- Email: [email protected]
Long-Term Outcome After Radical Prostatectomy for Patients With Lymph Node Positive Prostate Cancer in the Prostate Specific Antigen Era
Long-Term Outcome After Radical Prostatectomy for Patients With Lymph Node Positive Prostate Cancer in the Prostate Specific Antigen Era Stephen A. Boorjian, R. Houston Thompson, Sameer Siddiqui, Stephanie Bagniewski, Erik J. Bergstralh, R. Jeffrey Karnes, Igor Frank and Michael L. Blute* From the Department of Urology and Division of Biostatistics (SB, EJB), Mayo Medical School and Mayo Clinic, Rochester, Minnesota
Purpose: While the incidence of lymph node positive prostate cancer has decreased during the prostate specific antigen era, the optimal treatment of these patients remains in question. We examined the impact of lymph node metastases on the outcome of patients following radical prostatectomy and investigated prognostic factors that affect survival. Materials and Methods: We identified 507 men treated with radical prostatectomy between 1988 and 2001 who had lymph node positive disease. Of the 507 patients 455 (89.7%) were treated with adjuvant hormonal therapy. Median followup was 10.3 years (IQR 6.1–13.5). Postoperative survival rates were estimated using the Kaplan-Meier method and the impact of various clinicopathological factors on outcome was analyzed using Cox proportional hazard regression models. Results: Ten-year cancer specific survival for patients with positive lymph nodes was 85.8% with 56% of the men free from biochemical recurrence at last followup. On multivariate analysis pathological Gleason score 8 –10 (p ⫽ 0.004), positive surgical margins (p ⫽ 0.016), nondiploid tumor ploidy (p ⫽ 0.023) and 2 or greater positive nodes (p ⫽ 0.001) were adverse predictors of cancer specific survival. Tumor stage, year of surgery and total number of nodes removed did not significantly affect outcome. Adjuvant hormonal therapy decreased the risk of biochemical recurrence (p ⬍0.001) and local recurrence (p ⫽ 0.004) but it was not associated with systemic progression (p ⫽ 0.4) or cancer specific survival (p ⫽ 0.4). Conclusions: Radical prostatectomy may offer long-term survival to patients with lymph node positive prostate cancer. Gleason score, margin status, tumor ploidy and the number of involved nodes predict survival, while the role of adjuvant hormonal therapy continues to be defined. Key Words: prostate, prostatic neoplasms, lymph node dissection, prostate-specific antigen, prostatectomy
s part of the stage migration in prostate cancer during the PSA era,1 the incidence of positive lymph nodes in patients undergoing RRP has decreased to approximately 4% in recent series.2,3 While lymph node metastases adversely impact the prognosis of patients following surgery, extended postoperative survival in a subset of these men was reported.2–7 However, studies to date of patients with lymph node disease have been largely composed of pre-PSA and early PSA era cohorts, when the clinicopathological features of prostate cancer tend to be more advanced.1 While men with lymph node metastases continue to be identified, optimal treatment for these patients is not well established. Traditionally when lymph node disease was found at surgery, RRP was abandoned and patients were treated with hormones alone. However, several studies now show a trend toward improved survival for patients treated with prostatectomy and hormone therapy compared to patients treated with androgen ablation only.8 –10 These studies have been criticized for their retrospective design and for incompletely controlling for clinicopathological variables.
Nevertheless, a potential importance for addressing the primary tumor in patients with advanced prostate cancer has been suggested. Moreover, because most patients with positive lymph nodes are currently found to have microscopic disease on pathological examination of the nodes after RRP, the appropriate postoperative treatment of such patients similarly remains in question. In particular the optimal time to initiate hormone therapy continues to be debated.11,12 The identification of risk factors for disease progression in patients with nodal disease may help individualize treatment for these patients. We report the long-term outcome in men with lymph node positive prostate cancer treated with RRP at our institution during the PSA era. In addition, clinicopathological features were analyzed to identify predictors of disease progression.
A
MATERIALS AND METHODS After institutional review board approval was obtained we reviewed the records of 10,261 consecutive patients who underwent RRP and bilateral pelvic lymph node dissection between 1988 and 2001 at our institution. Patients found to have positive lymph nodes at surgery were identified. Patients who received hormonal treatment or radiotherapy before RRP were excluded from analysis.
Submitted for publication January 17, 2007. Study received institutional review board approval. * Correspondence: 200 First St. Southwest, Rochester, Minnesota 55905 (telephone: 507-284-3982; FAX: 507-284-4987; e-mail: blute. [email protected]).
0022-5347/07/1783-0864/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION
864
Vol. 178, 864-871, September 2007 Printed in U.S.A. DOI:10.1016/j.juro.2007.05.048
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER Surgical procedures were performed by different surgeons using standardized techniques. The 1997 TNM classification system was used for staging and the Gleason system was used for grading. DNA ploidy was assessed by flow cytometry. Adjuvant therapy was administered at the discretion of the treating physician and it consisted of treatment received within 90 days of RRP. Medical hormone deprivation therapy was generally intended to be lifelong. However, given the retrospective nature of this study, it is uncertain whether patients discontinued treatment after a period of ADT. Postoperative assessments, including physical examination and serum PSA measurement, were done quarterly for the initial 2 years, semiannually for an additional 2 years and annually thereafter. Radiographic evaluation was done as indicated clinically. BCR was defined as PSA 0.4 ng/ml or greater. Local recurrence was defined as cancer on biopsy of the prostatic bed or salvage radiation therapy to the prostatic bed without evidence of systemic recurrence. Systemic progression involved demonstrable metastatic deposits on radionuclide bone scan or on biopsies other than biopsy of the prostatic bed. Cause of death was identified from death certificates or physician correspondence. For patients followed elsewhere the prostatectomy registry at our institution monitors outcomes annually by correspondence. Of the 10,261 patients studied 246 (2.4%) were lost to followup. Statistical analyses were done using SAS®, version 8.2. Demographic comparisons were performed using the chisquare and rank sum tests as appropriate. Postoperative survival was estimated using the Kaplan-Meier method with patients censored at last followup or death if the end point of interest was not attained. The impact of various clinicopathological factors on outcome was analyzed using Cox proportional hazard regression models. The HR and
TABLE 1. Patient demographics Feature No. pts No. surgery era (%): 1988–1993 1994–1997 1998–2001 Median age at surgery (range) Median No. nodes removed (range) No. ng/ml preop PSA (%): Less than 10 10–19.9 20 or Greater No. pathological Gleason score (%): 6 or Less 7 8–10 No. pathological tumor stage (%): T2 T3/T4 No. seminal vesicle invasion (%): No Yes No. pos surgical margin (%): No Yes No. tumor DNA ploidy (%): Diploid Nondiploid
Pos Nodes
Neg Nodes
507
9,754
315 (9.1) 135 (3.7) 57 (1.8) 66.0 (47–79)
3,133 (90.9) 3,467 (96.3) 3,154 (98.2) 64.0 (34–84)
p Value ⬍0.001
11
(1–37)
8
(1–83)
142 150 215
(28) (29.6) (42.4)
7,227 1,755 772
(74.1) (18) (7.9)
109 267 131
(21.5) (52.7) (25.8)
6,521 2,698 535
(66.8) (27.7) (5.5)
73 434
(14.4) (85.6)
7,375 2,379
(75.6) (24.4)
170 337
(33.5) (66.5)
8,884 869
(91.1) (8.9)
192 315
(37.9) (62.1)
6,616 3,138
(67.8) (32.2)
247 247
(50) (50)
6,788 2,524
(72.9) (27.1)
0.01 ⬍0.001 ⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001 ⬍0.001
865
TABLE 2. Clinicopathological characteristics of patients by AHT Feature
AHT
No. AHT*
p Value
No. pts Median age at surgery (range) Median No. nodes removed (range) No. pos nodes (%): 1 2 or Greater No. ng/ml preop PSA (%): Less than 10 10–19.9 20 or Greater No. pathological Gleason score (%): 6 or Less 7 8–10 No. pathological tumor stage (%): T2 T3/T4 No. seminal vesicle invasion (%): No Yes No. pos surgical margin (%): No Yes No. tumor DNA ploidy (%): Diploid Nondiploid No. adjuvant radiation (%): No Yes
455 66.0 (47–79)
52 63.0 (47–76)
0.01
11.5 (1–37)
10.5 (4–25)
0.08 ⬍0.001
245 210
(53.8) (46.2)
45 7
(86.5) (13.5)
124 130 201
(27.3) (28.6) (44.2)
18 20 14
(34.6) (38.5) (26.9)
0.06
0.02 95 234 126
(20.8) (51.4) (27.7)
14 33 5
(27) (63.5) (9.6) 0.14
62 393
(13.6) (86.4)
11 41
(21.2) (78.8) 0.001
142 313
(31.2) (68.8)
28 24
(53.8) (46.2)
164 291
(36) (64)
28 24
(53.8) (46.2)
214 234
(47.8) (52.2)
33 13
(71.7) (28.3)
413 42
(90.8) (9.2)
49 3
(94.2) (5.8)
0.01 0.002 0.41
* Of 52 patients 30 (57.7%) were ultimately treated with hormone deprivation for PSA only recurrence in 21, local recurrence in 2 and systemic progression in 7.
