Mediastinoscopy is an exploratory surgical procedure

Transcription

1 Remediastinoscopy After Induction Chemotherapy in Non-Small Cell Lung Cancer Miquel Mateu-Navarro, MD, Ramón Rami-Porta, MD, Romà Bastus-Piulats, MD, Luis Cirera-Nogueras, MD, and Guadalupe González-Pont, MD Sections of Thoracic Surgery and Oncology, Department of Pathology, Hospital Mútua de Terrassa, University of Barcelona, Terrassa, Barcelona, Spain Background. This study was undertaken to evaluate the technical feasibility and the sensitivity, specificity, and accuracy of remediastinoscopy in restaging N2 bronchogenic carcinoma treated with neoadjuvant chemotherapy. Methods. Patients presenting mediastinal lymph node involvement at mediastinoscopy received three or four cycles of neoadjuvant chemotherapy with mitomycin, iphosphamide, and cisplatin or cisplatin and gemcitabine. If there was no disease progression, these patients underwent remediastinoscopy and, if no residual extracapsular involvement or N3 disease was found, a thoracotomy was then carried out. Results. Twenty-four patients underwent remediastinoscopy. In 12 (50%) remediastinoscopy was positive. The 12 remaining patients were operated on and the tumors resected: 5 pneumonectomies and 7 lobectomies. Lymphadenectomy specimens showed residual disease in mediastinal lymph nodes in 5 patients (pn2) and hilar lymph nodes in 1 patient (pn1). The other 6 patients were free of nodal disease, and 4 of them presented no involvement at lung level either. The sensitivity, specificity, and accuracy of remediastinoscopy were 0.7, 1, and 0.8, respectively. Conclusions. Remediastinoscopy is a technically feasible staging tool with high diagnostic accuracy that is useful in the selection of patients who can be served best by complete resection after neoadjuvant chemotherapy. (Ann Thorac Surg 2000;70:391 5) 2000 by The Society of Thoracic Surgeons Mediastinoscopy is an exploratory surgical procedure that allows obtaining biopsies of highest mediastinal (lymph node station 1), superior (lymph node station 2R and 2L) and inferior (lymph node station 4R and 4L) paratracheal, and subcarinal (lymph node station 7) nodal groups [1]. Extended cervical mediastinoscopy [2, 3] and left parasternal mediastinotomy allow for the surgical exploration of the subaortic region and the collection of biopsy specimens from aortopulmonary (5) and anterior mediastinal nodes (6) in left bronchogenic carcinoma (BC) [1]. Although these surgical explorations of the mediastinum are now beginning to be performed systematically in some studies [4, 5] and selectively in others [6, 7] in neoadjuvant chemotherapy clinical trials for stage IIIA (N2) BC, these explorations are rarely repeated in the evaluation of tumoral response to induction therapy. However, there is evidence that operation is most beneficial in patients who are downstaged to N1 or N0 and those remaining N2 have an ominous prognosis [8]. Therefore, the indication for operation in the latter group of patients is questionable. This study presents our preliminary experience of the use of remediastinoscopy to ascertain nodal downstaging after neoadjuvant chemotherapy in patients with nodal involvement initially diagnosed by mediastinoscopy. Accepted for publication Feb 2, Address reprint requests to Dr Mateu-Navarro, Section of Thoracic Surgery, Hospital Mutua de Terrassa, Calle Sant Antoni 8-14, Terrassa, Barcelona, Spain; mmn@comb.es. Material and Methods Preoperative Study All patients with diagnostic certainty or with suspected BC underwent fibrobronchoscopy and a study of disease extension, which included computed tomography (CT) of the chest and upper abdomen, abdominal ultrasound, and bone scan. If patients were deemed eligible for thoracotomy and their functional status permitted, surgical exploration of the mediastinum was performed systematically (mediastinoscopy for right BC or mediastinoscopy associated to left parasternal mediastinotomy or extended cervical mediastinoscopy for left BC). At mediastinoscopy, right and left superior and inferior paratracheal nodes and subcarinal and pretracheal nodes were routinely explored. If no nodes were found on the left paratracheal margin, a