- Email: [email protected]
Is systematic scalene node biopsy in pretreatment evaluation of locally advanced cervical carcinoma necessary?
Gynecologic Oncology 103 (2006) 1091 – 1094 www.elsevier.com/locate/ygyno
Is systematic scalene node biopsy in pretreatment evaluation of locally advanced cervical carcinoma necessary? Systematic dissection and histopathology of left scalene node biopsies in patients with locally advanced cervical carcinoma Vanessa Samouëlian a,b,⁎, Marie-Christine Baranzelli b , Fabrice Narducci a , Solphie Taïeb c , Véronique Cabaret b , Denis Querleu d , Eric Leblanc a a
b
Departement of Surgery, Centre Oscar Lambret, 3 rue Frédéric Combemale, BP307, 59020 Lille Cedex, France Departement of Anatomopathology, Centre Oscar Lambret, 3 rue Frédéric Combemale, BP307, 59020 Lille Cedex, France c Departement of Radiolology, Centre Oscar Lambret, 3 rue Frédéric Combemale, BP307, 59020 Lille Cedex, France d Departement of Surgery, Institut Claudius Regaud, 20-24 rue Pont St Pierre 31052 Toulouse Cedex, France Received 6 February 2006 Available online 4 August 2006
Abstract Objective. Cervical carcinomas mainly spread via lymphatics, stepwise from pelvic to aortic and scalenic lymph nodes. Metastatic nodes are the major prognostic factor in this disease. When scalenic nodes are involved, cervical cancer is considered to be disseminated. Since there is a major discrepancy in reported percentages of metastatic scalene nodes in the literature (0 to 50%), we proceeded to systematic pretreatment scalene node biopsy and then evaluated the validity of this procedure. Methods. From January 1998 to May 2003, 72 patients with locally advanced cervical carcinoma and no suspicious paraaortic or scalenic nodes (respectively on magnetic resonance imaging and clinically) had a systematic surgical pretreatment lymph node evaluation (retroperitoneal laparoscopic infrarenal paraaortic lymph node dissection and left scalenic lymph node biopsy). Scalene biopsy was examined using hematoxylin/ eosin stain and immunohistochemistry (KL1 antibodies). Results. Among the 72 patients, 20 were stage IB2, 4 were IIA, 14 were IIB, 4 were IIIA, 27 were IIIB, 1 was IVA and 2 had a recurrent cervical carcinoma. Fourteen women had histologically confirmed paraaortic metastases (11 macroscopic, 3 microscopic). No metastatic involvement of the scalene nodes was detected. Fifteen patients developed a recurrence within 12 months (3 to 19 months). None of the patients developed scalenic recurrence. Conclusion. Left scalene node biopsy does not appear to be mandatory in routine pretherapeutic lymph node evaluation of patients with advanced cervical carcinoma and no clinical suspicious nodes. It may be useful to prove disseminated disease in patients with suspicious clinical nodes or hot spots on PET-scan, if fine needle biopsy is unconclusive. © 2006 Elsevier Inc. All rights reserved. Keywords: Scalene node; Cervical carcinoma; Pretherapeutic evaluation; Immunohistochemistry
Introduction Cervical carcinomas mainly spread via lymphatic routes, stepwise from pelvic to aortic and scalenic lymph nodes. ⁎ Corresponding author. Hôpital Jeanne de Flandre, CHRU de Lille, 1 rue Eugène Avinée, 59037 Lille Cedex, France. Fax: +33 3 20 44 63 11. E-mail addresses: [email protected], [email protected] (V. Samouëlian). 0090-8258/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2006.06.029
Metastatic nodes are the main prognostic factor in this disease. Indeed, for any stage, when lymph nodes are found positive, 5year survival will drop 30 to 70% [1,2]. In locally advanced cervical carcinomas, pelvic and aortic nodes are metastatic in 27 to 66% and 5 to 33% of patients respectively [3,4]. Only a few reports of scalenic lymph node involvement are available in the literature, ranging from 0 to 50% of patients with metastatic aortic nodes [5,6]. When the scalenic nodes are involved, cervical cancer is considered to be disseminated. In
1092
V. Samouëlian et al. / Gynecologic Oncology 103 (2006) 1091–1094
such cases, aggressive surgery such as pelvic exenteration or extensive radiation is contraindicated. Some authors consider scalene lymph node sampling as part of pretreatment evaluation of patients with locally advanced cervical carcinomas [7]. Others conclude that scalenic evaluation is of limited value [8]. Nowadays, locally advanced cervical carcinomas are generally managed by concurrent chemoradiation. Since there is a major discrepancy in reported percentages of metastatic scalene nodes in the literature, we proceeded to systematic scalene node biopsy while installing the port site, at the time of surgical pretreatment nodal evaluation, to evaluate the validity of this procedure. In addition to standard pathological examination, scalenic nodes have been evaluated using immunohistochemistry techniques, which, to our knowledge, has never been reported before. Materials and methods From January 1998 to May 2003, 72 patients with locally advanced cervical carcinoma (stages IB2–IVA), no enlarged scalenic nodes on clinical examination nor enlarged aortic nodes on magnetic resonance imaging (MRI) were managed at the Centre Oscar