Robot-Assisted Laparoscopic Radical Prostatectomy in Patients with Prostate Cancer with High-Risk Features: Predictors of Favorable Pathologic Outcome

Abstract

Introduction:

Preoperative determination of pathologic outcomes in patients with high-risk prostate cancer is challenging because of limitations of existing nomograms. We aimed to assess whether certain preoperative clinical and pathologic characteristics correlate with pathologic outcome in high-risk prostate cancer patients who underwent robot-assisted laparoscopic radical prostatectomy (RALP).

Methods:

A retrospective evaluation of patients with high-risk disease (prostate-specific antigen [PSA] ≥10 ng/dL with high volume disease or Gleason score ≥8) who underwent RALP between December 2004 and September 2008 was conducted. Patients were grouped based on favorable pathology, including organ-confined disease and negative surgical margins (group 1), and unfavorable pathology, including positive surgical margins and extracapsular extension (group 2). Preoperative PSA levels, transrectal ultrasonography findings, and biopsy reports were compared to final pathology data.

Results:

Of 69 high-risk patients, 37 (54%) had favorable postoperative pathology (group 1) and 32 (46%) had unfavorable pathology (group 2). Mean PSA was 10.0 ng/dL (range, 4.1–20.3) (group 1) and 13.8 ng/dL (range, 3.1–39.9) (group 2). Mean PSA density was 0.28 (group 1) and 0.41 (group 2). Mean positive biopsy core was 33% (group 1) and 44% (group 2). Differences in PSA levels, PSA density, and percentage of positive cores were statistically significant (p < 0.05) between the groups. Bilateral disease and high-grade prostatic intraepithelial neoplasia were not statistically significant (p > 0.05).

Discussion:

Lower PSA level and PSA density, as well as fewer positive biopsy cores, were associated with favorable postoperative pathology. Continued surveillance of these patients will serve to determine whether these findings will assist in predicting which high-risk prostate cancer patients may likely benefit from RALP.

Introduction

A lthough the lifetime risk of developing prostate cancer in the United States is approximately 1 in 6, only 1 in 35 men will die from the disease.¹ Each year, prostate cancer is newly diagnosed in more than 180,000 men, and appropriate treatment for all of them is controversial. Although widespread use of prostate-specific antigen (PSA) testing has led to detection of cancer at earlier stages,² a significant number of men present with aggressive prostate cancer that can be locally advanced or metastatic.

The standard surgical treatment of early prostate cancer is radical prostatectomy, which has shown favorable cancer control rates in low-risk patients.³ The role of this procedure for patients with high-risk disease is unclear. According to the National Cancer Institute and others, patients with high-risk disease are more commonly offered treatment modalities such as radiotherapy, hormonal therapy, or active surveillance because of the concern that surgical treatment may not offer a significant survival benefit and may be associated with unnecessary morbidity.^4
–6 There is evidence to support multi-modality therapy in high-risk patients, including surgery, radiation therapy, hormonal therapy, and chemotherapeutic agents.⁷ Nevertheless, some patients with high-risk prostate cancer who undergo radical prostatectomy may be cured of their disease and may not require further treatment. An aggressive multimodality approach in these patients can subject them to unnecessary morbidity from these treatments. Unfortunately, determination of which patients will be cured by radical prostatectomy as monotherapy is currently not possible.

Various risk stratification systems have been developed using clinical and pathological variables to stratify patients with low, intermediate, and high-risk disease, with the theoretical advantage of allowing treatment to be tailored to the patients in each group. The search for models and nomograms that are efficacious for high-risk patients continues, as current models are thought to be most accurate for low-risk patients.⁸ Therefore, clinicians and patients continue to have uncertainty regarding the most appropriate treatment at the time of diagnosis of high-risk prostate cancer.

The aim of this study was to determine if preoperative variables are useful in predicting favorable pathologic outcomes in high-risk prostate cancer patients undergoing robot-assisted laparoscopic radical prostatectomy (RALP). To our knowledge, there have not been any similar studies specifically examining surgical outcomes after RALP in patients with high-risk prostate cancer.

