2, CLDN3, CLDN7, AR, TOP2a, TUBBIII). We identified 29 cases of discrepancy (29 / 84; 34.5 %) in tumor subtype, including 11 cases of luminal A and B breast cancer, which might potentially affect the choice of the treatment regimen. In 18 cases, there were some principal discrepancies in the tumor subtype that implied totally different treatment regimens. The proposed polygenic signature allows accurate identification of the tumor subtype in patients with metastatic breast cancer and choice of an optimal treatment strategy. Conclusion. We have developed a 100-gene signature including molecular subtypes of breast cancer (luminal A, luminal B, basal, claudin-like) and treatment-oriented clusters. Molecular tumor profiling using this polygenic signature is an accurate method for determining tumor subtype in patients with breast cancer, which enables a tailored approach to therapy.