Lambret, Lille, France. Aortic and left scalenic lymph nodes were routinely surgically assessed in this group of patients. Surgical staging included a diagnostic transperitoneal laparoscopic abdominopelvic exploration to preclude patients with obvious peritoneal carcinomatosis and/or fixed pelvic or paraaortic lymph nodes. If negative, a retroperitoneal laparoscopic infrarenal paraaortic lymph node dissection was performed, as described by Querleu et al. [9]. In the 72 patients, a left scalenic lymph node dissection followed. Prescalenic lymph node dissection was performed through a 5-cm-long left supra clavicular transverse incision above the middle third of the clavicle. Boundaries of dissection were the triangle marked inferiorly by the subclavian vein, medially by the internal jugular vein, and superolaterally by the inferior belly of the omohyoid muscle. The floor of the dissected tissue was made up of the anterior scalenus muscle and the phrenic nerve. The transverse cervical artery was respected. Scalenic nodes and their associated fat pads were resected en bloc between surgical vascular clips. When visible, the thoracic duct and/or its main branches were ligated with a prolene tie. The procedure was followed by the placement of a port-site catheter for concurrent chemoradiation using the left external or internal jugular vein. The entire left scalene biopsy, including lymph nodes and fat pad, was fixed in formol aldehyde and embedded in paraffin blocs. All scalene lymph nodes were first examined on one slide with standard hematoxylin/eosin (H&E) stain. We retrospectively performed immunohistochemistry using anti-keratin antibodies KL1 on all negative scalenic nodes. For this purpose, serial sections every 150 μm were cut from each block until paraffin block depletion. This was done for all scalenic nodes from patients who presented with paraaortic nodal involvement, distant metastases or recurrent disease. For patients without paraaortic involvement and without recurrence, immunohistochemistry was done in a standard fashion on three serial sections at 300-μm distance. Our positive control was an epithelial tissue, included in each run of KL1 immunostaining. All slides were reviewed and interpreted by the same pathologist specialized in gynecology oncology (MCB).
Results The median age of the 72 surgically staged patients was 46 years (25–70). Seventy had a primary and 2 had a recurrent cervical carcinoma. Of the 70 primary cervical carcinomas, 20 were stage IB2, 4 were IIA, 14 were IIB, 4 were IIIA, 27 were IIIB and 1 was IVA. Sixty-seven women had pathologically confirmed squamous cell carcinomas and five had adenocarcinomas.
Of the 72 patients included in this study, 14 had histologically confirmed paraaortic metastases. In 11 women, node metastasis (2 to 15 mm) was found, and in three, micrometastases (< 2 mm) were identified. All patients had normal preoperative MRI. Biopsy of the left scalene lymph nodes was performed on all 72 patients. The mean additional time necessary for the procedure was 12 min. No complication due to the procedure occurred intraoperatively, during the immediate postoperative course, or during follow-up. The average number of scalene nodes was 4.5 per patient (range: 1–12). No metastatic involvement of the scalene nodes was detected on the initial H&E slides. Consequently, immunohistochemistry on serial slides was performed on each of the 323 resected scalene nodes. One hundred and fourteen scalene nodes from 24 patients harboring metastatic paraaortic nodes and/or presenting with a recurrence were examined on serial cuts every 150 μm until paraffin bloc depletion. For the 209 others, immunohistochemistry was done on three cuts 300 μm apart. Immunohistochemical examination of all the scalene nodes never showed any metastatic involvement. The median follow-up was 36 months (16 to 80 months). Fifteen patients developed a recurrence after a median of 12 months (3 to 19 months). Five of those patients had positive paraaortic lymph nodes. Recurrence was locoregional in 12 cases and distant in three cases. In our series, none of the patients developed scalenic recurrence on either side. Discussion In this study, no metastatic scalene lymph nodes were identified among the 72 patients with locally advanced cervical carcinoma and no suspicion of nodal involvement clinically or on preoperative MRI. We sampled an average of 4 to 5 scalene lymph nodes per patient, which is consistent with literature data. Although only a few studies report on the number of scalenic nodes sampled per patient, this number has been found to range from 2 to 4.9
Table 1 Incidence of positive scalenic lymph nodes in patients with locally advanced cervical carcinoma and positive paraaortic lymph nodes Author
Incidence of metastatic scalenic nodes
Delgado, 1975 Buchsbaum, 1976 Buchsbaum, 1979 Brandt, 1981 Lee, 1981 Stehman, 1987 Trinci, 1988 Vasilev, 1989 Goff, 1999 Horowitz, 1999 Boran, 2003 Samouëlian, 2006 Combined series
1/9 (11.1%) 5/10 (50%) 8/23 (34.8%) 7/25 (28%) 3/7 (43%) 4/55 (7.3%) 2/20 (10%) 4/17 (23.5%) 5/14 (35.7%) 7/33 (21.2%) 3/28 (10.7%) 0/14 (0%) 49/255 (19.2%)
Review of the literature.
