On the date of surgery, the prostate specimen will be processed according the method developed by Michael J. Becich, M.D., Ph.D. at the University of Pittsburgh Medical Center. This method provides the highest quality of surgical pathology for the patient, while allowing sufficient tissue to be banked for further research. Only after all surgical pathology requirements are met will the tissue be released for research purposes.

Specimen receiving and initial processing: The operating room personnel call the surgical pathology gross room 15-20 minutes before anticipated completion. This ensures that, by the time the surgical pathology specimen is completely excised and placed in a sterile container, a pathology assistant from the Tissue Bank is there in the OR for specimen pick up. In instances when the processing is delayed by more than 5 minutes, the prostate is placed in iced saline or chilled tissue culture media. Because of the high level of protease activity in the prostate, coordination of this initial processing is critical. Delays of 15 minutes or more result in degradation of RNA and membrane proteins. Additionally, two vacutainers of blood, with sodium heparin as anti-coagulant, are drawn from existing IV's, collected during surgery and subsequently picked up by the pathology assistant.

Separation of Blood: Once received by the Tissue Bank personnel, an aliquot of the blood is taken as a whole blood sample, and the remaining blood is spun at 3500rpm for 5 minutes. The resulting serum, buffy coat, and red blood cell components are separated, labeled appropriately, catalogued in the database, and stored with the whole blood component in a -700C freezing unit within the Tissue Bank facilities for future research.

Gross Examination: The gross examination and tissue processing is performed by the pathology assistant or resident/fellow supplied by the Tissue Bank with the supervision and assistance of participating institution's Department of Pathology personnel. These personnel (including PA's and residents) will also be trained in the methodology in the unlikely event that Tissue Bank personnel are unavailable. This ensures that all patients enrolled in the study are banked. The specimen is immediately brought to the gross room, where it is weighed and gross measurements are recorded. A detailed description of the physical characteristics of the prostate, urethra, seminal vesicles, capsule and associated tissues is recorded. It is then placed under a sterile hood, and examined. Any asymmetry, bulges or firm areas suspicious for cancer are noted. The outer surface of the specimen is dried with sterile towels, and India ink applied to the surface of the prostate and seminal vesicles.

Organ Sectioning: At this stage, the proximal bladder neck margin and distal urethral margin are removed, and the specimen is oriented such that the posterior surface is face up. The seminal vesicles are removed as one segment of tissue, such that orientation is maintained, and this resultant segment of tissue contains both the seminal vesicles and the seminal vesicle/prostate junction.

A transverse cut is made through the center of the specimen for easier handling. The specimen is then serially sectioned proximal to distal in 1-2 mm thick slices. These are serially placed with the posterior surface lying superiorly. The cut surfaces of all slices are examined for tumor, which grossly appears as a slightly raised, firm, yellow or white areas distinct from the surrounding parenchyma. If the tumor is readily discernible and present in a solid, nodular form, the dimensions, location and extent are noted. In case the tumor is present in more than one slice, the slice which shows tumor extending closest to the capsular/surgical margin is initially selected for whole mount formalin fixation processing. If the tumor is multifocal and present in more than one lobe, slices which contain tumors closest to the margins in the different lobes are selected for formalin fixation processing. Once this decision is made, alternate slices are picked up for whole mount formalin fixation processing, excluding the most proximal and most distal slices as mentioned earlier.