sample of fatty tissue was taken for pathologic analysis. At left parasternal mediastinotomy or at extended cervical mediastinoscopy, both subaortic and anterior mediastinal nodal stations were routinely explored. These surgical explorations yielded the maximum certainty factor factor 3 for clinical classification [9]. The criteria for contraindication to curative resection resulting from the mediastinal exploration were invasion of upper paratracheal lymph nodes; extracapsular invasion at any location; contralateral lymph node involvement; direct invasion of the tissues or mediastinal structures; and diagnosed small cell carcinoma [10]. Moreover, extracapsular lymph node involvement was considered a criteria for incomplete resection [11] by The Society of Thoracic Surgeons /00/$20.00 Published by Elsevier Science Inc PII S (00)

2 392 MATEU-NAVARRO ET AL Ann Thorac Surg STAGING REMEDIASTINOSCOPY FOR LUNG CANCER 2000;70:391 5 Until 1993, patients who were not eligible for operation on the basis of the above-mentioned criteria received radiotherapy on the tumor and mediastinum as the only definitive therapeutic option. Neoadjuvant Chemotherapy In 1994, we started a neoadjuvant chemotherapy protocol for stage IIIA on the basis of preceding positive experiences [7, 12]. Those patients with N2 BC diagnosed at mediastinoscopy and with a Karnofsky index more than 70% received three or four cycles of neoadjuvant chemotherapy with cisplatin 50 mg/m 2, iphosphamide 3 g/m 2, and mitomicin 6 mg/m 2 every 21 days, or, in 1998, cisplatin 100 mg/m 2 on day 1 and gemcitabine 1 g/m 2 on days 1 and 8 every 21 days. Chemotherapy was started between days 7 and 10 after mediastinoscopy, with the exception of 1 patient who developed an infection of the surgical incision and in whom chemotherapy was delayed until the 20th postoperative day. Restaging After Neoadjuvant Chemotherapy At completion of induction chemotherapy, restaging was performed by means of posteroanterior and lateral chest roentgenograms, fiberoptic bronchoscopy, and CT of the chest and upper abdomen. Other diagnostic aids, such as CT of the brain or bone scan, were performed selectively upon clinical indication, if extrathoracic extension were suspected during induction therapy. Those patients who, under radiologic criteria, showed no progression of the disease underwent remediastinoscopy 60 to 196 days (mean, 122 days; standard deviation, 36 days) after the initial mediastinoscopy for two purposes: first, to evaluate from a pathologic standpoint the response to induction therapy; and second, to select which patients could undergo operation with complete resection. In practice, remediastinoscopy was specifically intended to screen out N3 disease and extranodal involvement. Remediastinoscopy is started by resecting the scar of the previous mediastinoscopy. Dissection as far as the pretracheal fascia is usually not difficult as at initial mediastinoscopy the paratracheal muscles are not sutured to the midline. When the pretracheal fascia is reached it is incised and torn by digital dissection following a craniocaudal direction. This maneuver allows the determination of the presence of peritracheal adhesions, which are stronger and much more abundant in number than those encountered at initial mediastinoscopy. Peritracheal digital palpation is sufficient to start inserting the mediastinoscope. Next, and to obtain new biopsy samples, an access route is gradually created by means of electrocautery, gaining access at least into those lymph node stations that had proved positive during the first mediastinoscopy and that need to be reanalyzed. If technically feasible, other nodal areas are also reached to rule out subclinical progression of the disease. Generally, the mediastinoscope is more easily inserted through the left side of the trachea, as there are less lymph nodes at this site, having been less manipulated than the right side at the time the initial mediastinoscopy was performed. All biopsy samples taken during the exploration are sent for intraoperative pathologic examination. If extracapsular residual disease or N3 disease is found, the operation is brought to an end and radiotherapy