0022-5347/96/1554-1357$03.00/0 THE J ~ ) ~ T ~ A I . OF UHOLI)(:Y Copyright 0 1996 by AMEHICAN UIWLWICAL ASSOCIATION, INC

Vol. 155, 1357-1360, Apnl 1996 Printed tn U S A .

RANDOMIZED STUDY OF NEOADJUVANT TESTICULAR ANDROGEN ABLATION THERAPY BEFORE RADICAL PROSTATECTOMY IN MEN

WITH CLINICALLY LOCALIZED PROSTATE CANCER BRUCE L. DALKIN,* FREDERICK R. AHMA", RAYMOND NAGLE AND CYNTHIA S. JOHNSON

From the Departments of Surgery I Urology, Pathology and Medicine /Hematology/ Medical Oncology, University of Arizona College of Medicine, and Tucson Veterans Aftairs Medical Center, Tucson, Arizona

ABSTRACT

Purpose: We determined whether 12 weeks of neoadjuvant testicular androgen ablation ther- apy using a luteinizing hormone-releasing hormone agonist could improve pathological outcomes in men undergoing radical retropubic prostatectomy for clinically localized (stages TlC, T2A and T2B) prostatic carcinoma.

Materials a n d Methods: A total of 56 participants was randomized to receive either monthly injections of a luteinizing hormone-releasing hormone agonist at 4-week intervals followed by radical retropubic prostatectomy (28) o r to undergo immediate radical retropubic prostatectomy alone (28). Operations were performed via a similar technique and all prostatic specimens were processed histologically in their entirety.

Results: There was no improvement in pathological outcome using a luteinizing hormone- releasing hormone agonist preoperatively compared to surgery alone. Of 28 men undergoing immediate radical retropubic prostatectomy 23 had organ-confined (17) o r specimen-confined (6) disease versus 22 of 28 who received luteinizing hormone-releasing hormone neoadjuvant ther- apy for 12 weeks preoperatively (16 with organ-confined and 6 with specimen-confined disease, p = 1.00). In addition, when the study population was analyzed by pretreatment prostate specific antigen (PSA) levels (10 ng./ml. or less, o r greater than 10 nglml.) there was also no difference in pathological outcome (p = 0.65 for PSA greater than 10 a n d p = 0.32 for PSA less than 10).

Conclusions: Neoadjuvant androgen ablation therapy for 12 weeks before radical prostatec- tomy in patients with clinically localized adenocarcinoma of the prostate does not result in improved pathological outcomes.

KEY WORDS: prostatic neoplasms, adenocarcinoma, gonadorelin, hormones

Concomitant with the increase in the number of men diag- nosed with prostate cancer in recent years has been an in- crease in the number of men undergoing radical prostatec- tomy for treatment of clinically localized disease.' Although recent studies suggest that there may be some improvement in surgical pathological outcomes, a population of men re- mains who have a poor pathological result.' Neoadjuvant therapies have resulted in surgical down staging in other malignancies. In prostate cancer a reversible systemic ther- apy associated with a high objective response is medical androgen ablation, which can be accomplished either with a luteinizing hormone-releasing hormone agonist alone or com- bined with an androgen receptor blocker. We evaluated in a randomized fashion the ability of neoadjuvant androgen ab- lation using 12 weeks of luteinizing hormone-releasing hor- mone agonist therapy before radical prostatectomy in men with clinically localized (stages TlC, T2A and T2B) adeno- carcinoma of the prostate to improve pathological outcomes.

