Intraoperative Frozen Section Analysis During Nerve Sparing Laparoscopic Radical Prostatectomy: Feasibility Study

0022-5347/03/1705-1843/0 THE JOURNAL OF UROLOGY® Copyright © 2003 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 170, 1843–1846, November 2003 Printed in U.S.A.

DOI: 10.1097/01.ju.0000092081.71167.34

INTRAOPERATIVE FROZEN SECTION ANALYSIS DURING NERVE SPARING LAPAROSCOPIC RADICAL PROSTATECTOMY: FEASIBILITY STUDY ¨ LLE FROMONT,* HERVE ´ BAUMERT, XAVIER CATHELINEAU, FRANC GAE ᝺ OIS ROZET, PIERRE VALIDIRE AND GUY VALLANCIEN From the Departments of Pathology (GF, PV) and Urology (HB, XC, FR, GV), Institut Montsouris, Paris, France

ABSTRACT

Purpose: Nerve sparing radical prostatectomy may allow preservation of potency but it can increase positive surgical margins. We used intraoperative frozen section (IFS) analysis to monitore the nerve sparing procedure in laparoscopic prostatectomy. Materials and Methods: A total of 100 patients with localized prostatic carcinoma underwent bilateral intrafascial nerve sparing laparoscopic prostatectomy with IFS. A wedge of tissue was cut from base to apex in the region of the neurovascular bundles (NVBs) and analyzed on frozen section. If carcinoma was detected at the inked margin, the corresponding NVB was resected. Definitive margin status was evaluated after permanent section analysis of IFS prostatectomy specimens and eventually NVB specimens. Results: IFS analysis was positive in 24 patients, as confirmed in all by permanent section of the wedges. Three of these patients had positive margins in the prostate specimen at another site. Of the 76 tumors with negative IFSs 1 had positive margins on permanent sections of the wedges and 8 had positive margins on the prostate specimen at another site. IFS led to a decrease in the overall positive margin status from 33% to 12% and from 26.1% to 7.9% in pT2 tumors. Tumor was found on NVB resection in 8 cases (33%). Conclusions: These results suggest that IFS analysis is a reliable method by which to monitor nerve sparing during laparoscopic prostatectomy. IFS could allow the surgeon to offer a nerve sparing procedure more frequently without compromising cancer control. KEY WORDS: prostate, prostatic neoplasms, prostatectomy, laparoscopy, frozen sections

Prostate carcinoma is one of the most common male cancers.1 Routine serum prostate specific antigen (PSA) screening in industrialized countries has enabled the detection of tumors at a lower stage in younger patients often concerned about potency preservation. Therefore, the actual issue of local definitive treatment for localized prostatic carcinoma, such as radical prostatectomy, is to achieve cancer control with the preservation of sexual function. Laparoscopic prostatectomy, which has been performed in our institution since 1998, is currently the standard surgical treatment for our patients with clinically localized prostate cancer. Laparoscopy has been shown to be comparable to retropubic approach in terms of cancer control and continence.2, 3 Moreover, the laparoscopic technique facilitates precise dissection around the prostate because of improved magnification of the pelvic area and minimal blood loss. Bilateral preservation of the neurovascular bundles (NVBs) significantly improves the potency outcome after retropubic and laparoscopic prostatectomy.4, 5 Conversely since it limits the extent of resection, the nerve sparing procedure can induce an increased risk of positive surgical margins and, therefore, of cancer recurrence. Positive margins in the region of the NVBs after aggressive nerve sparing prostatectomy may be due to extracapsular extension in pT3 tumors or to intracapsular incision through the tumor. A recent study showed an increased positive margin incidence after bilateral nerve sparing laparoscopic prostatectomy for pT2 tumors compared with no nerve sparing (18.6% vs 14%), although

the difference did not attain significance.5 We observed similar results with 20% vs 10% of positive margins in bilateral vs no nerve sparing laparoscopic prostatectomy for organ confined tumors.6 The nerve sparing procedure can be considered in all patients who are potent before surgery and who have clinically localized prostate cancer. However, none of the factors associated with an increased risk for extraprostatic tumor alone can be used to exclude a patient from nerve sparing prostatectomy.7 Therefore, there is a need for the surgeon to be helped perioperatively in the decision to perform NVB preservation. Two recent studies described intraoperative frozen section (IFS) to select candidates for nerve sparing during retropubic radical prostatectomy.8, 9 In the current study we tested the feasibility and the value of IFS to monitor the intrafascial nerve sparing procedure in laparoscopic radical prostatectomy.

