Endocrine DOI 10.1007/s12020-015-0578-x
VIEWPOINT
Classification of lung neuroendocrine tumors: lights and shadows Marco Volante1 • Gaia Gatti1 • Mauro Papotti1
Received: 27 November 2014 / Accepted: 11 March 2015 Ó Springer Science+Business Media New York 2015
Abstract Neuroendocrine tumors of the lung are classified into low-grade typical and intermediate-grade atypical carcinoids, and high-grade poorly differentiated neuroendocrine carcinomas of the large and small cell types. This scheme is strongly predictive of patients’ prognosis but relies on few and scarcely reproducible pathological parameters (namely mitotic count and assessment of the presence of necrosis), which have been demonstrated to affect the inter-observer agreement of the classification. Moreover, tumor and nodal staging schemes are not specific for lung carcinoids, at variance with neuroendocrine tumors of the gastro-entero-pancreatic system, despite these tumors have specific features that strongly differ from conventional lung cancer. Finally, there is no grading for lung neuroendocrine neoplasms and prognostication, as well as the definition of treatment modalities and clinical strategies, which are based on tumor histotypes, only. However, literature data indicate that the evaluation of Ki-67 proliferation index may be a reliable and useful tool to determine the biological and clinical behavior of neuroendocrine tumors, with special reference to carcinoids, both in pre-operative and surgical samples. Keywords Lung Carcinoid Neuroendocrine Carcinoma Classification
& Marco Volante [email protected] 1
Department of Oncology, University of Turin at San Luigi Hospital, Regione Gonzole 10, Orbassano, 10043 Turin, Italy
Lung neuroendocrine tumor diagnosis and classification: what is ‘‘under the sun’’ Neuroendocrine tumors of the lung are a heterogeneous group of neoplasms which includes four histological types representative of a spectrum of tumor differentiation and having considerably different clinical and biological behavior. On the one side of this spectrum, lung carcinoids are separated into low-grade typical and intermediate-grade atypical carcinoids, while high-grade poorly differentiated lung neuroendocrine neoplasms include both large and small cell neuroendocrine carcinomas, on the other. All four histological types are characterized by different cytological and architectural findings. However, according to the most recent World Health Organization classification, year 2004 [1], the differential diagnosis among them relies on the assessment of two parameters, only, namely the presence/absence of necrosis and mitotic index per 2 mm2, with a cut-off of \2 to segregate typical and atypical carcinoids and of C10 for high-grade poorly differentiated neuroendocrine carcinomas. Cell size, nuclear morphology, and architecture are additional characteristics useful to distinguish large from small cell neuroendocrine carcinomas. In 2015, the 7th edition of the WHO classification of Tumors of the Lung, Pleura, Thymus, and Heart is expected, based on the proposal of the International Association for the Study of Lung Cancer. A major change in the new classification will concern large cell neuroendocrine carcinoma, which will be moved into the group of neuroendocrine neoplasms and not considered, as in the current classification, a variant of large cell carcinomas. However, no other changes—with special reference to carcinoid
123
Endocrine
classification—will be present, based on the proven highly prognostic performance of the currently employed system. The use of immunohistochemistry is restricted to general neuroendocrine markers, such as chromogranin A and synaptophysin, which are required, especially in small biopsies, to prove the neuroendocrine nature of these neoplasms. At variance with the gastro-entero-pancreatic system, Ki-67 labeling index is not mandatory in the diagnostic flow of lung neuroendocrine tumors, but may be useful in distinguishing carcinoids from high-grade carcinomas, especially in small biopsies [2]. There is no specific staging for lung neuroendocrine neoplasms, but they are staged according to the AJCC/ IASLC TNM staging system following the same recommendations formulated for non-neuroendocrine lung cancer [3]. Based on the aforementioned, the classification system and the diagnostic workflow of lung neuroendocrine neoplasms are apparently easy and well settled. However, several gray zones and unresolved issues still exist in lung tumor classification and pathological characterization, which will be briefly outlined here below (Table 1).
