Prophylactic Pelvic Lymph Node Dissection in Patients with Penile Cancer Rosa S. Djajadiningrat, Erik van Werkhoven and Simon Horenblas* From the Departments of Urology and Biometrics (EvW), The Netherlands Cancer Institute, Amsterdam, The Netherlands
Abbreviations and Acronyms CT ¼ computerized tomography DSS ¼ disease specific survival ENE ¼ extranodal extension LND ¼ lymph node dissection PET ¼ positron emission tomography pLND ¼ pelvic LND Accepted for publication December 1, 2014. * Correspondence: Department of Urology, The Netherlands Cancer Institute, Plesmanlaan 121, NL-1066 CX Amsterdam, The Netherlands (telephone: þ31 20 512 2553; FAX: þ31 20 512 2554; e-mail: [email protected]).
Purpose: Pelvic lymph node involvement in penile cancer carries a poor prognosis. Therefore, there is controversy about the curative role of pelvic lymph node dissection. We analyzed the characteristics of tumor positive inguinal regions predictive for pelvic lymph node involvement in patients prophylactically treated with pelvic lymph node dissection. Materials and Methods: All chemona€ıve consecutive cases treated with prophylactic pelvic lymph node dissection at our institution since 2001 were included in analysis. A generalized estimating equation model was used to predict pelvic node involvement based on inguinal characteristics. Disease specific survival was calculated with the Kaplan-Meier method. Results: Included in study were 79 chemotherapy na€ıve patients without preoperative evidence of pelvic disease who were treated with prophylactic pelvic lymph node dissection. Pelvic nodes were positive in 24% of the patients. Inguinal extranodal extension, or 2 or more tumor positive nodes were predictive of tumor positive pelvic nodes. The 5-year disease specific survival rate in patients with pelvic involvement was 17%. Conclusions: Inguinal extranodal extension, or 2 or more inguinal tumor positive lymph nodes are predictive of pelvic tumor positivity in patients without evidence of pelvic involvement. However, disease specific survival remains poor in patients with pelvic node involvement who are treated with surgery only. Key Words: penis; carcinoma, squamous cell; lymph nodes; neoplasm metastasis; lymph node excision
THE most important prognostic factors in penile cancer are the presence and extent of lymph node involvement.1e5 Pelvic lymph node involvement carries an especially poor prognosis.2,4,6e9 Tumor positive pelvic nodes are reported in 19% to 48% of patients but only in those with proven inguinal metastasis. Established prognostic risk factors for pelvic node involvement are inguinal metastasis ENE, the number of groin metastases, metastasis size in the
1976
j
www.jurology.com
inguinal region, the lymph node ratio (number of positive nodes/total number removed), tumor differentiation grade and p53 expression in the inguinal specimen.7,8,10,11 LND has an important role in metastatic disease since the DSS rate in patients with 1 or 2 metastases in the groin is 75%.3,9 There is still controversy on the curative role of pLND since only 0% to 33% of patients with pelvic metastasis can be cured.2,4,7,10,12e14 Some studies
0022-5347/15/1936-1976/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.12.019 Vol. 193, 1976-1980, June 2015 Printed in U.S.A.
PROPHYLACTIC PELVIC LYMPH NODE DISSECTION FOR PENILE CANCER
suggest that pLND is beneficial in patients with a microscopic and/or single focus in the pelvic specimen.12,15 Therefore, the April 2010 update of the EAU (European Association of Urology) penile cancer guideline recommends pLND in patients in whom inguinal extranodal metastasis is present, or the node of Cloquet or more than 2 inguinal nodes are involved.16 Based on 2 articles the guidelines state that 23% of patients with more than 2 positive inguinal nodes are at risk for pelvic involvement and in those with more than 3 tumor positive inguinal nodes or ENE in at least 1 inguinal node the incidence increases to 56%.1,5,17 The NCCN GuidelinesÒ on penile cancer in the United States uses a cutoff of 2 or more nodes, as mentioned in the publications on the EAU guideline.1,18 Several groups have investigated the relationship between inguinal and pelvic lymph node involvement.8,10,11,19 These studies also included patients treated with chemotherapy and/or those with preoperative evidence of pelvic lymph node involvement. However, in daily practice the correct candidates for prophylactic pLND show inguinal tumor positivity but no preoperative sign of pelvic lymph node involvement on imaging. Therefore, we identified characteristics in the inguinal specimen predictive of pelvic lymph node involvement in patients prophylactically treated with pLND, ie those without preoperative evidence of pelvic lymph node involvement. We also assessed DSS in patients who underwent prophylactic pLND and were found to have tumor positive pelvic nodes.
