Salvage Lymph Node Dissection for Nodal Recurrence of Prostate Cancer after Radical Prostatectomy Derya Tilki,*,† Philipp Mandel,† Flora Seeliger, Alexander Kretschmer, € leyman Ergu € n, Michael Seitz‡ and Christian G. Stief‡ Alexander Karl, Su From the Department of Urology, Ludwig-Maximilians-University, Klinikum Grosshadern (DT, PM, FS, AK, AK, MS, CGS) and UroClinic Bogenhausen (MS), Munich, Martini-Clinic Prostate Cancer Center and Department of Urology, University € rzburg (SE), Hospital Hamburg-Eppendorf (DT), Hamburg and Institute of Anatomy and Cell Biology, University of Wu € rzburg, Germany Wu
Abbreviations and Acronyms ADT ¼ androgen deprivation therapy BCR ¼ biochemical recurrence BR ¼ biochemical response CR ¼ clinical recurrence CSS ¼ cancer specific survival CT ¼ computerized tomography LN ¼ lymph node PET ¼ positron emission tomography PSA ¼ prostate specific antigen RP ¼ radical prostatectomy sLND ¼ salvage LN dissection Accepted for publication August 22, 2014. Study received institutional review board approval. Supported by the German Research Foundation (DT, SE) and the Bavarian Academy of Sciences and Humanities (DT). * Correspondence and requests for reprints: Martini-Clinic Prostate Cancer Center, Martinistr. 52, 20246 Hamburg, Germany (telephone: þ49 40 7410 51300; e-mail: [email protected]). † Equal study contribution. ‡ Equal study contribution.
Purpose: We analyzed the impact of salvage lymph node dissection on the prognosis in patients with biochemical recurrence and positive lymph nodes on positron emission tomography/computerized tomography after radical prostatectomy. Materials and Methods: We retrospectively analyzed the records of 58 patients who underwent pelvic and/or retroperitoneal salvage lymph node dissection from June 2005 to February 2012. Biochemical response was defined as prostate specific antigen less than 0.2 ng/ml 40 days after salvage treatment. Biochemical recurrence in those with a biochemical response was defined as prostate specific antigen greater than 0.2 ng/ml and increasing. Kaplan-Meier curves were used to assess time to biochemical recurrence, clinical recurrence and cancer specific survival. Cox and binary logistic regressions were used to determine factors influencing clinical recurrence and biochemical response. Results: Median followup after salvage lymph node dissection was 39 months. A total of 13 patients (22.4%) achieved a biochemical response. Only 1 patient remained free of biochemical recurrence during followup. Clinical recurrence developed in 25 patients (48.1%) after salvage treatment. Six patients (10.3%) died of disease, including 4 with indeterminate extralymphatic findings on positron emission tomography/computerized tomography before salvage therapy. The 5-year cancer specific survival rate was 71.1%. Patients with a complete biochemical response showed a trend toward a longer time to clinical recurrence (p ¼ 0.20). Biochemical response did not influence cancer specific survival. Conclusions: Salvage lymph node dissection in patients with biochemical recurrence and positive lymph nodes on positron emission tomography/computerized tomography led to a biochemical response in a certain proportion. Most patients progressed to biochemical recurrence after salvage treatment but almost half showed no further clinical recurrence. Cancer specific mortality occurred predominantly in patients with prior suspicion of extralymphatic lesions. Salvage lymph node dissection may delay androgen deprivation therapy and clinical recurrence in select patients. Key Words: prostatic neoplasms, prostatectomy, lymph node excision, salvage therapy, prostate-specific antigen RADICAL prostatectomy is a treatment of choice for localized prostate cancer of all levels of risk.1,2 BCR is
484
j
www.jurology.com
detectable in up to 40% of patients after RP.3e5 The timing and mode of treatment of BCR are controversial.
0022-5347/15/1932-0484/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.08.096 Vol. 193, 484-490, February 2015 Printed in U.S.A.