MATERIAL AND METHODS

Patient p o p u l a t i o ~ . Patients diagnosed with clinically 10- calized adenocarcinoma of the prostate who had greater than a lo-year projected survival were recruited from the urology clinics at Arizona Health Sciences Center and Tucson Veter-

Accepted for publication September 1, 1995. Supported in part by a research grant from Zeneca Pharmaceuti-

cals. * Requests for re rints: De artment of SurgeryKJrologY. Univer-

sity of Arizona Coiege of dedicine, 1501 North Campbell h e . , Tucson, Arizona 85724.

ans AfTairs Medical Unit, and counseled regarding therapeu- tic options. If the patient elected radical retropubic prosta- tectomy as the primary treatment modality, they were given the option of entering into this study. To be eligible a partic- ipant had to have a prostate specific antigen (PSA) level of greater than 4.0 ng./ml. and a clinical stage TlC, T2A or T2B lesion. In addition, all patients had a bone scan that was not suspicious for metastatic disease.

Treatment arms. After obtaining informed consent as set forth by the University of Arizona human subjects commit- tee, patients were randomized to undergo radical retropubic prostatectomy alone or with 12 weeks of preoperative ther- apy with a luteinizing hormone-releasing hormone agonist. Patients randomized to the surgery only group underwent bilateral pelvic lymph node dissection and radical retropubic prostatectomy 6 to 8 weeks after biopsy. Those randomized to the luteinizing hormone-releasing hormone agonist plus sur- gery group received 3.6 ng. goserelin acetate subcutaneously every 28 days for 3 injections, followed 5 to 10 days later by pelvic lymphadenectomy and radical retropubic prostatec- tomy.

All operations were performed by or under the direction of 1 of us (B. L. D.) to evaluate an important variable in surgical outcome studies, that is surgical skill. The surgical technique was that of anatomical radical prostatectomy with 2 modifi- cations. The apical dissection was performed by incising the endopelvic fascia on both sides of the prostate, and identify- ing a plane between the dorsal venous complex and anterior urethra. A right-angle clamp was passed between the dorsal vein and urethra, and the dorsal vein was ligated with free

1357

1358 NEOADJUVANT TESTICULAR ANDROGEN ABLATION THERAPY BEFORE PROSTATECTOMY

ties. A curved Kocher clamp was placed on the anterior prostatic dorsal venous complex and the dorsal vein was transected. Using a tonsil clamp, a plane in the penurethral tissues between the lateral border of the urethra and the neurovasnrlar bundle was identified. A line tip, short, right- angle clamp was passed in this plane behind the urethra, which was then transected under direct vision using a NO. 15 blade anteriorly, and a No. 12 blade posterior and laterally. The Foley catheter was brought up in the field and transected. The posterior fascia was incised under direct vision, and a plane was developed between the anterior rectal wall and prostate in the midline. Right-angle clamps were then used to clamp and divide the neurovascular bundles far from the apex of the prostate. If a nerve sparing procedure was to be performed the neurovascular bundle was dissected off the lateral aspect of the prostate at this point. Nerve sparing surgery was considered only in patients with strong pre-operative erections, a pretreatment PSA of less than 10 ng./ml. and 1 side of the prostate negative for malignancy on biopsy. Only unilateral nerve sparing was performed.

The second modification involved use of a “peel” technique at the bladder neck. After completion of the seminal vesicle and vas deferens portions of the procedure, the plane be- tween the bladder neck and base of the prostate was identi- fied. Using sharp dissection this plane was established and incised for 360 degrees around the prostate. After completion of dissection down to mucosa, the mucosa was opened on the anterior side and this incision was extended around to com- plete the 360-degree incision, which allowed for easy identi- fication of the ureteral orifices as well as preserving native bladder neck fibers. The bladder neck tissue was evaluated by frozen section and no patient had evidence of prostate tissue or prostate cancer.