METHODS Patients To obtain a homogenous study population all patients with penile cancer treated with prophylactic pLND since 2001 at our institution were included in analysis. These cases were treated according to a contemporary management protocol consisting of standard preoperative imaging, primary tumor treatment, dynamic sentinel node biopsy and standard surgical templates. To ensure at least 2-year followup patients who underwent pLND later than March 2012 were excluded from study. Those who received neoadjuvant chemotherapy were also excluded since this could have affected the histopathological examination. The number of inguinal and pelvic tumor positive nodes, ENE, size of the largest pelvic metastasis and adjuvant radiotherapy were reported in a consecutive penile cancer database. Inguinal characteristics were analyzed for the prediction of pelvic tumor involvement. A subset of this patient series was described previously.14
Standard Management Until 2006 CT was often done to stage pelvic nodes. This was subsequently replaced by PET/CT. According to institutional guidelines the indication for prophylactic ipsilateral pLND was 2 or more tumor positive nodes
1977
and/or ENE in the inguinal specimen without clinical evidence of pelvic disease. The pelvic procedure was performed at the same time as the inguinal procedure if preoperative clinical inguinal characteristics met institutional criteria (1 stage). It was performed at a second surgery after the inguinal procedure if histopathology met the criteria (2 stages). Adjuvant radiotherapy to the pelvic region was administered if pelvic lymph nodes were involved. Exceptions included patients with wound healing problems, poor performance status, previous radiotherapy to the pelvic area or rapid disease progression.
Surgical Technique pLND consists of the removal of all nodes between certain boundaries, including the common iliac vessels proximal, the passage of lymphatic vessels to the groin distal, the ilioinguinal nerve lateral, and the bladder and prostate medial with the deepest part of the obturator fossa as the base. The obturator fossa is completely cleaned, especially the space behind the external iliac vessels, all the way to the sacrum (triangle of Marcille). There has been no change in the boundaries with time. Surgical complications of pLND only were graded according to the Clavien-Dindo grading system20 and prospectively maintained in the institutional database.
Histological Examination and Analysis All histopathology findings were reviewed by experienced pathologists at our tertiary reference center. Grade was assigned as well, moderately or poorly differentiated based on the amount of undifferentiated cells in the tumor on histopathological examination according to Broders.21 Lymphovascular invasion was defined as embolic tumor cells in thin-walled vessel-like structures. ENE was defined as tumor extension through the lymph node capsule into the perinodal fibrous-adipose tissue.
Statistical Analysis A generalized estimation equation model was used with an exchangeable correlation matrix to account for left and right pelvic node clustering in a patient. We used a univariable proportional hazards model to analyze pelvic characteristics predictive of DSS in patients with pelvic involvement. DSS was calculated using the Kaplan-Meier method. DSS was defined as the time from the date of pLND to the date of death from penile cancer, penile cancer metastasis or complications related to penile cancer treatment. At the last followup date survivors and patients who died of another cause were censored. All analysis was done with SPSSÒ, version 20.0 and R, version 2.15.2 (http://www.r-project.org/) with p
Abbreviations and Acronyms CT ¼ computerized tomography DSS ¼ disease specific survival ENE ¼ extranodal extension LND ¼ lymph node dissection PET ¼ positron emission tomography pLND ¼ pelvic LND Accepted for publication December 1, 2014. * Correspondence: Department of Urology, The Netherlands Cancer Institute, Plesmanlaan 121, NL-1066 CX Amsterdam, The Netherlands (telephone: þ31 20 512 2553; FAX: þ31 20 512 2554; e-mail: [email protected]).
Purpose: Pelvic lymph node involvement in penile cancer carries a poor prognosis. Therefore, there is controversy about the curative role of pelvic lymph node dissection. We analyzed the characteristics of tumor positive inguinal regions predictive for pelvic lymph node involvement in patients prophylactically treated with pelvic lymph node dissection. Materials and Methods: All chemona€ıve consecutive cases treated with prophylactic pelvic lymph node dissection at our institution since 2001 were included in analysis. A generalized estimating equation model was used to predict pelvic node involvement based on inguinal characteristics. Disease specific survival was calculated with the Kaplan-Meier method. Results: Included in study were 79 chemotherapy na€ıve patients without preoperative evidence of pelvic disease who were treated with prophylactic pelvic lymph node dissection. Pelvic nodes were positive in 24% of the patients. Inguinal extranodal extension, or 2 or more tumor positive nodes were predictive of tumor positive pelvic nodes. The 5-year disease specific survival rate in patients with pelvic involvement was 17%. Conclusions: Inguinal extranodal extension, or 2 or more inguinal tumor positive lymph nodes are predictive of pelvic tumor positivity in patients without evidence of pelvic involvement. However, disease specific survival remains poor in patients with pelvic node involvement who are treated with surgery only. Key Words: penis; carcinoma, squamous cell; lymph nodes; neoplasm metastasis; lymph node excision
THE most important prognostic factors in penile cancer are the presence and extent of lymph node involvement.1e5 Pelvic lymph node involvement carries an especially poor prognosis.2,4,6e9 Tumor positive pelvic nodes are reported in 19% to 48% of patients but only in those with proven inguinal metastasis. Established prognostic risk factors for pelvic node involvement are inguinal metastasis ENE, the number of groin metastases, metastasis size in the
1976
j
www.jurology.com
inguinal region, the lymph node ratio (number of positive nodes/total number removed), tumor differentiation grade and p53 expression in the inguinal specimen.7,8,10,11 LND has an important role in metastatic disease since the DSS rate in patients with 1 or 2 metastases in the groin is 75%.3,9 There is still controversy on the curative role of pLND since only 0% to 33% of patients with pelvic metastasis can be cured.2,4,7,10,12e14 Some studies
0022-5347/15/1936-1976/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.
http://dx.doi.org/10.1016/j.juro.2014.12.019 Vol. 193, 1976-1980, June 2015 Printed in U.S.A.
