Researchers from Belgium, France, the US and Poland have concluded that just because a tumour ‘responds’ to chemotherapy doesn’t automatically mean a better chance of patient survival.

They drew their conclusions from a meta-analysis of 25 randomised trials involving nearly 4000 patients with colorectal cancer. The data showed that, while different drugs produced different responses rates experimental fluoropyrimidines, for example, produced a better tumour response rate than bolus fluoropyrimidines the rate of survival did not always equate with the rate of tumour response.

The authors noted, for example, that a treatment which lowered by 50 per cent the odds of a tumour failing to respond would only decrease the odds of death by 6 per cent. They also note that, even in the most successful trials, response rates are small (under 30 per cent) and short lived (Lancet, 2000; 356: 373-8).