Relationship between obesity and race in predicting adverse pathologic variables in patients undergoing radical prostatectomy.

OBJECTIVES: To determine whether obesity is associated with more advanced prostate cancer (PCa) in radical prostatectomy patients and to explore the ethnic variability in body mass index (BMI) as a potential explanation for racial differences in PCa risk. METHODS: A multi-institutional retrospective analysis of the clinical and pathologic parameters was performed on data from 860 patients with PCa undergoing radical prostatectomy between 1992 and 1998. Patient height and weight was used to calculate the BMI, which categorized patients into obese (BMI 30 kg/m(2) or greater), overweight (BMI 25 to 30 kg/m(2)), and normal (BMI 25 kg/m(2) or less) groups. Age, serum prostate-specific antigen level, pathologic stage, and Gleason score for each group were compared. The distribution of the BMI in each of four ethnic groups was also determined. RESULTS: Of 860 patients, 171 (20%) were obese, 425 (49%) overweight, and 264 (31%) normal. The obese patients presented at a younger mean age (62 years, P = 0.001), had higher mean Gleason scores (6.7, P = 0.002), had a higher likelihood of Gleason score 7 or greater cancer (71%, P = 0.003), and had a lower chance of organ-confined cancer (46%, P = 0.050). The BMI was highest in blacks, followed by whites and Asians, and blacks had significantly higher grade cancers (P = 0.045). In multiple logistic regression analysis of the BMI and race, only BMI remained an independent predictor of Gleason grade. CONCLUSIONS: Obese patients with PCa present for radical prostatectomy at a younger age with higher grade and more pathologically advanced cancers. Blacks have higher grade cancers than other ethnic groups and, at the same time, have significantly higher BMIs. These findings suggest that obesity may in part account for the racial variability in PCa risk.

A long-term study of the efficacy of treatment of localized prostate cancer.

This paper presents a retrospective comparison of patients undergoing treatment for clinically localized prostate cancer. We reviewed the age, grade, and stage at diagnosis as well as the survival and recurrence rates of 222 patients treated for carcinoma of the prostate with either radical prostatectomy (RP) or radiotherapy (XRT) at four Army medical centers. Mean follow-up was 8.02 years (range 0.026-32.5 years). Stage and grade were similar in patients receiving either RP or XRT. Kaplan-Meier estimates showed that patients who underwent RP had a significantly greater disease-specific survival (p = 0.0001) and a significantly lower rate of distant metastases (p = 0.006) than did those who received XRT. There was no significant difference in the rate of local progression (p = 0.276) or in the mean time to local progression (XRT = 3.5, RP = 4.0 years) or to distant metastases (XRT = 3.79, RP = 4.52 years). Cox proportional hazards model incorporating age, stage, grade, and treatment type demonstrated that those patients who received XRT had more than two times the risk of dying of their disease than did those who underwent RP (risk ratio = 2.37; 95% confidence interval = 1.49-3.76). These data in similar groups of patients suggest that metastasis-free survival is improved in those who receive RP compared with XRT and that this translates into an enhanced survival advantage. Further study of larger groups of patients stratified by risk factors in randomized, prospective trials with longer follow-up will improve our ability to determine the best treatment for clinically localized prostate cancer.

Infrequent RAS oncogene mutations in human prostate cancer.

The RAS gene family includes three functional genes, H-RAS, K-RAS, and N-RAS, which have been most widely studied in human tumors. Point mutations most commonly occurring at codons 12, 13, or 61 of these genes allow the RAS protooncogene to be converted to a RAS oncogene. A variety of human tumors have been studied for RAS mutations to date, however, conflicting data has been reported regarding prostate cancer. Cell line studies and two American studies of clinical material have found a low incidence of RAS mutation in prostate cancer. The few mutations found were predominantly in the H-RAS gene. Conversely, a recent study of Japanese occult autopsy specimens found an approximate 25% incidence of K-RAS mutations. In this current study, DNA was extracted from 24 archival paraffin-embedded, formalin-fixed radical prostatectomy specimens. Twenty-one of the 24 cases had pathologic stage C disease, and paraffin blocks were selected having the most concentrated area of neoplasm. Twelve, seven, and five cases demonstrated moderate, well and poorly differentiated histologic grade respectively. Polymerase chain reaction (PCR) was used to amplify the K-RAS, N-RAS, and H-RAS 12, 13, 61 codons of these specimens and mutations were detected with mutation-specific oligonucleotide probe hybridization of southern and slot blots. No definite point mutations were detected. PCR's and hybridizations were performed three separate times by three investigators to confirm these results. PCR-generated mutation-specific positive controls and known negative controls were used and found to be important to interpret oligonucleotide hybridization assays. RAS gene mutations appear to be infrequent in clinical prostate carcinomas in American males.