indicated as definitive treatment. If, however, no extracapsular disease or N3 disease is found, the patient s position is changed and a thoracotomy is performed for lung resection followed by systematic nodal dissection. Results From 1994 to 1998, 303 mediastinoscopies were performed for staging purposes in patients with BC. Of these, 55 had N2 disease and 5 had N3. Twenty-five patients with N2 were not treated with chemotherapy because of a Karnofsky index less than 70%. Four patients did not accept neoadjuvant chemotherapy and were treated alternatively with radiotherapy. Twenty-six male patients with a mean age of 63 years (range, 52 to 73 years) who had been diagnosed as N2 by mediastinoscopy were eligible for chemotherapy. The characteristics of these patients are shown in Table 1. The most frequent histologic type found was squamous cell carcinoma. Tumors were localized to a large extent at the upper lobes. Eight tumors were localized at the hilum. At initial staging mediastinoscopy, the most frequently affected lymph node areas were the lower right paratracheal region (16 patients) and the subcarinal region (8 patients). There was single node involvement in 23 patients and double nodal involvement in 3 patients (refer to Table 1). Most patients showed good response to chemotherapy. During the course of induction chemotherapy, 2 patients had to be excluded from the study: 1 patient developed metastasis at the elbow (patient 16, Table 1) and the other patient died of pulmonary thromboembolism (patient 10, Table 1). After induction chemotherapy, CT of the chest revealed the persistence of radiologically pathologic mediastinal lymph nodes in 8 patients, of which 5 were true positives; of the 16 patients showing radiologic improvement on the postchemotherapy CT, 9 were true negatives. Independently of CT results, a remediastinoscopy was performed on the remaining 24 patients. In all of them it was possible to obtain another biopsy at the same nodal stations that had been positive at initial mediastinoscopy. Once a biopsy was taken of these nodal stations, no further exploration was carried out. It was found that there was persistent mediastinal disease in 12 (50%) patients. In 10 patients, lymph node involvement coincided at the same location as at initial mediastinoscopy; 1 patient was found to have N2 disease in a different nodal area; and another patient was shown to have contralateral mediastinal involvement (N3) that was nonexistent at the time of the first mediastinoscopy. Except for a surgical wound infection no other compli-

3 Ann Thorac Surg MATEU-NAVARRO ET AL 2000;70:391 5 STAGING REMEDIASTINOSCOPY FOR LUNG CANCER 393 Table 1. Patients Affected With Bronchogenic Carcinoma With Mediastinal Lymph Node Involvement Patient No. Pathology Location N2 Mediastinoscopy N. Adjuvant Chemotherapy Response CT Scan Remediastinoscopy Treatment pt pn 1 SCC LH 4L mic neg LP T0N0 2 SCC RUL 4R mic pos/4r RDT 3 ADC RH 4R 7 mic pos/4r RDT 4 ADC M, RLL 4R mic pos/4r RDT 5 ADC RUL 4R mic pos/4r RDT 6 LC RH 4R mic pos/4r RDT 7 SCC RUL 4R mic neg Rul T2N2 8 SCC LUL 7 mic neg LP T2N2 9 LC LH 7 mic pos/4r RDT 10 SCC RH 4R, 7 mic 11 SCC RUL 4R mic neg RP T2N2 12 SCC RH 4R mic pos/4r RDT 13 SCC RUL 4R mic neg RP T0N0 14 ADC RH 7 mic pos/7 RDT 15 LC LH 7 mic pos/4rn3 RDT 16 SCC RUL mic 17 SCC RUL 4R mic pos/4r RDT 18 SCC, ADC LUL 4L mic neg Lul T2N0 19 ADC RUL 4R mic pos/4r RDT 20 ADC RUL 4R mic pos/4r RDT 21 LC RUL 4R mic neg Rul S T0N0 22 SCC LUL 4R mic neg Lul T0N0 23 SCC LLL 7 gc neg Lll T0N0 24 SCC LUL 5 gc neg Lul S T2N2 25 ADC LUL 5 gc neg Lul T1N1 26 LC LUL 7, 10L gc neg LP T0N2 ADC adenocarcinoma; CTscan good response, bad response to chemotherapy; gc gencitabin and cisplatin; LC large cell carcinoma; LH left hilum; LLL left lower lobe; Lll left lower lobectomy; LP left pneumonectomy; LUL left upper lobe; Lul left upper lobectomy; mic mitomycin, iphosphamide, and cisplatin; ML middle lobe; N neoadjuvant; neg negative; pos positive; RDT radiotherapy; RH right hilum; RLL right lower lobe; RUL right upper lobe; RP right pneumonectomy; Rul right upper lobectomy; Rul or Lul S right upper lobectomy and sleeve resection; SCC squamous cell carcinoma. cations