SALVAGE LYMPH NODE DISSECTION AFTER RADICAL PROSTATECTOMY
Of patients with BCR 15% to 20% die of prostate cancer.6 BCR can be due to local or distant prostate cancer recurrence. Bone scan and CT are commonly used to assess BCR. The value of these modalities is low in patients with PSA less than 10 ng/ml.1 A consensus has not been reached on the value of PET/CT. It enables the visualization of metabolic to anatomical abnormalities and may help improve the detection of metastasis. It may detect tumor recurrence even in patients with low PSA.1,7,8 However, it can underestimate the extent of metastasis.8 The treatment options for BCR after RP proposed by guidelines of the EAU (European Association of Urology) and AUA (American Urological Association) consist of radiotherapy to the prostatic bed, complete or intermittent androgen deprivation, a combination of antiandrogens with 5a-reductase inhibitors and early chemohormonal approaches.1,2,7 Patients with isolated clinical LN recurrence have a favorable outcome compared to that in men with metastasis to other sites.9 sLND is considered a therapeutic option for BCR after RP and positive LNs on PET/CT, especially in patients with PSA less than 4 ng/ml, Gleason score less than 8 and CR in a single LN.8,10,11 In the current study we sought to add further information and knowledge to the existing data. We validated available findings in a cohort of patients with more aggressive disease.
PATIENTS AND METHODS
485
The decision to perform pelvic, retroperitoneal and pelvic plus retroperitoneal lymphadenectomy was made according to the PET/CT result. After lymphadenectomy surgical specimens were processed according to standard pathology protocols. LN analysis was performed by step sections (200 mm slices) to detect micrometastasis.
Followup Patient followup consisted of PSA testing 40 days postoperatively and every 3 to 6 months thereafter. Postoperative PET/CT was performed according to PSA and patient preference. Patients received ADT after sLND depending on PSA, further symptoms, followup PET/CT results and individual discussion between patients and surgeons. ADT included luteinizing hormone-releasing hormone agonists and antagonists. To evaluate the benefit of sLND we used the outcome variables of BR, BR interval to BCR, CR after sLND and CSS after sLND. BR was defined as PSA less than 0.2 ng/ml 40 days postoperatively, BCR was defined as 2 consecutive PSA values greater than 0.2 ng/ml and CR was defined as positive PET/CT after sLND in the presence of increasing PSA. Cause of death was determined by treating physicians, chart review corroborated by death certificates or death certificates alone.
Statistical Analysis To assess differences in respective subsamples we used the 2-group mean comparison t-test, and the Pearson chisquare and Kruskal-Wallis tests. Kaplan-Meier curves and the log rank test were used to evaluate BCR in patients with BR, time to CR and CSS. Univariate and multivariate binary logistic regressions were fitted for BR, and univariate and multivariate Cox regression models were fitted for CR. STATAÒ, version 11 for WindowsÒ was used for statistical analysis.
Patient Population The study was approved by the institutional review board. We retrospectively reviewed the records of 58 consecutive patients with prostate cancer and previous RP who underwent sLND from June 2005 to February 2012 at University Hospital of Munich-Grosshadern. All patients underwent 18F-fluoroethylcholine PET/CT and had pathological uptake in at least 1 LN. Six patients had extralymphatic findings on PET/CT with nonspecific uptake in at least 1 extralymphatic location before sLND, including uptake in the prostatic bed in 4 and suspicion of a solitary bone metastasis in 2. Pelvic and/or retroperitoneal sLND was performed after highlighting the absence of guideline recommendations for this surgical approach. No patient was castration resistant.
Surgical Procedure and Histological Evaluation The surgical field for pelvic lymphadenectomy included LNs along the internal and external iliac vessels, common iliac vessels, presacral region, aortocaval region up to the inferior mesenteric artery and, if positive on PET/CT, pararectal region. The lateral border of lymphadenectomy was the genitofemoral nerve. Retroperitoneal lymphadenectomy included removal of all lymphatic tissue along the abdominal great vessels from the origin of the iliac vessels to the cranial border of the upper renal pole.