95% CI associated with the presence and number of positive lymph nodes were estimated using the proportional hazards model. All tests were 2-sided with p ⱕ0.05 considered significant. RESULTS Of the 10,261 men (4.9%) who underwent RRP between 1988 and 2001, 507 had positive lymph nodes. The incidence of positive lymph nodes at RRP decreased significantly during the study period from 315 of 3,448 patients (9.1%) between 1988 and 1993 to 57 of 3,211 (1.8%) between 1998 and 2001 (p ⬍0.001). Table 1 lists clinicopathological features of patients with and without nodal disease. Patients with positive lymph nodes had a greater number of lymph nodes removed and higher preoperative PSA, and they were more likely to have nonorgan confined tumors. Of 507 patients with nodal metastases 199 (39.3%) underwent preoperative computerized tomography or magnetic resonance imaging and only 2 had radiographic evidence of nodal disease. Postoperatively 455 of 507 men (89.7%) received AHT. Table 2 lists patient clinicopathological features by AHT. The use of AHT remained relatively constant during the study period since 285 of 315 patients (90.5%) with nodal metastases treated between 1988 and 1993 received AHT vs 116 of 135 (85.9%) treated between 1994 and 1997, and 54 of 57 (94.7%) treated from 1998 to 2001. In addition, 44 of 507 patients (8.7%) received adjuvant radiation therapy, while 39 of 507 (7.7%) were treated with salvage radiation therapy. Four patients received other treat-
866
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
FIG. 1. Outcome following RRP for patients with positive lymph nodes. LR, local recurrence. SP, systemic progression
ment after RRP, of which all were given at systemic progression, including chemotherapy in 1, strontium in 1 and steroids in 2. At a median followup after RRP of 10.3 years (IQR 6.1– 13.5) 213 patients had BCR, 51 experienced local recurrence, 97 had systemic relapse and 200 had died, including 72 of prostate cancer. Figure 1 shows Kaplan-Meier survival curves for patients with lymph node positive disease. Tenyear event-free survival for patients with positive lymph
nodes was 56%, 89%, 80% and 86% for BCR, local recurrence, systemic progression and cancer death, respectively. Postoperative survival was then stratified for all patients who underwent RRP by the number of positive lymph nodes (figs. 2 to 5). Although a single metastatic node significantly increased the risk of BCR (HR 1.4, 95% CI 1.1–1.7, p ⬍0.001), increasing nodal involvement did not worsen the patient risk of BCR (fig. 2). On the other hand, while patients with a single positive node were not at increased risk
FIG. 2. Postoperative BCR-free survival according to degree of lymph node involvement
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
867
FIG. 3. Postoperative local recurrence-free survival according to degree of lymph node involvement
for local recurrence compared to patients without nodal disease (HR 1.1, 95% CI 0.8 –1.7, p ⫽ 0.56), increasing nodal burden did increase the risk of local recurrence (HR 1.8, 95% CI 1.2–2.6, p ⫽ 0.004, fig. 3). The risk of systemic progression similarly increased with increasing nodal burden (fig. 4). While patients with a single positive node were at 3-fold increased risk for systemic progression (HR 2.9, 95% CI 2.1– 4.0, p ⬍0.001), patients with 2 or
greater nodes were approximately twice as likely to experience systemic progression as patients with 1 positive node (HR 1.9, 95% CI 1.2–2.8, p ⫽ 0.003). For CSS 1 lymph node metastasis increased the risk of death almost 4-fold (HR 3.6, 95% CI 2.4 –5.4, p ⬍0.001), while the risk of prostate cancer death in patients with 2 or greater positive lymph nodes was increased 2-fold compared to that in patients with 1 positive node (HR 2.2, 95% CI 1.3–3.5, p ⫽ 0.001, fig. 5).
FIG. 4. Postoperative systemic progression-free survival according to degree of lymph node involvement
868
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
FIG. 5. Postoperative CSS according to degree of lymph node involvement
We next examined risk factors for disease progression in patients with positive lymph nodes. In a multivariate Cox model only increased preoperative PSA (p ⫽ 0.036) and nondiploid tumor ploidy (p ⫽ 0.006) were significant predictors of BCR, while pathological Gleason score 8 –10 was associated with a trend toward BCR (p ⫽ 0.078, table 3). A positive surgical margin was the only predictor of local recurrence (p ⫽ 0.038). Patients with 2 or greater positive nodes (p ⫽ 0.003) and patients with Gleason 8 –10 cancer (p ⫽ 0.009) were at significantly increased risk for systemic progression, while a positive surgical margin also tended to increase this risk (p ⫽ 0.052). Lymph node burden (p ⫽ 0.001), Gleason score 8 –10 (p ⫽ 0.004), nondiploid cancers (p ⫽ 0.023) and positive surgical margins (p ⫽ 0.016) predicted prostate cancer death. Pathological tumor
stage, year of surgery and the total number of nodes examined did not affect survival. AHT was associated with a decreased risk of BCR (p ⬍0.001) and local recurrence (p ⫽ 0.004) but it did not significantly impact systemic progression (p ⫽ 0.38) or CSS (p ⫽ 0.43). Ten-year event-free survival for patients who did not receive AHT was 17.9%, 77.5%, 88.1% and 100% for BCR, local recurrence, systemic progression and prostate cancer death, respectively. DISCUSSION We present what is to our knowledge the largest series to date of patients with lymph node positive disease treated during the PSA era. Experience with RRP for lymph node positive prostate cancer at our institution was previously
TABLE 3. Multivariate analysis of risk factors associated with outcomes after RRP in patients with lymph node positive disease BCR Risk Factor No. pos nodes (2 or greater vs 1) Tumor stage (pT3/4 vs pT2) Preop PSA Gleason score (8–10 vs 7 or less) Nondiploid vs diploid tumor Pos surgical margin Total No. nodes removed RRP yr AHT vs no AHT
HR (95% CI)
Local Recurrence
p Value (chi-square test)
HR (95% CI)
p Value (chi-square test)
Systemic Progression HR (95% CI)
p Value (chi-square test)
CSS HR (95% CI)
p Value (chi-square test)
1.2
(0.8–1.6)
0.37
1.7
(0.9–3.1)
0.11
1.9
(1.2–2.8)
0.003
2.2
(1.3–3.5)
0.001
1.1
(0.7–1.7)
0.79
1.4
(0.5–4.0)
0.55
1.1
(0.5–2.3)
0.79
2.2
(0.7–7.1)
0.20
1.1 1.3
(1.0–1.3) (1.0–1.8)
0.036 0.078
0.95 (0.8–1.2) 1.5 (0.8–2.8)
0.63 0.22
0.96 1.8
(0.8–1.1) (1.2–2.8)
0.63 0.009
0.96 (0.8–1.2) 2.0 (1.3–3.3)
0.66 0.004
1.5
(1.1–2.0)
0.006
1.7
(0.9–3.0)
0.096
1.2
(0.8–1.8)
0.34
1.8
(1.1–2.9)
0.023
1.2
(0.9–1.7)
0.25
2.1
(1.0–4.1)
0.038
1.6
(1.0–2.6)
0.052
2.1
(1.2–3.9)
0.016
1.0 (0.98–1.04)
0.42
1.0 (0.94–1.1)
0.99
0.99 (0.95–1.03)
0.69
0.98 (0.94–1.0)
0.50
1.0 (0.96–1.1) 0.2 (0.1–0.3)
0.70 ⬍0.001
1.0 0.3
0.57 0.004
1.0 0.7
0.88 0.38
1.0 (0.94–1.1) 1.8 (0.42–7.5)
0.50 0.43
(0.9–1.1) (0.1–0.7)
(0.9–1.1) (0.3–1.5)
