I was wrong to say "longest running." What I should have said is it has the longest follow up (20 years) in North America.
Klotz originally had two criteria for intervention:
1) Detection of any Gleason 4+3 or higher
2) PSADT<3 years
(and radiological evidence)
In the following, he discusses why he got rid of the second criterion.
LIMITATIONS OF PSA KINETICS AS A TRIGGER FOR INTERVENTION
In the early days of active surveillance, PSA kinetics was an appealing trigger for intervention. This was based on the simple observation that most patients with advanced prostate cancer had an elevated PSA, and, therefore, a stable PSA likely implied stable disease and vice versa. Until multiparametric MRI became available, men on active surveillance (AS) with poor PSA kinetics (doubling time <3 years) were offered treatment. In the PRIAS multi-institutional AS registry, 20% of men being treated had intervention based on a PSA doubling time less than 3 years . Indeed, in a report of the five men dying of metastatic prostate cancer in the Toronto cohort, all had a PSA doubling time of less than 2 years . However, PSA kinetics turned out to have a crucial limitation: lack of specificity. The liability of PSA, and the high prevalence of inflammatory causes of a sharp transient rise in PSA, severely limits the reliability of PSA kinetics as a trigger . In a study of PSA kinetics in a large surveillance cohort, false-positive PSA triggers (doubling time <3 years, or PSA velocity >2 ng/year) occurred in 50% of stable untreated patients, none of whom went on to progress, require treatment or die of prostate cancer . Vickers , in an overview of all of the studies of more than 200 patients examining the predictive value of PSA kinetics in localized prostate cancer, concluded that kinetics had no independent predictive value beyond the absolute value of PSA. Currently, PSA kinetics are used as a guide to identify patients at a higher risk, but not to drive the decision to treat.