Methods

A retrospective review was performed of patients undergoing RALP between December 2004 and September 2008 at a single institution, identifying patients with high-risk characteristics. High-risk disease was defined as preoperative PSA ≥10 ng/dL with high volume disease (≥50% of biopsy cores positive for cancer) or Gleason score (GS) ≥8 on transrectal ultrasonography (TRUS) biopsy or final pathology. All patients underwent RALP with bilateral pelvic lymphadenectomy utilizing the standard three-arm da Vinci™ Surgical System (Intuitive Surgical, Inc., Sunnyvale, CA) until January 2007, at which point procedures were performed with a four-arm da Vinci™ Surgical System. Patients with preoperative high-risk features underwent wide excision without neurovascular bundle preservation. Patients were then divided into two groups based on pathological analysis of the radical prostatectomy specimens. Group 1 included patients with favorable prognostic factors, defined as pathologic stage T2 or less and negative surgical margins. Group 2 included patients with unfavorable prognostic factors, defined as pathologic stage T3 or greater or one or more positive surgical margins. Data collected included clinical stage, PSA level, GS, number of biopsy cores positive for cancer, PSA density, and TRUS abnormalities. Pre- and postoperative histology and the use of adjuvant therapies were also evaluated.

Results

Sixty-nine patients underwent RALP for high-risk prostate cancer using the da Vinci™ Surgical System. Forty patients had high PSA levels (mean, 16.4 ng/dL; range, 10–39.9 ng/dL) with high volume disease, and 39 patients had high GSs (GS 8 [n = 24], GS 9 [n = 14], and GS 10 [n = 1]; Tables 1 –3). Ten patients had high PSA levels as well as high GSs. Pathology revealed pT2a (n = 5), pT2b (n = 1), pT2c (n = 38), T3a (n = 13), and pT3b (n = 12) (Table 3). Forty-five patients (65%) had negative and 24 (35%) had positive surgical margins. Extracapsular extension (ECE) was noted in 25 specimens (36%). Seminal vesicle invasion was noted in 12 specimens (17.4%). No pathologic lymph node involvement was noted.

Table 1.

Clinical Data for Patients with Prostate Cancer with High-Risk Characteristics

	Group 1	Group 2	p
Number of patients	37	32
Age (years)	59.8 (±6.5)	60.2 (±6.0)	NS
Median follow-up (months)	24	23	NS
Preoperative PSA (ng/dL)	10.0 (±4.8)	13.8 (±9.7)	<0.05
Preoperative TRUS vol (cm³)	41.9 (±21.5)	35.9 (±14.1)	NS
PSA density	0.28 (±0.18)	0.41 (±0.30)	<0.05
TRUS hypoechoic areas	20%	14%	NS

Group 1 = patients with favorable pathology; Group 2 = patients with unfavorable pathology; NS = not significant; PSA = prostate-specific antigen; TRUS = transrectal ultrasonography.

Table 2.

Pathologic Results from Transrectal Ultrasonography–Guided Biopsies

	Group 1	Group 2	p
Biopsy GS	7.24	7.00	NS
	GS 6 (9)	GS 6 (1)
	GS 7 (17)	GS 7 (13)
GS distribution	GS 8 (5)	GS 8 (9)
	GS 9 (6)	GS 9 (8)
	GS 10 (0)	GS 10 (1)
Number of biopsy cores	10.9	10.6	NS
Mean number of positive cores	3.64	4.64	NS
Percentage of positive cores	33%	44%	<0.05
Bilateral positive biopsies	40%	53%	NS
HGPIN	32%	43%	NS
Inflammation present	25%	20%	NS

GS = Gleason score; HGPIN = high-grade prostatic intraepithelial neoplasia.

Table 3.