Tissue Banking and Cell Culture: The slices adjacent to the slices selected for formalin fixation are further examined. If any tumor is seen which corresponds to the tumor seen in the slice selected for formalin fixation, touch preps are made for the entire left and right posterior quadrants of the prostate, and a 1 cm by 1 cm piece is cut through the full thickness of the slice. This piece is then halved. A very small piece of tissue (approx. 1mm x 1mm) is sectioned off of one of the halves and placed in Karnovsky's fixative for EM embedding, and a another small portion of tissue (approx. 3mm x 3mm) is embedded in OCT embedding media and frozen for future use. The remaining portion from this same half is sent for routine formalin fixation processing labeled "CULT" (culture tumor). This acts as a control to determine that cancerous tissue is actually taken for cell culture. The other half is processed in either of the following two ways: If there is adequate tumor seen grossly, this half piece is immediately cultured according to our protocol (described later). Additionally, if this tumor is close to the capsular margin, the margin is removed for formalin fixation and paraffin-embedding for clinical diagnosis. In case only suspicious areas not definite for tumor, or very insignificant amount of tumor are seen, this half piece is placed in MEM media in a cryostat cassette and slow frozen to -700C. This half piece can be stored indefinitely in MEM media, or until the final surgical pathology report is completed. If this tissue is not required for surgical reporting, it is quick thawed and then cultured according to our protocol. (described later). We have been successful in storing and retrieving tissue at a later date, with good culture growth with the method described above. We have also retrieved the tissue stored in this manner, and sent for routine formalin fixation and processing, with comparable quality of H&E sections. Similar 1 cm by 1 cm pieces of tissue are taken from a normal and hyperplastic areas. Each piece is then halved, and one half is sent for formalin fixation processing labeled CULN (culture normal), and CULH (culture hyperplastic) respectively. These serve as controls for the cell cultures of these areas. The other halves are then processed for short term in vitro cell culture. The remaining slices not selected for whole mount formalin fixation are then snap frozen in liquid nitrogen in autoclaved aluminum foils which are labeled with the corresponding alphabetical letter, and with regard to the orientation of the slices. These are then stored at -700C. Thus, all tissue harvested for research purposes is available for diagnostic purposes, if needed, until the surgical report is finalized. To date, the need for using tissue frozen at -700C for diagnostic purposes has not arisen. In one case, due to technical problems in the histology processing laboratory, all three whole mount slices were unavailable for evaluation. We retrieved the frozen tissue slices, and processed them following formalin fixation. The H&E sections prepared from these slices underwent the same diagnostic evaluation process as described above with the same satisfactory results. Once the case is signed out, the tissue are released for research purposes.

Processing of Remaining Specimen for Diagnostic Examination: The slices selected for formalin fixation are placed in a plastic box, filled with 10% neutral/phosphate buffered formalin. Fixation is done for 8 to 12 hours. Since positive surgical margins is as strong an independent predictor of progression as are Gleason grade and seminal vesicle invasion, extra care is taken in processing and evaluation of these. Moreover, the apices are a frequent site where tumor involves the margins. Both the proximal bladder neck margin and the distal urethral margin are sectioned into left and right segments. Each segment is processed by conization and submitted for pathologic examination. No attempt is made to obtain the urethral mucosa as it usually gets retracted following excision and/or denuded by catheterization. It is also a very unlikely site of involvement by prostate adenocarcinoma.

Interpretation of Margins: The histotechnologists are instructed to orient all the above sections on the microtome with minimal shaving, especially sections representing the bladder neck and apical margins. Any tumor on the inked margin (ink on tumor) seen in these sections is interpreted as a positive margin. The seminal vesicles are cut at the base. They are then bisected longitudinally in a plane parallel to the prostate and examined for any involvement by tumor. Routinely, two sections from the base and the distal tip of each seminal vesicle are submitted along with the surgical margin of vas deferens of the same side in cassettes labeled "LSD" and "RSV". Finally the empty container is weighed, and the weight of the specimen is calculated. Following fixation of whole mount sections for 6-8 hours, they are submitted in special large cassettes and labeled on the side annotated with the orientation of the whole mount sections.

Cell Disaggregation and Culture from Harvested Fresh Human Prostate Tissues: The tissue slices sampled for prostate epithelial cell isolation are sequentially digested with collagenase for 5-6 hours in "Chemically Defined Media" (CDM) consisting of Dulbecco's Modified Eagle Medium with amino acid supplements, glucose, 5% bovine (non-fetal) serum, protease inhibitors, and antibiotics after an initial exposure to Ca⁺⁺ free buffer containing 1mM EDTA. The cells are then incubated for 45 minutes in buffer to remove tissue fragments and debris. Following this step, the tissue undergoes sequential collagenase digestions, as the prostate epithelium is "released" in a non-uniform manner from the tissue fragments. The first digestion is for 1-2 hours, and the supernatant is discarded. The second and third collagenase digestions yield the highest amount of viable prostatic epithelial cells, and are immediately processed for cell attachment to gelatin-coated culture dishes containing CDM. The fourth and fifth sequential collagenase digestions with fresh enzyme also yield epithelium, but in smaller amounts than the second and third digestions. After isolation, cell attachment may take 4-16 hours, then serum is removed and cells are grown in serum-free CDM. This washing process may take 1-2 hours. Cultures containing attached prostatic epithelium and stromal elements are ready for experimentation 12-24 hours after receipt. The average yield per gram of prostate tissue is approximately 1x106 cells per 60mm plate after 3-5 days. Methodology is available to specifically enhance growth of epithelial cells devoid of stromal elements.

