How do you treat prostate cancer? The specialists can't agree whether it's best to treat tumours with surgery or radiotherapy, or simply leave them alone.

At the moment, we've got the worst of both worlds. In the UK, sufferers seek help for prostate cancer when it's too late for therapy. And in America, every man over 50 is supposed to have a yearly check-up; if a tumour is found, it's removed automatically, but so far there isn't any evidence to show this approach makes you live longer (JAMA, September 14, 1994).

Prostate cancer currently accounts for 14,000 new cases each year in Britain and is the third most common cancer among British men (The Lancet, May 22, 1993). It's also the most frequently occurring cancer among American men-165,000 new prostate cancer cases were diagnosed in 1993 (JAMA, May 26, 1993). A year later, this had supposedly grown to around 200,000 cases; prostate cancer was supposedly detected in 30 per cent of men over 50. Black rather than white males are more likely to contract it. The highest number of Black sufferers live in America.

Although it predominantly affects older men, it has been estimated that men dying of prostate cancer in America on average lose nine years of their life (National Cancer Institute 1989; No 89-2789).

The problem is that most doctors these days are scalpel happy. Surgery is the first-line and most dangerous treatment. Although the incidence of cancer hasn't really gone up, aggressive treatment like surgery has increased by 36 per cent, leading everyone to worry that prostate cancer is becoming epidemic.

The most famous study of this , entitled the SEER programme (Surveillance, Epidemiology and End Results), conducted by the Center for Evaluative Clinical Sciences in Hanover, New Hampshire, found that while prostate cancer had only increased modestly in America during the Eighties, the rates of prostatectomy were increasing by 35 per cent per year, and even more in different areas of the country. Nevertheless, according to the SEER study, this wasn't having any impact on survival rates.

A tendency to prostate cancer may be hereditary, although many environmental factors such as diet appear important in bringing on the disease (see box, p 3). If, say, a brother or father has prostate cancer, a patient is more than twice as likely to develop it than someone without any family links. And the risk increases according to the number of sufferers in the family. For example, two relatives would quadruple the risk.

Tests carried out over a number of years at Britain's Atomic Energy Authority also show nuclear workers exposed to some radioactive substances have an increased risk of contracting prostate cancer, although the effect of each substance is not known (BMJ, November 27, 1993).

Sexual activity or sexually transmitted diseases may also be associated with prostate cancer (The Lancet, October 9, 1993), as may vasectomy. Two US studies showed that vasectomized men had a 60 per cent increased risk of prostate cancer (JAMA, 1993; 269 (7): 878-82 and The Lancet, May 22, 1993). However, a medical panel reporting in the Journal of the American Medical Association rushed in to claim that the studies were found to be flawed from "detection bias" those men who had vasectomies were under the care of a urologist and may have been more likely to be examined and diagnosed with prostate cancer.