Active surveillance strategy for patients with localised prostate cancer: criteria for progression

Active surveillance - an initial observational strategy - offers a tailored management of patients with localised prostate cancer. The aim of the strategy is to appoint patients with potentially lethal prostate cancer to curatively intended treatment, while patients with slowly evolving tumours are spared from an unnecessary curative intervention.

All data included were derived from a single-institution active surveillance cohort of 317 patients which was followed prospectively at Rigshospitalet from 2002 until 2013. The patients were managed with serial PSA measurements, repeated biopsies, and regular digital rectal examinations. The programme recommended change of management from active surveillance to curatively intended treatment based on PSA doubling time, deteriorating histopathology in repeated prostatic biopsies, and increased clinical tumour category.

The programme entailed close monitoring during the first 5 years with 3-4 out-patient contacts annually. Altogether, 2-3 biopsy sessions were performed in most patients. Complications necessitating hospital admissions arose in almost 10% of the repeated biopsy sessions. The 5-year cumulative incidence of curatively intended treatment was estimated to be 39.5%. Active surveillance resulted in a 34.8% cost-reduction following 3.7 years compared to the estimated cost of immediate radical prostatectomy. The calculated PSA doubling times were associated with wide 95% confidence intervals, which resulted in a significant risk of being misclassified according to the definition of progression. The interobserver agreement of biopsy histopathology between expert uropathologist was substantial. Still, the pathologists' disagreement would have resulted in different treatment recommendations in up to 10% of the re-evaluated biopsies. Neither PSA doubling time nor increased clinical tumour category was associated with final histopathological findings following subsequent radical prostatectomy. Although the level of significance was only met in univariate analysis, biopsy progression was associated with defined final histopathological findings at radical prostatectomy that was perceived as unacceptable for a continued observational strategy.

The thesis has demonstrated that active surveillance is feasible and reduces the number of patients undergoing curative intended treatment. However, active surveillance necessitates close monitoring during the first 5 years. PSA doubling time is unreliable as a progression criterion, while progression on repeated biopsy in part seems to fulfil the requirements of a dependable progression criterion. The need for more accurate progression criteria in the management of prostate cancer patients on active surveillance is emphasised.