Management of Clinically Localized Prostate Cancer: Pathologic Processing to Robotic Prostatectomy

13TH INTERNATIONAL PROSTATE CANCER UPDATE Management of Clinically Localized Prostate Cancer: Pathologic Processing to Robotic Prostatectomy Ashutosh Tewari, MD Josephine Ford Cancer Center, Detroit, MI Additional Contributors Robert E. Donohue, MD, Jonathan L. Epstein, MD, Randall B. Meacham, MD, Mani Menon, MD, Cliff Vestal, MD, Timothy J. Wilt, MD, MPH Prostate cancer management has undergone significant changes in recent years. The existing modalities are being refined, and a better understanding of pathologic parameters is helping in decision making. Newer sources of energy can be used to kill cancer cells, and modern robots are lending a helping hand in surgical removal of prostate cancer. The 13th International Prostate Cancer Update meeting provided a unique opportunity to review these recent developments. Reviewed here are highlights of presentations addressing recent developments in the surgical management of prostate cancer. The first discussion involved the processing of surgical specimens and interpretation of information obtained from histopathologic analysis. The second topic discussed here focused on an ongoing randomized trial comparing surgery with watchful waiting in the management of clinically localized prostate cancer. The last detailed presentation covered robotic prostatectomy and compared it with conventional open radical prostatectomy. [Rev Urol. 2003;5(suppl 6):S33-S39] © 2003 MedReviews, LLC Key words: Prostate cancer • Histopathology • Randomized trial • Radical prostatectomy • Robotics rostate cancer is the most common solid tumor in men. It is estimated that 220,900 new cases will be diagnosed in the United States in 2003 and that 28,900 patients will die of this cancer.1 Surgical treatment of localized prostate cancer is offered to an estimated 70,000 patients each year. E. David Crawford, MD, recently organized the 13th International Prostate Cancer Update in Vail, Colorado, P VOL. 5 SUPPL. 6 2003 REVIEWS IN UROLOGY S33 Clinically Localized Prostate Cancer continued and invited various experts to summarize the current status of surgical and other nonradiation, minimally invasive modalities of treating prostate cancer. This article summarizes three presentations from the meeting. Jonathan L. Epstein, MD, provided an update on the recent developments regarding pathologic assessment of radical prostatectomy specimens. Timothy Wilt, MD, discussed the current status of an important randomized trial comparing radical prostatectomy with watchful waiting (VA, NCI, AHRQ Cooperative Study #407: Prostate Cancer Intervention Versus Observation Trial [PIVOT]), and Ashutosh Tewari, MD, and Mani Menon, MD, presented results of a nonrandomized prospective study comparing robotic prostatectomy with conventional prostatectomy. The three other presentations at this session included an elegant study by Robert Donohue, MD, that looked at the veteran population of Denver, Colorado, reporting long-term survival and the incidence of second malignancies in patients who had undergone radical prostatectomy; a discussion by Randall Meacham, MD, on recent advances in management of erectile dysfunction in patients undergoing surgical treatment of prostate cancer; and an update on cryotherapy by Cliff Vestal, MD, which concluded the session. Pathologic Assessment and Staging of Prostate Cancer Dr. Epstein, from Johns Hopkins Hospital, summarized current data on the prognostic value of radical prostatectomy specimen parameters.2 He focused on the following important areas. Specimen Handling and Reporting of Margins and Seminal Vesicles Dr. Epstein stated that, although thin shaving of the apical margin, a commonly performed procedure, allows S34 VOL. 5 SUPPL. 6 2003 analysis of the entire surface of the distal margin, it is fraught with a high incidence of false-positive margin status because of misclassification of tumors close to, but not extending through, the specimen. Therefore, his preferred technique is to transect the distal 1 cm of the prostate and process it as perpendicular margins. He also mentioned that seminal vesicles should be analyzed by taking a section through the base of the seminal vesicle where it joins the prostate. able tumor on gross examination.2 The question arises as to how to sample radical prostatectomy specimens that are clinically nonpalpable and not grossly visible on pathologic examination. Sampling