Tumor necrosis refers to areas of dead tissue within a tumor - cells that have been killed, usually because they outgrew the blood supply that feeds them. In many cancer types, the presence of necrosis signals that a tumor is growing aggressively and has outpaced its nutrient supply.
In endometrial cancer specifically, prior research showed that necrosis correlates with higher cell proliferation rates, advanced tumor stage, and reduced survival. The researchers who conducted this study suspected these necrotic regions weren't just passive dead zones, but were actively influencing the tumor environment in ways that drive cancer spread.
This study investigated exactly what is happening at the molecular level within endometrial tumors that contain necrosis, with the goal of identifying treatment targets specific to these aggressive cancers.
The researchers analyzed two series of endometrial cancer patient samples. Series I included 57 patients whose tumors were analyzed using DNA microarrays - a technology that can measure the expression levels of thousands of genes simultaneously. Series II included 286 patients used to validate the clinical findings.
Using a statistical approach called SAM (Significance Analysis of Microarrays), the team identified which genes were expressed differently in tumors that had necrosis compared to those that did not. They then performed Gene Set Enrichment Analysis (GSEA), which asks whether groups of genes known to be involved in specific biological processes - like hypoxia responses or inflammation - were collectively over-represented among the differentially expressed genes.
To identify potential drug treatments, the researchers used Connectivity Map analysis, a database that matches gene expression patterns against the gene signatures produced by thousands of known drugs, to find compounds that might counteract the necrosis-associated gene signature.
Necrosis was found in 58-61% of all endometrial cancer tumors across both patient series. Its presence was strongly linked to the most dangerous tumor features: non-endometrioid histological subtype (the more aggressive Type II cancers), Grade 3 (highest grade) tumors, and advanced FIGO Stage III or IV disease.
Crucially, tumors with necrosis showed significantly higher rates of vascular invasion - the process by which cancer cells enter blood vessels to spread to other organs. This mechanistic link between necrosis and metastasis was consistent across both patient series.
Patients whose tumors contained necrosis had significantly reduced survival in both study series. The presence or absence of necrosis was so informative that a 13-gene predictor built from the necrosis-associated gene signature could correctly classify tumors as necrosis-present or necrosis-absent with 87% accuracy.
The microarray analysis identified 38 genes differentially expressed between tumors with and without necrosis - 25 were upregulated and 13 were downregulated. Several of these genes are well-known markers of hypoxia, the condition of low oxygen that develops when tumors outgrow their blood supply: IL6, CXCL8 (also known as IL8), SERPINE1, and genes involved in glycolysis (the alternative energy metabolism cells switch to when oxygen is scarce).
The necrosis gene signature score correlated strongly (correlation coefficients 0.52-0.72) with five independent published hypoxia signatures when mapped to the same dataset. When endometrial cancer cell lines were grown in low-oxygen conditions in the laboratory, they showed significantly elevated necrosis signature scores compared to cells grown in normal oxygen - confirming that hypoxia directly activates this gene program.
The key transcription factor activated by hypoxia, HIF1A (hypoxia-inducible factor 1-alpha), appeared to be a central driver. HIF1A acts as a master switch that turns on dozens of genes when oxygen is low - including many of the genes upregulated in necrotic tumors in this study. The inflammatory transcription factor NF-kB was also implicated, connecting low oxygen levels to inflammatory signaling.
Angiogenesis is the formation of new blood vessels, a process tumors hijack to grow and spread. Several angiogenesis-related genes were significantly upregulated in necrotic tumors: MMP1, MMP3 (enzymes that break down tissue barriers), SFRP2, CXCL8, and SERPINE1. The endothelial cell gene signature - markers of blood vessel cells - was also significantly elevated in tumors with necrosis, consistent with active blood vessel formation.
The wound response gene signature, which reflects the tissue remodeling processes active during wound healing, was also strongly associated with necrosis. This makes biological sense: tumors with necrosis are essentially experiencing tissue injury, triggering repair responses that end up promoting cancer progression rather than normal healing.
The combination of hypoxia, angiogenesis, and inflammation creates a self-reinforcing cycle of aggressiveness. Hypoxic tumor cells die, creating necrosis. Dying cells release inflammatory signals that recruit immune cells and stimulate new blood vessel formation. New vessels allow the tumor to grow further, creating more hypoxic zones that produce more necrosis. This cycle may explain why necrosis is such a strong predictor of poor outcomes.
Using Connectivity Map analysis, the researchers identified drugs whose gene expression signatures were the inverse of the necrosis signature - candidates that might reverse the necrosis-associated gene program. Two compounds, Emetine and Cephaeline, were among the top matches. These drugs have been reported to inhibit HIF1A activation, consistent with HIF1A being a key driver of the necrosis gene signature.
More therapeutically relevant, the analysis also identified signatures of PI3K/mTOR inhibitors as negatively correlated with the necrosis signature. The PI3K/mTOR pathway is already a validated drug target in advanced endometrial cancer, and this finding provides additional rationale for using these inhibitors specifically in patients with tumor necrosis.
The study therefore suggests that three pathways - HIF1A, NF-kB, and PI3K/mTOR - are especially promising targets for treating the most aggressive endometrial cancers that show necrosis. Drugs targeting these pathways could potentially cut off the oxygen-sensing, inflammatory, and growth-signaling processes that make necrotic tumors so dangerous.
This study establishes tumor necrosis as far more than a passive sign of aggressive cancer. It is an active molecular environment defined by hypoxia, inflammation, and angiogenesis - each feeding into the others in a cycle that promotes metastasis and treatment resistance.
The 38-gene necrosis signature provides a molecular tool that could be applied to clinical biopsy samples to identify patients with necrosis-associated biology even in cases where traditional pathology assessment is ambiguous. A condensed 13-gene version predicts necrosis status with 87% accuracy, a level of performance that could be practical in clinical settings.
Identifying necrosis at diagnosis could guide treatment decisions - flagging patients who would most benefit from therapies targeting HIF1A, NF-kB, or PI3K/mTOR. Because these molecular programs are not unique to endometrial cancer, the findings may also be relevant to other cancers where tumor necrosis is commonly observed.