MATERIAL AND METHODS

Surgical technique. A total of 100 laparoscopic radical prostatectomies with bilateral intrafascial nerve sparing were performed between October 2001 and June 2002. The intrafascial nerve sparing technique was offered to potent patients who had clinically localized prostate cancer. Patients were informed about the possibility of additional bundle resection. The nerve sparing was done in antegrade fashion. After dissecting the bladder neck and grasping the previously dissected seminal vesicles, intrafascial nerve sparing was performed. The prostatic fascia was opened longitudinally from the anterolateral part of the prostate at the 2 and 10 o’clock

Accepted for publication June 20, 2003. * Corresponding author: Department of Pathology, Institut Montsouris, 42, Blvd. Jourdan, 75014 Paris, France (telephone: 33 0 1 56 61 68 01; FAX: 33 0 1 56 61 66 43; e-mail: [email protected]). 1843

1844

INTRAOPERATIVE FROZEN SECTIONS DURING LAPAROSCOPIC PROSTATECTOMY

FIG. 1. IFS technique. Samples were obtained along bundle course (a, arrows) and inked (b)

positions. This maneuver opening allowed the dissection of the prostate in contact with the capsule. When the correct layer was found, the surgeon pushed away the NVB. Only few capsular vessels had to be coagulated and cut. This dissection allowed prostate removal without any periprostatic tissue. The prostate was then extracted with an endoscopy bag through a small median incision and sent to the pathologist. During IFS analysis the incision was closed and anastomosis was performed. At the end of anastomosis, which was a mean of 20 minutes in duration, the results of IFS were obtained. When frozen section analysis revealed a positive margin, the homolateral NVB was removed. Bundle resection was easily done without disassembling the anastomosis by retracting the bladder to the opposite side of the positive margin. IFSs. Immediately upon reception of the surgical specimen 2 tissue wedges were cut for frozen section in the posterolateral areas along the course of the spared nerve bundles. Samples were obtained from the right and left sides in a vertical plane from base to apex (fig. 1, a). Each wedge was about 0.5 to 0.6 cm wide on the surface and 2 to 4 cm long depending on the prostate size. The surface of each wedge was then inked (fig. 1, b). At least 4, 10 ␮m frozen sections were analyzed, followed by hematoxylin and eosin staining. Positive surgical margins were defined as tumor at the inked surface. Tumor close to but not at the ink was considered negative margins. If carcinoma was detected at the inked margin, the side, length and location (ie distance from the apical extremity) were provided to the surgeon. The process required 10 to 15 minutes. If carcinoma was detected at the inked margin, the corresponding NVB was resected and send for pathological study. The bundle was then pinched on a support and fixed in 10% formalin. The remaining prostate specimen was then carefully sutured along the margins of wedge resection to obtain valuable analysis of the surgical specimen. All IFSs were analyzed by 2 pathologists. Pathological evaluation. Tissues remaining from wedge resection after frozen section were fixed in formalin and examined on permanent sections at the same time as the prostate specimen. After 2 days of fixation in 10% formalin the prostate surface was inked and the whole specimen was cut in 3 to 4 mm sections according to the Stanford procedure.10 Pathological staging was performed using the 1997 TNM classification and tumors were graded according to the Gleason system. For each specimen the maximal dimension of the largest tumor focus was measured directly on glass slides. Positive surgical margins on the prostate specimen were separated into those immediately adjacent to wedge resection (in the posterolateral areas along the course of the bundles) and those distant from the wedge resection. Positive surgical margins were measured and classified according to location. For NVB resection the whole bundle was cut trans-

versally from the basal to the apical area. Positive definitive surgical margins (DMs) were defined as positive margins on permanent sections of wedge resection that were not seen on frozen section or on the prostate specimen in an area other than that submitted for IFS analysis. Therefore, a positive IFS margin leading to wide NVB resection was considered a negative definitive margin. Significant differences between groups were analyzed using chi-square or Fisher’s exact test for categorical data and or the Student t test for continuous data with p ⬍0.05 considered significant. RESULTS