must take into account that in 15% to 25% of stage T1c tumors the dominant lesion is located primarily within the transition zone anteriorly. If a pathologist samples only the posterior aspect of the prostate, 92% of tumors that are Gleason score 7 or greater, If a pathologist samples only the posterior aspect of the prostate, 92% of tumors that are Gleason score 7 or greater, 86% of cases with positive margins, and 84% of cases with extraprostatic extension will be detected. Almost all cases of seminal vesicle invasion first involve the base of the seminal vesicle, directly through the ejaculatory duct or by extension out of the prostate into periprostatic soft tissue and back into the seminal vesicles. Processing of the Specimen Because of cost issues, the strain on laboratory resources, and the time required to process an entire specimen, 80% of pathologists have adopted various methods of partial sampling. Dr. Epstein recommended that if pathologists can identify tumor grossly in the radical prostatectomy specimen, they should process only those sections that have gross tumor, as well as use the previously mentioned method of margin and seminal vesicle evaluation.3 Use of this method can detect 96% of cases with positive margins and 91% of cases with established extraprostatic extension. However, in the current era of prostate-specific antigen (PSA) testing for detection of prostate cancer, lesions are often not grossly visible. In a recent study conducted at Johns Hopkins, only 26% of patients with stage T1c disease had identifi- REVIEWS IN UROLOGY 86% of cases with positive margins, and 84% of cases with extraprostatic extension will be detected. If, in addition to the posterior sections, midanterior sections from the right and left midgland are sampled and these sections show significant tumor anteriorly, the entire ipsilateral anterior portion of that side is also submitted. In this situation, the clinician can detect 98% of tumors that are Gleason score 7 or greater, 100% of cases with positive margins, and 96% of cases with extraprostatic extension. Pelvic Lymph Node Assessment Because of improved patient selection and earlier detection of prostate cancer through the widespread use of PSA testing, the incidence of pelvic lymph node metastases at the time of radical prostatectomy has decreased over the last few decades to 1% to 2%. The handling of lymphadenectomy specimens at the time of surgery is controversial and depends on the philosophy of the urologist. Every effort should be made to identify lymph node metastasis if surgery will be aborted based on the results. A practical approach is to preoperative- Clinically Localized Prostate Cancer ly identify clinical parameters associated with very low risk of lymph node metastases, thus avoiding the need for frozen sections in these cases. In the remaining cases, the pathologist should freeze two or three small, firm lymph nodes from each side. In Dr. Epstein’s experience, this practice will identify more than half of small metastases. However, if the prostate will be removed regardless of microscopic node status, the lymph nodes should not be frozen unless they have a Gleason score of 8 or higher, because, in Dr. Epstein’s experience, these patients can not be cured. Dr. Epstein also emphasized that, because of the lack of convincing evidence regarding prognostic utility of size, largest diameter, and DNA status of lymph nodal metastasis, he prefers only to report on the number of positive nodes. Definition and Implications of Seminal Vesicle Invasion Seminal vesicle invasion is defined as tumor infiltrating the muscular coat of the seminal vesicle. Controversy continues regarding whether seminal vesicle invasion can occur without extraprostatic extension. Dr. Epstein restricts the diagnosis of seminal vesicle invasion to patients who have invasion of structures exterior to the prostate. By definition, once seminal vesicle invasion has taken place, the tumor is no longer confined to the prostate, and there must be extraprostatic extension. If tumor extends through the ejaculatory duct into the seminal vesicles but not into the periprostatic soft tissue, the patient is classified as not showing “capsular penetration." The finding of seminal vesicle invasion in a radical prostatectomy specimen is related to poor prognosis. In men with seminal vesicle invasion and negative results on analysis of pelvic lymph nodes, 5-year biochemical progression-free rates range