The prognostic relevance of classifying neuroendocrine

Transcription

1 Large Cell Neuroendocrine Carcinoma and Large Cell Carcinomas With Neuroendocrine Morphology of the Lung: Prognosis After Complete Resection and Systematic Nodal Dissection Joseph Zacharias, FRCS (CTh), Andrew G. Nicholson, MRCPATH, George P. Ladas, MD, and Peter Goldstraw, FRCS Department of Thoracic Surgery and Histopathology, Royal Brompton Hospital, London, United Kingdom Background. Large cell neuroendocrine carcinoma (LCNEC) and large cell carcinoma with neuroendocrine morphology of the lung are both currently classified as subtypes of large cell carcinomas according to the World Health Organization IASLC classification system for lung and pleural tumors. Prognosis is reported as similar to that of small cell carcinomas. There is no consensus on management of this subset and adjuvant chemotherapy is recommended by some for early stage LCNEC to impact long-term prognosis. We retrospectively reviewed a cohort of patients at our institution who had this type of tumor to determine factors that might influence survival. Methods. Twenty-one cases of LCNEC and large cell carcinoma with neuroendocrine morphology were identified in the files of the Royal Brompton Hospital between 1986 and All patient data were reviewed, and complete follow-up was achieved with 20 of these patients. Results. Of the 21 patients identified, 20 underwent Accepted for publication July 26, Address reprint requests to Mr Goldstraw, Department of Thoracic Surgery, Royal Brompton Hospital, Sydney St, London SW3 6NP, UK; p.goldstraw@rbh.nthames.nhs.uk. resection with systematic nodal dissection in 18. There was no in-hospital mortality. Of those patients fully staged by systematic nodal dissection, 9 were stage I, 5 were stage II and 4 were stage III. Median follow-up was 25 months (range, 2 to 120 months). At the time of review, 11 patients were alive and free of disease. One patient was alive and free of disease when lost to follow-up. Nine patients had died, 7 related and 2 unrelated to disease. The 5-year actuarial survival for the entire group was 47%. The actuarial survival of accurately staged, stage I patients at 5 years was 88%. The actuarial survival of patients in stage II and III was 28% at 5 years. Conclusions. LCNEC and large cell carcinoma with neuroendocrine morphology are aggressive tumors, but patients with completely resected disease after systematic nodal dissection have a better prognosis than previously described. Patients with more advanced disease have a poor prognosis. (Ann Thorac Surg 2003;75:348 52) 2003 by The Society of Thoracic Surgeons The prognostic relevance of classifying neuroendocrine tumors in the lung has changed significantly since Travis and colleagues [1] introduced the concept of large cell neuroendocrine carcinoma (LCNEC) as a subgroup of neuroendocrine tumors. LCNEC, being distinguishable from typical carcinoids, atypical carcinoidsm and small cell carcinoma. This is supported by later studies [2, 3], confirming that prognosis varies between the different subgroups, and also that such a classification system is reproducible [4]. However in 1999 the World Health Organization IASLC classification of lung and pleural tumors classified large cell neuroendocrine carcinoma on morphologic grounds as a type of large cell carcinoma, recognizing LCNECs as tumors with both neuroendocrine morphology and immunohistochemical evidence of neuroendocrine differentiation. Those tumors with neuroendocrine morphology that lack immunohistochemical evidence of neuroendocrine differentiation are termed large cell carcinomas with neuroendocrine morphology (LCCNM) [5]. Prognostic data on LCNECs are limited and most series are small [1, 2, 6], despite collections of cases from multiple institutions [3] with even less data on the clinical relevance of subclassification as LCCNM [7]. Use of the term large cell carcinoma with neuroendocrine features has been suggested to encompass both these tumor types as clinical behavior is similar [7]. The purpose of this study was to report our experience of these tumors, in particular to assess the importance of such classification in relation to prognosis. Material and Methods A retrospective review of resection specimens from nonsmall cell tumors with neuroendocrine morphology was undertaken from the case records of the Royal Brompton Hospital between 1986 and Twenty-five cases showed morphologic evidence of neuroendocrine differentiation as described by Travis and colleagues [1]. Of these, 9 had previously been classified as atypical carcinoids, but a mitotic count of greater than 10 per 2 mm 2 led to reclassification as either LCNEC or LCCNM. Three cases were rejected because they were mixed tumors 2003 by The Society of Thoracic Surgeons /03/$30.00 Published by Elsevier Science Inc PII S (02)