In all cases a consensus was achieved by the pathologists concerning margin status. Table 1 lists patient clinical and pathological characteristics. IFS analysis was positive in 24 of the 100 patients (fig. 2, a). Of these patients 100% had positive margins on wedge resection permanent sections. Of these 24 patients 15 (62.5%) also had positive margins on the prostate specimen immediately adjacent to the wedge resection and 3 (12.5%) had positive margins on the prostate specimen in an area other than that submitted for IFS, ie at the apex in 2 and at the bladder neck in 1. Only 1 of the 76 patients with negative IFSs had positive margins on the permanent sections of wedge resection (fig. 2, b). In 8 of these patients we found positive margins elsewhere on the prostate specimen, ie apical in 5, basal in 1 and lateral in 2. Table 2 lists results according to the TNM classification. The proportion of positive IFSs in pT3 tumors was significantly higher than in pT2 tumors 58.7% or 7 of 12 vs 19.4% or 17 of 88, p ⬍0.01). Definitive positive surgical margins were also more frequent in pT3 than in pT2 disease (41.7 or 5 of 12 vs 7.9% or 7 of 88, p ⬍0.005). Without IFS analysis the total rate of positive surgical margins would have been 33% (total DM positive plus IFS positive and DM negative), and 26.1% and 83.3% for pT2 and pT3 tumors, respectively. With IFS analysis the total rate of positive surgical margins significantly decreased to 12%, and to 7.9% and 41.7% for pT2 and pT3 tumors, (p ⬍0.001, ⬍0.005 and p ⬍0.05, respectively). Patients with positive IFSs had a significantly higher Gleason score and tumor size than those with negative IFSs.

TABLE 1. Study population characteristics Median age ⫾ SD Median PSA ⫾ SD (ng/ml) Median tumor size ⫾ SD (cm) No. TNM stage: pT2 pT3 No. Gleason score: 5–6 7 8–9

59 ⫾ 6.2 6.6 ⫾ 3.4 2⫾1 88 12 54 43 3

1845

INTRAOPERATIVE FROZEN SECTIONS DURING LAPAROSCOPIC PROSTATECTOMY

FIG. 2. IFS results revealed positive IFS with tumor at inked surface (a, arrow), negative IFS with tumor close to but not at ink (b) and tumor (arrow) involving adipose tissue in resected NVB (c). H & E, reduced from ⫻20 (a and b) and ⫻40 (c). TABLE 2. IFS and DM results according to pathological stage IFS pos: DM neg DM pos IFS neg: DM neg DM pos Totals: DM neg DM pos (%)

No. pT2

No. pT3

Total No.

16 1

5 2

65 6

2 3

24 21 3 76 67 9

81 7 (7.9)

7 5 (41.7)

88 12 (12)

Preoperative PSA and tumor size were increased in patients with positive definitive margins compared with patients with negative margins (table 3). We found tumoral glands in 8 of the 24 resected NVBs (33%). In 4 cases residual tumor was extensive, involving adipose tissue without reaching the surgical margins (fig. 2, c). DISCUSSION

Several studies have demonstrated that bilateral preservation of the NVBs has a significant influence on the potency outcome after surgery.4, 5 However, it has been shown that wide preservation of the bundles could lead to an increased positive margin rate at the posterolateral location with an impact on the disease-free survival advantage.11 The preoperative selection of potent patients who could benefit from the bilateral nerve sparing procedure is related to the prediction of extraprostatic tumor and, therefore, it is difficult to assess in individuals. The decision to preserve or excise the bundles may be guided during dissection by the presence or absence of periprostatic adherence to the prostate capsule. However, it must be pointed out that periprostatic adherence can be related not only to extraprostatic cancer extension, but also to extensive lesions of prostatis and periprostatis, which can cause misleading impressions. Therefore, standardized IFS analysis could help surgeons guide the decision to perform the nerve sparing procedure. The IFS procedure during retropubic laparoscopic prostatectomy was reported in 2 recent studies, including 488 and 1019 patients. Since 1998, laparoscopic prostatectomy has been performed at our institution with results on cancer control comparable to those observed with the retropubic approach.2 We have previously observed that for organ confined tumors bilateral intrafascial nerve preservation led to a positive surgical margin rate of 20% compared with 10% without preservation.6 Therefore, we analyzed the feasibility of IFS

