S. Egawa

A Comparison of the Morphological Features of Cancer Arising in the Transition Zone and in the Peripheral Zone of the Prostate

To determine the characteristics of transition zone and peripheral zone prostate cancer, we examined a series of 42 stage A and 54 stage B radical prostatectomy specimens with particular attention to the number of separate foci of cancer, zone of origin, volume and grade of each focus, and presence of severe intraductal dysplasia (high grade prostatic intraepithelial neoplasia), extra-capsular extension and seminal vesicle invasion associated with cancer in each zone. We found that there were fundamental differences between transition zone and peripheral zone cancers, and that the features that characterize these tumors were apparent in stages A and B disease. Although the total tumor burden was similar in stages A (3.98 cc) and B (4.56 cc) disease, stage A cancer tended to be multifocal (3.1 tumors per prostate) and more diffuse. While 81% of stage A prostate specimens contained a tumor of transition zone origin and 93% had cancer of peripheral zone origin, transurethral resection of the prostate sampled a transition zone cancer in 77% and a peripheral zone cancer in 31% (8% had both types). Stage B cancer tended to be more focal (2.3 cancers per prostate). All stage B prostate specimens contained a peripheral zone cancer and 43% had a transition zone cancer as well. In only 1 stage B cancer patient was the transition zone tumor the palpable or index cancer. In stages A and B disease, peripheral zone tumors were less well differentiated (median Gleason sum 6 and 7) than transition zone tumors (5 and 5, respectively) and more likely to extend through the capsule (44% versus 11%). Seminal vesicle invasion arose from 19% of the peripheral zone but none of the transition zone cancers. Peripheral zone tumors were almost always (93%) associated with high grade prostatic intraepithelial neoplasia, while none of the transition zone cancers was so associated. For peripheral zone disease there was a moderate correlation between volume and grade (tau = 0.46, p less than 0.001) so that the larger the tumor the higher the Gleason sum but within transition zone disease this correlation was poor (tau = 0.23) and not statistically significant (p greater than 0.05). Extracapsular extension occurred at a smaller volume with peripheral zone cancer (mean 3.86, minimum 0.06 cc) than transition zone cancer (mean 4.98, minimum 0.39 cc). Cancer that arises in the transition zone appears to have a different histogenesis, is associated with more favorable pathological features and may have less malignant potential than tumors that arise in the peripheral zone.

The Distribution of Residual Cancer in Radical Prostatectomy Specimens in Stage a Prostate Cancer

To assess the volume and distribution of residual cancer after transurethral resection of the prostate in stage A cancer patients 42 step-sectioned radical prostatectomy specimens were examined, and the volume, location, grade and extracapsular extension of the residual tumor were recorded. A total of 13 patients had stage A1 tumors (5% or less tumor in the transurethral resection specimen and a Gleason sum of 7 or less) and 29 had stage A2 disease. Residual cancer was present in the radical prostatectomy specimen in 41 patients (98%) with a mean volume of 1.28 cc. The location of residual cancer, that is multifocal (76%), peripheral (81%) and distal to the verumontanum (66%), makes complete removal or even identification of residual tumor (restaging) by repeat transurethral resection improbable. Of the stage A1 cancer patients 4 (30%) had more than 1 cc residual tumor volume, extracapsular extension or seminal vesicle invasion. On the other hand, 14 of the stage A2 cancer patients (48%) had less than 1 cc residual tumor completely confined to the gland. Foci of residual cancer were found in the transition zone in 67% and in the peripheral zone in 90% of the patients. The grade of the residual peripheral zone cancer was significantly higher than that of the transition zone cancer in the same gland (p = 0.0004). Eight of 13 instances of extracapsular extension and all 5 of seminal vesicle invasion were directly attributable to peripheral zone cancer. These observations imply that the greatest threat to patients with stage A prostate cancer may be a separate, associated cancer in the peripheral zone rather than the primary transition zone cancer incidentally removed at transurethral resection.

Detection of residual prostate cancer after radiotherapy by sonographically guided needle biopsy

Detection of persistent or recurrent prostate cancer by digital rectal examination (DRE) after definitive radiotherapy is difficult. With the availability of transrectal ultrasonography (TRUS), the detection of prostate cancer has improved substantially. Since 1987 we have used TRUS to evaluate the prostate after definitive radiotherapy. A hypoechoic lesion suggestive of cancer was identified in 45 of 56 patients (80%) studied. Sonographically directed transrectal needle biopsies were performed in 27 of these (60%), and 16 (59%) were positive for cancer. The presence of a palpable nodule suggestive of cancer (present in 7 patients) was not predictive of a positive biopsy specimen. In 14 patients ultrasound-guided and digitally-guided biopsies were performed at the same time; 8 (57%) of the ultrasound-guided biopsy specimens were positive compared with only 4 (29%) of the digitally-guided biopsy specimens. In all 7 patients with an elevated serum level of prostate-specific antigen (PSA) an ultrasound-guided biopsy result was positive. Random biopsies of sonographically normal (isoechoic) areas of the prostate were performed in 8 patients, but only 2 specimens (25%) were positive for cancer. Ultrasound-guided transrectal biopsy of hypoechoic lesions was a safe and effective technique for identifying residual cancer in the irradiated prostate, regardless of the palpable findings. In the presence of an elevated PSA level, such biopsies invariably identified residual cancer. The use of TRUS, ultrasound-guided biopsy, and the measurement of PSA, in addition to DRE, may aid in the detection of residual cancer after definitive radiotherapy.