Unclear clinical significance of pre-neoplastic lesions Pre-neoplastic neuroendocrine lesions include neuroendocrine cell hyperplasia and tumorlets. Neuroendocrine cell hyperplasia is defined by a linear or nodular proliferation of neuroendocrine cells confined within the bronchial basal membrane, whereas tumorlets may show signs of stromal invasion. In both cases, however, the largest dimension should not exceed 5 mm, which is the unique criterion to differentiate these lesions from carcinoids. Tumorlets and neuroendocrine cell hyperplasia may be not infrequently observed in lung parenchyma at the periphery
of carcinoids [4] or in patients affected by chronic diseases leading to respiratory failure. Moreover, they may be the hallmark of a rare idiopathic condition, the so-called diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH) [5]. Irrespective of the clinical situation, the pathogenetic link between neuroendocrine cell hyperplasia/tumorlets and carcinoids, their distinction (based currently only on size of the lesion) and the real clinical significance as pre-invasive lesions of the former, is questioned. In fact, no molecular nor clinical data are available to demonstrate or disprove a true progression from neuroendocrine cell hyperplasia/tumorlets to carcinoids, and rare cases of tumorlets with nodal metastases have been described [6].
Low reproducibility of lung neuroendocrine neoplasm classification As already stated above, two parameters are fundamental for lung neuroendocrine neoplasm classification, namely mitotic index and necrosis. However, the assessment of these two parameters may be affected by high subjectivity. In a study on 110 lung neuroendocrine tumors, up to 44 % were re-classified after revision by a panel of pathologist experienced in the field of lung cancer [7]. More recently, carcinoid classification was demonstrated to have a fair inter-observer agreement (Kappa: 0.32), even between experienced pathologists [8]. The low number of mitoses in a rather large microscopic area, the fact that mitoses are not homogeneously distributed over the tissue section and that they may be confused with apoptotic cells, crushed cells, and granulocytes are proposed to explain this low reproducibility. In addition, punctate necrosis may be difficult to identify. The distinction between large and small cell highgrade neuroendocrine carcinomas also shows a low reproducibility, due to problems related to sampling procedure and the resulting presence of artifacts, and to the fact that
Table 1 Lights and shadows in lung neuroendocrine tumor classification Cons
Item
Pros
Histological classification is not assessable in small biopsies/cytological samples
Current four-ties classification scheme
Worldwide accepted
Lung cancer overall staging is prognostic also for carcinoids A carcinoid-specific staging might be a confounding factor
Carcinoid-specific staging
Mean carcinoid size is lower than non-neuroendocrine lung cancer (inappropriate size cut-offs)
Histological classification is the most relevant prognostic factor at multivariate analysis in most studies
Carcinoid-specific grading
Not fully reproducible (low inter-observer agreement for mitotic index and necrosis)
Familiar to clinicians Relevant for prognosis
pTN stage is not prognostic for typical carcinoids
Ki-67 cut-off values are not uniform nor consolidated
123
Proliferation index is more easy to evaluate in small biopsies Homogenization with neuroendocrine tumors of the GEP area is desirable
Endocrine
some tumors present transitional cell characteristics between small and large cell neuroendocrine carcinomas [9, 10]. The use of mitosis-specific antibodies, such as phosphohistone H3 [11], has been proposed to improve the mitotic index determination indistinguishing typical from atypical carcinoids. Moreover, Ki-67 proliferation index has been shown to outperform mitotic count in terms of reproducibility of carcinoid classification [12].