Pathologic Results from Prostatectomy Specimens

	Group 1	Group 2	p
Final pathology Gleason	7.24	7.84	NS
	GS 6 (9)	GS 6 (1)
	GS 7 (9)	GS 7 (11)
GS distribution	GS 8 (13)	GS 8 (11)
	GS 9 (6)	GS 9 (8)
	GS 10 (0)	GS 10 (1)
	T2a (4)	T2a pos. margin (1)
	T2b (1)	T2c pos. margin (6)
Pathologic stage (n)	T2c (32)	T3a (13)
		T3b (12)
Positive margins	0%	75%	<0.01
Adjuvant therapies	5.4%	56%	<0.01

A favorable pathological outcome, defined as organ-confined disease (pT2c or less) and negative surgical margins, was present in 37 patients (53.6%) (group 1). The mean age of patients in group 1 was 59.8 years (range, 47–73) and median follow-up was 24 months (range, 6–49 months). Thirty-two patients (46.4%) had an unfavorable pathologic outcome, defined as pathologic T3 disease or pathologic T2 disease with at least one positive surgical margin (group 2). The mean age in group 2 was 60.2 years (range, 47–70) and median follow-up was 23 months (range, 8–46). Data from each group were analyzed and compared based on preoperative clinical parameters, TRUS findings, and prostate biopsy specimen reports (Tables 1 and 2).

PSA analysis

Both groups exhibited PSA elevation when compared with normal age-specific PSA ranges. Group 1 had a mean PSA of 10.0 ± 4.8 ng/dL (range, 4.1–20.3), whereas group 2 had a mean PSA of 13.8 ± 9.7 ng/dL (range, 3.1–39.9). The difference in PSA values was statistically significant (p < 0.05) between the two groups.

PSA density

Both groups exhibited elevated PSA density (>0.1). In group 1, the mean PSA density was 0.28 ± 0.18 (range, 0.10–0.86), and in group 2 the mean PSA density was 0.41 ± 0.32 (range, 0.07–1.49). The difference in PSA density between the two groups was statistically significant (p < 0.05).

Prostate biopsy specimens

Patients in group 1 had an average of 10.9 biopsy cores with an average of 3.6 positive cores (33%). In this group, 15 (40%) patients had bilateral disease based on preoperative TRUS biopsy specimens. In addition, 12 (32%) patients had high-grade prostatic intraepithelial neoplasia (HGPIN) present. Patients in group 2 had an average of 10.6 biopsy cores with an average of 4.6 positive cores (44%). In this group, 17 (53%) patients had bilateral disease and 14 (43%) had HGPIN on preoperative TRUS biopsy specimens. The difference in positive cores between the two groups was statistically significant (p < 0.05).

Adjuvant therapy

Postoperative PSA nadir was <0.1 ng/dL in 100% of patients within group 1. Two patients (5.4%) within this group had biochemical failure within the first year after surgery. These patients received adjuvant radiation therapy with short-term hormonal therapy. Within group 2, 18 patients (56%) received adjuvant therapy because of unfavorable pathology or biochemical failure (Table 3).

Discussion

The widespread availability of PSA testing has resulted in the detection of prostate cancer at earlier clinical stages.² Radical prostatectomy remains the standard surgical therapy for early stage, localized prostate cancer, and is associated with favorable disease-specific outcomes. For patients who are found to have locally advanced disease, treatment may include multimodality therapy with a combination of surgery, radiation therapy, hormonal therapy, or chemotherapy. Patients with high-risk prostate cancer are often excluded from selecting surgical excision as an option to be spared from the morbidity of surgery, especially if adjuvant radiation therapy is to be administered.

In a recent article by Yossepowitch et al⁹ a cohort of 4708 patients who underwent open radical prostatectomy at Memorial Sloan Kettering Cancer Center were evaluated. Several high-risk populations were defined using eight different criteria as previously defined by this group. Depending on which definition was used, patients at high risk comprised 3% to 38% of the study population. The proportion of patients with ECE ranged from 35% to 71% and seminal vesicle invasion ranged from 10% to 33%. In this high-risk population, 22% to 63% had organ-confined disease based on postoperative pathology. The authors concluded that because of the numerous variations in classification of high-risk prostate cancer, many classified in this category will have organ-confined disease and therefore should not be perceived as poor candidates for radical prostatectomy.