V. Samouëlian et al. / Gynecologic Oncology 103 (2006) 1091–1094
[5,10]. In our series, the entire scalenic fat pad was included into the paraffin blocs. Techniques for immunopathological assessment of lymph nodes are only briefly described in the reported studies, and none of them used immunohistochemistry. The serial section and immunohistochemical study of the samples, with an ultrastaging of the high risk patients in our series, confirmed the absence of scalenic nodal involvement. As expected, 19.4% (14/72) of the patients had histopathologically documented metastatic paraaortic lymph nodes. This percentage is concordant with the data found in the literature about the prevalence of metastatic paraaortic lymph nodes in women with locally advanced cervical carcinoma and no evidence of metastatic disease on preoperative MRI [4,11]. The overall incidence of scalenic metastases in patients with locally advanced cervical carcinoma and metastatic paraaortic lymph nodes is 49/241 (20.3%) when compiling of the available literature date (Table 1) [6–8,10,12–18]. The percentage of positive scalenic lymph nodes in this population varies from 7.3% to 50% [6,17]. The authors propose several explanations for the wide range of scalenic node positivity reported in the literature. First of all, most of the studies are clinical retrospective studies, including a small number of patients (7 to 55). On the other hand, important clinical differences are observed among the patients. Some of the patients with advanced stages of cervical carcinoma and documented paraaortic node, have clinically suspicious lymph nodes or proof of distant disease while others have no clinical suspicion of metastasis [14]. When limiting published data to those specifying clinical status of the scalenic nodes, the overall incidence of metastatic scalene lymph nodes rises to 7/11 (63.6%) when nodes are clinically suspicious, whereas it drops to 21/158 (13.3%) when scalene lymph nodes are clinically non-suspicious (Table 2). Only two of the reviewed studies specified the clinical status of paraaortic lymph nodes [7,10]. Among the 11 patients harboring scalenic metastases reported in these two studies, all but one had clinically evident paraaortic spread. All patients had histologically confirmed paraaortic metastatic nodes. Vasilev reports the only case of metastatic scalenic nodes in a patient with clinically non-suspicious paraaortic and scalenic nodes. However, this patient had histopathologically confirmed paraaortic metastasis. Regardless of their clinical status, scalenic lymph nodes were metastatic in 43.5% (10/23) of patients with grossly involved paraaortic lymph nodes. This percentage drops to 22.7% (5/22) when considering only patients with grossly involved paraaortic lymph nodes and clinically non-suspicious scalenic lymph nodes. Finally, only 3.7% (1/27) of patients with clinically nonsuspicious paraaortic or scalenic nodes, but with histologically confirmed paraaortic disease, harbored scalenic metastases. Only the latter population is comparable to our series. Considering the literature data, we confirm that scalene lymph node metastasis is rare in patients presenting with locally advanced cervical carcinoma and clinically non-suspicious paraaortic or scalenic lymph nodes. The results of immunohistochemistry and serial section assessment of scalenic nodes reported in this paper verifies that scalene nodes do not harbor
1093
Table 2 Histologic scalene node positivity versus clinical detection of scalenic and paraaortic nodes in advanced cervical carcinoma with histologically positive paraaortic nodes (review of the literature) Incidence of metastatic scalenic nodes Clinically suspicious scalene nodes
Horowitz, 1999 5/9 (55.6%) Trinci, 1988 2/2 (100%) Total 7/11 (63.6%) Clinically non-suspicious scalene Boran, 2003 3/28 (10.7%) Brandt, 1981 7/25 (28.0%) nodes regardless clinical status of metastatic paraaortic nodes Delgado, 1975 1/9 (11.1%) Horowitz, 1999 2/24 (8.3%) Stehman, 1987 4/55 (7.3%) Vasilev, 1989 4/17 (23.5%) Total 21/158 (13.3%) Including this series: 21/230 (9.1%) Clinically non-suspicious scalene Horowitz, 1999 2/11 (18.2%) nodes and suspicious paraaortic Vasilev, 1989 3/11 (27.3%) nodes Total 5/22 (22.7%) Clinically non-suspicious scalene Horowitz, 1999 0/17 (0%) and paraaortic nodes Vasilev, 1989 1/6 (16.7%) Total 1/23 (4.3%) Including this series: 1/37 (2.7%) Suspicious paraaortic nodes regardless Horowitz, 1999 7/16 (43.8%) of clinical status of scalene nodes Vasilev, 1989 3/11 (27.3%) Total 10/23 (43.5%)