Lack of a carcinoid-specific staging Although lung carcinoids are staged as non-neuroendocrine lung cancer and staging overall has been shown to be prognostic in lung neuroendocrine neoplasms [3], some carcinoid-specific features suggest that this classification may be not completely representative of the real clinical situation. In fact, TNM staging cut-off values of 3, 5, and 7 cm are poorly representative of carcinoids, since both typical and atypical carcinoids are often less than 3 cm in largest diameter. This is also clear from the original paper by Travis and co-workers [3] which showed that the number of carcinoid cases in the SEER database having a diameter below/equal to or above five cm was 1223 and 41, respectively. Moreover, literature data—which confirmed a prognostic significance of the TNM classification—also showed an overlapping survival of carcinoid patients in stages I and II [13, 14]. In fact, in both papers, either in the complete group of carcinoids [13] or in typical and atypical ones considered separately [14], a significantly different prognosis was observed only for patients in stage III (and IV) as compared to stages I and II, but not between stages I and II. Moreover, in two other papers, the prognostic impact of the TNM staging was lost at multivariate analysis which retains histological type as the major prognostic indicator [4, 15]. The impact of nodal involvement as a prognostic factor in carcinoids is also controversial. In fact, the presence of lymph node metastases has been demonstrated to be and independent negative prognostic factor by some Authors [13] but not by others [4, 18]. Moreover, it is a strong negative prognostic indicator in atypical carcinoids [16, 17] but no data have been generated specifically in the group of typical ones, which indeed in some series present a survival rate significantly higher that atypical ones despite a similar rate of cases with positive nodal status [4, 18].
Lack of grading for lung neuroendocrine neoplasms In pulmonary neuroendocrine neoplasms, no formal grading system has been created, as tumor grade is actually part of the histological classification. Morphology is undoubtedly able
to predict the clinical behavior of tumors. However, grading systems have been proposed in the literature to integrate histotype and to implement the prediction of patients’ outcome. The potential impact of such an approach is limited to carcinoids, since both small cell and large cell neuroendocrine carcinomas typically die quickly and they are considered ‘‘high grade’’ tumors by definition. Similar to grading of neuroendocrine neoplasms of the gastro-enteropancreatic tract, a major role for these systems is played by Ki-67 proliferation index, which has been demonstrated to be a strong prognostic indicator in lung neuroendocrine neoplasms [12, 19]. Moreover, the homogenization of a lung carcinoid grading approach with that of neuroendocrine tumors of the gastro-entero-pancreatic system would be also a benefit for designing novel clinical trials, based on the fact that lung carcinoids are usually randomly included in trials recruiting other types of neuroendocrine cancers and therefore not fully comparable. However, a definite grading for lung carcinoids is still a matter of debate. In fact, different prognostic cut-offs of Ki-67 have been proposed to segregate G1 and G2 tumors, from 3 % (as in the case of neuroendocrine tumors of the gastrointestinal tract) [20], to 4 % [21, 22], to 5 % [8]. Moreover, Ki-67 proliferation index has been proposed either alone or integrated with mitotic count and necrosis, as recently proposed by the group of Rindi and coworkers [21]. In addition, although not extensively investigated in lung neuroendocrine neoplasms so far, the assessment of Ki-67 is affected by either technical issues or poor reproducibility of scoring [19]. Finally, the real prognostic role of Ki-67 proliferation index, independently from carcinoid histotypes, has been called into question by some other Authors [23].
Doubtful clinical meaning of additional pathologic parameters Apart from morphological (or phenotypical) parameters included in classification or proposed grading systems, some other pathological characteristics have been analyzed as potential predictors of biological and clinical behavior in lung neuroendocrine neoplasms, with special reference to carcinoids. In a study by Tsuta and co-workers, several pathological parameters were associated to either the presence of lymph node metastases (blood and lymphovascular invasion, nuclear features) or survival (pleural invasion and active fibroblast proliferation) [24]. Vascular invasion has been associated to worse prognosis in another single study [25]. Moreover, invasion of the lung parenchyma has been also associated to atypical carcinoid histotypes and poor prognosis [26]. However, most of the features above have been investigated in single studies lacking multivariate statistical analysis, and are associated
123
Endocrine
to the atypical carcinoid histotypes; further studies are therefore needed to validate these findings and possibly insert these parameters in future classifications/diagnostic grids for pathology reporting of lung neuroendocrine neoplasms. Acknowledgments This study was partially supported by a grant from the Italian Association for Cancer Research (AIRC) (grant nr. IG 13567 to MV). Conflict of interest All Authors declare the absence of any potential conflict of interest.