were encountered and the 12 patients who did not undergo thoracotomy were discharged 24 hours after operation. The 12 patients in whom the remediastinoscopy had been negative underwent operation. Lung resection and systematic nodal dissection was possible in all 12 patients. There were 5 pneumonectomies and 7 lobectomies, of which two were sleeve lobectomies (one right upper and one left upper) (Table 1). Systematic nodal dissection of the right side was extended to include the paratracheal, tracheobronchial, subcarinal, paraesophageal, and pulmonary ligament regions, as well as the hilar and interlobar nodes. On the left side, the dissection was extended to include the subaortic and anterior mediastinal regions, as well as the regions included in the right side, with the exception of the paratracheal region. This region is only examined if lymph nodes are identified at first sight or upon palpation. No mobilization of the aortic arch was performed on any of the patients. The mean number of lymph nodes excised was 18.4 (range, 8 to 28 nodes; standard deviation, 7.2 nodes). With regard to the T-factor, pathologic staging was as follows: in 5 patients no tumor was found in the resected specimen (pt0) and in 1 patient nonmeasurable tumoral remains were evident (ptx). The tumor persisted in 6 other patients, 5 pt2 and 1 pt1, which before induction chemotherapy were ct2. As far as the N-factor is concerned, 5 patients were pn2 (1 patient had two nodal stations involved: lower left paratracheal and subaortic; and 4 patients had single nodal station involvement: one each, in right upper paratracheal, right lower paratracheal, subcarinal, and anterior mediastinal stations) and 1 patient was pn1. No hilar or mediastinal nodal involvement was found in any of the 6 remaining patients. None of the patients with pn2 disease had pn1 disease simultaneously. When evaluating the response of mediastinal lymph nodes to neoadjuvant chemotherapy, CT sensitivity was 0.41, specificity was 0.75, and accuracy was Remediastinoscopy sensitivity was 0.70, specificity was 1, and accuracy was Comment Neoadjuvant chemotherapy followed by operation in locally advanced BC appears to have a favorable prognostic effect when compared to operation alone [7, 12]. In

4 394 MATEU-NAVARRO ET AL Ann Thorac Surg STAGING REMEDIASTINOSCOPY FOR LUNG CANCER 2000;70:391 5 these patients the clinical classification, especially in clinical trial protocols, must yield the maximum certainty factor [9]. With reference to mediastinal staging, the surgical exploration of the mediastinum (mediastinoscopy, parasternal mediastinotomy, and extended cervical mediastinoscopy) provides the possibility of a pathologic diagnosis, which yields the maximum clinical classification certainty. Its sensitivity ranges between 66% and 81% depending on the histologic type and tumor location, with a specificity of 100% [13]. In contrast, a clinical classification using imaging diagnostic aids, most commonly CT (with a sensitivity and specificity [14] of around 69% and 71%, respectively, and results of which are based solely on the evaluation of lymph node size), does not provide, in our opinion, sufficient information to be able to reach suitable therapeutic decisions in such patients. In keeping with this view, a systematic surgical exploration of the mediastinum is performed on all patients that are suitable for thoracotomy. At present, those with N2 involvement receive neoadjuvant chemotherapy and, if no progression of the disease occurs, they undergo a second exploratory mediastinoscopy. The main purpose of this repeat mediastinoscopy is to obtain an objective assessment of the effect that the neoadjuvant chemotherapy has had and also to rule out the presence of conditions that could render any future operation incomplete, namely extracapsular involvement and presence of N3 not apparent by radiologic criteria. This allows the selection of patients with a higher probability to undergo complete operation. Complete operation appears to be the most significant prognostic factor associated with patients who present long-term survival [15]. Remediastinoscopy has been performed for various indications: assessment