occult micrometastasis and that ultrastaging does not provide additional information on the natural history of advanced cervical cancer. Some authors of the reviewed studies recommended routine scalene biopsy in all patients with positive paraaortic nodes [5–7,12–14,18]. More recent series, including ours, suggest that left scalene node biopsy is not mandatory in routine pretherapeutic lymph node evaluation of clinically node negative patients with cervical carcinoma [10]. This result allows for a better selection of patients who should undergo scalenic biopsy. Although we did not observe any complications related to the procedure, it does remain a surgical procedure with potential morbidity or mortality, as reported by Skinner [19]. Left scalene node biopsy may be useful for proof of disseminated disease in patients with advanced cervical carcinoma and obviously enlarged aortic or scalenic nodes or after the finding of abnormal fixation in the scalene area at PET scan, if fine needle biopsy is unconclusive [20]. References [1] Tanaka Y, Sawada S, Murata T. Relationship between lymph node metastases and prognosis in patients irradiated postoperatively for carcinoma of the uterine cervix. Acta Radiol Oncol 1984;23:455–9. [2] Tinga DJ, Timmer PR, Bouma J, Aalders JG. Prognostic significance of single versus multiple lymph node metastases in cervical carcinoma stage IB. Gynecol Oncol 1990;39:175–80. [3] Graham JB, Sotto LSJ, Paoloucek FP. Carcinoma of the cervix. Philadelphia: WB Saunders; 1962. [4] Lagasse LD, Creasman WT, Shingleton HM, Ford JH, Blessing JA. Results and complications of operative staging in cervical cancer: experience of the Gynecologic Oncology Group. Gynecol Oncol 1980;9:90–8. [5] Bortolozzi G, Belloni C, Maggioni A, Mangioni C. Considerations of the
1094
[6]
[7] [8]
[9]
[10]
[11]
[12]
V. Samouëlian et al. / Gynecologic Oncology 103 (2006) 1091–1094 clinical significance of diagnostic prescalene node biopsy in carcinoma of the cervix. Eur J Gynaecol Oncol 1981;2:24–7. Buchsbaum HJ, Lifshitz S. The role of scalene lymph node biopsy in advanced carcinoma of the cervix uteri. Surg Gynecol Obstet 1976;143: 246–8. Vasilev SA, Schlaerth JB. Scalene lymph node sampling in cervical carcinoma: a reappraisal. Gynecol Oncol 1990;37:120–4. Delgado G, Smith JP, Ballantyne AJ. Scalene node biopsy in carcinoma of the cervix. Pelvic and para-aortic lymphadenectomy. Cancer 1975;35: 784–6. Querleu D, Dargent D, Ansquer Y, Leblanc E, Narducci F. Extraperitoneal endosurgical aortic and common iliac dissection in the staging of bulky or advanced cervical carcinomas. Cancer 2000;88(15):1883–91. Horowitz NS, Tamimi HK, Goff BA, Koh WJ, Schmidt RA, Greer BE, et al. Pretreatment scalene node biopsy in gynecologic malignancy: prudent or passe? Gynecol Oncol 1999;75:238–41. Michel G, Morice P, Castaigne D, Leblanc M, Rey A, Duvillard P. Lymphatic spread in stage Ib and II cervical carcinoma: anatomy and surgical implications. Obstet Gynecol 1998;91:360–3. Boran N, Kayikcioglu F, Tulunay G, Kose MF. Scalene lymph node dissection in locally advanced cervical carcinoma: is it reasonable or unnecessary? Tumori 2003;89:173–5.
[13] Brandt III B, Lifshitz S. Scalene node biopsy in advanced carcinoma of the cervix uteri. Cancer 1981;47:1920–1. [14] Buchsbaum HJ. Extrapelvic lymph node metastases in cervical carcinoma. Am J Obstet Gynecol 1979;133:814–24. [15] Goff BA, Muntz HG, Paley PJ, Tamimi HK, Koh WJ, Greer BE. Impact of surgical staging in women with locally advanced cervical cancer. Gynecol Oncol 1999;74:436–42. [16] Lee RB, Weisbaum GS, Heller PB, Park RC. Scalene node biopsy in primary and recurrent invasive carcinoma of the cervix. Gynecol Oncol 1981;11:200–6. [17] Stehman FB, Bundy BN, Hanjani P, Fowler WC, Abdulhay G, Whitney CW. Biopsy of the scalene fat pad in carcinoma of the cervix uteri metastatic to the periaortic lymph nodes. Surg Gynecol Obstet 1987;165: 503–6. [18] Trinci M, Raffetto N, Petrozza V, Melis M, Biagini C. Pretreatment scalene node biopsy in cervical carcinoma. Eur J Gynaecol Oncol 1988;9:308–12. [19] Skinner DB. Scalene-lymph-node biopsy. Reappraisal of risks and indications. N Engl J Med 1963;268:1324–9. [20] Tran BN, Grigsby PW, Dehdashti F, Herzog TJ, Siegel BA. Occult supraclavicular lymph node metastasis identified by FDG-PET in patients with carcinoma of the uterine cervix. Gynecol Oncol 2003;90:572–6.