2 Ann Thorac Surg ZACHARIAS ET AL 2003;75: LCNEC AND LCCNM OF THE LUNG Table 1. Characteristics of the Patients With Large Cell Neuroendocrine Carcinoma and Large Cell Carcinoma of Neuroendocrine Morphology M 61 Left lower lobectomy T3N0M0 DWD 1.5 a LCNEC M 79 Left lower lobectomy T2NxM0 DFO 2 LCNEC F 64 Left upper lobectomy & rib resection T3N2M0 DWD 2 LCNEC M 46 Right upper lobectomy & rib resection T3N0M0 DWD 2.2 LCNEC M 54 Right pneumonectomy T3N0M0 ADF 12 LCNEC M 69 No resection T4N1M0 DWD 14 LCCNM M 67 Right lower lobectomy T1NxM0 DWD 15 LCNEC M 70 Left lower lobectomy T2N0M0 ADF 18 LCCNM F 65 Left wedge excision of tumour T4N0M0 DWD 20 LCNEC M 57 Right upper lobectomy T2N0M0 DWD 22 LCCNM M 75 Left pneumonectomy T3N1M0 DFO 25 LCNEC F 45 Left segmentectomy of lingula T2N0M0 ADF 26 LCNEC M 62 Left lower lobectomy T2N1M0 ADF 32 LCCNM F 67 Right upper lobectomy T2N0M0 ADF 37 LCNEC M 54 Left upper lobectomy T2N0M0 ADF b 37 LCNEC M 76 Left upper lobectomy T2N0M0 ADF 41 LCNEC F 62 Left upper lobectomy T2N2M0 ADF 49 LCNEC M 59 Right upper lobectomy T2N0M0 ADF 60 LCNEC F 71 Right upper bilobectomy T3N0M0 ADF 62 LCCNM M 49 Right lower lobectomy T1N0M0 ADF 120 LCCNM F 66 Right lower lobectomy T2N0M0 ADF 120 LCNEC 349 GENERAL THORACIC a Months. b Lost to follow-up. ADF alive disease free; DFO died from other disease; DWD died with disease; F female; LCCNM large cell carcinoma of neuroendocrine morphology; LCNEC large cell neuroendocrine carcinoma; M male. with the second component as small cell carcinoma. A fourth case was rejected because no material was available for immunohistochemical analysis. An immunohistochemical panel comprising CD56 (dilution 1:100 Novocastra; UK), chromogranin (dilution 1/20; Dako; UK), and synaptophysin (dilution 1/100; Dako; UK) was subsequently applied to look for immunohistochemical evidence of neuroendocrine differentiation with subsequent subdivision into either LCNEC (neuroendocrine marker positive) or LCCNM (neuroendocrine marker negative). The notes of these patients were reviewed and clinical data comprising sex, age, symptoms at presentation, preoperative staging investigations, type of resection, postoperative staging, morbidity and mortality, and follow-up details were collected. Statistical Analysis Statistical analysis was carried out on the follow-up details using an SPSS package (SPSS Inc, Chicago, IL). We used the Kaplan-Meier survival curves for calculating actuarial survival of the accurately staged patients (n 18 patients). We also compared actuarial survival between patients (n 9 patients) proven to be in Stage I by systematic nodal dissection (SND) and patients (n 11 patients) not in stage I. Actuarial survival was also compared between the LCNEC group (15 patients) and those classified as LCCNM (6 patients). We used the log rank test to see if the survival was significantly different between the groups. Results Pathology The pathologic data are presented in Table 1. Of the 21 tumors, 15 patients showed immunohistochemical evidence of neuroendocrine differentiation and were classified as LCNEC, with the remainder classified as LCCNM. Mitotic counts ranged form 21 to 140 (average, 66) per 2 mm 2. Extensive necrosis was present in 20 of 21 patients. The architecture was either solid islands of tumor with peripheral palisading or cribriform. Clinical Data Of the 21 identified patients (7 females, 14 males), the average age at presentation was 62.7 years. Of the resected tumors, 9 were on the right side and 11 on the left side. At time of presentation, 5 patients were asymptomatic and tumors were picked up incidentally on a chest x-ray. In the other 16 patients, symptoms at presentation included chest pain in 6, hemoptysis in 