of second primary tumors or recurrent tumors [16 18]; in patients where the first mediastinoscopy was incomplete [19]; in patients with a delayed commencement of the treatment [20]; and in patients of neoadjuvant chemotherapy [18]. In all these patients, it was technically possible to perform remediastinoscopy and no complications were encountered. Our initial experience in the assessment of response to neoadjuvant chemotherapy in N2 BC has been published previously [21]. The persistence of lymph node involvement after neoadjuvant chemotherapy has a discouraging prognosis [8]. Because of this, we consider the surgical reassessment of the mediastinum to take biopsies and to verify the condition of lymph nodes after induction to be important in selecting those patients who responded to induction therapy. From a technical point, remediastinoscopy is somewhat more complex than initial mediastinoscopy. This is because of the presence of a larger number of peritracheal adhesions that make the exploration more cumbersome. Despite this added difficulty and by means of digital dissection and electrocautery, it was possible in all patients to perform new biopsies of, at least, all those lymph node groups that had proved positive at the first mediastinoscopy. These remediastinoscopies made it possible for us to reassess those adenopathies that were initially affected. From our experience we have learned that remediastinoscopy is not a perfect technique. It yields falsenegative findings; but, on the other hand, it enabled us to exclude from operation 12 of the 24 patients that had not presented progression of the disease after neoadjuvant chemotherapy. Eleven of these 12 patients had persistent lymph node disease with extracapsular involvement, and in 1 patient a new N3 had developed in lymph nodes that had had a negative biopsy at initial mediastinoscopy. In these 12 patients, surgical resection would have been incomplete. On the other hand, in the 12 patients that in fact underwent thoracotomy, it was possible to perform lung resection and systematic nodal dissection. Regarding the five false negatives, which were shown to have residual disease upon examination of the mediastinal lymphadenectomy specimen, it could be said that neoadjuvant chemotherapy did not accomplish a complete eradication of lymph node disease. It was, however, capable of reversing a clinically evident N2 (by mediastinoscopy) into an N2 that could not be identified by remediastinoscopy, and which could only be demonstrated upon pathologic examination of the sample obtained from the mediastinal lymphadenectomy. At least in theory, these patients would have a better prognosis [22]. We lack sufficient follow-up to confirm this interpretation in our series. When analyzing the value of CT in mediastinal restaging after neoadjuvant chemotherapy for N2 BC, we see that the presence or absence of lymph nodes of a radiologically pathologic size (more than 1 cm in diameter) is of little value and can be very misleading. This is because in 40% of our case experiences, the increase in lymph node size was totally unrelated to the actual neoplastic disease. We conclude that in the restaging of N2 bronchogenic carcinoma treated with neoadjuvant chemotherapy, remediastinoscopy is a technically feasible exploratory procedure that does not pose any complications secondary to the complexity associated to the presence of peritracheal adhesions. Remediastinoscopy has higher sensitivity, specificity, and accuracy than CT and, therefore, allows for a more accurate selection of patients eligible to undergo complete resection. We are very thankful to Dr Erino A. Rendina, Department of Thoracic Surgery, University La Sapienza, Rome, Italy, for reviewing the manuscript and for his thoughtful and stimulating comments. References 1. Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111: Ginsberg RJ, Rice TW, Goldberg M, et al. Extended cervical mediastinoscopy: a single staging procedure for bronchogenic carcinoma of the left upper lobe. J Thorac Cardiovasc Surg 1987;94: Lopez L, Varela A, Freixinet J, et al. Extended cervical mediastinoscopy: prospective study of fifty cases. Ann Thorac Surg 1994;57:555 8.

