Welcome to the CJU website » LOG IN


Delay in surgical therapy for clinically localized prostate cancer and biochemical recurrence after radical prostatectomy
Division of Urology, University Health Network, University of Toronto, Toronto,
Jun  2003 (Vol.  10, Issue  3, Pages( 1891 - 1898)


Text-Size + 

  • BACKGROUND: In Canada, waiting times for cancer care have been increasing, particularly for patients with genitourinary malignancies. We examined whether delay from diagnosis for patients undergoing surgery for clinically localized prostate cancer affects cancer cure rates. METHODS: We conducted a historical cohort study among 645 patients who underwent radical prostatectomy between 1987 and 1997, using biochemical recurrence (PSA elevation) and metastasis as endpoints. We examined whether patients who underwent surgery ≥ months (delayed surgery group) from the date of diagnosis had reduced recurrence-free survival, compared to patients who had surgery <3 months (early surgery group) from the date of diagnosis, adjusting for grade, stage and PSA level at diagnosis.


    The crude 10-year recurrence-free and metastasis-free survival rates for all patients were 71.1% (95% C.I.: 64.9% - 77.3%) and 95.3% (95% C.I.: 91.3% - 99.3%), respectively. Of the 645 patients, 189 (29.3%) had surgery ≥ months after diagnosis. The median time from the date of diagnosis to surgery was 68 days (range 15 to 951 days). The 10-year recurrence-free survival was higher for patients who underwent early surgery (74.6%, 95% C.I.: 67.9% - 81.4%) compared to patients in the delayed surgery group (61.3%, 95% C.I.: 46.7% - 76.0%, p=0.05). The crude and adjusted hazard ratios for developing biochemical recurrence for patients in the delayed surgery group were 1.58 (95% C.I.: 1.0 ? 2.4, p=0.04) and 1.46 (95% C.I.: 0.9 ? 2.3, p=0.09), respectively, compared to patients who underwent early surgery.


    There may exist a possible relationship between delays from diagnosis for radical prostatectomy and prostate cancer cure rates. These findings may have many biases that could not be properly accounted in this retrospective analysis and larger cohort analyses will be required to confirm these findings.