ESGO 2013

Understanding the causes of endometrial cancer

Prof Martin Widschwendter - University College London Hospital, UK

Martin, good to have you here. What is it you’ve been cooking up with endometrial cancer?

We are interested in the earliest changes of endometrial cancer. We know that endometrial cancer is not a genetic disease, unlike some breast and ovarian cancers; endometrial cancer is a disease which is triggered by the environment, reproductive factors, obesity, for instance, and other factors and we speculated that all these factors lead to so-called epigenetic alterations which regulate gene activity.

Now there’s a belief that they’re regulated also by hormones, or at least the hormones are involved, aren’t they?

Yes. So what we know is that oestrogen triggers proliferation of the endometrium whereas progesterone suppresses the proliferation, the growth of the endometrium. What we were interested in whether this regulation is abnormal in women who develop endometrial cancer. So we’ve looked into so-called epigenetic alterations and we’ve found a phenomenon which is called DNA methylation. DNA methylation is a change of the DNA which supresses a gene and the gene which we’ve found to be supressed in the earliest changes of women who develop endometrial cancer is called HAND2.

Did you find that HAND2 was involved in it, then?

Yes, we’ve discovered that HAND2 is involved in the earliest stages of endometrial cancer. We’ve analysed hundreds of samples from women collected across Europe and we’ve also employed an animal modeller where we’ve artificially switched off this HAND2 gene in the endometrium of these mice and they all developed these pre-neoplastic, pre-cancerous changes.

So, are you saying that the environmental influences including hormones and things like nulliparity can influence whether that gene is switched on and causes cancer?

Absolutely. Our data are the first to provide sufficient evidence that these epidemiological factors like obesity, for instance, do refer their carcinogenic potential via modulating a gene which is DNA methylation of this particular gene, HAND2, in the endometrium. Once this gene is silenced the progesterone, which is the hormone which would supress the growth, can not do its action any more and hence that’s the first step for developing endometrial cancer.

So you’ve got something that is the first step but is that any use to you, practically?

Absolutely. We’ve done two proof of principle studies to see whether there is any immediate clinical use. What we’ve done is we’ve recruited patients with post-menopausal bleeding, they came to our clinic and we’ve applied these tests, which is a vaginal fluid sample and we’ve analysed DNA methylation of HAND2 and almost identified with 100% sensitivity and specificity those women who actually have got an endometrial cancer and separated them from those who did not have endometrial cancer.

And you’ve got another application, haven’t you? Something concerning hyperplasia?

Yes, absolutely. The second clinical approach which we were testing is to see whether we can predict the response of progesterone in women who present with abnormal bleeding due to hyperplastic endometrium and what we found is that HAND2 methylation clearly is associated with a lack of response to progesterone. So we know that women who have HAND2 methylated in their endometrium do not respond to progesterone and hence we don’t need to treat these women with progesterone.

How easy is it to do these tests in practical terms?

One of the beauties of these tests is that they are very, very cost effective. One of these tests costs about £10 at the most. So we’re currently doing them not in an NHS environment, we’re doing them in a clinical trial environment but I’m fairly sure that if we provide more evidence that that could lead to a clinical application fairly swiftly.

So do you see doctors putting these tests into action soon?

Not yet because we have just published these data, they need to be validated in larger cohorts but I’m pretty sure within the next two or three years those findings can lead to immediate clinical applications.

What kind of impact could this have on endometrial cancer?

One of the immediate consequences could be that we could develop a test to early diagnose endometrial cancer. In particular diagnosing endometrial cancer in women who present with post-menopausal bleeding, we know that women who present with post-menopausal bleeding only have a 10% risk to actually have endometrial cancer, 90% of women who present with post-menopausal bleeding do not have an endometrial cancer. So we could immediately see whether these women require immediate treatment or not at all.

So, as of now, and you’ve only just published this work, what’s the big message that doctors should take note of?

Currently, as I said, this has no clinical implication but what we have shown, really for the first time, is that epigenetic alterations are really crucial in cancer development. This breaks up with that dogma which was that genetic alterations are crucial for cancer development. We have provided now evidence to demonstrate that epigenetic alterations, respectively DNA methylation of HAND2, is crucially involved in carcinogenesis as such.

Could this translate into screening methods, do you think?

Yes, we have a got a big clinical trial or a grant from the European Union where we test this currently in breast and ovarian cancer patients using blood or respectively serum and try to identify epigenetic alterations, DNA methylation alterations, which come from the tumour, in this case breast and ovarian cancer. So this is a trial which we do in conjunction with a big European consortium and we hope to have the answer within the next three years.

I hope to hear all about it Martin. Thank you very much.

A pleasure.