Pathological processing. The prostate, including the semi- nal vesicles, was first coated on the outside with india ink to define the surgical margin. Serial 5 mm. sections were made across the prostate in a plane vertical to the rectal surface. After fresh specimens were snap-frozen from select areas, the remainder of the prostate was futed in 10% formalin. Each 5 mm. section was divided into 4 quadrants and submitted with 1 quadrant to each cassette. The apical section was further subdivided and processed separately. The entire prostate was sectioned and stained with hematoxylin and eosin, including sections taken through the seminal vesicles. Each section was examined, and the most common and next most common Gleason histological grades were recorded and reported as a Gleason score. The tumors were outlined on a form corresponding to each sagittal section. From this outline the largest tumor volume was estimated and reported in cubic centimeters. The entire surgical margin was examined for the presence or absence of tumor, as well as for any evidence of perforation of the capsule and lymph node in- volvement. These data were used to stage the disease patho- logically as organ-confined (all tumor present within the capsule of the prostate without evidence of seminal vesicle or pelvic lymph nodes disease), specimen-confined (carcinoma through the prostatic capsule but within the inked margins, also without seminal vesicle or pelvic lymph node involve- ment), margin positive (carcinoma through the prostatic cap- sule at the inked margin, also without evidence of seminal vesicle or pelvic lymph node involvement), seminal vesicle invasion (carcinoma involvement in the tissue planes sur- rounding the seminal vesicles) and pelvic lymph node disease (microscopic evidence of carcinoma in the obturatorhliac lymph node packets).

For the purpose of our study, patients with positive mar- gins, seminal vesicle invasion or pelvic lymph node involve- ment were considered to have a poor pathological outcome. Of note, no patient with margin positive disease had an iatrogenic positive margin, that is all patients had obvious

@n, rather than having a capsular incision resulting in a positive margin. Finally, all patients regardless of patholog- ical outcome had a postoperative nadir PSA level that was undetectable.

Statistical methods. Fisher’s exact tests were used to ana- lyze the association between treatment (lutehizing hormone- releasing hormone plus radical retropubic prostatectomy ver- sus radical retropubic prostatectomy alone) and pathological outcome (favorable versus unfavorable). A favorable patho- logical outcome was defined as organ-confined or organ- confined plus specimen-confined. All other pathological out- comes were considered unfavorable. Therefore, analysis was performed using 2 x 2 tables.

RESULTS

Of 61 men enrolled into the study 56 completed the proto- col. Five patients (2 randomized to receive preoperative lu- teinizing hormone-releasing hormone and 3 to undergo rad- ical retropubic prostatectomy alone) failed to undergo surgery because of a preoperative PSA level less than 4.0 ng./ml. in 1 and failed preoperative cardiac clearance in 2, while 1 refused surgical intervention and in 1 surgery was aborted midway through the operation because of bleeding. The pretreatment patient characteristics of age, clinical stage of tumor, number of positive cores on biopsy, histolog- ical grade of disease and pre-biopsy PSA level are given in table 1. The 2 arms of the study were balanced for the aforementioned pretreatment criteria.

Table 2 shows the pathological staging outcomes for the overall population. Using organ-confined disease as the only favorable pathological outcome, there was no statistically significant difference between the 2 treatment groups (p = 1.00). Using organ-confined or specimen-confined disease as favorable pathological outcomes, there again was no statis- tically significant difference between the 2 treatment groups (p = 1.00).

The pathological outcomes for men based on a pretreat- ment PSA level of 10 ng./ml. or less are shown in table 2. There was no statistically significant difference between the 2 groups (p = 1.00), with organ-confined disease as the only favorable pathological outcome or with organ-confined or specimen-confined disease as the favorable outcomes (p = 0.32). Table 2 also shows the pathological staging results for men whose PSA level was greater than 10 ng./ml. before biopsy. Again, there was no statistically significant differ-

TABLE 1. Clinical prognostic valuables in 56 randomized patients No. Pts.