3, nonspecific flu-like symptoms in 2, dyspnea in 2, night sweats in 1, a persistent cough in 1, and carcinoid syndrome in 1. Of the 6 patients who presented with chest pain as a feature, only 2 had chest wall involvement on subsequent investigations. The diagnosis of nonsmall cell lung cancer was made at bronchoscopy in 7 patients before referral. In none of the cases was LCNEC diagnosed before referral. A computed tomography guided biopsy was carried out in 4 patients. The diagnosis of nonsmall cell lung cancer was made in 3 patients, but biopsy failed to show malignant cells in 1 patient. Mediastinoscopy was carried out in 7 patients with mediastinal glands on computed tomographic scans and all were negative for malignant cells. All 21 patients proceeded to thoracotomy. One patient was found to be unresectable at the time of operation because of the involvement of mediastinal structures. Of the remaining 20 patients, 2 underwent pneumonec-

3 350 ZACHARIAS ET AL Ann Thorac Surg LCNEC AND LCCNM OF THE LUNG 2003;75: Fig 1. Kaplan-Meier survival estimates for all 18 accurately staged patients with large cell neuroendocrine carcinoma. tomy, 16 had lobectomy, 1 had a segmental resection, and 1 had a wedge resection. Eighteen patients had systematic nodal dissection at the time of resection. Two patients were operated on before the practice of regular systematic nodal dissection, although they did not have lymph node sampling. Data are summarized in Table 1. Of the 21 patients identified, 18 patients were considered to have been accurately staged by their preoperative and intraoperative evaluation. The TNM details are given in Table 1. Stage distribution using the 1997 TNM system for classification of lung cancers [8] showed 1 patient at stage Ia, 8 patients at stage Ib, 0 at stage IIa, 5 at stage IIb, 2 at stage IIIa, and 2 at stage IIIb. Two patients who did not have any lymph node sampling were considered to be inaccurately staged and were not included in the initial survival calculations. Fig 2. Disease-free survival for 9 early stage patients (stage Ia & Ib) with large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology compared with 11 patients with non-early stage disease (log rank test significant; p ). (SND systematic nodal dissection.) free at 36 months. Two patients died from other causes. Recurrence was seen in 7 patients. All of these patients had died from recurrence of disease at the time of review. In the group that presented with recurrence, no patients were deemed fit for a repeat operation because there was mediastinal involvement (3 of 7) or brain involvement (4 of 7). Four patients with recurrence were treated with multiagent chemotherapy using a combination of agents including Adriamycin, Etopside Ifosphamide, Doxorubicin, Carboplatin, VP16, Mitomycin-C, Vinblastine, and Cisplatin in cycles. In the presence of brain metastasis, 2 patients had cranial irradiation. Follow-Up Data The follow-up for this group ranged from 1.5 months to 120 months. The median follow-up time was 25 months. One patient was lost to follow-up at 37 months, at which time he was alive and disease free. Of the 20 with complete follow-ups 11 were alive and disease free, 7 were dead from their disease, and 2 had died from other causes. The Kaplan-Meier survival curve for the 18 accurately staged patients with complete resection was 47% at 5 years as shown in Figure 1. Patients not having SND were included in the second calculation that looked at a difference between recurrence-free survival in patients with accurate early stage disease (stage Ia and Ib; 9 of 20) compared with patients in other stages (11 of 20), and found a significant difference in survival (Fig 2). Actuarial survival of patients with pathologic diagnosis of LCCNM and LCNEC were found to be similar at 63% and 52%, respectively (Fig 3). The log rank test was used to look for significance between the groups. One patient received elective Navelbine and Cisplatin chemotherapy postoperatively and was alive and disease Fig 3. Kaplan-Meier survival estimates for all patients comparing survival between large cell neuroendocrine carcinoma (LCNEC) (15 patients) and large cell carcinoma with neuroendocrine morphology (LCCNM) (6 patients). (Log rank test not significant; p 0.54.)