Is systematic scalene node biopsy in pretreatment evaluation of locally advanced cervical carcinoma necessary? Systematic dissection and histopathology of left scalene node biopsies in patients with locally advanced cervical carcinoma Vanessa Samouëlian a,b,⁎, Marie-Christine Baranzelli b , Fabrice Narducci a , Solphie Taïeb c , Véronique Cabaret b , Denis Querleu d , Eric Leblanc a a
b
Departement of Surgery, Centre Oscar Lambret, 3 rue Frédéric Combemale, BP307, 59020 Lille Cedex, France Departement of Anatomopathology, Centre Oscar Lambret, 3 rue Frédéric Combemale, BP307, 59020 Lille Cedex, France c Departement of Radiolology, Centre Oscar Lambret, 3 rue Frédéric Combemale, BP307, 59020 Lille Cedex, France d Departement of Surgery, Institut Claudius Regaud, 20-24 rue Pont St Pierre 31052 Toulouse Cedex, France Received 6 February 2006 Available online 4 August 2006
Abstract Objective. Cervical carcinomas mainly spread via lymphatics, stepwise from pelvic to aortic and scalenic lymph nodes. Metastatic nodes are the major prognostic factor in this disease. When scalenic nodes are involved, cervical cancer is considered to be disseminated. Since there is a major discrepancy in reported percentages of metastatic scalene nodes in the literature (0 to 50%), we proceeded to systematic pretreatment scalene node biopsy and then evaluated the validity of this procedure. Methods. From January 1998 to May 2003, 72 patients with locally advanced cervical carcinoma and no suspicious paraaortic or scalenic nodes (respectively on magnetic resonance imaging and clinically) had a systematic surgical pretreatment lymph node evaluation (retroperitoneal laparoscopic infrarenal paraaortic lymph node dissection and left scalenic lymph node biopsy). Scalene biopsy was examined using hematoxylin/ eosin stain and immunohistochemistry (KL1 antibodies). Results. Among the 72 patients, 20 were stage IB2, 4 were IIA, 14 were IIB, 4 were IIIA, 27 were IIIB, 1 was IVA and 2 had a recurrent cervical carcinoma. Fourteen women had histologically confirmed paraaortic metastases (11 macroscopic, 3 microscopic). No metastatic involvement of the scalene nodes was detected. Fifteen patients developed a recurrence within 12 months (3 to 19 months). None of the patients developed scalenic recurrence. Conclusion. Left scalene node biopsy does not appear to be mandatory in routine pretherapeutic lymph node evaluation of patients with advanced cervical carcinoma and no clinical suspicious nodes. It may be useful to prove disseminated disease in patients with suspicious clinical nodes or hot spots on PET-scan, if fine needle biopsy is unconclusive. © 2006 Elsevier Inc. All rights reserved. Keywords: Scalene node; Cervical carcinoma; Pretherapeutic evaluation; Immunohistochemistry
Introduction Cervical carcinomas mainly spread via lymphatic routes, stepwise from pelvic to aortic and scalenic lymph nodes. ⁎ Corresponding author. Hôpital Jeanne de Flandre, CHRU de Lille, 1 rue Eugène Avinée, 59037 Lille Cedex, France. Fax: +33 3 20 44 63 11. E-mail addresses: [email protected], [email protected] (V. Samouëlian). 0090-8258/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2006.06.029
Metastatic nodes are the main prognostic factor in this disease. Indeed, for any stage, when lymph nodes are found positive, 5year survival will drop 30 to 70% [1,2]. In locally advanced cervical carcinomas, pelvic and aortic nodes are metastatic in 27 to 66% and 5 to 33% of patients respectively [3,4]. Only a few reports of scalenic lymph node involvement are available in the literature, ranging from 0 to 50% of patients with metastatic aortic nodes [5,6]. When the scalenic nodes are involved, cervical cancer is considered to be disseminated. In
1092
V. Samouëlian et al. / Gynecologic Oncology 103 (2006) 1091–1094
such cases, aggressive surgery such as pelvic exenteration or extensive radiation is contraindicated. Some authors consider scalene lymph node sampling as part of pretreatment evaluation of patients with locally advanced cervical carcinomas [7]. Others conclude that scalenic evaluation is of limited value [8]. Nowadays, locally advanced cervical carcinomas are generally managed by concurrent chemoradiation. Since there is a major discrepancy in reported percentages of metastatic scalene nodes in the literature, we proceeded to systematic scalene node biopsy while installing the port site, at the time of surgical pretreatment nodal evaluation, to evaluate the validity of this procedure. In addition to standard pathological examination, scalenic nodes have been evaluated using immunohistochemistry techniques, which, to our knowledge, has never been reported before. Materials and methods From January 1998 to May 2003, 72 patients with locally advanced cervical carcinoma (stages IB2–IVA), no enlarged scalenic nodes on clinical examination nor enlarged aortic nodes on magnetic resonance imaging (MRI) were managed at the Centre Oscar Lambret, Lille, France. Aortic and left scalenic lymph nodes were routinely surgically assessed in this group of