during laparoscopic radical prostatectomy and the impact of this technique on the incidence of the final margin status. We first observed that IFS did not increase operative time since during the procedure, which lasted approximately 15 minutes, surgeons had the ability to perform the anastomosis. The comparison of IFS results with those of permanent sections of the wedges showed 100% specificity and 96% sensitivity with only 1 false-negative result. We also observed that the IFS procedure is reproducible and can be performed by different pathologists who achieve the same analysis of margin status. Furthermore, in all cases of positive IFSs additional resection of the corresponding bundle was easily performed by the laparoscopic approach without any complications. The incidence of positive IFSs was 24%, which is higher than that reported by others for retropubic prostatectomy using a comparable pathological procedure.8, 9 The difference observed could have been related to the modification of the nerve sparing technique, which was of the intrafascial type in our group of patients. In addition, our group included 12% of tumors with extraprostatic extension, which showed a high rate of positive IFSs, ie 58.7%. Some of these tumors were classified as pT3 based on secondary NVB resection showing tumor infiltration in adipose tissue. In pT2 tumors the rate of positive IFS was 19.3%, related to intraprostatic or intracapsular excision without adipose tissue infiltration in the resected bundles. Because in most patients with positive IFSs negative final surgical margins were achieved after bundle resection, IFS led to a significant decrease in the overall positive margin rate from 33% to 12% and for organ confined tumors from 26.1% to 7.9%. It is important to point out that the 26.1% incidence of positive margins before IFS is comparable to the 20% observed in our previous series of bilateral nerve sparing laparoscopic prostatectomies without IFS.6 At the end we achieved a final margin incidence (total of 12% and 7.9% in pT2 tumors), similar to or better than that reported in recent series of radical prostatectomies using the retropubic or laparoscopic approach in groups of 1,000 patients, including nerve sparing and no nerve sparing procedures.3, 12 In the current study the incidence of positive DMs was associated with preoperative PSA and tumor size, in agreement with previous observations.13, 14 We found tumor tissue in 33% of additionally resected bundles. The absence of residual tumor in most patients has also been observed in previous studies of IFS analysis during retropubic prostatectomy with an incidence of respectively 0%8 and 20%.9 Because it has been shown that tumor pene-

TABLE 3. Tumor characteristics according to IFS and DM status IFS Pos No. Gleason score: 5–6 7 8–9 Median PSA ⫾ SD (ng/ml) Median tumor size ⫾ SD (cm)

IFS Neg

p Value

DM Pos

DM Neg

⬍0.05 8 14 2 7.1 ⫾ 3.5 3 ⫾ 0.9

46 29 1 6.4 ⫾ 3.4 2⫾1

Not significant 0.03

p Value Not significant

5 6 1 7.5 ⫾ 5.2 3.5 ⫾ 0.9

49 37 2 6.2 ⫾ 2.9 2 ⫾ 0.9

0.02 ⬍0.001

1846

INTRAOPERATIVE FROZEN SECTIONS DURING LAPAROSCOPIC PROSTATECTOMY

tration in the bundle is inferior to 2 mm in most patients,4 it is possible that some tumor cells may be hidden in the tissue and not seen on permanent sections. Alternatively because of the difference in density between tumor and periprostatic tissues, it is also possible that the section layer could have reached the limit of the tumor without leaving residual cancer tissue. This hypothesis could also explain the fact that some patients with organ confined tumor and positive surgical margins had no cancer recurrence. With the extension of the number of cases included in IFS analysis a major prospective interest would be to predict which patients, including those with positive IFSs, would have residual tumor in the resected bundle. To date most studies have demonstrated an impact of margin status on cancer recurrence, even in pT2 tumors.3, 15 Goharderakhshan et al reported that the risk of cancer recurrence was similar in patients with negative IFSs, and in patients with positive IFSs and additional resection.9 In the current study at least 2 years of followup are now required to evaluate the clinical outcome of our patients. Other studies have already shown the benefit of bilateral nerve sparing on potency outcome compared with unilateral or no nerve sparing with similar observations in retropubic and laparoscopic procedures.4, 5 In our patients preoperative potency, defined as the ability to achieve unassisted intercourse, was assessed. A minimum of 12 months of followup is now required to evaluate the impact of bilateral intrafascial nerve sparing on postoperative sexual function.