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER reported in patients in the pre-PSA and early PSA eras.4,9 However, because PSA screening has impacted the clinical and pathological features of patients diagnosed with prostate cancer,1 we updated our experience to report the longterm survival of these patients and evaluate potential prognostic features. We found that CSS depends on the degree of lymph node involvement, aspects of primary tumor pathology and surgical technique. The impact of nodal burden is consistent with previous results from our group4 and others,2–3,5 demonstrating decreased survival with increasing lymph node disease. However, the reported significance of primary tumor pathological features in patients with positive lymph nodes is inconsistent.2–3,5 For example, we noted a significant impact of Gleason score on the risk of systemic progression and CSS but found that tumor stage was not predictive of outcome. However, only 14% of our patients had organ confined disease, which may have obscured the assessment of its independent predictive value. Tumor stage was not a significant predictor of outcome for patients with nodal disease in several previous studies.2,5 Interestingly we found that positive surgical margins were significantly associated not only with local recurrence, which might be expected, but also with prostate cancer death. Again, the impact of surgical margins on outcome for patients with lymph node metastases, which was also reported previously,2 may reflect the high proportion of patients in our study with positive margins (62%). Nevertheless, this finding supports the importance of addressing the primary tumor in the setting of systemic disease, potentially for local control and removal of a source of cancer cells that may continue to metastasize. In fact, a benefit of RRP vs hormonal therapy alone was reported for decreasing the incidence of local complications, such as hematuria and urethral stricture, in patients with node positive prostate cancer.13 Meanwhile, the therapeutic value of pelvic lymph node dissection continues to be investigated. It was suggested that lymphadenectomy may be curative for patients with minimal nodal metastases.2,6 –7 Masterson et al noted improved BCR-free survival with increased nodes removed in patients without histological evidence for nodal metastases.7 They speculated that pelvic lymphadenectomy may offer a therapeutic effect through the removal of micrometastases not detected by routine histological evaluation. As a corollary to this hypothesis, prostate cancer cells were detected in archival tissue from 79% of patients with high risk prostate cancer without histological evidence of metastases in 1 study.14 However, DiMarco et al from our institution did not find a similar relationship between the extent of lymph node dissection and outcome in node negative cases.15 We noted that the total number of nodes removed did not affect the outcome in patients with lymph node positive disease. We believe that the treatment of patients with advanced prostate cancer should involve a multimodal approach. While the fact that almost 90% of our patients received AHT may be criticized as a confounder of the impact of RRP alone on lymph node positive prostate cancer, we assert that this reflects a practice that is not uncommon among urologists today. We found that AHT decreased the risk of BCR and local recurrence but did not
869
significantly impact systemic progression or CSS. However, our overall 10-year systemic progression-free rate of 80% in patients with lymph node positive disease compares favorably with the results of Daneshmand et al, who reported a 10-year clinical recurrence-free survival of 65% in their series of 235 patients with node positive prostate cancer, in which 69% received no adjuvant treatment.3 Likewise, our 5-year CSS rate of 94% compares with the 74% 5-year CSS noted by Bader et al in 88 patients with lymph node metastases who underwent RRP without AHT.5 Our 10-year survival data are similar to the updated results from the treatment arm of the ECOG randomized, prospective trial of immediate vs delayed hormonal therapy in men with lymph node positive prostate cancer after RRP.11 In that series 15% of men who received immediate hormone therapy after RRP died of prostate cancer and 40% experienced clinical progression. The appropriate timing of ADT in prostate cancer has been debated.11,12 The lack of an effect of AHT on CSS in our study limits our ability to conclude that immediate therapy is best. In addition, a recent trial of immediate vs deferred ADT in patients with newly diagnosed prostate cancer showed no significant difference in prostate cancer mortality or symptom-free survival.12 However, the ECOG study demonstrated that at a median followup of 11.9 years patients with lymph node positive disease treated with immediate ADT after RRP had significant improvement in overall survival, CSS and progression-free survival.11 The argument for early treatment of patients is based in part on the concept that ADT may be most effective against small tumor volume.16 Early delivery of AHT may arrest stromal-epithelial interaction, which was suggested to contribute to the growth of micrometastases into clinically evident lesions.17 A future challenge is to individualize the timing and selection of systemic treatments, including AHT, in patients with advanced prostate cancer based on an assessment of the patient risk of progression. One such potential predictor of the patient response to AHT is DNA tumor ploidy.18 We found that men with nondiploid tumors were at significantly increased risk for BCR and prostate cancer death, suggesting that they may benefit from systemic treatments such as chemotherapy or investigational agents. Our analysis was in part limited by the fact that the boundaries of lymph node dissection were not standardized. Indeed, the extent of lymph node dissection correlates with the incidence of positive nodes.5– 6,19 In addition, at our institution lymph nodes are submitted en bloc from each side of the pelvis, rather than as separate packets, which has been shown to yield increased nodal counts.20 Therefore, we cannot make conclusions from our data about the optimal anatomical extent of dissection that should be performed. Although our study represents a retrospective, single center experience, the decreasing incidence of lymph node positive prostate cancer during the PSA era makes future prospective, randomized trials of these patients unlikely. Therefore, large, retrospective series with long-term followup remain an important tool for risk stratification and assessment of treatment outcomes.
870
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
CONCLUSIONS RRP with pelvic lymph node dissection may offer long-term survival to patients with node positive prostate cancer. Pathological Gleason score, surgical margin status, tumor ploidy and the number of involved nodes significantly predict survival. The value of AHT for patients with lymph node metastases continues to be defined.
12.
13.
14.
Abbreviations and Acronyms ADT AHT BCR CSS ECOG PSA RRP
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
androgen deprivation adjuvant hormonal therapy biochemical recurrence cancer specific survival Eastern Cooperative Oncology Group prostate specific antigen radical retropubic prostatectomy
15.
16.