Radical Retropubic Radical Retropubic Prostatectomy +

Prostatectomy Alone Luteinizing Hormone- (28 pts.) Releasing Hormone

(28 DtS . )

Clinical stage: Tlc T2a T2b

No. pos. cores: 2 or Less 3-4 More than 4 Not known

Histology (grade): Low (2-4) Moderate (5-7) High ( & I O J

PSA level (ng./ml. 1: 4.1 - 10 10.1 - 20

16 12 0

11 8 2 7

6 21

1

18 9

17 8 3

10 8 3 7

8 16 4

16 9

More than 20 1 3 Mean age was 64.7 years (range 48 to 76) in the surgery only ~ o u p and 65.5

years lrange 50 to 76) in the-surgery plus luteini’zin-g hormone-releasing extension of tumor through the capsule to the positive mar- hormone group.

NEOADJUVANT TESTICULAR ANDROGEN ABLATION THERAPY BEFORE PROSTATECTOMY 1359

_~---______ T.ABLE 2. Putho[ogical results ____ No, pts. No. With Organ- No. With Specimen- M a ~ ~ ; ~ ~ ~ ~ ~ ~ o s , No. With P O ~ Pelvic

Confined C a Confined Ca Seminal Ves,cles Lvmph Nodes . -

0i:erall" Radical retropuhic prostatectomy alone 28 17 6 Radical retropubic prostatectomy + adjuvant lutein- 28 16 6

wing hormone-releasing hormone PSA 10 ng. lml. or less?

4 1 5 1

Radical retropubic prostatectomy alone 18 13 Radical retropubic prostatectorny + adjuvant lutein- 16 11

izing hormone-releasing hormone

4 1 2 2

0 1

PSA more than 10 ng. 1ml.f Radical retropubic prostatectomy alone 10 4 2 3 1 Radical retropubic prostatectomy + adjuvant lutein- 12 5 4 3 0

izing hormone-releasing hormone * Favorable outcome: organ-confined and organ-confined plus specimen-confined p = 1.00. t Favorable outcome: organ-confined p = 1.00 and organ-confined plus specimen-confined p = 0.32. i Favorable outcome: organ-confined p = 1.00 and organ-confined plus specimen-confined p = 0.65.

ence between the 2 groups with organ-confined tumor as the only favorable pathological outcome ( p = 1.00), or with organ- confined or specimen-confined disease as favorable patholog- ical outcomes (p = 0.65).

When clinical stage Tlc lesions were evaluated, there was no benefit to neoadjuvant therapy. In the untreated group 12 of 16 patients (757~) had a favorable (organ-confined plus specimen-confined) outcome versus 14 of 17 (82%,) in the luteinizing hormone-releasing hormone group (p = 0.688). Similar results occurred with stage T2 lesions, with the un- treated group actually doing somewhat better (11 of 12, or 91%, with a favorable pathological outcome versus 8 of 11, or 73%, in the luteinizing hormone-releasing hormone treated group, p = 0.317).

In our pre-study projections we predicted a poor patholog- ical outcome rate of 50%, in the control population. Predicting a 50% improvement in pathological outcome mandated 58 patients in each arm of the study (80% power to detect a 50% improvement with p <0.05). Mid study data analysis re- vealed a poor pathological outcome of only 18% in the control group. In addition, we essentially saw no difference in patho- logical outcomes for the control versus luteinizing hormone- releasing hormone treated groups (table 2). From a statistical significance standpoint, there was no validity in continuing the study. Even when subdividing the population to review only the stage T lc cases, a difference in pathological outcome of 7% (poor pathological outcome rate of 25% in controls versus 18% in luteinizing hormone-releasing hormone treated patients) would require 1,080 patients to show sta- tistical significance to this 7% difference. The requirement of 1,080 patients for the study to support a relatively slight improvement in pathological outcome speaks against any clinical significance to a study of that size. Therefore, our sample size is adequate to support the questions raised in the study design, as well as our stated conclusions.

