Natural history of prostate cancer in a large cohort of untreated patients in the United Kingdom

Reviewer: Ryan Smith , MD
Last Modified: May 15, 2005

Presenter: M.W. Kattan
Presenter's Affiliation: Transatlantic Prostate Group
Type of Session: Scientific


  • The natural history of untreated prostate cancer is not well understood
  • This is especially true, as no studies have addressed this in the PSA era.  Also, studies lack centralized review of pathology to determine Gleason score, adequate sample size, or long term follow up
  • This study followed a cohort of localized prostate cancer during the PSA era to determine patterns of failure

Materials and Methods

  • A cohort of patients was derived from cancer registries from six different regions in Great Britain
  • After registration, data was obtained from hospitals
  • The cohort of patients were diagnosed from 1990-1996, were all <76 years old, had a baseline PSA and which was require to be <100, and no surgery or radiation within 6 months of diagnosis, which they used to define as watchful waiting
  • Out of 36,201 potential patients, only a total of 1680 patients met all criteria, including having available records, baseline PSAs, without hormonal treatment, etc.
  • All had a central review of the pathology to obtain the Gleason scores
  • Endpoints were treatment, progression, and death
  • Median follow up was 89 months


  • Over half of the patients were diagnosed by TURP
  • 215 patients develped metastatic disease and 509 patients had required hormones because of progression
  • 300 died of other events before they developed metastatic disease, required hormones, or died of prostate cancer
  • In total, just over 30% died of prostate cancer and just over 30% died of other causes
  • In patients who did not die in 12 years from other causes, over 90% either developed metastatic disease, required hormones or other therapy, or died of prostate cancer
  • In a nomogram created by the authors based on the data, Gleason score was by far the most important predictor for progression of disease 

Author's Conclusions

  • This is the first prospective study in the PSA era that has been done following untreated prostate cancer patients
  • Clinically localized prostate cancer managed conservatively have a high rate of progression or secondary treatment in their lifetime

Clinical/Scientific Implications
This study has determined that many patients with prostate cancer will eventually be affected by their disease.  A couple of points should be noted that make the patient population different from that seen currently in the United States.  First, PSA screening was not widespread in Great Britain until the early to mid 1990s, so many of the patients may not have been detected by PSA screening alone.  Also, the majority of these patients were identified by TURP, which means they are lower stage if T1a or T1b, but higher stage if they were having symptoms that initiated the TURP.  Also, there was the question as to whether or not 6 months is a good time to follow patients without treatment.  Many patients simply take that long to make a decision as to whether to be treated or not, without specifically choosing watchful waiting.  What can be derived from this and other data is that patients with a lifespan less than five years do not necessarily need treatment, and mortality estimates can be made from nomograms such as these, taking into account PSA, T stage, and above all, Gleason score.