17.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Han M, Partin AW, Piantadosi S, Epstein JI and Walsh PC: Era specific biochemical recurrence-free survival following radical prostatectomy for clinically localized prostate cancer. J Urol 2001; 166: 416. Palapattu GS, Allaf ME, Trock BJ, Epstein JI and Walsh PC: Prostate specific antigen progression in men with lymph node metastases following radical prostatectomy: results of long-term followup. J Urol 2004; 172: 1860. Daneshmand S, Quek ML, Stein JP, Lieskovsky G, Cai J, Pinski J et al: Prognosis of patients with lymph node positive prostate cancer following radical prostatectomy: longterm results. J Urol 2004; 172: 2252. Cheng L, Zincke H, Blute ML, Bergstralh EJ, Scherer B and Bostwick DG: Risk of prostate carcinoma death in patients with lymph node metastasis. Cancer 2001; 91: 66. Bader P, Burkhard FC, Markwalder R and Studer UE: Disease progression and survival of patients with positive lymph nodes after radical prostatectomy. Is there a chance of cure? J Urol 2003; 169: 849. Allaf ME, Palapattu GS, Trock BJ, Carter HB and Walsh PC: Anatomical extent of lymph node dissection: impact on men with clinically localized prostate cancer. J Urol 2004; 172: 1840. Masterson TA, Bianco FJ Jr, Vickers AJ, DiBlasio CJ, Fearn PA, Rabbani F et al: The association between total and positive lymph node counts, and disease progression in clinically localized prostate cancer. J Urol 2006; 175: 1320. Cadeddu JA, Partin AW, Epstein JI and Walsh PC: Stage D1 (T1–3, N1–3, M0) prostate cancer: a case-controlled comparison of conservative treatment versus radical prostatectomy. Urology 1997; 50: 251. Ghavamian R, Bergstralh EJ, Blute ML, Slezak J and Zincke H: Radical retropubic prostatectomy plus orchiectomy versus orchiectomy alone for pTxN⫹ prostate cancer: a matched comparison. J Urol 1999; 161: 1223. Grimm MO, Kamphausen S, Hugenschmidt H, StephanOdenthal M, Ackermann R and Vogeli TA: Clinical outcome of patients with lymph node positive prostate cancer after radical prostatectomy versus androgen deprivation. Eur Urol 2002; 41: 628. Messing EM, Manola J, Yao J, Kiernan M, Crawford D, Wilding G et al: Immediate versus deferred androgen deprivation treatment in patients with node- positive prostate cancer after radical prostatectomy and pelvic lymphadenectomy. Lancet Oncol 2006; 7: 472.
18.
19.
20.
Studer UE, Whelan P, Albrecht W, Casselman J, de Reijke T, Hauri D et al: Immediate or deferred androgen deprivation for patients with prostate cancer not suitable for local treatment with curative intent: European Organisation for Research and Treatment of Cancer (EORTC) Trial 20891. J Clin Oncol 2006; 24: 1868. Frazier HA 2nd, Robertson JE and Paulson DF: Does radical prostatectomy in the presence of positive pelvic lymph nodes enhance survival? World J Urol 1994; 12: 308. Ferrari AC, Stone NN and Eyler JN: Prospective analysis of prostate-specific markers in pelvic lymph node of patients with high-risk prostate cancer. J Natl Cancer Inst 1997; 89: 1498. DiMarco DS, Zincke H, Sebo TJ, Slezak J, Bergstralh EJ and Blute ML: The extent of lymphadenectomy for pTXN0 prostate cancer does not affect prostate cancer outcome in the prostate specific antigen era. J Urol 2005; 173: 1121. Isaacs JT: Relationship between tumor size and curability of prostatic cancer by combined chemo-hormonal therapy in rats. Cancer Res 1989; 49: 6290. Sokoloff MH, Rinker-Schaeffer CW, Chung LWK and Brendler CB: Adjunctive therapy for men with high risk localized and locally advanced prostate cancer: targeting disseminated tumor cells. J Urol 2004; 172: 2539. Zincke H, Bergstralh EJ, Larson-Keller JJ, Farrow GM, Myers RP, Leiber MM et al: Stage D1 prostate cancer treated by radical prostatectomy and adjuvant hormonal treatment: Evidence for favorable survival in patients with DNA diploid tumors. Cancer 1992; 70: 311. Heidenreich A, Varga Z and von Knobloch R: Extended pelvic lymphadenectomy in patients undergoing radical prostatectomy: high incidence of lymph node metastasis. J Urol 2002; 167: 1681. Bochner BH, Herr HW and Reuter VE: Impact of separate versus en bloc pelvic lymph node dissection on the number of lymph nodes retrieved in cystectomy specimens. J Urol 2001; 166: 2295.
EDITORIAL COMMENT These authors present a strong argument for performing randomized, prospective, controlled trials. Results in patients with N⫹ disease undergoing prostatectomy who received immediate ADT are similar to those in the early treatment arm of the randomized ECOG study and to prior Mayo Clinic reports (references 9 and 11 in article). However, this study could not distinguish between the benefits of early vs deferred ADT in this scenario because 1) only 10% of all patients were not treated with immediate ADT, 2) those who did not receive immediate ADT were significantly younger and at far lower risk for progression than those who did (table 2) and 3) as opposed to the ECOG study, in which deferred ADT was withheld almost exclusively until osseous metastases occurred, the majority of patients in this series receiving delayed ADT received it upon biochemical failure only. Thus, not only were the 2 groups dramatically different in risk for progression, but early ADT was compared with almost early ADT as opposed to very delayed ADT. It should be noted that the 17.9% of patients with BCR-free survival who did not receive immediate ADT at 10 years is similar to that in the deferred arm of the ECOG study. Unfortunately the duration of ADT for many treated patients is unknown and the morbidity of long-term ADT was not reported. Despite the ECOG study the precise time to administer ADT in men with N⫹ prostate cancer remains debated and the optimal duration of therapy remains uncer-
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER tain. Unfortunately this series cannot answer these questions. Edward M. Messing Department of Urology University of Rochester James P. Wilmot Cancer Center Rochester, New York REPLY BY AUTHORS While we recognize that prospective randomized clinical trials represent the optimal way to establish treatment patterns, nevertheless the decreasing incidence of lymph node positive prostate cancer noted in the more recent PSA era (1.8% of RRPs performed at Mayo Clinic between 1998 and 2001) makes future prospective randomized trials of these patients difficult. Indeed, the only randomized trial of hormone therapy in this setting involved only 98 patients (reference 11 in article) despite accrual during the early PSA era when the incidence of lymph node metastases in patients undergoing surgery for prostate cancer was significantly higher (315 men were treated with RRP at our institution during that time for lymph node positive disease). Therefore, we believe that large, retrospective series with long-term followup remain an important tool for risk stratification and assessment of treatment outcomes. Moreover, we did not intend to evaluate the impact of timing of hormone initiation or duration on patient outcome following surgery for lymph node positive disease. Instead our goal was to present the outcomes of our approach to
871
patients with advanced prostate cancer during the PSA era, which has been to advocate aggressive local therapy plus early systemic therapy. Interestingly our results are remarkabley similar to the treatment arm of the randomized ECOG study (reference 11 in article). Furthermore, recent data have demonstrated that prostate cancer cells may disseminate to the bone marrrow of patients before the development of clinically evident metastases, and this dissemination may be decreased with androgen ablation.1 Therefore, administration of hormone therapy at the time that histological lymph node disease is identified may in fact be relatively late in the metastatic process, particularly when considering that adjuvant hormone therapy has the potenital to arrest the stromal-epithelial interaction that has been suggested to contribute to the growth of micrometastases into clinically evident lesions (reference 17 in article). Lymph node metastases are a marker of late stage prostate cancer, with a high likelihood of subsequent disseminated disease. Therefore, we believe that the debate about when to initiate hormone therapy in these patients following RRP is really one of late (adjuvant) versus later (at osseous metastases). As such, future randomized trials should be designed to determine the optimal duration of androgen deprivation, as has been studied in patients treated with radiation thearpy. 1.