patients. Surgical staging included a diagnostic transperitoneal laparoscopic abdominopelvic exploration to preclude patients with obvious peritoneal carcinomatosis and/or fixed pelvic or paraaortic lymph nodes. If negative, a retroperitoneal laparoscopic infrarenal paraaortic lymph node dissection was performed, as described by Querleu et al. [9]. In the 72 patients, a left scalenic lymph node dissection followed. Prescalenic lymph node dissection was performed through a 5-cm-long left supra clavicular transverse incision above the middle third of the clavicle. Boundaries of dissection were the triangle marked inferiorly by the subclavian vein, medially by the internal jugular vein, and superolaterally by the inferior belly of the omohyoid muscle. The floor of the dissected tissue was made up of the anterior scalenus muscle and the phrenic nerve. The transverse cervical artery was respected. Scalenic nodes and their associated fat pads were resected en bloc between surgical vascular clips. When visible, the thoracic duct and/or its main branches were ligated with a prolene tie. The procedure was followed by the placement of a port-site catheter for concurrent chemoradiation using the left external or internal jugular vein. The entire left scalene biopsy, including lymph nodes and fat pad, was fixed in formol aldehyde and embedded in paraffin blocs. All scalene lymph nodes were first examined on one slide with standard hematoxylin/eosin (H&E) stain. We retrospectively performed immunohistochemistry using anti-keratin antibodies KL1 on all negative scalenic nodes. For this purpose, serial sections every 150 μm were cut from each block until paraffin block depletion. This was done for all scalenic nodes from patients who presented with paraaortic nodal involvement, distant metastases or recurrent disease. For patients without paraaortic involvement and without recurrence, immunohistochemistry was done in a standard fashion on three serial sections at 300-μm distance. Our positive control was an epithelial tissue, included in each run of KL1 immunostaining. All slides were reviewed and interpreted by the same pathologist specialized in gynecology oncology (MCB).
Results The median age of the 72 surgically staged patients was 46 years (25–70). Seventy had a primary and 2 had a recurrent cervical carcinoma. Of the 70 primary cervical carcinomas, 20 were stage IB2, 4 were IIA, 14 were IIB, 4 were IIIA, 27 were IIIB and 1 was IVA. Sixty-seven women had pathologically confirmed squamous cell carcinomas and five had adenocarcinomas.
Of the 72 patients included in this study, 14 had histologically confirmed paraaortic metastases. In 11 women, node metastasis (2 to 15 mm) was found, and in three, micrometastases (< 2 mm) were identified. All patients had normal preoperative MRI. Biopsy of the left scalene lymph nodes was performed on all 72 patients. The mean additional time necessary for the procedure was 12 min. No complication due to the procedure occurred intraoperatively, during the immediate postoperative course, or during follow-up. The average number of scalene nodes was 4.5 per patient (range: 1–12). No metastatic involvement of the scalene nodes was detected on the initial H&E slides. Consequently, immunohistochemistry on serial slides was performed on each of the 323 resected scalene nodes. One hundred and fourteen scalene nodes from 24 patients harboring metastatic paraaortic nodes and/or presenting with a recurrence were examined on serial cuts every 150 μm until paraffin bloc depletion. For the 209 others, immunohistochemistry was done on three cuts 300 μm apart. Immunohistochemical examination of all the scalene nodes never showed any metastatic involvement. The median follow-up was 36 months (16 to 80 months). Fifteen patients developed a recurrence after a median of 12 months (3 to 19 months). Five of those patients had positive paraaortic lymph nodes. Recurrence was locoregional in 12 cases and distant in three cases. In our series, none of the patients developed scalenic recurrence on either side. Discussion In this study, no metastatic scalene lymph nodes were identified among the 72 patients with locally advanced cervical carcinoma and no suspicion of nodal involvement clinically or on preoperative MRI. We sampled an average of 4 to 5 scalene lymph nodes per patient, which is consistent with literature data. Although only a few studies report on the number of scalenic nodes sampled per patient, this number has been found to range from 2 to 4.9
Table 1 Incidence of positive scalenic lymph nodes in patients with locally advanced cervical carcinoma and positive paraaortic lymph nodes Author
Incidence of metastatic scalenic nodes
Delgado, 1975 Buchsbaum, 1976 Buchsbaum, 1979 Brandt, 1981 Lee, 1981 Stehman, 1987 Trinci, 1988 Vasilev, 1989 Goff, 1999 Horowitz, 1999 Boran, 2003 Samouëlian, 2006 Combined series
1/9 (11.1%) 5/10 (50%) 8/23 (34.8%) 7/25 (28%) 3/7 (43%) 4/55 (7.3%) 2/20 (10%) 4/17 (23.5%) 5/14 (35.7%) 7/33 (21.2%) 3/28 (10.7%) 0/14 (0%) 49/255 (19.2%)
Review of the literature.