4. 5. 6.

7. 8.

9.

10.

11.

CONCLUSIONS

This study demonstrates that IFS during laparoscopic radical prostatectomy can be performed without increasing operative time and it can accurately determine posterolateral margin status. This technique could allow enable the more frequent performance of a bilateral nerve sparing procedure without increasing the final margin incidence. REFERENCES

1. Landis, S. H., Murray, T., Bolden, S. and Wingo, P. A.: Cancer statistics, 1998. CA Cancer J Clin, 48: 6, 1998 2. Fromont, G., Guillonneau, B., Validire, P. and Vallancien, G.: Laparoscopic radical prostatectomy: preliminary pathological evaluation. Urology, 60: 661, 2002 3. Guillonneau, B., El-Fettouh, H., Baumert, H., Cathelineau, X., Doublet, J. D., Fromont, G. et al: Laparoscopic radical prosta-

12.

13.

14.

15.

tectomy: oncological evaluation after 1,000 cases at the Montsouris Institute. J Urol, 169: 1261, 2003 Quinlan, D. M., Epstein, J. I., Carter, B. S. and Walsh, P. C.: Sexual function following radical prostatectomy: influence of preservation of neurovascular bundles. J Urol, 145: 998, 1991 Katz, R., Salomon, L., Hoznek, A., de la Taille, A., Vordos, D., Cicco, A. et al: Patient reported sexual function following laparoscopic radical prostatectomy. J Urol, 168: 2078, 2002 Guillonneau, B., El Fettouh, H., Fromont, G., Validire, P. and Vallancien, G.: Pathological results of neurovascular bundle preservation during laparoscopic radical prostatectomy. J Urol, suppl., 167: 343, abstract 1361, 2002 Sokoloff, M. H. and Brendler, C. B.: Indications and contraindications for nerve-sparing radical prostatectomy. Urol Clin North Am, 28: 535, 2001 Cangiano, T. G., Litwin, M. S., Naitoh, J., Dorey, F. and deKernion, J. B.: Intraoperative frozen section monitoring of nerve sparing radical retropubic prostatectomy. J Urol, 162: 655, 1999 Goharderakhshan, R. Z., Sudilovsky, D., Carroll, L. A., Grossfeld, G. D., Marn, R. and Carroll, P. R.: Utility of intraoperative frozen section analysis of surgical margins in region of neurovascular bundles at radical prostatectomy. Urology, 59: 709, 2002 McNeal, J. E., Villers, A. A., Redwine, E. A., Freiha, F. S. and Stamey, T. A.: Capsular penetration in prostate cancer. Significance for natural history and treatment. Am J Surg Pathol, 14: 240, 1990 Smith, R. C., Partin, A. W., Epstein, J. I. and Brendler, C. B.: Extended followup of the influence of wide excision of the neurovascular bundle(s) on prognosis in men with clinically localized prostate cancer and extensive capsular perforation. J Urol, 156: 454, 1996 Hull, G. W., Rabbani, F., Abbas, F., Wheeler, T. M., Kattan, M. W. and Scardino, P. T.: Cancer control with radical prostatectomy alone in 1,000 consecutive patients. J Urol, 167: 528, 2002 Jones, E. C.: Resection margin status in radical retropubic prostatectomy specimens: relationship to type of operation, tumor size, tumor grade, and local tumor extension. J Urol, 144: 89, 1990 Wieder, J. A. and Soloway, M. S.: Incidence, etiology, location, prevention and treatment of positive surgical margins after radical prostatectomy for prostate cancer. J Urol, 160: 299, 1998 Grossfeld, G. D., Chang, J. J., Broering, J. M., Miller, D. P., Yu, J., Flanders, S. C. et al: Impact of positive surgical margins on prostate cancer recurrence and the use of secondary cancer treatment: data from the CaPSURE database. J Urol, 163: 1171, 2000