PATHOLOGY

Incidence and follow-up of patients with focal prostate carcinoma in 2
screening rounds after an interval of 4 years
Postma R, de Vries SH, Roobol MJ, Wildhagen MF, Schroder FH, van der
Kwast TH
Department of Urology, Erasmus Medical Center, Rotterdam, The Netherlands
Cancer 2005; 103: 708-16

• Background: Focal carcinoma detected by needle biopsy has been a common finding since prostate-specific antigen (PSA)-based screening was introduced. Clinicopathologic features in patients with focal prostate carcinoma who underwent radical prostatectomy (RP) or who were treated with watchful waiting (WW) were analyzed to detect clinical predictors for disease progression during follow-up.
• Methods: Patients were selected from the European Randomized Screening study for Prostate Cancer. Focal carcinoma on sextant biopsy was defined as < or = 3.0 mm involvement by tumor in 1 biopsy core lacking Gleason pattern 4 or 5. PSA doubling time was used in the WW group as a marker of disease progression.
• Results: The proportion of patients with focal prostate carcinoma increased significantly from 16% in the first screening round to 29% in the second screening round. One hundred eighteen men underwent RP, and 108 men were treated with WW. The median tumor volume was 0.13 mL. PSA level and prostate volume were predictive for tumor volume in a multivariate regression analysis. A PSA density cut-off level of < 0.1 ng/mL/cm3 predicted organ-confined tumor ( < 0.5 mL) in 94% of patients.
  Positive surgical margins were predictive for PSA recurrence. Four patients in the RP group had PSA recurrence at follow-up. PSA doubling times < 2 years, < 3 years, and < 4 years were noted in 4.9%, 14.6%, and 22.0% of patients in the WW group, respectively.
• Conclusions: The median tumor volume was small (0.13 mL). A comparison between PSA recurrence in the RP group and PSA doubling time in the WW group showed a significantly more favorable outcome after RP if a PSA doubling time of < 3 years or < 4 years was chosen as a marker for disease progression in the WW group. A WW policy with delayed curative intent may be recommended in patients ages 55-75 years with focal carcinoma and PSA density < 0.1 ng/mL/cm3.
  
  
• Editorial Comment
  “Minimal” or “insignificant” refers to a low-grade, organ confined cancer < 0.5 mL in a surgical specimen of radical prostatectomy. It must be emphasized that it does not mean “latent”, “dorment” or “indolent” cancer but a low-volume incipient neoplasia that can progress either as a latent or clinical carcinoma. Unfortunately there is no known marker for biologic behavior of prostatic cancer.
  Postma et al. propose some criteria on needle biopsies to predict these minimal cancers: focal carcinoma on sextant biopsy defined as < 3 mm in length in only one core, no Gleason grade (pattern) 4 or 5, and PSA density cut-off level of < 0.10 ng/mL. Using these criteria, the authors predicted 94% of patients harboring low-grade, organ confined < 0.5 mL tumors.
  We have used the criteria proposed by Epstein et al. (1): needle biopsies with prostate carcinoma in fewer than 3 cores (from a 6-core biopsy sample), absence of Gleason grade (pattern) 4 or 5, no more than 50% prostate carcinoma involvement in any of these cores, stage T1c and PSA density < 0.15 ng/mL. Using these criteria, in 1994, the authors found that 79% of tumors with volume < 0.5 mL were organ confined and did not qualify as high-grade lesions at the time of radical prostatectomy. Bastian et al. (2) from Johns Hopkins, applied these criteria in a series of 237 men with extended neddle biopsies who had undergone radical prostatectomy for T1c disease between December 2000 and August 2003. According to the Epstein needle biopsy criteria, low-grade, organ-confined prostate carcinoma was detected in 91.6% of all patients.
  The histologic criteria used by Bastian et al. and Postma et al. differ in two aspects: number of cores with carcinoma and the extent of core involvement. Postma et al. are more restrictive considering just one core involved and report the linear length of cancer. Epstein et al. consider a maximum of 2 cores involved and the percentage of involvement of the core. It seems to us that linear length is superior to the percentage of involvement: percentage varies according to the length of the core, unless is established a minimum length of the core for percentage evaluation.

References
1. Epstein JI, Walsh PC, Carmicahel M, Brendler CB: Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA. 1994; 271: 368-74.
2. Bastian PJ, Mangold LA, Epstein JI, Partin AW: Characteristics of insignificant clinical T1c prostate tumors. A contemporary analysis. Cancer. 2004; 101: 2001-5.

Dr. Athanase Billis
Full-Professor of Pathology
State University of Campinas, Unicamp
Campinas, São Paulo, Brazil