DISCUSSION

From our data,z as well as those of other^,:^-^ it is apparent that a significant percentage of men undergoing radical pros- tatectomy for clinically localized carcinoma of the prostate continue to have a less than optimal pathological outcome. Labrie et a1 reported on 161 patients diagnosed with stage B (134) or C (27) prostate cancer who were randomized to receive 3 months of neoadjuvant combination therapy with flutamide plus a luteinizing hormone-releasing hormone ag- mist followed by radical prostatectomy versus radical pros- tatectomy alone.6 Overall, there was an improved patholog-

the most common presentation of a patient with prostate cancer included a normal digital rectal examination and an elevated serum PSA level. This clinical scenario (stage Tlc lesion) was not part of the study by Labrie et a16 and has not been investigated using neoadjuvant therapy.

Three other similar randomized neoadjuvant studies exist. A Canadian study using 3 months of cyproterone showed decreased positive margin rates from 64.1% in the control group to 33.7% in the treated group.7 Soloway et a1 evaluated stage T2b lesions using a luteinizing hormone-releasing hor- mone agonist and flutamide for 3 months, and the positive margin rate was 48% in the control group and 18% in the treated group.R Finally, a Swedish cooperative group used a luteinizing hormone-releasing hormone analogue for 3 months and noted decreased positive margin rates from 46% in the control group to 24% in the treated group.9

Our results differ from these other studies in several ways. The rate of poor pathological outcomes (positive margins plus seminal vesicle or pelvic nodal involvement) in our control group was substantially lower (18%) than that in the other studies (64.1%, 48% and 46% ). Histological processing of the specimen cannot account for the difference, since our thor- ough evaluation could only increase the rate of positive mar- gins.

Our study was comprised mostly (59% )of stage T lc lesions, a clinical stage not well evaluated in the other randomized studies. The margin positive rate with stage Tlc lesions in our control group was 7% (1 of 13 patients, excluding those with seminal vesicle and lymph node involvement). Regard- less of the biological reason, stage T l c lesions may be more likely to be locally contained. Therefore, attempts a t neoad- juvant treatments to improve pathological outcomes would be fruitless. Another possible explanation lies in the concept of iatrogenic positive margins or surgical incision of the pros- tatic capsule resulting in a positive margm. We had no iat- rogenic positive margins in our study, a point not addressed in other neoadjuvant trials. Capsular incision with a positive margin has been reported commonly,10, accounting for 87% of positive apical margins in 1 study.'" Possibly, suboptimal surgical technique resulted in a greater positive margin rate for the control group in the other studies. Potentially, neoad- juvant treatment may decrease the incidence of iatrogenic positive margins (if, as noted previously, it is significant), a factor not investigated to date.

The improved pathological outcomes using neoadjuvant therapy with stage T2 lesions in other studies is perpleldng. I t is difficult to believe that malignant cells in extraprostatic regions regress to an intraprostatic location with neoadju- vant treatment. Androgen ablation can result in cell death, and possibly neoadjuvant therapy affects the histopatholog-

leal outcome using neoadjuvant androgen ablation therapy. "he positive margin rates were 33.86 in the control group and only 7.8% in the treated group. Importantly, in our series

1360 NEOADJUVANT TESTICULAR ANDROGEN ABLATION THERAPY BEFORE PROSTATECTOMY

ical evaluation of the specimen, making the cancer and its extraprostatic extension more difficult to assess.

Importantly, pathological stage is only a short-term end point. Although pathological outcomes may be improved us- ing neoadjuvant androgen ablation, the natural biological potential may not be affected. More appropriate long-term end points, such as biochemical failure (detectable PSA level) or disease-specific survival, must show improvement with neoadjuvant androgen ablation before recommending its clinical use.

CONCLUSIONS

The purpose of this study was to assess whether neoadju- vant androgen ablation using a luteinizing hormone-releas- ing hormone agonist alone improved pathological outcomes in men with clinically localized adenocarcinoma of the pros- tate. There was no benefit to neoadjuvant androgen ablation therapy in our population, nor were we able to identify a subset of patients, based on PSA or clinical stage, who bene- fited from such treatment. We cannot recommend the use of luteinizing hormone-releasing hormone agonist therapy be- fore radical prostatectomy in men with clinically localized adenocarcinoma of the prostate in an effort to improve patho- logical outcomes.