from 5% to 60% (mean, 34%).4 Extraprostatic Extension Because the prostate lacks a discrete capsule, the term “extraprostatic extension" has replaced “capsular penetration" to describe tumor that has extended out of the prostate into periprostatic soft tissue. In some cases, because of the desmoplastic response, it can be difficult to judge whether the tumor has extended out of the gland or is within the fibrous tissue of the not correlate with progression. Margins of Resection The pathologic definition of positive margins of resection seems straightforward, that is, according to Dr. Epstein, “tumor extending to the surface of the prostate which the surgeon has cut across." Dr. Epstein has demonstrated that, as long as tumor is not actually cut across and at the ink, close margins (< 0.1 mm) should not be designated positive margins because they are not associated with an increased risk of postoperative As long as tumor is not actually cut across and at the ink, close margins (< 0.1 mm) should not be designated positive margins because they are not associated with an increased risk of postoperative progression. prostate. This can be resolved by looking at the adjacent edge of the prostate on scanning magnification to see whether the normal rounded contour of the gland has been altered by a protuberance corresponding to extension of tumor into the periprostatic tissue. In palpable cancers (T2), extraprostatic extension preferentially occurs posteriorly and posterolaterally by perineural invasion. In transition zone cancers found in stage T1b and some T1c tumors, extraprostatic extension occurs anteriorly, usually by direct stromal invasion without prominent perineural invasion. Dr. Epstein designates cases in which only a few glands are seen outside the prostate as “focal" (less than one high-power field on no more than two sections), whereas cases with more extensive extraprostatic spread are designated “established extraprostatic extension." Other methods of quantifying the extent of extraprostatic extension based on the depth of penetration, the width of penetration, or the length multiplied by the width do progression. One cause of a positive resection margin is transection of intraprostatic tumor (capsular incision). These cases should be designated as pathologic stage T2X. The percentage of positive margins resulting from capsular incision ranges from as low as 1.3% to as high as 71%.2 Capsular incision most frequently occurs posterolaterally, where less experienced surgeons inadvertently cut into the prostate in an attempt to preserve the neurovascular bundle. Capsular incisions can also occur at the apex, where they contribute to 45% to 87% of all positive margins. The high incidence of capsular incision at the apex might result from an artifact of sectioning the apex. At the apex, if Dr. Epstein sees tumor extending to the inked edge of the prostate in an area where benign prostate gland is not cut across, a positive margin is diagnosed, not capsular incision. Dr. Epstein believes that most positive margins resulting from capsular incision are not true positive margins because they are not associated with VOL. 5 SUPPL. 6 2003 REVIEWS IN UROLOGY S35 Clinically Localized Prostate Cancer continued a higher risk of progression after radical prostatectomy. On the other hand, positive margins can also result from a failure to excise widely extraprostatic extension of tumor. One of the most frequent sites of extraprostatic extension is posterolaterally in the region of the neurovascular bundle. However, Dr. Epstein its palpable nature, 15% to 24% of radical prostatectomies performed for stage T1c disease reveal predominantly anterior or transition zone tumor. Although there is some overlap, there is a tendency to see a relative increase in anterior positive margins in stage T1c tumors compared with stage T2 tumors. For both Although the decrease in positive margins might be attributable in part to improvements in surgical techniques, most of the decrease probably relates to more careful patient selection and stage migration. believes that there is no statistically significant difference in the incidence of positive margins that occur in nerve-sparing and non–nerve-sparing radical prostatectomy. Although with nerve-sparing techniques some of the positive margins occur posterolaterally in the region of nerve sparing, it is uncommon (0%–7%) for the sole positive margin to be in the area of the nerve preservation.2 Dr. Epstein