Association of p27Kipl Levels With Recurrence and Survival in Patients With Stage C Prostate Carcinoma

Richard J. Cote 0 Yan Shi 0 Susan Groshen 0 An-Chen Feng 0 Carlos Cordon-Cardo 0 Donald Skinner 0 Gary Lieskovosky 0 0 Oxford University Press 916 REPORTS - Background: There are few biologic determinants that are prognostic for patients with localized prostate cancer. We examined whether cellular levels of the cyclin-kinase inhibitor p27Kip1 (also known as p27) in prostate tumors could be used to predict progression of this disease. Methods: Levels of p27 in tumor cell nuclei were assessed by immunohistochemical analysis of tissue sections from the primary tumors of 96 patients with stage C prostate carcinoma who had been treated by radical prostatectomy. Tumors were classified into one of the following three groups on the basis of the percentage of tumor cells showing nuclear p27 reactivity: low (0%10%), moderate (11%50%), and high (>50%). The Mantel-Haenszel test, KaplanMeier analysis, and the logrank test were used to calculate the probability that nuclear p27 levels were associated with tumor grade and substage, with a serum prostate-specific antigen (PSA) recurrence (defined as the finding of a detectable level [0.4 ng/ mL or greater] of serum PSA following radical prostatectomy), with the recurrence of clinically evident disease, and with survival. All reported P values are two-sided. Results: Luminal cells and basal cells of normal prostate glands showed high levels of nuclear p27 immunoreactivity in all tissue sections examined. Fifty-three tumors showed high p27 reactivity, 31 showed moderate reactivity, and 12 showed low or no detectable reactivity. Decreased levels of p27 were associated with tumor grade (P = .004). Tumor levels of p27 were not associated with preoperative prostate-specific antigen levels (P = .360) or with tumor substage (P = .320). However, decreased p27 reactivity was significantly associated with an increased probability of recurrence (P = .004) and decreased survival (P = .010). The median recurrence-free interval for patients with tumors showing high, moderate, or low p27 reactivity was 13.7 years, 8.4 years, and 4.7 years, respectively. Median survival times were more than 14 years, more than 13.5 years, and 8.1 years for patients in the high, moderate, and low p27 reactivity groups, respectively. Conclusion: Levels of nuclear p27 immunoreactivity in the primary tumor can be used to predict recurrence and survival among patients with localized prostate cancer. [J Natl Cancer Inst 1998;90:91620] The management of cancer depends on an accurate assessment of the tumors biologic potential. This is particularly true of prostate cancer, now the most commonly diagnosed human cancer. The ability to identify those tumors most likely to progress would greatly facilitate management of the disease. While preoperative serum prostate-specific antigen (PSA) determination has been useful in this regard, it has not shown the specificity for tailoring therapy in individual patients (1). Analysis of genes and proteins involved in cell cycle regulation has yielded important prognostic and treatment information in many tumor systems. Recently, cell cycle inhibitors have been demonstrated to play a potentially important role in tumor progression (2). Loss of expression of one of these, the cyclin-kinase inhibitor p27Kip1, a member of the cip/kip family (also known as p27) (3), has been shown recently to be associated with progression of several tumor types (46). We undertook this study to assess the potential role of p27 expression in prostate cancer progression. Patients, Materials, and Methods Patient Population From July 20, 1982, through April 10, 1989, a total of 101 patients who underwent radical retropubic prostatectomy with bilateral pelvic lymph node dissection at the USC/Norris Comprehensive Cancer Center for the treatment of adenocarcinoma of the prostate had pathologic stage C (pT3, N0, M0) disease (7). We report on 96 of these participants (46 79 years old; median age 4 65 years). Subjects were excluded from the study if their tissue was unavailable for immunohistochemical analysis. Tumors were graded according to the Gleason system (8), and all patients had no evidence of lymph node metastasis as judged by histologic examination. The substage of disease was categorized as described by Gibbons et al. (9): C1invasion through capsule without involvement of surgical margin or seminal vesicles, C2a positive surgical margin without seminal vesicle involvement, and C3involvement of seminal vesicle(s). This study was approved by the USC/Norris Institutional Review Board. The median follow-up time for the 96 subjects was 9.5 years. All subjects were evaluated at 1, 2, and 6 months after surgery, followed by 6-month intervals until postoperative year 5, and annually thereafter. Subject evaluation included a complete physical examination and determination of serum prostatic acid phosphatase levels and, after 1987, of serum PSA levels. Bone scans were performed as clinically indicated or when PSA levels were increased to 0.4 ng/mL or greater. Biopsy specimens were obtained to document local cancer recurrence or metastatic disease as clinically indicated. Clinical recurrence of cancer was determined histologically for locally recurrent disease; metastatic disease was confirmed histologically or by bone scan. PSA recurrence was defined as a detectable serum PSA level (0.4 ng/mL) on two consecutive tests determined at least 30 days apart. To determine serum PSA levels, the Hybritech enzyme-linked immunoabsorbent assay was used according to the manufacturers instructions (Hybritech, San Diego, CA) for most of the study period; for this assay, detectable is defined as a PSA level of 0.4 ng/mL or greater. More recently, a modification of this assay (TOSOH, Foster City, CA) has been used. Time to clinical recurrence was defined as the time from prostatectomy to clinical recurrence or until last follow-up, if the subject had not experienced a clinical cancer recurrence; time to PSA recurrence was defined as the time from prostatectomy to the first detectable PSA level or until last follow-up, if the subject had not experienced a PSA recurrence. Participants who died prior to clinical cancer recurrence or a PSA recurrence were censored at the time of death. Survival was calculated as the time from prostatectomy to death due to any cause; patients who are still alive were censored at the date of last contact. Monoclonal Antibody and Immunohistochemistry Sections (5-mm-thick) from formalin-fixed, paraffin-embedded, primary tumor tissue were mounted on poly-L-lysine-coated slides. Because we found that stored slides gave poor results, all assays were carried out on tissue sections that were less than 1 week old from the time of mounting. The slides were deparaffinized with iodinexylene, washed with absolute ethanol, and rehydrated with 95% ethanol. Endogenous peroxidase was quenched in 3% hydrogen peroxide in absolute methanol. Antigen retrieval was perf (...truncated)