Heart damage occurs in about 1 in 25 patients, a figure described as "unacceptably high" (Clin Breast Cancer, 2002; Suppl 2: S75-9).

Herceptin can also cause a litany of nasty side effects, which can occur in up to 40 per cent of cases (see box, page 6). So the claim that Herceptin is a nice, clean "targeted" cancer drug doesn't really stack up.

"Wonder-drug" endorsement
But does it work? Well, if the drug's spokesmen are anything to go by, Herceptin is nothing short of miraculous. In May last year, researchers announced at a cancer conference that the drug could reduce the risk of breast cancer recurrence by over 50 per cent within one year. One of the lead researchers in the UK, Professor Ian Smith, commented: "This is the biggest treatment development in breast cancer, in terms of the magnitude of its effect, for at least the last 25 years, perhaps as big as anything we've seen."

Prof Smith is head of the Breast Unit at the Royal Marsden Hospital, Britain's premier cancer hospital. Herceptin couldn't hope for a better endorsement-hence the media snowstorm of "wonder-drug" stories, desperate women selling their houses, and the PR blitzkrieg to coerce NHS hospitals to prescribe the drug for early-stage breast cancer.

Herceptin is relatively new to Britain, but it's been around for about eight years in the USA. The American experience is instructive; they've found that the drug is much less effective on its own than when combined with conventional chemotherapy drugs. With that approach, Herceptin's claimed "targeted" advantages are wiped out, condemning cancer patients to the myriad toxic side effects of standard chemotherapy.

Initially, the results appeared to be worth it. In October 2005, two Herceptin research papers were published in the New England Journal of Medicine, authored by about 50 doctors, reporting on how their patients responded to Herceptin plus chemotherapy. It was impressive-looking research, comparing about 4500 treated patients against different control groups (NEJM, 2005; 353: 1673-84; NEJM, 2005; 353: 1659-72).

An accompanying editorial described the results as "simply stunning", and "maybe even a cure" for breast cancer. The results showed that Herceptin plus conventional chemotherapy reduces the death rate by 33 per cent, and the cancer recurrence rate by more than 50 per cent. Those were the figures given to the press. So good were the results that the US authorities decided to stop some of the trials early, as the case for Herceptin appeared to be proven.

However, let's look at the figures more closely. The three-year, disease-free survival rate was found to be 75.4 per cent for those on conventional treatment, and 87.1 per cent for those on Herceptin. If you subtract these numbers from 100, you get 24.6 per cent versus 12.9 per cent-which is where the 50 per cent improvement in recurrence figure comes from. But another way of looking at it is that only 12 per cent of Herceptin users may be helped by the drug to remain disease-free after three years.

Hardly dramatic, hardly wonder drug, hardly cure.
What about life expectancy, the most meaningful statistic of all? Here the small print is even less impressive. In the trial that compared Herceptin to no treatment, death rates after one year "were not significantly different", with just over 2 per cent deaths in the untreated group, and just under 2 per cent in the Herceptin group. A significant number of women in that group were reported to have experienced "severe cardiotoxicity".