RESULTS Baseline Characteristics Median followup was 39 months after sLND. Table 1 and the supplementary table (http://jurology.com/) list preoperative and postoperative patient characteristics by subgroup. Of the patients 87.9% underwent pelvic LN dissection at RP with a mean of 8 LNs removed. Mean PSA at sLND was 9.8 ng/dl. Six patients (10.3%) also had indeterminate extralymphatic findings on PET/CT, which were classified as unspecific before sLND. Subsequent PET/CT after sLND was performed in 35 patients due to persistent PSA or BCR. Bone lesions were found in 37.1% of patients with positive PET/CT. In addition, pulmonary lesions, local recurrence and positive LNs were found in 5.7%, 17.1% and 60% of patients, respectively. sLND Pathological Results A total of 23 patients (39.7%) underwent pelvic only, 3 (5.2%) underwent retroperitoneal only and 32 (55.2%) underwent pelvic plus retroperitoneal
SALVAGE LYMPH NODE DISSECTION AFTER RADICAL PROSTATECTOMY
486
Table 1. Histopathological results and followup data on patients treated with sLND for prostate cancer LN recurrence after RP Prostate Ca Death
No. pts (%) No. postop ADT (%) Mean No. LNs removed (range) LNs: No. histologically pos on sLND (%) Mean No. pos Mean LN density No. sLND site (%): Pelvic only Retroperitoneal only Pelvic þ retroperitoneal No. postop BR (%) No. followup CR (%) No. followup Ca specific death (%)
Suspicious Extralymphatic Lesions
Overall
No
Yes
p Value*
No
Yes†
58 (100) 39 (67.2) 18.6 (1e88)
52 (89.7) 35 (67.3) 19.2 (1e88)
6 (10.3) 4 (66.7) 13.7 (4e35)
0.98 0.42
52 (89.7) 35 (67.3) 19.1 (1e88)
6 (10.3) 4 (66.7) 14.5 (2e35)
0.98 0.50
45 (77.6) 6.3 0.38
40 (76.9) 5.9 0.36
5 (83.3) 9.4 0.55
0.39 0.20 0.15
40 (76.9) 5.9 0.36
5 (83.3) 9.4 0.54
0.39 0.20 0.15
23 3 32 13 25 6
19 3 30 12 23
4 0 2 1 2
0.15 0.55 0.26 0.72 0.13 e
21 2 29 12
2 1 3 1
(39.7) (5.2) (55.2) (22.4) (48.1) (10.3)
(36.5) (5.8) (57.7) (23.1) (46.0) e
(66.7) (33.3) (16.7) (100) e
(40.4) (3.8) (55.8) (23.1) e 2 (3.8)
p Value*
(33.3) (16.7) (50.0) (16.7) e 4 (66.7)
0.74 0.18 0.79 0.72 e
Abbreviations and Acronyms ADT ¼ androgen deprivation therapy BCR ¼ biochemical recurrence BR ¼ biochemical response CR ¼ clinical recurrence CSS ¼ cancer specific survival CT ¼ computerized tomography LN ¼ lymph node PET ¼ positron emission tomography PSA ¼ prostate specific antigen RP ¼ radical prostatectomy sLND ¼ salvage LN dissection Accepted for publication August 22, 2014. Study received institutional review board approval. Supported by the German Research Foundation (DT, SE) and the Bavarian Academy of Sciences and Humanities (DT). * Correspondence and requests for reprints: Martini-Clinic Prostate Cancer Center, Martinistr. 52, 20246 Hamburg, Germany (telephone: þ49 40 7410 51300; e-mail: [email protected]). † Equal study contribution. ‡ Equal study contribution.