states that nerve sparing should be avoided if there is ipsilateral induration. A review of radical prostatectomies performed by Dr. Epstein revealed that the incidence of positive margins has decreased dramatically over the years—from 41% in the 1980s to 5.8% in 1999.2 Although the decrease in positive margins might be attributable in part to improvements in surgical techniques, most of the decrease probably relates to more careful patient selection and stage migration. of these stages, the apical region is still the most common site of positive margins. The other frequent sites of positive margins are posterior and posterolateral. It is uncommon for the bladder neck to be the sole site of a positive margin. Usually, when the bladder neck margin is involved, other areas show positive margins as well.2 Sites of Positive Surgical Margins Because stage T1a and T1b tumors tend to predominate within the transition zone, extraprostatic extension preferentially occurs anteriorly and apically, which are also the most common sites of positive margins.2 Although most stage T2 disease is posterior in location, accounting for ctomy ranges from 81% to 83% for margin-negative disease and falls to 58% to 64% for margin-positive disease.2 Several studies have demonstrated that apically positive margins do not correlate independently with progression. In contrast, nonapical positive margins are associated with a significantly increased risk of pro- S36 VOL. 5 SUPPL. 6 2003 Relationship of Margin Positivity to Progression Patients with positive margins have a significantly increased risk of progression compared with patients with negative margins. Data from 2 studies have shown that the probability of a patient remaining progressionfree at 5 years after radical prostate- gression compared with negative margins. Several studies have suggested that margins positive at the bladder base indicate a higher risk of progression. Gleason Grade Dr. Epstein then discussed tumor grade and highlighted the importance of Gleason score 7 with 4+3 pattern and also emphasized the evolving prognostic importance of tertiary patterns.2 Tumor Volume Dr. Epstein stated that, once other routinely assessed variables are taken into account, tumor volume might not be an independent prognostic parameter. Furthermore, measurement of tumor volume in the prostate is not straightforward. Prostate cancers are not easily appreciated grossly and tend to have irregular shapes with infiltrative growth patterns. Using computer-assisted image analysis systems might not be practical for most institutions. Contrary views, from Stanford, have demonstrated that tumor volume measured by an image analysis system is an independent prognostic parameter even when factoring the percentage of Gleason pattern 4/5, vascular invasion, lymph node metastases, seminal vesicle invasion, extraprostatic extension, and Because of better patient selection and earlier detection of prostate cancer through PSA testing, the incidence of pelvic lymph node metastases at the time of radical prostatectomy has decreased. REVIEWS IN UROLOGY positive margin status.5 It should be noted, however, that many of the radical prostatectomies in this study were performed before the PSA screening era and involved high tumor volumes (mean, 4.66 cm3).2 Perhaps the independent prognostic significance of tumor volume in this study was driven by the broad spec- Clinically Localized Prostate Cancer 1.00 Risk of PSA Recurrence (n = 429) oc or epe, mar-, 3+4 0.75 (n = 26) fepe, mar-, 4+3, or fepe, mar+, 3+4 (n = 97) 0.50 eepe, mar-, 4+3 or eepe, mar+, 3+4 (n = 18) 0.25 Figure 1. Results of multivariate analysis regarding impact of various histopathologic parameters on PSA recurrence. oc, organ-confined; epe, extraprostatic extension; mar, margin; fepe, focal extraprostatic extension; eepe, extensive extraprostatic extension. epe, mar+, 4+3 P < 0.00001 0.00 0 2 4 6 8 10 12 14 16 Years Following Radical Prostatectomy trum of large tumors, which invariably have a poor outcome, and by the presence of smaller tumors that invariably are eliminated. Several surrogates of tumor volume have been investigated, including “percentage of specimen" involved by cancer and “maximum tumor diameter." There are conflicting data regarding whether tumor volume provides independent information beyond that routinely recorded from a radical prostatectomy specimen. Until more data suggest that tumor volume provides novel prognostic information, Dr. Epstein