Morgan TM, Lin DW, Ellis WJ, Gallaher I, Kinnunen M, Lakely B et al: Disseminated tumor cells in prostate cancer: implications for systemic progression and tumor dormancy. J Urol, suppl., 2007; 77: 220, abstract 657.
Purpose: While the incidence of lymph node positive prostate cancer has decreased during the prostate specific antigen era, the optimal treatment of these patients remains in question. We examined the impact of lymph node metastases on the outcome of patients following radical prostatectomy and investigated prognostic factors that affect survival. Materials and Methods: We identified 507 men treated with radical prostatectomy between 1988 and 2001 who had lymph node positive disease. Of the 507 patients 455 (89.7%) were treated with adjuvant hormonal therapy. Median followup was 10.3 years (IQR 6.1–13.5). Postoperative survival rates were estimated using the Kaplan-Meier method and the impact of various clinicopathological factors on outcome was analyzed using Cox proportional hazard regression models. Results: Ten-year cancer specific survival for patients with positive lymph nodes was 85.8% with 56% of the men free from biochemical recurrence at last followup. On multivariate analysis pathological Gleason score 8 –10 (p ⫽ 0.004), positive surgical margins (p ⫽ 0.016), nondiploid tumor ploidy (p ⫽ 0.023) and 2 or greater positive nodes (p ⫽ 0.001) were adverse predictors of cancer specific survival. Tumor stage, year of surgery and total number of nodes removed did not significantly affect outcome. Adjuvant hormonal therapy decreased the risk of biochemical recurrence (p ⬍0.001) and local recurrence (p ⫽ 0.004) but it was not associated with systemic progression (p ⫽ 0.4) or cancer specific survival (p ⫽ 0.4). Conclusions: Radical prostatectomy may offer long-term survival to patients with lymph node positive prostate cancer. Gleason score, margin status, tumor ploidy and the number of involved nodes predict survival, while the role of adjuvant hormonal therapy continues to be defined. Key Words: prostate, prostatic neoplasms, lymph node dissection, prostate-specific antigen, prostatectomy
s part of the stage migration in prostate cancer during the PSA era,1 the incidence of positive lymph nodes in patients undergoing RRP has decreased to approximately 4% in recent series.2,3 While lymph node metastases adversely impact the prognosis of patients following surgery, extended postoperative survival in a subset of these men was reported.2–7 However, studies to date of patients with lymph node disease have been largely composed of pre-PSA and early PSA era cohorts, when the clinicopathological features of prostate cancer tend to be more advanced.1 While men with lymph node metastases continue to be identified, optimal treatment for these patients is not well established. Traditionally when lymph node disease was found at surgery, RRP was abandoned and patients were treated with hormones alone. However, several studies now show a trend toward improved survival for patients treated with prostatectomy and hormone therapy compared to patients treated with androgen ablation only.8 –10 These studies have been criticized for their retrospective design and for incompletely controlling for clinicopathological variables.
Nevertheless, a potential importance for addressing the primary tumor in patients with advanced prostate cancer has been suggested. Moreover, because most patients with positive lymph nodes are currently found to have microscopic disease on pathological examination of the nodes after RRP, the appropriate postoperative treatment of such patients similarly remains in question. In particular the optimal time to initiate hormone therapy continues to be debated.11,12 The identification of risk factors for disease progression in patients with nodal disease may help individualize treatment for these patients. We report the long-term outcome in men with lymph node positive prostate cancer treated with RRP at our institution during the PSA era. In addition, clinicopathological features were analyzed to identify predictors of disease progression.
A
MATERIALS AND METHODS After institutional review board approval was obtained we reviewed the records of 10,261 consecutive patients who underwent RRP and bilateral pelvic lymph node dissection between 1988 and 2001 at our institution. Patients found to have positive lymph nodes at surgery were identified. Patients who received hormonal treatment or radiotherapy before RRP were excluded from analysis.
Submitted for publication January 17, 2007. Study received institutional review board approval. * Correspondence: 200 First St. Southwest, Rochester, Minnesota 55905 (telephone: 507-284-3982; FAX: 507-284-4987; e-mail: blute. [email protected]).
0022-5347/07/1783-0864/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION
864
Vol. 178, 864-871, September 2007 Printed in U.S.A. DOI:10.1016/j.juro.2007.05.048
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER Surgical procedures were performed by different surgeons using standardized techniques. The 1997 TNM classification system was used for staging and the Gleason system was used for grading. DNA ploidy was assessed by flow cytometry. Adjuvant therapy was administered at the discretion of the treating physician and it consisted of treatment received within 90 days of RRP. Medical hormone deprivation therapy was generally intended to be lifelong. However, given the retrospective nature of this study, it is uncertain whether patients discontinued treatment after a period of ADT. Postoperative assessments, including physical examination and serum PSA measurement, were done quarterly for the initial 2 years, semiannually for an additional 2 years and annually thereafter. Radiographic evaluation was done as indicated clinically. BCR was defined as PSA 0.4 ng/ml or greater. Local recurrence was defined as cancer on biopsy of the prostatic bed or salvage radiation therapy to the prostatic bed without evidence of systemic recurrence. Systemic progression involved demonstrable metastatic deposits on radionuclide bone scan or on biopsies other than biopsy of the prostatic bed. Cause of death was identified from death certificates or physician correspondence. For patients followed elsewhere the prostatectomy registry at our institution monitors outcomes annually by correspondence. Of the 10,261 patients studied 246 (2.4%) were lost to followup. Statistical analyses were done using SAS®, version 8.2. Demographic comparisons were performed using the chisquare and rank sum tests as appropriate. Postoperative survival was estimated using the Kaplan-Meier method with patients censored at last followup or death if the end point of interest was not attained. The impact of various clinicopathological factors on outcome was analyzed using Cox proportional hazard regression models. The HR and
TABLE 1. Patient demographics Feature No. pts No. surgery era (%): 1988–1993 1994–1997 1998–2001 Median age at surgery (range) Median No. nodes removed (range) No. ng/ml preop PSA (%): Less than 10 10–19.9 20 or Greater No. pathological Gleason score (%): 6 or Less 7 8–10 No. pathological tumor stage (%): T2 T3/T4 No. seminal vesicle invasion (%): No Yes No. pos surgical margin (%): No Yes No. tumor DNA ploidy (%): Diploid Nondiploid
Pos Nodes
Neg Nodes
507
9,754
315 (9.1) 135 (3.7) 57 (1.8) 66.0 (47–79)
3,133 (90.9) 3,467 (96.3) 3,154 (98.2) 64.0 (34–84)
p Value ⬍0.001
11
(1–37)
8
(1–83)
142 150 215
(28) (29.6) (42.4)
7,227 1,755 772
(74.1) (18) (7.9)
109 267 131
(21.5) (52.7) (25.8)
6,521 2,698 535
(66.8) (27.7) (5.5)
73 434
(14.4) (85.6)
7,375 2,379
(75.6) (24.4)
170 337
(33.5) (66.5)
8,884 869
(91.1) (8.9)
192 315
(37.9) (62.1)
6,616 3,138
(67.8) (32.2)
247 247
(50) (50)
6,788 2,524
(72.9) (27.1)
0.01 ⬍0.001 ⬍0.001
⬍0.001
⬍0.001
⬍0.001
⬍0.001 ⬍0.001
865
TABLE 2. Clinicopathological characteristics of patients by AHT Feature
AHT
No. AHT*
p Value
No. pts Median age at surgery (range) Median No. nodes removed (range) No. pos nodes (%): 1 2 or Greater No. ng/ml preop PSA (%): Less than 10 10–19.9 20 or Greater No. pathological Gleason score (%): 6 or Less 7 8–10 No. pathological tumor stage (%): T2 T3/T4 No. seminal vesicle invasion (%): No Yes No. pos surgical margin (%): No Yes No. tumor DNA ploidy (%): Diploid Nondiploid No. adjuvant radiation (%): No Yes
455 66.0 (47–79)
52 63.0 (47–76)
0.01
11.5 (1–37)
10.5 (4–25)
0.08 ⬍0.001
245 210
(53.8) (46.2)
45 7
(86.5) (13.5)
124 130 201
(27.3) (28.6) (44.2)
18 20 14
(34.6) (38.5) (26.9)
0.06
0.02 95 234 126
(20.8) (51.4) (27.7)
14 33 5
(27) (63.5) (9.6) 0.14
62 393
(13.6) (86.4)
11 41
(21.2) (78.8) 0.001
142 313
(31.2) (68.8)
28 24
(53.8) (46.2)
164 291
(36) (64)
28 24
(53.8) (46.2)
214 234
(47.8) (52.2)
33 13
(71.7) (28.3)
413 42
(90.8) (9.2)
49 3
(94.2) (5.8)
0.01 0.002 0.41
* Of 52 patients 30 (57.7%) were ultimately treated with hormone deprivation for PSA only recurrence in 21, local recurrence in 2 and systemic progression in 7.