PROPHYLACTIC PELVIC LYMPH NODE DISSECTION FOR PENILE CANCER
suggest that pLND is beneficial in patients with a microscopic and/or single focus in the pelvic specimen.12,15 Therefore, the April 2010 update of the EAU (European Association of Urology) penile cancer guideline recommends pLND in patients in whom inguinal extranodal metastasis is present, or the node of Cloquet or more than 2 inguinal nodes are involved.16 Based on 2 articles the guidelines state that 23% of patients with more than 2 positive inguinal nodes are at risk for pelvic involvement and in those with more than 3 tumor positive inguinal nodes or ENE in at least 1 inguinal node the incidence increases to 56%.1,5,17 The NCCN GuidelinesÒ on penile cancer in the United States uses a cutoff of 2 or more nodes, as mentioned in the publications on the EAU guideline.1,18 Several groups have investigated the relationship between inguinal and pelvic lymph node involvement.8,10,11,19 These studies also included patients treated with chemotherapy and/or those with preoperative evidence of pelvic lymph node involvement. However, in daily practice the correct candidates for prophylactic pLND show inguinal tumor positivity but no preoperative sign of pelvic lymph node involvement on imaging. Therefore, we identified characteristics in the inguinal specimen predictive of pelvic lymph node involvement in patients prophylactically treated with pLND, ie those without preoperative evidence of pelvic lymph node involvement. We also assessed DSS in patients who underwent prophylactic pLND and were found to have tumor positive pelvic nodes.
METHODS Patients To obtain a homogenous study population all patients with penile cancer treated with prophylactic pLND since 2001 at our institution were included in analysis. These cases were treated according to a contemporary management protocol consisting of standard preoperative imaging, primary tumor treatment, dynamic sentinel node biopsy and standard surgical templates. To ensure at least 2-year followup patients who underwent pLND later than March 2012 were excluded from study. Those who received neoadjuvant chemotherapy were also excluded since this could have affected the histopathological examination. The number of inguinal and pelvic tumor positive nodes, ENE, size of the largest pelvic metastasis and adjuvant radiotherapy were reported in a consecutive penile cancer database. Inguinal characteristics were analyzed for the prediction of pelvic tumor involvement. A subset of this patient series was described previously.14
Standard Management Until 2006 CT was often done to stage pelvic nodes. This was subsequently replaced by PET/CT. According to institutional guidelines the indication for prophylactic ipsilateral pLND was 2 or more tumor positive nodes
1977
and/or ENE in the inguinal specimen without clinical evidence of pelvic disease. The pelvic procedure was performed at the same time as the inguinal procedure if preoperative clinical inguinal characteristics met institutional criteria (1 stage). It was performed at a second surgery after the inguinal procedure if histopathology met the criteria (2 stages). Adjuvant radiotherapy to the pelvic region was administered if pelvic lymph nodes were involved. Exceptions included patients with wound healing problems, poor performance status, previous radiotherapy to the pelvic area or rapid disease progression.
Surgical Technique pLND consists of the removal of all nodes between certain boundaries, including the common iliac vessels proximal, the passage of lymphatic vessels to the groin distal, the ilioinguinal nerve lateral, and the bladder and prostate medial with the deepest part of the obturator fossa as the base. The obturator fossa is completely cleaned, especially the space behind the external iliac vessels, all the way to the sacrum (triangle of Marcille). There has been no change in the boundaries with time. Surgical complications of pLND only were graded according to the Clavien-Dindo grading system20 and prospectively maintained in the institutional database.
Histological Examination and Analysis All histopathology findings were reviewed by experienced pathologists at our tertiary reference center. Grade was assigned as well, moderately or poorly differentiated based on the amount of undifferentiated cells in the tumor on histopathological examination according to Broders.21 Lymphovascular invasion was defined as embolic tumor cells in thin-walled vessel-like structures. ENE was defined as tumor extension through the lymph node capsule into the perinodal fibrous-adipose tissue.
Statistical Analysis A generalized estimation equation model was used with an exchangeable correlation matrix to account for left and right pelvic node clustering in a patient. We used a univariable proportional hazards model to analyze pelvic characteristics predictive of DSS in patients with pelvic involvement. DSS was calculated using the Kaplan-Meier method. DSS was defined as the time from the date of pLND to the date of death from penile cancer, penile cancer metastasis or complications related to penile cancer treatment. At the last followup date survivors and patients who died of another cause were censored. All analysis was done with SPSSÒ, version 20.0 and R, version 2.15.2 (http://www.r-project.org/) with p
