The congratulations offered by Dr Harding to those health authorities yet to make Paclitaxel available to their patients with ovarian cancer are misplaced. The evidence was summarised in Effective Health Care and is set out more fully in the research evidence volume of the Guidance on Commissioning Cancer Services - Improving Outcomes in Gynaecological Cancers, published by the NHS Executive at the end of July.

While the evidence supporting the use of Paclitaxel may be inconvenient, it is nonetheless clear, and provides sufficient guidance for policy and practice in 1999. There is evidence from four main studies, GOGIII, OV10, GOG132 and ICON3. Of these, ICON3 has not long finished recruitment, and data from this trial is not yet reliable enough to use. This fact is acknowledged by Dr Harding, who then fails to resist the temptation to draw conclusions from this preliminary data.

GOGIII and OV10 clearly show significant improvements in overall survival and progression-free survival. The other study (GOG132) appears negative (in fact it is statistically inconclusive) but suffers from the fact that many women randomised not to receive Paclitaxel 'crossed over' and in fact did receive it. Thus, this study is not easy to assess.

Meanwhile, the evidence justifies Paclitaxel as standard therapy in ovarian cancer. Women in the UK with this disease deserve the possibility of a standard treatment, justified by evidence, and used internationally. No doubt the full results of ICON3 will emerge over the next two to three years.

If the evidence base shifts at some time in the future then of course HAs will be justified in reconsidering.

Current evidence strongly suggests they should look for ways to ensure that this treatment can be made available to women suffering from ovarian cancer, rather than trying to second-guess the final results of one trial in order to defer their decision.

Professor Bob Haward

Chair, National Cancer Group

Rachel Richardson

Research Fellow

NHS Centre for Reviews and Dissemination