REFERENCES

1. Catalona, W. J., Richie. J. P., Ahmann, F. R., Hudson, M. A, Scardino, P. T., Flanigan, R. C., deKernion, J. B.. RatlifF, T. L., Kavoussi, L. R., Dalkin. B. L., Waters, W. B., MacFarlane, M. T. and Southwick, P. C.: Comparison of digital rectal ex- amination and serum prostate specific antigen in the early detection of prostate cancer: results of multicenter clinical trial of 6,630 men. J. Urol., 151: 1283, 1994.

2. Ehreth, J. T., Miller, J. I., McBeath, R. B., Hansen, K K, Ahmann, F. R., Dalkin, B. L. and SchX, M., Jr.: Prostate- specific antigen obtained under optimal conditions determines

extracapsular adenocarcinoma of the prostate. Brit. J . Urol., 76: 21,1995.

3. Hudson, M. A., Bahnson, R. R. and Catalona, W. J.: Clinical use of prostate specific antigen in patients with prostate cancer. J. Urol., 142 1011, 1989.

4. partin, A. W., Carter, H. B., Chan, D. W., Epstein, J. I., Oesterling, J. E., Rock, R. C., Weber, J. P. and Walsh, P. C.: prostate specific antigen in the staging of localized prostate cancer: influence of tumor differentiation, tumor volume and benign hyperplasia. J. Urol., 143: 747, 1990.

5. Lange, P. H., Emole, C. J., Lightner, D. J., Fraley, E. E. and Vessella, R.: The value of serum prostate specific antigen determinations before and after radical prostatectomy. J. Urol., 141: 873, 1989.

6. Labrie, F., Cusan, L., Gomez, J. L., Diamond, P. and Suburu, R.: Down-staging of early stage prostate cancer before radical prostatectomy the tirst randomized trial of neoadjuvant com- bination therapy with flutamide and a luteinizing hormone- release hormone agonist. Urol. Symposium, 44: 29, 1994.

7. Goldenberg, S. L., Klotz. L. H.. Jewett, M. A S., Srigley, J., Barkin, J..Mador, D., Fradet, Y., Chin, J. L., Paquin, J. M. and Laplank, S.: Randomized, controlled study of neoadjuvant reversible androgen withdrawal therapy with cyprotemne ac- etate in the surgical management of localized prostate cancer. J. Urol., part 2, lb3: 254A, abstract 103, 1995.

8. Soloway, M. S., Shari& R., Wajsman, Z., McLeod, D., Wood, D. P., Jr. and h a s - B a e z for the Lupron Depot Neoadjuvant Prostate Cancer Study Group: Randomized prospective study comparing radical prostatectomy alone versus radical prosta- tectnmy preceded by androgen blockade in clinical stage B2 (T2bNxMO) prostate cancer. J. Urol., 154: 424, 1995.

9. Pedersen, K. V., Lundberg, S., Hugosson, J., Aus, G., Schelin, S., Ahlgren, G. and Abrahamsson, P A : Neoadjuvant hormonal treatment with triptorelin versus no treatment prior to radical prostatectomy: a prospective randomized multicenter study. J. Urol., part 2,163: 39lA, abstract 651, 1995.

10. Ackerman, D. h, Barry, J. M., Wicklund, R. A., Olson, N. and Lowe, B. A: Analysis of risk factors associated with prostate cancer extension to the surgical margin and pelvic node me- tastases at radical prostatedomy. J. Urol., 150: 1845, 1993.

11. Voges, G. E., McNeal, J. E., Redwine, E. A, Freiha, F. S. and Stamey, T. A: Morphologic analysis of surgical margins with positive findings in prostatectomy for adenocarcinoma of the prostate. Cancer, 69: 520, 1992.