Progestins (synthetic progesterone compounds) are the primary hormonal treatment for endometrial cancer and endometrial hyperplasia, particularly important for young women who wish to preserve fertility. When a tumor responds to progestin therapy, it can sometimes be treated without hysterectomy. However, a significant proportion of tumors develop progesterone resistance, defined as loss of response to progestin treatment over time.
The molecular mechanisms underlying progesterone resistance in endometrial cancer are incompletely understood. The progesterone receptor (PGR) is the primary mediator of progestin action, but resistance can develop through reduced PGR expression, altered PGR signaling, or activation of compensatory growth pathways that override the tumor-suppressive effects of progesterone. Understanding which genes change when resistance develops is the first step toward identifying potential therapeutic targets.
This study developed a cell line model of acquired progesterone resistance by repeatedly exposing the Ishikawa endometrial cancer cell line to medroxyprogesterone acetate (MPA) over time until the cells no longer responded. The resulting IshikawaPR cell line was used as a controlled biological system to identify gene expression changes associated with the transition from progesterone-sensitive to progesterone-resistant disease.
Gene expression differences between the progesterone-resistant IshikawaPR cells and parental progesterone-sensitive Ishikawa cells were measured using microarray technology, which simultaneously measures the expression of thousands of genes across both cell lines. Differentially expressed genes (DEGs) were identified using standard statistical thresholds for fold-change and statistical significance.
The analysis identified 821 DEGs: 453 upregulated and 368 downregulated in the resistant cells. To identify which of these DEGs were most likely to be functionally important, the researchers used protein-protein interaction (PPI) network analysis, which maps the known physical and functional interactions between gene products (proteins) and identifies hub genes - those with many connections whose disruption would have broad network effects.
A secondary analysis focused specifically on genes co-expressed with PGR (the progesterone receptor), since these are most likely to be directly relevant to progesterone-mediated transcriptional control. Additionally, Gene Ontology (GO) and KEGG pathway enrichment analyses were performed to identify biological processes and molecular pathways systematically altered in resistant cells.
PPI network analysis identified 8 hub genes - the most highly connected nodes in the interaction network among DEGs: HSPA1A, EEF1A2, AR, POU5F1, C3, SYK, LPAR1, and NMU. Hub genes are considered functionally central because their proteins interact with many other proteins; changes in hub gene expression are likely to have cascading effects throughout the network rather than isolated effects.
Among the hub genes, AR (androgen receptor) is particularly notable in the context of hormonal resistance. Androgen signaling through AR can directly antagonize progesterone receptor function and has been implicated in progesterone resistance in other reproductive cancers. LPAR1 (lysophosphatidic acid receptor 1) is a lipid signaling receptor associated with cell invasion and survival that represents a potential therapeutic target.
Seven PGR co-expressed genes were identified: ANO1, SOX17, CGNL1, DACH1, RUNDC3B, SH3YL1, and CRISPLD1. These genes normally change their expression coordinately with PGR in response to progesterone signaling. Their dysregulation in resistant cells suggests that the progesterone transcriptional program is disrupted beyond just loss of PGR expression itself.
Pathway enrichment analysis revealed three major categories of biological change in progesterone-resistant cells: alterations in lipid metabolism, changes in immune and inflammatory response, and remodeling of the extracellular environment. Each of these categories has known connections to cancer progression and therapy resistance.
Lipid metabolism changes are consistent with the altered energy requirements of cells that have developed resistance mechanisms, and with the identification of LPAR1 as a hub gene - lipid signaling is clearly a recurrent theme. In cancers, lipid metabolism reprogramming supports membrane synthesis for rapid cell division and modulates cell survival signaling pathways.
Immune and inflammatory gene expression changes suggest that progesterone-resistant cells may interact differently with the surrounding immune environment than sensitive cells. Progesterone normally has immunomodulatory effects in reproductive tissues, and loss of progesterone sensitivity may alter the inflammatory microenvironment in ways that favor tumor persistence. The complement component C3 identified as a hub gene is directly relevant to innate immune activation.
Several identified hub genes represent druggable targets with existing compounds. SYK (spleen tyrosine kinase) has inhibitors already approved or in clinical trials for hematological malignancies, and its presence in the PPI network of progesterone-resistant endometrial cancer genes suggests it may be worth investigating as a combination therapy partner with progestins.
AR as a hub gene opens the possibility that anti-androgen therapies, already established in prostate cancer, might have utility in endometrial cancer - particularly in cases where progesterone resistance has developed through an androgen receptor-mediated mechanism. Clinical trials of AR inhibitors in endometrial cancer are conceptually supported by this finding.
The identification of a coordinated network of resistance-associated genes, rather than a single pathway or gene, is biologically important. It suggests that effective strategies to overcome progesterone resistance may need to target multiple nodes simultaneously rather than relying on inhibition of any single molecule. This is consistent with the general principle that cancer drug resistance involves multiple redundant survival mechanisms.
Cell line studies are an important first step in understanding cancer biology but have significant limitations: cell lines grown in culture differ metabolically and molecularly from tumors in patients, and results from a single cell line model may not generalize to the diversity of progesterone-resistant endometrial cancers seen clinically. The IshikawaPR model reflects one specific path to resistance, and other mechanisms may operate in patient tumors.
Validation of the identified hub genes and PGR co-expressed genes in patient tumor tissue - comparing biopsies from progesterone-sensitive and progesterone-resistant patients - is the essential next step. If the same molecular signatures identified in cell culture appear in patient-derived resistant tumors, the clinical relevance of these findings would be substantially strengthened.
This work establishes a comprehensive molecular landscape of progesterone resistance and provides specific candidate genes and pathways for targeted functional studies. Future experiments should use gene knockdown and overexpression to test whether specific hub genes are causally involved in resistance, and clinical correlative studies should assess whether baseline expression of resistance-associated genes predicts which patients will respond to progestin therapy.