5 Ann Thorac Surg MATEU-NAVARRO ET AL 2000;70:391 5 STAGING REMEDIASTINOSCOPY FOR LUNG CANCER Strauss GM, Heridan JE, Sherman DD, et al. Neoadjuvant chemotherapy and radiotherapy followed by surgery in stage IIIA non-small cell carcinoma of the lung: report of a Cancer and Leukemia Group B phase II study. J Clin Oncol 1992;10: Burkes RL, Ginsberg RJ, Shepherd FA. Induction chemotherapy with mitomycin, vindesine and cisplatin for stage III unresectable non small-cell lung cancer. Results of the Toronto phase II trial. J Clin Oncol 1992;10: Martini N, Kris MG, Flehinger BJ. Preoperative chemotherapy for stage IIIA (N2) lung cancer. The Sloan-Kettering experience with 136 patients. Ann Thorac Surg 1993;55: Roth JA, Fossela F, Komaki R. A randomized trial comparing preoperative chemotherapy and surgery with surgery alone in resectable stage IIIA non small cell lung cancer. J Natl Cancer Inst 1994;86: Rice TW, Adelstein DJ, Ciezki JP. Short course induction chemotherapy with paclitaxel for stage III non-small cell lung cancer. Ann Thorac Surg 1998;66: UICC International Union Against Cancer. In: Sobin LH, Wittekind Ch, eds. TNM Classification of malignant tumours, 5th edition. New York: Wiley-Liss, 1997: Pearson FG, Delarue NC, Ilves R, Todd TRJ, Cooper JD. Significance of positive superior mediastinal nodes identified at mediastinoscopy in patients with resectable cancer of the lung. J Thorac Cardiovasc Surg 1982;83: Mountain CF. Biologic, physiologic, and technical determinants in surgical therapy for lung cancer. In: Straus MJ, ed. Lung cancer. Clinical diagnosis and treatment, 2nd ed. New York: Grune & Straton, 1983: Rosell R, Gómez-Codina J, Camps C, et al. A randomized trial comparing preoperative chemotherapy plus surgery with surgery alone in non-small-cell lung cancer. N Engl J Med 1994;330: Funatsu T, Matsuhara Y, Hatakenaka R, Kosaba S, Yasuda Y, Ikeda S. The role of mediastinoscopic biopsy in preoperative assessment of lung cancer. J Thorac Cardiovasc Surg 1992; 104: Dilleman B, Deniffe G, Verschakelen J, Decramer M. Value of computed tomography and mediastinoscopy in the preoperative evaluation of mediastinal nodes in non-small cell lung cancer. Eur J Cardiothorac Surg 1994;8: Roth JA, Atkinson EN, Fosella F. Long term follow-up of patients enrolled in a randomized trial comparing perioperative chemotherapy and surgery with surgery alone in resectable stage IIIA non small cell lung cancer. Lung Cancer 1998;21: Meerschaut D, Vermassen F, Brutel de la Riviere A, Knaepen PJ, Van den Bosch JM, Vanderschueren R. Repeat mediastinoscopy in the assessment of new and recurrent lung neoplasm. Ann Thorac Surg 1992;53: Lewis RJ, Sisler GE, Mackenzie JW. Repeat mediastinoscopy. Ann Thorac Surg 1984;37: Pauwels M, Van Schil P, De Backer W, Van den Brande F, Eyskens E. Repeat mediastinoscopy in the staging of lung cancer. Eur J Cardiothorac Surg 1998;14: Olsen PS, Stentoft P, Ellefsen B, Petterson G. Remediastinoscopy in the assessment of resectability of lung cancer. Eur J Cardiothoracic Surg 1997;11: Palva T, Palva A, Kärjä J. Re-mediastinoscopy. Arch Otolaryngol 1975;101: Mateu-Navarro M, Rami-Porta R. Cuesta-Palomero M. Mediastinoscopy and remediastinoscopy in the preoperative assessment of bronchogenic carcinoma. Am J Respir Crit Care Med 1996;153:A Shields TW. The significance of ipsilateral mediastinal lymph node metastasis in non small cell carcinoma of the lung: a commentary. J Thorac Cardiovasc Surg 1990;99:48 53.