V. Samouëlian et al. / Gynecologic Oncology 103 (2006) 1091–1094
[5,10]. In our series, the entire scalenic fat pad was included into the paraffin blocs. Techniques for immunopathological assessment of lymph nodes are only briefly described in the reported studies, and none of them used immunohistochemistry. The serial section and immunohistochemical study of the samples, with an ultrastaging of the high risk patients in our series, confirmed the absence of scalenic nodal involvement. As expected, 19.4% (14/72) of the patients had histopathologically documented metastatic paraaortic lymph nodes. This percentage is concordant with the data found in the literature about the prevalence of metastatic paraaortic lymph nodes in women with locally advanced cervical carcinoma and no evidence of metastatic disease on preoperative MRI [4,11]. The overall incidence of scalenic metastases in patients with locally advanced cervical carcinoma and metastatic paraaortic lymph nodes is 49/241 (20.3%) when compiling of the available literature date (Table 1) [6–8,10,12–18]. The percentage of positive scalenic lymph nodes in this population varies from 7.3% to 50% [6,17]. The authors propose several explanations for the wide range of scalenic node positivity reported in the literature. First of all, most of the studies are clinical retrospective studies, including a small number of patients (7 to 55). On the other hand, important clinical differences are observed among the patients. Some of the patients with advanced stages of cervical carcinoma and documented paraaortic node, have clinically suspicious lymph nodes or proof of distant disease while others have no clinical suspicion of metastasis [14]. When limiting published data to those specifying clinical status of the scalenic nodes, the overall incidence of metastatic scalene lymph nodes rises to 7/11 (63.6%) when nodes are clinically suspicious, whereas it drops to 21/158 (13.3%) when scalene lymph nodes are clinically non-suspicious (Table 2). Only two of the reviewed studies specified the clinical status of paraaortic lymph nodes [7,10]. Among the 11 patients harboring scalenic metastases reported in these two studies, all but one had clinically evident paraaortic spread. All patients had histologically confirmed paraaortic metastatic nodes. Vasilev reports the only case of metastatic scalenic nodes in a patient with clinically non-suspicious paraaortic and scalenic nodes. However, this patient had histopathologically confirmed paraaortic metastasis. Regardless of their clinical status, scalenic lymph nodes were metastatic in 43.5% (10/23) of patients with grossly involved paraaortic lymph nodes. This percentage drops to 22.7% (5/22) when considering only patients with grossly involved paraaortic lymph nodes and clinically non-suspicious scalenic lymph nodes. Finally, only 3.7% (1/27) of patients with clinically nonsuspicious paraaortic or scalenic nodes, but with histologically confirmed paraaortic disease, harbored scalenic metastases. Only the latter population is comparable to our series. Considering the literature data, we confirm that scalene lymph node metastasis is rare in patients presenting with locally advanced cervical carcinoma and clinically non-suspicious paraaortic or scalenic lymph nodes. The results of immunohistochemistry and serial section assessment of scalenic nodes reported in this paper verifies that scalene nodes do not harbor
1093
Table 2 Histologic scalene node positivity versus clinical detection of scalenic and paraaortic nodes in advanced cervical carcinoma with histologically positive paraaortic nodes (review of the literature) Incidence of metastatic scalenic nodes Clinically suspicious scalene nodes
Horowitz, 1999 5/9 (55.6%) Trinci, 1988 2/2 (100%) Total 7/11 (63.6%) Clinically non-suspicious scalene Boran, 2003 3/28 (10.7%) Brandt, 1981 7/25 (28.0%) nodes regardless clinical status of metastatic paraaortic nodes Delgado, 1975 1/9 (11.1%) Horowitz, 1999 2/24 (8.3%) Stehman, 1987 4/55 (7.3%) Vasilev, 1989 4/17 (23.5%) Total 21/158 (13.3%) Including this series: 21/230 (9.1%) Clinically non-suspicious scalene Horowitz, 1999 2/11 (18.2%) nodes and suspicious paraaortic Vasilev, 1989 3/11 (27.3%) nodes Total 5/22 (22.7%) Clinically non-suspicious scalene Horowitz, 1999 0/17 (0%) and paraaortic nodes Vasilev, 1989 1/6 (16.7%) Total 1/23 (4.3%) Including this series: 1/37 (2.7%) Suspicious paraaortic nodes regardless Horowitz, 1999 7/16 (43.8%) of clinical status of scalene nodes Vasilev, 1989 3/11 (27.3%) Total 10/23 (43.5%)