For high-risk patients with organ-confined disease, radical prostatectomy as monotherapy may offer a potential cure without the need for additional treatment. In fact, surgery may potentially offer certain patients curative therapy when other modalities may fail. For patients without organ-confined disease, surgical excision of the prostate may offer survival and recurrence-free periods equal to or greater than the other treatment modalities utilized, without resulting in a compromised quality of life that can been seen in patients with bulky locally advanced disease.

With the advent of robot-assisted surgery and its increasing utilization, more patients are being offered a minimally invasive approach for radical prostatectomy, with the benefits of decreased morbidity and a quicker recovery. Although long-term data on PSA progression are not currently available for this evolving technique, short-term data have been published and are encouraging.^10
–12 Some surgeons note that advantages of robotic assistance include decreased blood loss and faster recovery for patients. These opinions have led some investigators to question whether the role of RALP for high-risk prostate cancer needs reevaluation.

In our review, we assessed preoperative clinical and pathologic features of high-risk patients who underwent RALP. Our goal was to characterize variables and analyze the data to determine if there was a correlation with a favorable pathologic outcome after RALP. The definition of high-risk disease that we utilized included GS ≥8 on either TRUS biopsy or final pathology, or PSA ≥10 ng/dL with high volume disease based on preoperative biopsies, as adopted from the D'Amico criteria for risk stratification.¹³ Of patients with elevated PSA but GS <8, the mean positive core percentage was 52% and PSA density was >0.3. These data indicate high volume disease and are associated with a greater risk for an unfavorable pathologic outcome based on the D'Amico criteria. Favorable pathologic features were defined as stage pT2 or less with negative margins, whereas unfavorable features were defined as stage pT3 or higher or positive margins. The rationale for this division arises from the increased likelihood of biochemical recurrence noted from validated nomograms in the setting of ECE or positive margins.^14

–17 Patients without evidence of ECE but with positive surgical margins were included in the unfavorable pathology group, as these patients have been shown to have similar long-term disease-free survival as those with ECE and negative margins. As such, some investigators have advocated for positive margins to be classified similarly to stage pT3.¹⁸

In our retrospective analysis, we found that higher preoperative PSA levels, higher PSA density, and a greater percentage of positive TRUS biopsy cores correlated with unfavorable pathologic findings. The presence of bilateral disease or HGPIN in the biopsy specimens were seen more commonly in the unfavorable pathology cohort, however was not statistically significant (Table 2). Thus, within this high-risk cohort, patients with lower PSA levels or PSA density, or fewer positive cores, had a higher likelihood of having favorable pathology after RALP. In the setting of this study, patients with favorable pathology appear to have benefited from undergoing surgical excision, as 35 patients (94.6%) remain without evidence of disease and have not required adjuvant therapies.

There are a number of limitations within this study. The review is retrospective in nature, which limits its broader applicability. This degree of positive margins is a result of the overall high-risk nature of the group; however, there is controversy as to whether focal positive margins, especially at the apex of the prostate, portend a poorer long-term prognosis.¹⁹ The small cohort of patients and short follow-up is yet another limitation despite the fact that all patients were followed for a minimum of 6 months. It must be noted that selecting appropriate candidates with clinically localized high-risk prostate cancer who are interested in undergoing RALP is a challenging task. Our patients in each group will continue to be closely followed as we plan to report intermediate and long-term follow-up in the future.