95% CI associated with the presence and number of positive lymph nodes were estimated using the proportional hazards model. All tests were 2-sided with p ⱕ0.05 considered significant. RESULTS Of the 10,261 men (4.9%) who underwent RRP between 1988 and 2001, 507 had positive lymph nodes. The incidence of positive lymph nodes at RRP decreased significantly during the study period from 315 of 3,448 patients (9.1%) between 1988 and 1993 to 57 of 3,211 (1.8%) between 1998 and 2001 (p ⬍0.001). Table 1 lists clinicopathological features of patients with and without nodal disease. Patients with positive lymph nodes had a greater number of lymph nodes removed and higher preoperative PSA, and they were more likely to have nonorgan confined tumors. Of 507 patients with nodal metastases 199 (39.3%) underwent preoperative computerized tomography or magnetic resonance imaging and only 2 had radiographic evidence of nodal disease. Postoperatively 455 of 507 men (89.7%) received AHT. Table 2 lists patient clinicopathological features by AHT. The use of AHT remained relatively constant during the study period since 285 of 315 patients (90.5%) with nodal metastases treated between 1988 and 1993 received AHT vs 116 of 135 (85.9%) treated between 1994 and 1997, and 54 of 57 (94.7%) treated from 1998 to 2001. In addition, 44 of 507 patients (8.7%) received adjuvant radiation therapy, while 39 of 507 (7.7%) were treated with salvage radiation therapy. Four patients received other treat-
866
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
FIG. 1. Outcome following RRP for patients with positive lymph nodes. LR, local recurrence. SP, systemic progression
ment after RRP, of which all were given at systemic progression, including chemotherapy in 1, strontium in 1 and steroids in 2. At a median followup after RRP of 10.3 years (IQR 6.1– 13.5) 213 patients had BCR, 51 experienced local recurrence, 97 had systemic relapse and 200 had died, including 72 of prostate cancer. Figure 1 shows Kaplan-Meier survival curves for patients with lymph node positive disease. Tenyear event-free survival for patients with positive lymph
nodes was 56%, 89%, 80% and 86% for BCR, local recurrence, systemic progression and cancer death, respectively. Postoperative survival was then stratified for all patients who underwent RRP by the number of positive lymph nodes (figs. 2 to 5). Although a single metastatic node significantly increased the risk of BCR (HR 1.4, 95% CI 1.1–1.7, p ⬍0.001), increasing nodal involvement did not worsen the patient risk of BCR (fig. 2). On the other hand, while patients with a single positive node were not at increased risk
FIG. 2. Postoperative BCR-free survival according to degree of lymph node involvement
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
867
FIG. 3. Postoperative local recurrence-free survival according to degree of lymph node involvement
for local recurrence compared to patients without nodal disease (HR 1.1, 95% CI 0.8 –1.7, p ⫽ 0.56), increasing nodal burden did increase the risk of local recurrence (HR 1.8, 95% CI 1.2–2.6, p ⫽ 0.004, fig. 3). The risk of systemic progression similarly increased with increasing nodal burden (fig. 4). While patients with a single positive node were at 3-fold increased risk for systemic progression (HR 2.9, 95% CI 2.1– 4.0, p ⬍0.001), patients with 2 or
greater nodes were approximately twice as likely to experience systemic progression as patients with 1 positive node (HR 1.9, 95% CI 1.2–2.8, p ⫽ 0.003). For CSS 1 lymph node metastasis increased the risk of death almost 4-fold (HR 3.6, 95% CI 2.4 –5.4, p ⬍0.001), while the risk of prostate cancer death in patients with 2 or greater positive lymph nodes was increased 2-fold compared to that in patients with 1 positive node (HR 2.2, 95% CI 1.3–3.5, p ⫽ 0.001, fig. 5).
FIG. 4. Postoperative systemic progression-free survival according to degree of lymph node involvement
868
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
FIG. 5. Postoperative CSS according to degree of lymph node involvement
We next examined risk factors for disease progression in patients with positive lymph nodes. In a multivariate Cox model only increased preoperative PSA (p ⫽ 0.036) and nondiploid tumor ploidy (p ⫽ 0.006) were significant predictors of BCR, while pathological Gleason score 8 –10 was associated with a trend toward BCR (p ⫽ 0.078, table 3). A positive surgical margin was the only predictor of local recurrence (p ⫽ 0.038). Patients with 2 or greater positive nodes (p ⫽ 0.003) and patients with Gleason 8 –10 cancer (p ⫽ 0.009) were at significantly increased risk for systemic progression, while a positive surgical margin also tended to increase this risk (p ⫽ 0.052). Lymph node burden (p ⫽ 0.001), Gleason score 8 –10 (p ⫽ 0.004), nondiploid cancers (p ⫽ 0.023) and positive surgical margins (p ⫽ 0.016) predicted prostate cancer death. Pathological tumor
stage, year of surgery and the total number of nodes examined did not affect survival. AHT was associated with a decreased risk of BCR (p ⬍0.001) and local recurrence (p ⫽ 0.004) but it did not significantly impact systemic progression (p ⫽ 0.38) or CSS (p ⫽ 0.43). Ten-year event-free survival for patients who did not receive AHT was 17.9%, 77.5%, 88.1% and 100% for BCR, local recurrence, systemic progression and prostate cancer death, respectively. DISCUSSION We present what is to our knowledge the largest series to date of patients with lymph node positive disease treated during the PSA era. Experience with RRP for lymph node positive prostate cancer at our institution was previously
TABLE 3. Multivariate analysis of risk factors associated with outcomes after RRP in patients with lymph node positive disease BCR Risk Factor No. pos nodes (2 or greater vs 1) Tumor stage (pT3/4 vs pT2) Preop PSA Gleason score (8–10 vs 7 or less) Nondiploid vs diploid tumor Pos surgical margin Total No. nodes removed RRP yr AHT vs no AHT
HR (95% CI)
Local Recurrence
p Value (chi-square test)
HR (95% CI)
p Value (chi-square test)
Systemic Progression HR (95% CI)
p Value (chi-square test)
CSS HR (95% CI)
p Value (chi-square test)
1.2
(0.8–1.6)
0.37
1.7
(0.9–3.1)
0.11
1.9
(1.2–2.8)
0.003
2.2
(1.3–3.5)
0.001
1.1
(0.7–1.7)
0.79
1.4
(0.5–4.0)
0.55
1.1
(0.5–2.3)
0.79
2.2
(0.7–7.1)
0.20
1.1 1.3
(1.0–1.3) (1.0–1.8)
0.036 0.078
0.95 (0.8–1.2) 1.5 (0.8–2.8)
0.63 0.22
0.96 1.8
(0.8–1.1) (1.2–2.8)
0.63 0.009
0.96 (0.8–1.2) 2.0 (1.3–3.3)
0.66 0.004
1.5
(1.1–2.0)
0.006
1.7
(0.9–3.0)
0.096
1.2
(0.8–1.8)
0.34
1.8
(1.1–2.9)
0.023
1.2
(0.9–1.7)
0.25
2.1
(1.0–4.1)
0.038
1.6
(1.0–2.6)
0.052
2.1
(1.2–3.9)
0.016
1.0 (0.98–1.04)
0.42
1.0 (0.94–1.1)
0.99
0.99 (0.95–1.03)
0.69
0.98 (0.94–1.0)
0.50
1.0 (0.96–1.1) 0.2 (0.1–0.3)
0.70 ⬍0.001
1.0 0.3
0.57 0.004
1.0 0.7
0.88 0.38
1.0 (0.94–1.1) 1.8 (0.42–7.5)
0.50 0.43
(0.9–1.1) (0.1–0.7)
(0.9–1.1) (0.3–1.5)