4 Ann Thorac Surg ZACHARIAS ET AL 2003;75: LCNEC AND LCCNM OF THE LUNG Comment 351 The World Health Organization IASLC (International Association for Staging of Lung Cancer) classification of lung and pleural tumors have recently classified LCNEC as a subgroup of large cell carcinoma on the basis of morphologic and immunohistochemical evidence of neuroendocrine differentiation. Those showing morphologic evidence of neuroendocrine differentiation, but with negative immunohistochemistry are classified as LCCNM [4]. However, the prognostic significance of recognizing subgroups of large cell carcinomas with neuroendocrine features has yet to be proven with some groups showing a prognosis similar to that of other nonsmall cell lung cancers [9], whereas others findings showed a worse prognosis in those with neuroendocrine features [10]. Most series relate to LCNEC series, which show the prognosis lies between that of an atypical carcinoid and small cell carcinoma, although much closer to that of small cell lung cancer (SCLC) [2, 3, 5, 6]. In terms of the pathology, the histologic findings are similar to those seen in other series. Our results support the findings of Iyoda and colleagues [7] who found no difference in the clinical behavior between LCCNM and LCNEC (Fig 3). However, we also found that the accurately staged patients in stage 1 (9 of 18) had a 3-year survival of 88% (n 6), maintained at 5 years (n 4). In the group that did not have SND or complete resection in an advanced stage, the 3-year survival rate was 28% (Fig 2). These data contrast with other published series (eg, Dressler and colleagues showed a poor prognosis similar to that of small cell lung cancers with an overall 5-year survival of 18% [2]). In this study, patients with stage I disease showed 5-year survival of 20% (n 17), whereas stage I survival was 33% (n 13) in a Spanish multicenter study [3]. For patients with stage II or III disease in the Spanish study, none survived longer than 18 months after resection, whereas in our study those patients with advanced disease had a 3-year survival of 28% (Fig 2). The markedly better survival in our series may be caused by several factors. First, it is possible that stage migration accounted for the markedly superior survival results in our series. SND was performed in 18 of 21 patients, and it is known to provide improved staging at thoracotomy. The IASLC recommends SND for all cancer resections [11]. There is no mention of the exact extent of mediastinal nodal dissection in the other series that have published long-term follow-up in LCNEC [2 4, 6, 10]. Therefore potential understaging in the absence of SND may have an effect on prognostic data. Although not investigated in a randomized fashion, evidence from other studies of neuroendocrine lung cancers suggests SND to be of benefit. Martini and colleagues [12] showed excellent results after complete resection and SND with bronchial carcinoids with lymph node involvement. Shepherd and colleagues [13] also found that micrometastases are very likely in SCLCs, as clinical TNM stage correlated with final pathologic stage in less than 50% of patients if SND was carried out, and there was an apparent survival advantage in the accurately staged group. Finally, recent evidence suggests a possible independent survival benefit from SND in other forms of nonsmall cell lung cancer with lymph node metastasis [14]. Second, survival in our series may have been influenced by the extent of complete anatomical resections. In previously published trials the extent of wedge resections in the LCNEC group was up to 30% [2, 4], and this may have had an adverse effect on outcome. Two patients in our study were excluded from the initial survival calculations as no lymph