Determination of specific patients with high-risk prostate cancer who are likely to exhibit disease recurrence and metastasis is especially difficult. Existing cancer prediction models such as Partin²⁰ and Kattan²¹ nomograms have been well validated for low risk disease, but equivalent efficacy in the high-risk population has not been demonstrated.²² Certain patients with high-risk disease are likely to be cured with surgery alone. The advent of robotic assistance allows appropriately selected high-risk patients to elect for RALP with the potential for cure. The reduced morbidity of the laparoscopic approach allows patients a quicker recovery in the event that adjuvant therapy is necessary. Our study has provided us with some insight in predicting favorable pathologic outcomes in patients with GS ≥8 but with lower PSA levels, as suggested by Partin and Kattan nomograms. In addition, fewer positive cores, or lower PSA density, were also associated with more favorable results as suggested by the D'Amico criteria for risk stratification. The findings of this study highlight the necessity for additional clinical information that can aid in determining which high-risk patients are best suited for radical prostatectomy.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

American Cancer Society. Cancer Statistics, 2008. www.cancer.org. 2009 March 30.

Brawley

. Prostate carcinoma incidence and patient mortality: The effects of screening and early detection. Cancer, 1997; 80:1857–1863.

Pound

, Partin

, Eisenberger

, Chan

, Pearson

, Walsh

. Natural history of progression after PSA elevation following radical prostatectomy. JAMA, 1999; 281:1642–1645.

National Cancer Institute Prostate Cancer (PDQ): Treatment. www.cancer.gov/cancertopics/pdq/treatment/prostate/Patient/page5. March 20 .

Cooperberg

, Grossfeld

, Lubeck

, Carroll

. National practice patterns and time trends in androgen ablation for localized prostate cancer. J Nat Cancer Inst, 2003; 95:930–931.

Mendenhall

, Henderson

, Mendenhall

. Definitive radiotherapy for prostate cancer. Am J Clin Oncol, 2008; 31:496–503.

Jones

, Koeneman

. Local-regional prostate cancer. Urol Oncol Semin Orig Investig, 2008; 26:516–521.

Nguyen

, Reuther

, Stephenson

, Klein

, Jones

. The specific definition of high risk prostate cancer has minimal impact on biochemical relapse-free survival. J Urol, 2009; 181:75–80.

Yossepowitch

, Eggener

, Bianco

Jr.

et al. Radical prostatectomy for clinically localized, high risk prostate cancer: Critical analysis of risk assessment methods. J Urol, 2007; 178:493–499.

10.

Trabulsi

, Guillonneau

. Laparoscopic radical prostatectomy. J Urol, 2005; 173:1072–1079.

11.

Patel

, Thaly

, Shah

. Robotic radical prostatectomy: Outcomes of 500 cases. BJU Int, 2007; 99:1109–1112.

12.

Mikhail

, Orvieto

, Billatos

et al. Robotic-assisted laparoscopic prostatectomy: First 100 patients with one year of follow-up. Urology, 2006; 68:1275–1279.

13.

D'Amico

, Whittington

, Malkowicz

et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA, 1998; 280:969–974.

14.

Stephenson

, Scardino

, Eastham

et al. Preoperative nomogram predicting the 10 year probability of prostate cancer recurrence after radical prostatectomy. J Nat Can Inst, 2006; 98:715–717.

15.

Cheng

, Darson

, Bergstralh

et al. Correlation of margin status and extraprostatic extension with progression of prostate carcinoma. Cancer, 1999; 86:1775–1782.

16.

Wieder

, Soloway

. Incidence, etiology, location, prevention and treatment of positive surgical margins after radical prostatectomy for prostate cancer. J Urol, 1998; 160:299–315.

17.

Blute

, Bostwick

, Seay

et al. Pathologic classification of prostate carcinoma: The impact of margin status. Cancer, 1998; 82:902–908.

18.

Freedland

, Aronson

, Presti

Jr.

et al. Should a positive surgical margin following radical prostatectomy be pathologic stage T2 or T3? Results from the SEARCH database. J Urol, 2003; 169:2142–2146.

19.

Eastham

, Kuroiwa

, Ohori

et al. Prognostic significance of location of positive margins in radical prostatectomy specimens. Urology, 2007; 70:965–969.

20.

Makarov

, Trock

, Humphreys

et al. Updated nomogram to predict pathologic stage of prostate cancer given prostate-specific antigen level, clinical stage, and biopsy Gleason score (Partin tables) based on cases from 2000 to 2005. Urology, 2007; 69:1095–1101.

21.