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER reported in patients in the pre-PSA and early PSA eras.4,9 However, because PSA screening has impacted the clinical and pathological features of patients diagnosed with prostate cancer,1 we updated our experience to report the longterm survival of these patients and evaluate potential prognostic features. We found that CSS depends on the degree of lymph node involvement, aspects of primary tumor pathology and surgical technique. The impact of nodal burden is consistent with previous results from our group4 and others,2–3,5 demonstrating decreased survival with increasing lymph node disease. However, the reported significance of primary tumor pathological features in patients with positive lymph nodes is inconsistent.2–3,5 For example, we noted a significant impact of Gleason score on the risk of systemic progression and CSS but found that tumor stage was not predictive of outcome. However, only 14% of our patients had organ confined disease, which may have obscured the assessment of its independent predictive value. Tumor stage was not a significant predictor of outcome for patients with nodal disease in several previous studies.2,5 Interestingly we found that positive surgical margins were significantly associated not only with local recurrence, which might be expected, but also with prostate cancer death. Again, the impact of surgical margins on outcome for patients with lymph node metastases, which was also reported previously,2 may reflect the high proportion of patients in our study with positive margins (62%). Nevertheless, this finding supports the importance of addressing the primary tumor in the setting of systemic disease, potentially for local control and removal of a source of cancer cells that may continue to metastasize. In fact, a benefit of RRP vs hormonal therapy alone was reported for decreasing the incidence of local complications, such as hematuria and urethral stricture, in patients with node positive prostate cancer.13 Meanwhile, the therapeutic value of pelvic lymph node dissection continues to be investigated. It was suggested that lymphadenectomy may be curative for patients with minimal nodal metastases.2,6 –7 Masterson et al noted improved BCR-free survival with increased nodes removed in patients without histological evidence for nodal metastases.7 They speculated that pelvic lymphadenectomy may offer a therapeutic effect through the removal of micrometastases not detected by routine histological evaluation. As a corollary to this hypothesis, prostate cancer cells were detected in archival tissue from 79% of patients with high risk prostate cancer without histological evidence of metastases in 1 study.14 However, DiMarco et al from our institution did not find a similar relationship between the extent of lymph node dissection and outcome in node negative cases.15 We noted that the total number of nodes removed did not affect the outcome in patients with lymph node positive disease. We believe that the treatment of patients with advanced prostate cancer should involve a multimodal approach. While the fact that almost 90% of our patients received AHT may be criticized as a confounder of the impact of RRP alone on lymph node positive prostate cancer, we assert that this reflects a practice that is not uncommon among urologists today. We found that AHT decreased the risk of BCR and local recurrence but did not
869
significantly impact systemic progression or CSS. However, our overall 10-year systemic progression-free rate of 80% in patients with lymph node positive disease compares favorably with the results of Daneshmand et al, who reported a 10-year clinical recurrence-free survival of 65% in their series of 235 patients with node positive prostate cancer, in which 69% received no adjuvant treatment.3 Likewise, our 5-year CSS rate of 94% compares with the 74% 5-year CSS noted by Bader et al in 88 patients with lymph node metastases who underwent RRP without AHT.5 Our 10-year survival data are similar to the updated results from the treatment arm of the ECOG randomized, prospective trial of immediate vs delayed hormonal therapy in men with lymph node positive prostate cancer after RRP.11 In that series 15% of men who received immediate hormone therapy after RRP died of prostate cancer and 40% experienced clinical progression. The appropriate timing of ADT in prostate cancer has been debated.11,12 The lack of an effect of AHT on CSS in our study limits our ability to conclude that immediate therapy is best. In addition, a recent trial of immediate vs deferred ADT in patients with newly diagnosed prostate cancer showed no significant difference in prostate cancer mortality or symptom-free survival.12 However, the ECOG study demonstrated that at a median followup of 11.9 years patients with lymph node positive disease treated with immediate ADT after RRP had significant improvement in overall survival, CSS and progression-free survival.11 The argument for early treatment of patients is based in part on the concept that ADT may be most effective against small tumor volume.16 Early delivery of AHT may arrest stromal-epithelial interaction, which was suggested to contribute to the growth of micrometastases into clinically evident lesions.17 A future challenge is to individualize the timing and selection of systemic treatments, including AHT, in patients with advanced prostate cancer based on an assessment of the patient risk of progression. One such potential predictor of the patient response to AHT is DNA tumor ploidy.18 We found that men with nondiploid tumors were at significantly increased risk for BCR and prostate cancer death, suggesting that they may benefit from systemic treatments such as chemotherapy or investigational agents. Our analysis was in part limited by the fact that the boundaries of lymph node dissection were not standardized. Indeed, the extent of lymph node dissection correlates with the incidence of positive nodes.5– 6,19 In addition, at our institution lymph nodes are submitted en bloc from each side of the pelvis, rather than as separate packets, which has been shown to yield increased nodal counts.20 Therefore, we cannot make conclusions from our data about the optimal anatomical extent of dissection that should be performed. Although our study represents a retrospective, single center experience, the decreasing incidence of lymph node positive prostate cancer during the PSA era makes future prospective, randomized trials of these patients unlikely. Therefore, large, retrospective series with long-term followup remain an important tool for risk stratification and assessment of treatment outcomes.
870
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER
CONCLUSIONS RRP with pelvic lymph node dissection may offer long-term survival to patients with node positive prostate cancer. Pathological Gleason score, surgical margin status, tumor ploidy and the number of involved nodes significantly predict survival. The value of AHT for patients with lymph node metastases continues to be defined.
12.
13.
14.
Abbreviations and Acronyms ADT AHT BCR CSS ECOG PSA RRP
⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽
androgen deprivation adjuvant hormonal therapy biochemical recurrence cancer specific survival Eastern Cooperative Oncology Group prostate specific antigen radical retropubic prostatectomy
15.
16.