does not believe that it is necessary to calculate tumor volume routinely at radical prostatectomy. Multivariate Analysis to Predict Progression In studies at Johns Hopkins, Gleason score, extraprostatic extension status, and margin status all contributed to the optimal prediction of prognosis after radical prostatectomy. The power of using all of these pathologic parameters is demonstrated in Figure 1. The clinician can use the routine information provided within the pathology report to accurately stratify men according to their risk of progression after radical prostatectomy based on grade, margins, and organ-confined status. Treatment Options for Localized Prostate Cancer Dr. Timothy J. Wilt, presented an update on VA, NCI, AHRQ Cooperative Study #407: Prostate Cancer Intervention Versus Observation Trial (PIVOT).6,7 In the absence of definitive information from randomized controlled trials, patients seek information and recommendations from many sources. Acceptable treatment options for men with localized prostate cancer include radical prostatectomy, radiation therapy, cryotherapy, early androgen suppression therapy, and watchful waiting. This study aims to answer the following question: Does radical prostatectomy for clinically localized prostate cancer reduce allcause and disease-specific morbidity and mortality compared with expectant management? The inclusion criteria of PIVOT included an age less than 75 years; cancer stage T1-2, Nx, M0; and all histologic grades. Diagnosis must have been made within the previous 12 months, and the patient should be a radical prostatectomy candidate. The exclusion criteria included a life expectancy of less than 10 years; prior prostate cancer therapy (except transurethral resection of the prostate); nonlocalized prostate cancer on imaging studies or clinical examination; PSA level greater than 50 ng/mL; positive bone scan results; and previous androgen suppression therapy. The primary endpoint of the study is all-cause mortality; secondary endpoints are cancer-specific mortality and morbidity, metastatic disease, treatment-related mortality and morbidity, and quality of life. The study is powered at 90% to detect a 30% difference in all-cause mortality and at 85% to detect a 25% difference in cancer-specific mortality and morbidity. A total of 731 patients have been Main Points • For accurate analysis of prostate cancer, it has been recommended that surgeons transect the distal 1 cm of the prostate, to be processed as perpendicular margins, and sample a seminal vesicle by taking a section through the base where it joins the prostate. • Neither close margins nor positive margins resulting from capsular incision should be designated true “positive margins," because they are not associated with an increased risk of postoperative progression of prostate cancer. • It remains to be determined whether radical prostatectomy for clinically localized prostate cancer reduces all-cause and diseasespecific morbidity and mortality more than does watchful waiting; investigations are under way. • Robotic prostatectomy seems thus far to be a beneficial, and likely safer, alternative to conventional retropubic prostatectomy, with favorable oncologic and functional results. VOL. 5 SUPPL. 6 2003 REVIEWS IN UROLOGY S37 Clinically Localized Prostate Cancer continued 1.0 .8 .6 .4 Surg Code VIP .2 VIP-censored RRP 0.0 Probability of Return of Erection Probability of Return of Continence 1.0 .8 .6 .4 Surg Code VIP .2 VIP-censored RRP 0.0 RRP-censored –0.2 RRP-censored –0.2 –100 0 100 200 300 400 500 600 0 700 100 Figure 2. Mean probability of return of continence after radical retropubic prostatectomy (RRP) and robotic Vattikuti Institute Prostatectomy (VIP). Reproduced with permission from Tewari et al.12 Probability of Return of Intercourse .6 .5 .4 .3 Surg Code .2 VIP VIP-censored .1 RRP 0.0 RRP-censored –0.1 200 300 400 500 600 700 Follow-up (d) randomized in PIVOT. Mean age of the patients at baseline was 67 years; 35% were younger than 65 years. Thirty-two percent of subjects are African American. At baseline, 50% of patients had T1c cancer and 7% had a tumor Gleason score of 8 to 10. Associated medical conditions were seen in more than 40% of patients, of whom 1% had some other cancers. Dr. Wilt emphasized that the baseline characteristics were comparable between patients who were randomized and those who S38 VOL. 5 SUPPL. 