Once again in medicine, statistics have triumphed over common sense.
A wealth of epidemiological data now supports the position (still contested by many enthusiasts) that hormone replacement therapy (HRT) causes at least a 30 per cent increase in breast cancer. However, none of the studies has indicated which kind of breast cancer an HRT user is likely to develop slow growing or fast growing, ductal carcinoma in situ, invasive ductal or lobular, or anything else in between.
Susan Gapstur, a cancer epidemiologist at Northwestern University Medical School in Chicago, Illinois, and her colleagues decided to find out. For 11 years, their research project, the Iowa Women's Health Study, followed a random sample of more than 37,000 postmenopausal women to tally who got cancer and which kind.
The study found that women who had used hormones for five years or less were 1.8 times more likely to get invasive carcinoma with "favourable histology" that is, invasive, but slow growing cancer compared to women who'd never used hormones. Those using HRT for more than five years were 2.6 times more likely to develop these tumours. This is the relatively rare, slow growing, and according to the British Medical Journal (1999, 318: 1645), "highly curable" variety of breast cancer (see News, p 10).
Current users of HRT more than quadrupled their risk of developing this type of cancer for five or fewer years and more than doubled it for more than five years, respectively (JAMA, 1999; 281: 2091-7).
Although not prepared to completely go out on a limb, the authors certainly were shimmying up the tree when they implied that their results may mean that women taking HRT get a better type of breast cancer that is, a less aggressive, smaller, better differentiated and generally more friendly sort.
If you've got to get breast cancer, they seemed to be saying, this is definitely the one to have.
Call me old fashioned, but to me, cancer is cancer, and there is no such thing as a good cancer, any more than there is such a thing as a good war. Favourable prognoses are meaningless with a disease that is the second biggest killer in the West. Times columnist John Diamond was one of those with a 92 per cent chance of survival. Recently, he's been told by his doctors that the chemotherapy and surgery failed to arrest the cancer, and there is nothing much they can do for him now but buy a bit of time.
The way I read this study, it is unequivocably bad news: HRT quadruples your risk of cancer if you're currently on it.
To the medical community, however, the study represents no less than a discovery of a way to lick breast cancer. According to Gapstur and Co, "If HRT use selectively increases the risk of the less commonly occurring tumours with a good prognosis, then the overall risks and benefits of hormone use in the
population should be reexamined." What they appear to be saying is that by taking HRT, you're more likely to get a type of breast cancer we can treat and therefore less likely to get the dangerous type. And an editorial in the journal concluded that since this cancer only occurred in 5 per cent of the study population, it doesn't really count, anyway. Ergo: HRT does not cause "real" cancer.
This is the same old mistake doctors always make of considering medicine infallible. Cancer is only highly curable so long as we have a foolproof detection system and foolproof treatment. Only when medical researchers come up with a screening tool with a better track record than mammograms, and a treatment that works better than chemotherapy, can we start talking about a better class of cancer.