17.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Han M, Partin AW, Piantadosi S, Epstein JI and Walsh PC: Era specific biochemical recurrence-free survival following radical prostatectomy for clinically localized prostate cancer. J Urol 2001; 166: 416. Palapattu GS, Allaf ME, Trock BJ, Epstein JI and Walsh PC: Prostate specific antigen progression in men with lymph node metastases following radical prostatectomy: results of long-term followup. J Urol 2004; 172: 1860. Daneshmand S, Quek ML, Stein JP, Lieskovsky G, Cai J, Pinski J et al: Prognosis of patients with lymph node positive prostate cancer following radical prostatectomy: longterm results. J Urol 2004; 172: 2252. Cheng L, Zincke H, Blute ML, Bergstralh EJ, Scherer B and Bostwick DG: Risk of prostate carcinoma death in patients with lymph node metastasis. Cancer 2001; 91: 66. Bader P, Burkhard FC, Markwalder R and Studer UE: Disease progression and survival of patients with positive lymph nodes after radical prostatectomy. Is there a chance of cure? J Urol 2003; 169: 849. Allaf ME, Palapattu GS, Trock BJ, Carter HB and Walsh PC: Anatomical extent of lymph node dissection: impact on men with clinically localized prostate cancer. J Urol 2004; 172: 1840. Masterson TA, Bianco FJ Jr, Vickers AJ, DiBlasio CJ, Fearn PA, Rabbani F et al: The association between total and positive lymph node counts, and disease progression in clinically localized prostate cancer. J Urol 2006; 175: 1320. Cadeddu JA, Partin AW, Epstein JI and Walsh PC: Stage D1 (T1–3, N1–3, M0) prostate cancer: a case-controlled comparison of conservative treatment versus radical prostatectomy. Urology 1997; 50: 251. Ghavamian R, Bergstralh EJ, Blute ML, Slezak J and Zincke H: Radical retropubic prostatectomy plus orchiectomy versus orchiectomy alone for pTxN⫹ prostate cancer: a matched comparison. J Urol 1999; 161: 1223. Grimm MO, Kamphausen S, Hugenschmidt H, StephanOdenthal M, Ackermann R and Vogeli TA: Clinical outcome of patients with lymph node positive prostate cancer after radical prostatectomy versus androgen deprivation. Eur Urol 2002; 41: 628. Messing EM, Manola J, Yao J, Kiernan M, Crawford D, Wilding G et al: Immediate versus deferred androgen deprivation treatment in patients with node- positive prostate cancer after radical prostatectomy and pelvic lymphadenectomy. Lancet Oncol 2006; 7: 472.
18.
19.
20.
Studer UE, Whelan P, Albrecht W, Casselman J, de Reijke T, Hauri D et al: Immediate or deferred androgen deprivation for patients with prostate cancer not suitable for local treatment with curative intent: European Organisation for Research and Treatment of Cancer (EORTC) Trial 20891. J Clin Oncol 2006; 24: 1868. Frazier HA 2nd, Robertson JE and Paulson DF: Does radical prostatectomy in the presence of positive pelvic lymph nodes enhance survival? World J Urol 1994; 12: 308. Ferrari AC, Stone NN and Eyler JN: Prospective analysis of prostate-specific markers in pelvic lymph node of patients with high-risk prostate cancer. J Natl Cancer Inst 1997; 89: 1498. DiMarco DS, Zincke H, Sebo TJ, Slezak J, Bergstralh EJ and Blute ML: The extent of lymphadenectomy for pTXN0 prostate cancer does not affect prostate cancer outcome in the prostate specific antigen era. J Urol 2005; 173: 1121. Isaacs JT: Relationship between tumor size and curability of prostatic cancer by combined chemo-hormonal therapy in rats. Cancer Res 1989; 49: 6290. Sokoloff MH, Rinker-Schaeffer CW, Chung LWK and Brendler CB: Adjunctive therapy for men with high risk localized and locally advanced prostate cancer: targeting disseminated tumor cells. J Urol 2004; 172: 2539. Zincke H, Bergstralh EJ, Larson-Keller JJ, Farrow GM, Myers RP, Leiber MM et al: Stage D1 prostate cancer treated by radical prostatectomy and adjuvant hormonal treatment: Evidence for favorable survival in patients with DNA diploid tumors. Cancer 1992; 70: 311. Heidenreich A, Varga Z and von Knobloch R: Extended pelvic lymphadenectomy in patients undergoing radical prostatectomy: high incidence of lymph node metastasis. J Urol 2002; 167: 1681. Bochner BH, Herr HW and Reuter VE: Impact of separate versus en bloc pelvic lymph node dissection on the number of lymph nodes retrieved in cystectomy specimens. J Urol 2001; 166: 2295.
EDITORIAL COMMENT These authors present a strong argument for performing randomized, prospective, controlled trials. Results in patients with N⫹ disease undergoing prostatectomy who received immediate ADT are similar to those in the early treatment arm of the randomized ECOG study and to prior Mayo Clinic reports (references 9 and 11 in article). However, this study could not distinguish between the benefits of early vs deferred ADT in this scenario because 1) only 10% of all patients were not treated with immediate ADT, 2) those who did not receive immediate ADT were significantly younger and at far lower risk for progression than those who did (table 2) and 3) as opposed to the ECOG study, in which deferred ADT was withheld almost exclusively until osseous metastases occurred, the majority of patients in this series receiving delayed ADT received it upon biochemical failure only. Thus, not only were the 2 groups dramatically different in risk for progression, but early ADT was compared with almost early ADT as opposed to very delayed ADT. It should be noted that the 17.9% of patients with BCR-free survival who did not receive immediate ADT at 10 years is similar to that in the deferred arm of the ECOG study. Unfortunately the duration of ADT for many treated patients is unknown and the morbidity of long-term ADT was not reported. Despite the ECOG study the precise time to administer ADT in men with N⫹ prostate cancer remains debated and the optimal duration of therapy remains uncer-
OUTCOME AFTER PROSTATECTOMY FOR LYMPH NODE POSITIVE PROSTATE CANCER tain. Unfortunately this series cannot answer these questions. Edward M. Messing Department of Urology University of Rochester James P. Wilmot Cancer Center Rochester, New York REPLY BY AUTHORS While we recognize that prospective randomized clinical trials represent the optimal way to establish treatment patterns, nevertheless the decreasing incidence of lymph node positive prostate cancer noted in the more recent PSA era (1.8% of RRPs performed at Mayo Clinic between 1998 and 2001) makes future prospective randomized trials of these patients difficult. Indeed, the only randomized trial of hormone therapy in this setting involved only 98 patients (reference 11 in article) despite accrual during the early PSA era when the incidence of lymph node metastases in patients undergoing surgery for prostate cancer was significantly higher (315 men were treated with RRP at our institution during that time for lymph node positive disease). Therefore, we believe that large, retrospective series with long-term followup remain an important tool for risk stratification and assessment of treatment outcomes. Moreover, we did not intend to evaluate the impact of timing of hormone initiation or duration on patient outcome following surgery for lymph node positive disease. Instead our goal was to present the outcomes of our approach to
871
patients with advanced prostate cancer during the PSA era, which has been to advocate aggressive local therapy plus early systemic therapy. Interestingly our results are remarkabley similar to the treatment arm of the randomized ECOG study (reference 11 in article). Furthermore, recent data have demonstrated that prostate cancer cells may disseminate to the bone marrrow of patients before the development of clinically evident metastases, and this dissemination may be decreased with androgen ablation.1 Therefore, administration of hormone therapy at the time that histological lymph node disease is identified may in fact be relatively late in the metastatic process, particularly when considering that adjuvant hormone therapy has the potenital to arrest the stromal-epithelial interaction that has been suggested to contribute to the growth of micrometastases into clinically evident lesions (reference 17 in article). Lymph node metastases are a marker of late stage prostate cancer, with a high likelihood of subsequent disseminated disease. Therefore, we believe that the debate about when to initiate hormone therapy in these patients following RRP is really one of late (adjuvant) versus later (at osseous metastases). As such, future randomized trials should be designed to determine the optimal duration of androgen deprivation, as has been studied in patients treated with radiation thearpy. 1.
Morgan TM, Lin DW, Ellis WJ, Gallaher I, Kinnunen M, Lakely B et al: Disseminated tumor cells in prostate cancer: implications for systemic progression and tumor dormancy. J Urol, suppl., 2007; 77: 220, abstract 657.