occult micrometastasis and that ultrastaging does not provide additional information on the natural history of advanced cervical cancer. Some authors of the reviewed studies recommended routine scalene biopsy in all patients with positive paraaortic nodes [5–7,12–14,18]. More recent series, including ours, suggest that left scalene node biopsy is not mandatory in routine pretherapeutic lymph node evaluation of clinically node negative patients with cervical carcinoma [10]. This result allows for a better selection of patients who should undergo scalenic biopsy. Although we did not observe any complications related to the procedure, it does remain a surgical procedure with potential morbidity or mortality, as reported by Skinner [19]. Left scalene node biopsy may be useful for proof of disseminated disease in patients with advanced cervical carcinoma and obviously enlarged aortic or scalenic nodes or after the finding of abnormal fixation in the scalene area at PET scan, if fine needle biopsy is unconclusive [20]. References [1] Tanaka Y, Sawada S, Murata T. Relationship between lymph node metastases and prognosis in patients irradiated postoperatively for carcinoma of the uterine cervix. Acta Radiol Oncol 1984;23:455–9. [2] Tinga DJ, Timmer PR, Bouma J, Aalders JG. Prognostic significance of single versus multiple lymph node metastases in cervical carcinoma stage IB. Gynecol Oncol 1990;39:175–80. [3] Graham JB, Sotto LSJ, Paoloucek FP. Carcinoma of the cervix. Philadelphia: WB Saunders; 1962. [4] Lagasse LD, Creasman WT, Shingleton HM, Ford JH, Blessing JA. Results and complications of operative staging in cervical cancer: experience of the Gynecologic Oncology Group. Gynecol Oncol 1980;9:90–8. [5] Bortolozzi G, Belloni C, Maggioni A, Mangioni C. Considerations of the
1094
[6]
[7] [8]
[9]
[10]
[11]
[12]
V. Samouëlian et al. / Gynecologic Oncology 103 (2006) 1091–1094 clinical significance of diagnostic prescalene node biopsy in carcinoma of the cervix. Eur J Gynaecol Oncol 1981;2:24–7. Buchsbaum HJ, Lifshitz S. The role of scalene lymph node biopsy in advanced carcinoma of the cervix uteri. Surg Gynecol Obstet 1976;143: 246–8. Vasilev SA, Schlaerth JB. Scalene lymph node sampling in cervical carcinoma: a reappraisal. Gynecol Oncol 1990;37:120–4. Delgado G, Smith JP, Ballantyne AJ. Scalene node biopsy in carcinoma of the cervix. Pelvic and para-aortic lymphadenectomy. Cancer 1975;35: 784–6. Querleu D, Dargent D, Ansquer Y, Leblanc E, Narducci F. Extraperitoneal endosurgical aortic and common iliac dissection in the staging of bulky or advanced cervical carcinomas. Cancer 2000;88(15):1883–91. Horowitz NS, Tamimi HK, Goff BA, Koh WJ, Schmidt RA, Greer BE, et al. Pretreatment scalene node biopsy in gynecologic malignancy: prudent or passe? Gynecol Oncol 1999;75:238–41. Michel G, Morice P, Castaigne D, Leblanc M, Rey A, Duvillard P. Lymphatic spread in stage Ib and II cervical carcinoma: anatomy and surgical implications. Obstet Gynecol 1998;91:360–3. Boran N, Kayikcioglu F, Tulunay G, Kose MF. Scalene lymph node dissection in locally advanced cervical carcinoma: is it reasonable or unnecessary? Tumori 2003;89:173–5.
[13] Brandt III B, Lifshitz S. Scalene node biopsy in advanced carcinoma of the cervix uteri. Cancer 1981;47:1920–1. [14] Buchsbaum HJ. Extrapelvic lymph node metastases in cervical carcinoma. Am J Obstet Gynecol 1979;133:814–24. [15] Goff BA, Muntz HG, Paley PJ, Tamimi HK, Koh WJ, Greer BE. Impact of surgical staging in women with locally advanced cervical cancer. Gynecol Oncol 1999;74:436–42. [16] Lee RB, Weisbaum GS, Heller PB, Park RC. Scalene node biopsy in primary and recurrent invasive carcinoma of the cervix. Gynecol Oncol 1981;11:200–6. [17] Stehman FB, Bundy BN, Hanjani P, Fowler WC, Abdulhay G, Whitney CW. Biopsy of the scalene fat pad in carcinoma of the cervix uteri metastatic to the periaortic lymph nodes. Surg Gynecol Obstet 1987;165: 503–6. [18] Trinci M, Raffetto N, Petrozza V, Melis M, Biagini C. Pretreatment scalene node biopsy in cervical carcinoma. Eur J Gynaecol Oncol 1988;9:308–12. [19] Skinner DB. Scalene-lymph-node biopsy. Reappraisal of risks and indications. N Engl J Med 1963;268:1324–9. [20] Tran BN, Grigsby PW, Dehdashti F, Herzog TJ, Siegel BA. Occult supraclavicular lymph node metastasis identified by FDG-PET in patients with carcinoma of the uterine cervix. Gynecol Oncol 2003;90:572–6.