Purpose: We analyzed the impact of salvage lymph node dissection on the prognosis in patients with biochemical recurrence and positive lymph nodes on positron emission tomography/computerized tomography after radical prostatectomy. Materials and Methods: We retrospectively analyzed the records of 58 patients who underwent pelvic and/or retroperitoneal salvage lymph node dissection from June 2005 to February 2012. Biochemical response was defined as prostate specific antigen less than 0.2 ng/ml 40 days after salvage treatment. Biochemical recurrence in those with a biochemical response was defined as prostate specific antigen greater than 0.2 ng/ml and increasing. Kaplan-Meier curves were used to assess time to biochemical recurrence, clinical recurrence and cancer specific survival. Cox and binary logistic regressions were used to determine factors influencing clinical recurrence and biochemical response. Results: Median followup after salvage lymph node dissection was 39 months. A total of 13 patients (22.4%) achieved a biochemical response. Only 1 patient remained free of biochemical recurrence during followup. Clinical recurrence developed in 25 patients (48.1%) after salvage treatment. Six patients (10.3%) died of disease, including 4 with indeterminate extralymphatic findings on positron emission tomography/computerized tomography before salvage therapy. The 5-year cancer specific survival rate was 71.1%. Patients with a complete biochemical response showed a trend toward a longer time to clinical recurrence (p ¼ 0.20). Biochemical response did not influence cancer specific survival. Conclusions: Salvage lymph node dissection in patients with biochemical recurrence and positive lymph nodes on positron emission tomography/computerized tomography led to a biochemical response in a certain proportion. Most patients progressed to biochemical recurrence after salvage treatment but almost half showed no further clinical recurrence. Cancer specific mortality occurred predominantly in patients with prior suspicion of extralymphatic lesions. Salvage lymph node dissection may delay androgen deprivation therapy and clinical recurrence in select patients. Key Words: prostatic neoplasms, prostatectomy, lymph node excision, salvage therapy, prostate-specific antigen RADICAL prostatectomy is a treatment of choice for localized prostate cancer of all levels of risk.1,2 BCR is
484
j
www.jurology.com
detectable in up to 40% of patients after RP.3e5 The timing and mode of treatment of BCR are controversial.
0022-5347/15/1932-0484/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.08.096 Vol. 193, 484-490, February 2015 Printed in U.S.A.
SALVAGE LYMPH NODE DISSECTION AFTER RADICAL PROSTATECTOMY
Of patients with BCR 15% to 20% die of prostate cancer.6 BCR can be due to local or distant prostate cancer recurrence. Bone scan and CT are commonly used to assess BCR. The value of these modalities is low in patients with PSA less than 10 ng/ml.1 A consensus has not been reached on the value of PET/CT. It enables the visualization of metabolic to anatomical abnormalities and may help improve the detection of metastasis. It may detect tumor recurrence even in patients with low PSA.1,7,8 However, it can underestimate the extent of metastasis.8 The treatment options for BCR after RP proposed by guidelines of the EAU (European Association of Urology) and AUA (American Urological Association) consist of radiotherapy to the prostatic bed, complete or intermittent androgen deprivation, a combination of antiandrogens with 5a-reductase inhibitors and early chemohormonal approaches.1,2,7 Patients with isolated clinical LN recurrence have a favorable outcome compared to that in men with metastasis to other sites.9 sLND is considered a therapeutic option for BCR after RP and positive LNs on PET/CT, especially in patients with PSA less than 4 ng/ml, Gleason score less than 8 and CR in a single LN.8,10,11 In the current study we sought to add further information and knowledge to the existing data. We validated available findings in a cohort of patients with more aggressive disease.
PATIENTS AND METHODS
485
The decision to perform pelvic, retroperitoneal and pelvic plus retroperitoneal lymphadenectomy was made according to the PET/CT result. After lymphadenectomy surgical specimens were processed according to standard pathology protocols. LN analysis was performed by step sections (200 mm slices) to detect micrometastasis.
Followup Patient followup consisted of PSA testing 40 days postoperatively and every 3 to 6 months thereafter. Postoperative PET/CT was performed according to PSA and patient preference. Patients received ADT after sLND depending on PSA, further symptoms, followup PET/CT results and individual discussion between patients and surgeons. ADT included luteinizing hormone-releasing hormone agonists and antagonists. To evaluate the benefit of sLND we used the outcome variables of BR, BR interval to BCR, CR after sLND and CSS after sLND. BR was defined as PSA less than 0.2 ng/ml 40 days postoperatively, BCR was defined as 2 consecutive PSA values greater than 0.2 ng/ml and CR was defined as positive PET/CT after sLND in the presence of increasing PSA. Cause of death was determined by treating physicians, chart review corroborated by death certificates or death certificates alone.
Statistical Analysis To assess differences in respective subsamples we used the 2-group mean comparison t-test, and the Pearson chisquare and Kruskal-Wallis tests. Kaplan-Meier curves and the log rank test were used to evaluate BCR in patients with BR, time to CR and CSS. Univariate and multivariate binary logistic regressions were fitted for BR, and univariate and multivariate Cox regression models were fitted for CR. STATAÒ, version 11 for WindowsÒ was used for statistical analysis.