Transrectal Ultrasonography in Stage a Prostate Cancer: Detection of Residual Tumor after Transurethral Resection of Prostate

To evaluate the ability of transrectal ultrasonography to detect residual cancer in the prostate gland after transurethral resection in patients with stage A cancer, we studied 38 patients with stage A disease (11 stage A1 and 27 stage A2) in whom transrectal ultrasonography was done at least 3 weeks after resection. Each patient underwent radical prostatectomy, and residual cancer was present in 97% of the specimens (peripheral zone cancer in 95% and transition zone cancer in 61%). At sonography we identified hypoechoic areas suggestive of cancer in 10 patients (26%). In the pathological specimen residual cancer was present at the hypoechoic area in 8 of these cases (positive predictive value 80%). In a retrospective review of the sonograms we identified 25 hypoechoic lesions greater than 5 mm. in diameter, including 15 that corresponded to cancer in the radical prostatectomy specimens (positive predictive value 60%). Granulomas due to the transurethral resection were found in 92% of the radical prostatectomy specimens but none appeared hypoechoic on ultrasound. A total of 103 separate cancers was identified in the whole mount step sections of the radical prostatectomy specimens (2.7 cancers per patient). Of the 103 separate cancers 54 were less than 0.1 cc in volume and none of these could be identified in the retrospective review of the sonograms, 37 were 0.1 to 1.0 cc and 5 of these (14%) appeared hypoechoic, and 12 were greater than 1.0 cc and 10 of these (83%) appeared hypoechoic. Hypoechoic lesions greater than 5 mm. in diameter in the transition zone proved to be cancer in 47% of the cases, while 88% of similar lesions in the peripheral zone proved to be cancer. We conclude that suspicious-appearing hypoechoic lesions suggestive of cancer, whether in the peripheral zone or the transition zone, should be biopsied before expectant management of stage A prostate cancer is considered. Transrectal ultrasonography is useful for restaging after transurethral resection and for evaluating the extent of residual cancer in stages A1 and A2 prostate cancer.

Androgen sensitivity and gene expression in ras + myc-induced mouse prostate carcinomas

We established an androgen-sensitive cell line (BR31-5) from a ras + myc-induced mouse prostate carcinoma and used this cell line together with a previously reported transplantable androgen-independent mouse prostate carcinoma to investigate patterns of expression for apoptosis-related genes in an androgen-deprived environment. Single cell suspensions derived from the BR31-5 cell line were inoculated into the flank of intact or castrated adult male C57BL/6 mice and tumors were harvested 12 days post-inoculation for Northern blotting. A transplantable androgen-independent prostate cancer was also inoculated into intact or castrated mice and tumors harvested 21 days later. Tumor volume analyses showed that BR31-5 carcinomas were androgen-sensitive. Northern blotting showed that mRNA levels for two apoptosis-related genes, transforming growth factor-beta 1 and c-myc, were significantly elevated to a similar extent in carcinomas grown in castrated hosts compared to intact hosts for both the androgen-sensitive BR31-5 and androgen-independent carcinomas. Levels of mRNA for tissue type plasminogen activator, shown previously to be elevated in androgen-independent carcinomas following growth in castrates, were also increased in BR31-5 carcinomas under similar androgen-deprived conditions but to a lesser extent. Interestingly, testosterone repressed prostate mRNA No. 2 levels shown previously to be similar in both the intact and castrated groups for androgen-independent carcinomas were significantly increased in the castrated group compared to the intact group for BR31-5 carcinomas. Therefore, specific patterns of expression for apoptosis-related genes may be able to discriminate androgen-sensitive and androgen-independent prostate cancer under androgen-deprived conditions.

Sonographic monitoring of prostate cancer after definitive radiotherapy

Thirty patients who had transrectal ultrasonography before and after definitive radiotherapy were studied retrospectively to determine the effects of radiotherapy on the sonographic appearance of prostate cancer. Before therapy one or more discrete hypoechoic areas characteristic of cancer were present in 29 (97%) of the patients. In 10 patients (34%) the hypoechoic areas disappeared six to twenty-seven months (mean 11.4) after radiotherapy, but in 3 of these the hypoechoic lesion subsequently reappeared. Six months after radiotherapy a hypoechoic lesion could still be seen in the original area in 79 percent of 19 patients studied. Sonography showed persistent lesions in 65 percent of 17 patients at twelve months, in 79 percent of 14 patients at twenty-four months, and in 75 percent of 8 patients at thirty-six months. In 9 of the 29 patients (31%), there was a measurable increase in the size of the lesion, but overall, the size (maximum diameter) of the hypoechoic lesion had decreased by a mean of 41 percent when evaluated twelve months after radiotherapy. Previous studies from our laboratory have shown that persistent prostate cancer after definitive radiotherapy retains its hypoechoic appearance, and this study indicates that these characteristic hypoechoic lesions can be monitored by transrectal ultrasound.

Evidence for the Involvement of Genetic Differences and Mesenchymal Factors in the Progression of Oncogene-Induced Prostate Cancer in Reconstituted Mouse Prostate

Carcinogenesis is a multi-step process involving the accumulation of alterations within the cellular genome (reviewed in 1,2). These alterations can occur in protooncogenes converting them into activated oncogenes, and it has been suggested that the multi-step nature of carcinogenesis may partly reflect the consecutive activation of several dominantly acting oncogenes (3-7). It seems reasonable to presume that prostatic cancer follows this pattern. In support of this notion, oncogene activities have been reported to be present in human prostatic adenocarcinoma (8-10), in cell lines derived from cancerous prostatic tissues (11), and in Dunning R3327 rat prostatic adenocarcinomas (12,13). Recently, utilizing the mouse prostate reconstitution model system, it was demonstrated that activated ras or myc oncogenes singly were able to induce distinct dysplastic and hyperplastic lesions, respectively (14). The two oncogenes in combination induced rapidly growing, poorly differentiated malignant adenocarcinomas. Thus, the activation of the ras and myc oncogenes can induce a step-like progression of carcinogenesis in this experimental animal model.