node sampling was undertaken. Both these patients were considered preoperatively to be stage I but had died within 18 months of the resection from recurrence of disease. This may reflect other comorbid factors, which mitigated against SND and the extent of resection in this small subgroup. Despite this we believe that the outcome is markedly better in the group with anatomical resection and SND justifies this approach. Besides, our series had far fewer patients with locally advanced stage when compared with the group presented by Rusch and colleagues [6]. In their study, 52% of patients were in stage IIIa, and this would have skewed the prognosis. We had 50% of our patients in stage I, which was comparable with other published trials [2, 3]. A 25% pickup of asymptomatic patients could have helped the high incidence of early stage tumors in our series. Because of the neuroendocrine phenotype and the aggressive nature of the disease, postoperative chemotherapy or radiotherapy has been used in all studies published about LCNEC. The group receiving adjuvant chemotherapy or radiotherapy ranged from 40% to 60% in the published trials. There did not appear to be a survival advantage in the adjuvant therapy group in any of the trials [2 4, 6]. No standardized treatment protocol was followed. There is a suspicion that these tumors are chemotherapy resistant. Lai and colleagues [15] did find a majority of lung cancers with neuroendocrine markers to demonstrate the multi-drug resistance gene (MDR1) expression, which suggests the presence of chemotherapy resistance to the tumor. Due to the lack of clear advantages of adjuvant therapy postoperatively, a majority of our patients were not considered for postoperative chemotherapy or radiotherapy. In our series only one patient in stage I received elective postoperative chemotherapy and he was alive and well at 37 months. A recent publication by Iyoda and colleagues [16] showed a comparable survival of Stage I patients with LCNEC at 62% more than 5 years. They found that in a subgroup of 5 patients who had stage I disease and adjuvant chemotherapy, survival was 100% at 5 years. In patients not in stage I, survival with and without adjuvant chemotherapy was comparable at less than 20% at more than 5 years. In view of the small numbers in all the published groups, the question of adjuvant chemotherapy has not been adequately answered. On the evidence of our series we believe that complete resection and SND is the best treatment for early stage LCNEC and LCCNM. In conclusion, this series shows that LCNEC and LCCNM are rare and aggressive tumors. Complete resection with SND is associated with a 47% survival at 5 GENERAL THORACIC

5 352 ZACHARIAS ET AL Ann Thorac Surg LCNEC AND LCCNM OF THE LUNG 2003;75: years for the whole group and an 88% 5-year survival for stage I disease. This is better than what has been reported in previously published series. In the reviewed literature there seems to be conflicting evidence for adjuvant therapy in this group of patients. A multicenter prospective randomized control trial is needed to answer the role of adjuvant chemotherapy in early stage LCNEC and LCCNM. Patients with more advanced disease and those in whom intrathoracic nodal staging has been less reliable have a less favorable prognosis with a 5 year survival of 28%. We would like to thank Dr Steven Bourke and Therese Small for help with the statistical analysis of the results. References 1. Travis WD, Linnoila RI, Tsokos MG, et al. Neuroendocrine tumours of the lung with proposed criteria for large cell neuroendocrine carcinoma. An ultrastructural, immunohistochemical, and flow cytometric study of 35 cases. Am J Surg Pathol 1991;15(6): Dresler CM, Ritter JH, Patterson GA, et