Clinical and molecular studies of endocrine resistance in breast cancer

• Background: The vast majority of breast cancer patients have tumors that express the predictive biomarker estrogen receptor a (ER/ESR1) and are offered endocrine therapy. However, around one-third of the patients experience a relapse in their disease, and in most, the ER expression is retained. Several mechanisms for endocrine resistance have been suggested, and novel tests and treatments have been introduced in the clinical setting. Mutations in the phosphatidylinositol bisphosphate 3-kinase (PI3K) catalytic subunit encoded by the gene PIK3CA is the most prevalent aberration in ER- positive and human epidermal growth factor receptor 2 (HER2/ERBB2)-negative breast cancer and PI3K inhibitor treatment is utilized in the metastatic setting. The tests for the inhibitor treatment initially involved the 11 known hotspot regions, but extensive analysis of the whole gene is prevailing. Further, ER degraders for ESR1 mutant breast cancer are approved for patients with advanced breast cancer. In the adjuvant and advanced setting, inhibitors of cyclin-dependent kinase 4/6 (CDK4/6) are utilized, however resistance to these therapies are also evident, and the benefit of treatment is greater for high-risk patients. The only clinically utilized marker for endocrine therapy is ER, thus calling for novel tools for therapy prediction. Methods: This thesis investigated a unique cohort of patients with endocrine- resistant and endocrine-sensitive disease. Patients with an ER-positive/HER2- negative primary breast cancer and an ER-positive relapse within five years from primary diagnosis and during ongoing endocrine therapy were defined as resistant. Patients with an ER-positive/HER2-negative primary breast cancer and no relapse or disease progression after at least 10 years of follow-up were defined as endocrine-sensitive. Patient and tumor information were assessed via medical records, and archived glass slides and tumor blocks were collected. From the sensitive patients' primary tumors and the resistant patients' primary and relapse tumors, RNA was extracted from formalin-fixed paraffin-embedded (FFPE) blocks. RNA was hybridized onto Clariom D microarray and analyzed by Transcriptome Analysis Console (TAC) with downstream gene expression analysis with Gene Set Enrichment Analysis (GSEA). From resistant patients' primary and relapse tumors, DNA was extracted and analyzed by a 370 gene panel sequencing. Further, tumor-free lymph nodes were utilized as germline controls. Results: In study I, the PIK3CA mutations of the resistant patients were investigated separately. The analysis showed mutations in half of the patients, and a higher frequency of mutations in the relapse tumors than in the primary tumors. Moreover, of the patients with mutations, around 30% harbored mutations solely outside the known hotspot regions. Investigating the clinicopathological traits, the nodal involvement was significantly associated with the presence of PIK3CA mutations. In study II, the gene expression and clinicopathological assessments of the resistant and sensitive patients were explored. Of the primary tumors, the resistant patients showed significantly higher tumor grades, higher intrinsic subtype risk scores, and an upregulation of cell-cycle gene sets. The gene sets could further be evaluated in an external cohort, supporting the findings of a worse prognosis for the patients harboring a higher expression in the same cell-cycle gene sets. In turn, the resistant patients' relapse tumors showed upregulations of metabolism-associated gene sets and downregulations of estrogen-response gene sets as compared to their primary tumors. In study III, the entire DNA panel sequencing was analyzed, together with the gene expression and clinicopathological assessments. The most frequent alterations included the previously studied PIK3CA gene, and also TP53. Single nucleotide variations (SNVs) and copy number variations (CNVs) were evident in both the patients' primary and relapse tumors. Alterations in BRCA1 and BRCA2 were prevalent in both tumor groups and were associated with a worse overall survival (OS). Several European Society For Medical Oncology (ESMO) Scale for Clinical Actionability of molecular Targets (ESCAT)-listed genes showed a higher total copy number in relapses. The tumor mutational burden of the recurrent tumor was higher in distant metastases as compared to breast/regional recurrences. Moreover, a worse OS was seen for patients who experienced an amplification in their primary tumor in over 60 individual genes, such as CDK6, PARP1, and PIK3CA, for which targetable treatment may be possible. Further, the number of CNVs was significantly associated with the intrinsic subtype of both primary and relapse tumors, demonstrating the highest median number in the basal subtype. CNVs moreover demonstrated significant correlations to the gene expression, corroborating their respective results. Conclusions: The studies of this thesis showcase the clinicopathologic, transcriptomic, and genomic landscape of a unique cohort of endocrine-resistant patients. At primary diagnosis, their respective clinical and molecular profiles differentiated them from endocrine-sensitive patients. Further, the genomic results demonstrate a need for extending existing mutational testing for currently used inhibitor treatment. Several SNVs and CNVs were identified as potential targets for further diagnostic and treatment investigations. Moreover, these results may aid in further explorations of response prediction and help decipher the mechanisms of endocrine resistance. List of scientific papers I. Schagerholm, C., Robertson, S., Toosi, H., Sifakis, EG., Hartman, J. PIK3CA mutations in endocrine-resistant breast cancer. Sci Rep 14, 12542 (2024). https://doi.org/10.1038/s41598-024-62664-1 II. Schagerholm Stanev, C., Sifakis, EG., Hases, L., Chen, X., Williams, C., Robertson, S., Hartman, J. Transcriptomic profiles of endocrine- resistant breast cancer. BMC Cancer 25, 1556 (2025). https://doi.org/10.1186/s12885-025-14826-1 III. Schagerholm Stanev, C., Robertson, S., Toosi, H., Sifakis, EG., Lagergren, J., Hartman, J. Genomic characterization of ER-positive primary tumors and corresponding relapses reveals targetable alterations.[Submitted]

Publikations- och innehållstyp

vet (ämneskategori)

dok (ämneskategori)