6 2003 400 500 600 700 Figure 3. Mean probability of return of erection following radical retropubic prostatectomy (RRP) and robotic Vattikuti Institute Prostatectomy (VIP). Reproduced with permission from Tewari et al.12 Figure 4. Mean probability of return of intercourse following radical retropubic prostatectomy (RRP) and robotic Vattikuti Institute Prostatectomy (VIP). Reproduced with permission from Tewari et al.12 .7 100 300 Follow-up (d) Follow-up (d) 0 200 were screened but failed to be randomized. This study differs significantly from a European randomized study in that it includes a significant proportion of African American patients (32% vs 0% in the European study), and the majority of patients have cancer detected by PSA testing (76% vs 5% in the European study). Accrual has recently completed, and results sare expected in 2010.6,7 Robotic Prostatectomy Drs. Tewari and Menon8-12 presented REVIEWS IN UROLOGY an update on an ongoing prospective comparison between robotic prostatectomy (Vattikuti Institute Prostatectomy [VIP]) and radical retropubic prostatectomy (RRP). The study design was a single-institution, prospective, nonrandomized comparison of baseline, intra-operative, postoperative, histopathologic, and functional outcomes in 300 RRP and VIP patients. Although preoperative parameters, operative time, and pathological stages were comparable, there were significant differences in the measured outcomes. The blood loss was 910 mL with RRP and 150 mL with VIP (P < .001), and transfusion was much more common in the RRP group (67% vs 0%; P < .001). There was a four-fold increased incidence of complications with RRP (20% vs 5%; P < .05). The discharge hemoglobin level was lower with RRP (10 g vs 13 g; P < .005). The hospital stay was longer for patients who received RRP (3.5 days vs 1.2 days; P < .05); 93% of VIP patients and 0% of RRP patients were discharged within 24 hours (P < .001). The duration of catheterization was 2 times longer for RRP patients (15.8 days vs 7 days; P < .05). Tumor at inked mar- Clinically Localized Prostate Cancer gin was more frequent in the RRP group (23% vs 5%; P < .05). VIP patients achieved 50% return of continence much more quickly than did RRP patients (160 days vs 44 days; P < .05) (Figure 2). VIP patients also had a significantly shorter time to 50% return of erection (180 days vs 440 days; P < .05) (Figure 3) and 50% return of intercourse (340 days) (Figure 4). At 700 days, RRP patients had not yet achieved return of intercourse (P < .05). In this study, the VIP procedure seemed safer, was associated with less blood loss, and required shorter hospitalization and catheterization. The oncologic and functional results were favorable in the VIP patients. 7. References 8. 1. 2. 3. 4. 5. 6. American Cancer Society. Cancer Facts & Figures. Atlanta, GA: American Cancer Society; 2003. Epstein JI. Pathologic assessment of the surgical specimen. Urol Clin North Am. 2001;28:567–594. Hall GS, Kramer CE, Epstein JI. Evaluation of radical prostatectomy specimens. A comparative analysis of sampling methods. Am J Surg Pathol. 1992;16:315–324. Ohori M, Abbas F, Wheeler TM, et al. Pathological features and prognostic significance of prostate cancer in the apical section determined by whole mount histology. J Urol. 1999;161:500–504. Stamey TA, McNeal JE, Yemoto CM, et al. Biological determinants of cancer progression in men with prostate cancer. JAMA. 1999; 281:1395–1400. Wilt TJ, Brawer MK. The Prostate Cancer 9. 10. 11. 12. Intervention Versus Observation Trial (PIVOT). Oncology (Huntingt). 1997;11:1133–1140, 1143. Wilt TJ. Can randomized treatment trials in early stage prostate cancer be completed? Clin Oncol. 1998;10:141–143. Tewari A, Menon M. Vattikuti Institute prostatectomy: surgical technique and current results. Curr Urol. 2003;4:119–123. Menon M, Tewari A, Baize B, et al. Prospective comparison of radical retropubic prostatectomy and robot-assisted anatomic prostatectomy: the Vattikuti Urology Institute experience. Urology. 2002;60:864–868. Menon M, Tewari A. Robotic radical prostatectomy and the Vattikuti Urology Institute technique: an interim analysis of results and technical points. Urology. 2003;61:15–20. Tewari A, Peabody J, Sarle R, et al. Technique of da Vinci robot-assisted anatomic radical prostatectomy. Urology. 2002;60:569–572. Tewari A, Srivasatave A, Menon M, et al. A prospective comparison of radical retropubic and robot-assisted prostatectomy: experience in one institution. BJU Int. 2003;92:205-210. VOL. 5 SUPPL. 6 2003 REVIEWS IN UROLOGY S39