Angiogenesis and the mammography paradox

Michael Retsky, William Hrushesky, and Isaac Gukas have an interesting article relating to angiogenesis in the evolution of tumours published in BMC Cancer today.[1] It’s only an hypothesis, but it covers an area of cancer biology that I have been meaning to blog about for a long time now.

Angiogenesis is the process by which new blood vessels are created. It’s important in cancer biology: the growth of tumours is restricted by the amount of oxygen and nutrients available to the tumour cells. These are in-turn restricted by the amount of blood getting to the tumour, and tumours tend to outgrow their blood supply, hitting a barrier to further growth. But tumours undergo evolution — with so much cell division and death in tumours, a lot of variation is created, and there are a lot of opportunities for natural selection. A common adaptation seen in more advanced tumours is disregulated angiogenesis: new blood vessels are created throughout the tumour, providing unlimited oxygen and nutrients, and thus removing that barrier to growth.

Previously, Retsky and other members of the Harvard research group of the late Judah Folkman developed an hypothesis to explain an interesting paradox — the “mammography paradox”. It was noticed that after regular breast screening programs were introduced, more women in their forties were dying of breast cancer than had been predicted. Specifically, women who had undergone surgery were dying. Instead of saving lives, in a few cases cancer screening and the removal of tumours was expediting the end.

Retsky et al‘s 2005 hypothesis[2] proposed that angiogenesis was behind the paradox. They suggested that the act of removing the primary tumour somehow caused barely detectable distant metastases (“micrometastases”) to come out of a dormant state that they would otherwise have remained in for years. These micrometastases are in a stable state of dormancy because they have grown to the limit which their supply of nutrients and oxygen can support, and have yet to stumble upon the angiogensis adaptation. Retsky et al also made a couple of suggestions for why surgery should cause these micrometastases to become active. Firstly, they pointed to previous research showing that some tumours release anti-angiogenic factors into the blood, and they therefore suggest that the primary tumours are themselves remotely suppressing angiogenesis in the micrometastases. Secondly, following any surgery the body will release pro-angiogenic factors into the blood in order to aid wound healing, and these will have the side-effect of activating the dormant micrometastases.

So, under the proposed model, mammography was increasing breast cancer deaths because tumours were being detected and removed at a younger age and at an earlier stage. Because the surgery was being performed at an earlier stage, doctors were less worried about metastasis, when in-fact they should have been just as worried as before. Because the doctors were less worried about metastasis, they were giving less rigorous programs of adjuvant chemotheraphy after the surgery. And because the chemotherapy was less rigorous, micrometastases were able to survive and indeed thrive, and so the patients experienced relapse.

Based on their earlier hypothesis, Retsky et al had an idea for a novel cancer therapy. The therapy they were seeking would be some kind of anti-angiogenic treatment which targeted the micrometastases, providing the same kind of brake on angiogenesis that they believed was being provided by the primary tumour. At the same time, the treatment must not interfere with wound healing. Additionally, if it is true that the dormancy of the micrometastases is dependent on anti-angiogenic secretions from the primary tumour, any replacement anti-angiogenic treatment would have to be long-term, and so toxicity and side-effects could not be tolerated.

In their new article in BMC Cancer, Retsky et al believe they have found what they are looking for. Their proposed anti-angiogenic treatment is Endostatin. I briefly wrote about Endostatin in May 2007. The molecule, which has been shown to have anti-angiogenic effects in the lab, has been proposed as the explanation for another interesting observation: people with Down’s syndrome tend not to get solid tumours. People with Down’s have an extra copy of chromosome 21, and Endostatin is encoded by a gene on that chromosome. Those with Down’s have higher levels of Endostatin circulating in the blood, but do not have problems with wound healing after surgery. Endostatin, then, looks like a good candidate: it’s anti-angiogenic in laboratory animals, correlates with lower cancer rates in Down’s, is clearly not toxic, and will not interfere with wound healing.

Back in May 2007, I ended that throwaway post with a dismissal of Endostatin’s efficacy as a potential cancer treatment, based on trials in a some advanced tumours — Wikipedia‘s article on it indicates only that in phase II trials it didn’t work. Nevertheless, Retsky et al think that we shouldn’t give up on Endostatin just yet. Those earlier trials of Endostatin were as a treatment for advanced tumours, a very different problem to that of controlling dormant metastases. The hypotheses behind Endostatin, though wanting strength in places, are just too satisfactory to give up so easily.


  1. ^  Michael W Retsky, William JM Hrushesky, Isaac D Gukas (2009). Hypothesis: primary antiangiogenic method proposed to treat early stage breast cancer BMC Cancer, 9 DOI: 10.1186/1471-2407-9-7
  2. ^  Michael Retsky, Romano Demicheli, and William J.M. Hrushesky. 2005. Does surgery induce angiogenesis in breast cancer? Indirect evidence from relapse pattern and mammography paradox. International Journal of Surgery 3(3):179-187. doi:10.1016/j.ijsu.2005.08.002

Leave a comment

Your email address will not be published. Required fields are marked *