Radical prostatectomy for low-risk localised prostate cancer has no significant effect on all-cause or prostate cancer specific mortality compared with observation, a randomised controlled trial has shown.

The PIVOT study, carried out between 1994 and 2002 and published in the NEJM, compared radical prostatectomy with observation in 731 men who had received a diagnosis of clinically localised prostate cancer in the early era of PSA testing.

Entry criteria required that the men were: fit for radical surgery; with clinically localised prostate cancer of any grade (i.e. not only low Gleason grade, low-risk cancers) diagnosed within the previous 12 months; PSA < 50; age ≤ 75 years; a negative bone scan for metastatic disease; and a life expectancy ≥ 10 years from the time of randomisation.

The primary outcome measure was all-cause mortality, with prostate cancer specific mortality a secondary outcome measure.

Of 13,000 men diagnosed with prostate cancer during the study, just over 5,000 met the entry criteria. However, the study was only able to recruit 15% of these patients, reflecting the difficulty of carrying out randomised trials of this nature in cancer patients.

A total of 364 men were randomised to the radical prostatectomy group and 367 to the observation arm. Mean age of participants was 67 years and the median PSA 7.8 ng/ml. Almost half (48%) had Gleason scores > 7. Forty per cent of men were considered low risk, 34% intermediate risk, 21% high risk (5% had missing data).

Overall, 79% of those allocated to prostatectomy underwent this procedure, while 10% of men in the observation arm also underwent surgery (within a median time of 61 days, suggesting the decision was based on unhappiness with the outcome of randomisation rather than disease progression).

During a median 10 years follow up, 47% of the patients in the prostatectomy group died, compared with 49.9% in the observation group, an absolute risk difference of 2.9% which was not statistically significant. In terms of prostate cancer specific mortality, 5.8% of patients in the prostatectomy arm died, compared with 8.4% in the observation arm. Again this 2.6% absolute risk difference was not statistically significant.

In subgroup analysis, radical prostatectomy did appear to offer an all-cause survival advantage over observation in patients with a PSA >10 ng/ml (p=0.04), and possibly in those with intermediate- or high-risk tumours (p=0.07).

The treatment of early stage localised prostate cancer has long been controversial. Given the potential side effects of radical treatment in terms of erectile dysfunction and incontinence, it is vital that we continually aim to target these treatments at those patients most likely to benefit in terms of survival. The lifetime risk for a man developing prostate cancer is 17%, but given the risk of dying from the disease is only 3% it would be reasonable to surmise that conservative management may be appropriate for many men.

This study adds weight to the argument that low-risk prostate cancer i.e. low volume Gleason score ≤ 6 disease with PSA ≤ 10 ng/ml should ideally be managed with active surveillance rather than radical treatment, saving patients from the risks and side effects of treatment, as well as producing cost savings to the wider healthcare community.

Dr Jonathan Rees