al. Clinicopathologic analysis of 40 patients with large cell neuroendocrine carcinoma of the lung. Ann Thorac Surg 1997;63: Garcia-Yuste M, Matilla JM, et al. Prognostic factors in neuroendocrine lung tumors: a Spanish multicentre study. Ann Thorac Surg 2000;70: Travis WD, Gal AA, Colby TV, Klimstra DS, Falk R, Koss MN. Reproducibility of neuroendocrine lung tumor classification. Hum Pathol 1998;29(3): Travis WD, Corrin B, Shimosato Y, Brambilla E. The histological typing of lung and pleural tumors. WHO/IASLC classification of lung and pleural tumours, 3rd ed. Berlin: Springer Verlag, Rusch VW, Klimstra DS, Venkatraman ES. Molecular markers help characterize neuroendocrine lung tumors. Ann Thorac Surg 1996;62: Iyoda A, Hiroshima K, Toyozaki T, Haga Y, Fujisawa T, Ohwada H. Clinical characterization of pulmonary large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology. Cancer 2001;91(11): Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111: Linnoila RI, Piantadosi S, Ruckdeschel JC. Impact of neuroendocrine differentiation in non-small cell lung cancer. The LCSG experience. Chest 1994;106(Suppl):367S 71S. 10. Cooper WA, Thourani VH, Gal AA, et al. The surgical spectrum of pulmonary neuroendocrine neoplasms. Chest 2001;119: Goldstraw P. Report on the International Workshop on Intrathoracic Staging. London, October Lung Cancer 1997;18: Martini N, Zaman MB, Bains MS, et al. Treatment and prognosis in bronchial carcinoids involving regional lymph nodes. J Thorac Cardiovasc Surg 1994;107: Shepherd FA, et al. Surgical treatment for limited small cell lung cancer. J Thorac Cardiovasc Surg 1991;101: Izbicki JR, Passlick B, Pantel K, et al. Effectiveness of radical systematic mediastinal lymphadenectomy in patients with resectable non-small cell lung cancer: results of a prospective randomized trial. Ann Surg 1998;227(1): Lai S-L, Goldstein LJ, Gottesman MM, et al. MDR1 gene expression in lung cancer. J Natl Cancer Inst 1989;81: Iyoda A, Hiroshima K, Toyozaki T, et al. Adjuvant chemotherapy for large cell carcinoma with neuroendocrine features. Cancer 2001;92: Internet Discussion Forum Each month, we select an article from The Annals of Thoracic Surgery for discussion within the Surgeon s Forum of the CTSNet Discussion Forum Section. The articles chosen rotate among the six dilemma topics covered under the Surgeon s Forum, which include: General Thoracic Surgery, Adult Cardiac Surgery, Pediatric Cardiac Surgery, Cardiac Transplantation, Lung Transplantation, and Aortic and Vascular Surgery. Once the article selected for discussion is published in the online version of The Annals, we will post a notice on the CTSNet home page ( with a FREE LINK to the full-text article. Readers wishing to comment can post their own commentary in the discussion forum for that article, which will be informally moderated by The Annals Internet Editor. We encourage all surgeons to participate in this interesting exchange and to avail themselves of the other valuable features of the CTSNet Discussion Forum and Web site. For February, the article chosen for discussion under the Adult Cardiac Surgery Dilemma Section of the Discussion Forum is: Stroke After Cardiac Surgery: A Risk Factor Analysis of 16,184 Consecutive Adult Patients Jan Bucerius, MD, Jan F. Gummert, MD, PhD, Michael A. Borger, MD, PhD, Thomas Walther, MD, PhD, Nicolas Doll, MD, Jörg F. Onnasch, MD, Sebastian Metz, MD, Volkmar Falk, MD, PhD, and Friedrich W. Mohr, MD, PhD Tom R. Karl, MD The Annals Internet Editor UCSF Children s Hospital Pediatric Cardiac Surgical Unit 505 Parnassus Ave, Room S-549 San Francisco, CA Phone: (415) Fax: (212) karlt@surgery.ucsf.edu 2003 by The Society of Thoracic Surgeons Ann Thorac Surg 2003;75: /03/$30.00 Published by Elsevier Science Inc