Patient Population The study was approved by the institutional review board. We retrospectively reviewed the records of 58 consecutive patients with prostate cancer and previous RP who underwent sLND from June 2005 to February 2012 at University Hospital of Munich-Grosshadern. All patients underwent 18F-fluoroethylcholine PET/CT and had pathological uptake in at least 1 LN. Six patients had extralymphatic findings on PET/CT with nonspecific uptake in at least 1 extralymphatic location before sLND, including uptake in the prostatic bed in 4 and suspicion of a solitary bone metastasis in 2. Pelvic and/or retroperitoneal sLND was performed after highlighting the absence of guideline recommendations for this surgical approach. No patient was castration resistant.
Surgical Procedure and Histological Evaluation The surgical field for pelvic lymphadenectomy included LNs along the internal and external iliac vessels, common iliac vessels, presacral region, aortocaval region up to the inferior mesenteric artery and, if positive on PET/CT, pararectal region. The lateral border of lymphadenectomy was the genitofemoral nerve. Retroperitoneal lymphadenectomy included removal of all lymphatic tissue along the abdominal great vessels from the origin of the iliac vessels to the cranial border of the upper renal pole.
RESULTS Baseline Characteristics Median followup was 39 months after sLND. Table 1 and the supplementary table (http://jurology.com/) list preoperative and postoperative patient characteristics by subgroup. Of the patients 87.9% underwent pelvic LN dissection at RP with a mean of 8 LNs removed. Mean PSA at sLND was 9.8 ng/dl. Six patients (10.3%) also had indeterminate extralymphatic findings on PET/CT, which were classified as unspecific before sLND. Subsequent PET/CT after sLND was performed in 35 patients due to persistent PSA or BCR. Bone lesions were found in 37.1% of patients with positive PET/CT. In addition, pulmonary lesions, local recurrence and positive LNs were found in 5.7%, 17.1% and 60% of patients, respectively. sLND Pathological Results A total of 23 patients (39.7%) underwent pelvic only, 3 (5.2%) underwent retroperitoneal only and 32 (55.2%) underwent pelvic plus retroperitoneal
SALVAGE LYMPH NODE DISSECTION AFTER RADICAL PROSTATECTOMY
486
Table 1. Histopathological results and followup data on patients treated with sLND for prostate cancer LN recurrence after RP Prostate Ca Death
No. pts (%) No. postop ADT (%) Mean No. LNs removed (range) LNs: No. histologically pos on sLND (%) Mean No. pos Mean LN density No. sLND site (%): Pelvic only Retroperitoneal only Pelvic þ retroperitoneal No. postop BR (%) No. followup CR (%) No. followup Ca specific death (%)
Suspicious Extralymphatic Lesions
Overall
No
Yes
p Value*
No
Yes†
58 (100) 39 (67.2) 18.6 (1e88)
52 (89.7) 35 (67.3) 19.2 (1e88)
6 (10.3) 4 (66.7) 13.7 (4e35)
0.98 0.42
52 (89.7) 35 (67.3) 19.1 (1e88)
6 (10.3) 4 (66.7) 14.5 (2e35)
0.98 0.50
45 (77.6) 6.3 0.38
40 (76.9) 5.9 0.36
5 (83.3) 9.4 0.55
0.39 0.20 0.15
40 (76.9) 5.9 0.36
5 (83.3) 9.4 0.54
0.39 0.20 0.15
23 3 32 13 25 6
19 3 30 12 23
4 0 2 1 2
0.15 0.55 0.26 0.72 0.13 e
21 2 29 12
2 1 3 1
(39.7) (5.2) (55.2) (22.4) (48.1) (10.3)
(36.5) (5.8) (57.7) (23.1) (46.0) e
(66.7) (33.3) (16.7) (100) e
(40.4) (3.8) (55.8) (23.1) e 2 (3.8)
p Value*
(33.3) (16.7) (50.0) (16.7) e 4 (66.7)
0.74 0.18 0.79 0.72 e