Cells not immediately needed or used for experimentation are cryopreserved at -120^oC following a cell-preparation and trypsinizing protocol. Cell characterizations of our cultures were performed to verify their epidermal derivation. The collagenase digestion of prostate surgical specimens results in the isolation of cells of two types; epithelial and fibroblasts. The two cell types can be distinguished on the basis of morphology and immunocytochemistry. The polyhedron-shaped epithelial cells maintain tight junctions with neighboring epithelial cells in culture. Fibroblasts, in comparison, attach to the dish scattered singly with elongated cell processes. Immunohistochemical staining demonstrated that epithelial cells stained positive for cytokeratin AE1 and AE3, as well as PSA. Fibroblasts were negative for cytokeratin but positive for vimentin. Growth factors such as IGFI, IGFII and EGF increase the total cell number of both normal and tumor prostate cells as measured by cell proliferation assays. Tumor cells, however, respond differently to the profile of growth factors than the benign cells. The differentiating factor, retinoic acid, inhibits prostate cell proliferation in a dose-dependent manner. When both insulin and retinoic acid are added together, an increase in the cell number occurred in the range of the effect of adding insulin alone. These combination experiments may help to explain why compounds which appear promising in vitro are disappointing in clinical trials. This set of experiments will be repeated for different growth factors and also on cells derived from tumor specimens to determine if they respond in the same manner to combinations of growth factors and chemopreventive agents. To verify the nature of the tumor cells compared to benign prostate cells, electron microscopy of both types of cells from the same donor show that the benign cells had regular nuclei and tight junctions between neighboring cells. EM of the cancer cells displayed multilobed nuclei with many instances of apoptosis in nearby cells. Antibody studies showed E-cadherin was expressed 50% less in tumor cells compared to benign cells.

Grid-mapping for accurate sampling of whole-organ cross-sections: Following the rendering of pathological diagnosis, the paraffin-embedded whole mount sections taken during sectioning are examined from the H&E stained slides, and areas of malignancy, normal, benign prostatic hypertrophy, and prostatic intraepithelial neoplasia are outlined and stored on a digital image capture program. Special care is taken to note specimen orientation, and anatomical landmarks associated with the structure of the prostate. Since these paraffin-embedded sections are mirror images of the cross-sections frozen for use by the bank, accurate sampling of pathologic tissue is ensured.

Diagnostic reporting: Once complete gross and histologic examination of the prostate is completed, a final surgical pathology report with diagnosis is generated. This report is formatted following the structure of a unique synoptic reporting format developed by Dr. Michael Becich. This synoptic report (see Appendix A) serves to appropriately organize information for more accurate and standardized pathology reporting, as well as providing a template for database information collection.

Database information: Only after a surgical pathology diagnosis is rendered can the tissue banked according to the above protocol be released for research. At this time, the tissue bank personnel compile a patient profile from information existing in hospital medical records, computer databases, etc.. This information will include patient name, social security number, birthdate, race, pre-op biopsy results, pre-op PSA values, clinical history, family history, medications, secondary medical conditions, Gleason grading of tumor, tumor size, tumor volume, margin assessment, histological grading, CT scan results (pre-op), bone scan results (pre-op), and surgery date. Additional patient follow-up can be accessed for disease progression as needed, and current plans for follow-up information inclusion include 3-month, 6-month, 1-year, and subsequently yearly updates to include a format similar to pre-op and surgical information.. This information will be stored on a database to which only Dr. Becich and his Tissue Bank director will have access. This will ensure the utmost amount of patient confidentiality, and integrity of research.

Researchers will prepare written requests for tissue/blood, and submit them to Dr. Becich and the Tissue Bank along with any specifications for patient demographics, tumor characteristics, or clinical information needed that correspond to their research objectives. It is expected that each investigator will have applied for and received IRB approval for their specific study. This information will be requested by the Tissue Bank personnel at the time the investigator makes the tissue request. The Tissue Bank personnel will utilize the database only to identify qualifying patient tissue, and patient-identifying information will not be released to any of the investigators.