Final Report Summary - CHAMELEON (Cellular Hypoxia Alters DNA MEthylation through Loss of Epigenome OxidatioN) The early causes of cancer are well-known: due to coincidence or carcinogenic factors, a single cell acquires a genetic mutation, followed by a rapid expansion of the abnormal cell. These genetic mutations disturb the normal cell function, but are beneficial for the growth and survival of cancer cells. Apart from these genetic changes, tumor cells also differ epigenetically, determining if and how genes are used. Although such epigenetic changes don’t affect the genetic code, they can strongly disturb the gene function in a similar way, to the benefit of cancer cells. But until now, the origins of these epigenetic changes mostly remained unknown. We here investigated one frequent epigenetic alteration: DNA hypermethylation, or the excessive addition of methyl groups to DNA. Hypermethylation silences the expression of tumor suppressor genes, thereby enabling aberrant behavior of cells and the excessive growth of tumors. We show that these epigenetic alterations are caused by the environment of the tumor, more specifically by oxygen shortage due to the abnormal tumor vasculature – which we call ‘hypoxia’. Oxygen is required by the enzymes that normally remove the methyl groups from the DNA. When oxygen is lacking, these enzymes are dysfunctional, causing hypermethylation. This accumulation of methylation inactivates associated tumor suppressor genes, linking the hypoxic tumor microenvironment to epigenetic changes in cancer cells. Particular intriguing is that up to 50% of the methylation events in solid tumors could be explained by tumor hypoxia, a mechanism observed in all 8 solid tumor types investigated (breast, bladder, colorectal, head and neck, kidney, lung and uterine tumors), indicating that DNA hypermethylation can serve as a broadly applicable biomarker for tumor hypoxia. Uncovering the link between oxygen shortage and tumor growth was the result of the analyses of over 3,000 patient tumors. As a next step, we verified another assumption: would interfering in the tumor oxygen supply affect tumor hypermethylation? Using mice, we proved that normalizing the blood supply by angiogenesis inhibitors (i.e. inhibitors that stop tumors from growing their own blood vessels) is sufficient to inhibit this hypermethylation. DNA methylation thus serves as a dynamic read-out of changes in oxygen supply to the tumor. Moreover, we have linked DNA hypermethylation to tumor immunotolerance (e.g. making the tumor invisible to the host’s immune system) and to the resistance to anti-angiogenic therapies. Altogether, these understandings can impact cancer management. First of all, we developed a blood-based test with DNA methylation as read-out to detect circulating tumor DNA (ctDNA) in plasma that is released by tumors cells into the bloodstream, allowing us to characterize and monitor tumors non-invasively. Using this test, we have shown that changes in DNA methylation (before versus during therapy) can predict resistance to anti-angiogenic therapy in hepatocellular carcinoma. Specifically, we observed that patients developing resistance show an increase in methylation 1 month after starting the treatment, whereas responding patients fail to show changes or are even characterized by a reduction in DNA methylation. Thus, our optimized readily assessible blood-based methylation test can be used to predict and to monitor the response to anti-angiogenic treatment in cancer treatment, allowing us to make more informed treatment decisions. Secondly, we shed new light on existing anti-angiogenic therapies, as a better perfusion helps to deliver chemotherapy to the tumor, but also inhibits new epigenetic aberrations. This is proven to be therapeutically beneficial, as it reduces relapse-aggressiveness.