References 1. W.D. Travis, E. Brambilla, H.K. Muller-Hermelink, C.C. Harris, Pathology and genetics of tumours of the lung, pleura, thymus and heart (IARC Press, Lyon, 2004) 2. G. Pelosi, J. Rodriguez, G. Viale, J. Rosai, Typical and atypical pulmonary carcinoid tumor overdiagnosed as small-cell carcinoma on biopsy specimens: a major pitfall in the management of lung cancer patients. Am. J. Surg. Pathol. 29, 179–187 (2005) 3. W.D. Travis, D.J. Giroux, K. Chansky, J. Crowley, H. Asamura, E. Brambilla, J. Jett, C. Kennedy, R. Rami-Porta, V.W. Rusch, P. Goldstraw, International Staging Committee and Participating Institutions: the IASLC Lung Cancer Staging Project: proposals for the inclusion of broncho-pulmonary carcinoid tumors in the forthcoming (seventh) edition of the TNM Classification for Lung Cancer. J. Thorac. Oncol. 3, 1213–1223 (2008) 4. P. Ferolla, N. Daddi, M. Urbani, A. Semeraro, R. Ribacchi, P. Giovenali, S. Ascani, V. De Angelis, L. Crino`, F. Puma, G. Daddi, Regional multidisciplinary group for the diagnosis and treatment of neuroendocrine tumors, CRO, Umbria region cancer network, Italy: tumorlets, multicentric carcinoids, lymph-nodal metastases, and long-term behavior in bronchial carcinoids. J. Thorac. Oncol. 4, 383–387 (2009) 5. A.A. Nassar, D.E. Jaroszewski, R.A. Helmers, T.V. Colby, B.M. Patel, F. Mookadam, Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia: a systematic overview. Am. J. Respir. Crit. Care Med. 184, 8–16 (2011) 6. S.M. Liu, H.H. Wu, C.J. Wu, C.L. Kuo, L.R. Mo, Adrenocorticotropin-producing pulmonary tumorlets with lymph node metastasis. Pathol. Int. 53, 883–886 (2003) 7. B.G. Skov, M. Krasnik, S. Lantuejoul, T. Skov, E. Brambilla, Reclassification of neuroendocrine tumors improves the separation of carcinoids and the prediction of survival. J. Thorac. Oncol. 3, 1410–1415 (2008) 8. D.R. Swarts, R.J. van Suylen, M.A. den Bakker, M.F. van Oosterhout, F.B. Thunnissen, M. Volante, A.M. Dingemans, M.R. Scheltinga, G.P. Bootsma, H.M. Pouwels, B.E. van den Borne, F.C. Ramaekers, E.J. Speel, Interobserver variability for the WHO classification of pulmonary carcinoids. Am. J. Surg. Pathol. 38, 1429–1436 (2014) 9. M.A. den Bakker, S. Willemsen, K. Gru¨nberg, L.A. Noorduijn, M.F. van Oosterhout, R.J. van Suylen, W. Timens, B. Vrugt, A. Wiersma-van Tilburg, F.B. Thunnissen, Small cell carcinoma of the lung and large cell neuroendocrine carcinoma interobserver variability. Histopathology 56, 356–363 (2010) 10. A.M. Marchevsky, A.A. Gal, S. Shah, M.N. Koss, Morphometry confirms the presence of considerable nuclear size overlap between ‘‘small cells’’ and ‘‘large cells’’ in high-grade pulmonary neuroendocrine neoplasms. Am. J. Clin. Pathol. 116, 466–472 (2001)
123
11. K. Tsuta, D.C. Liu, N. Kalhor, I.I. Wistuba, C.A. Moran, Using the mitosis-specific marker anti-phosphohistone H3 to assess mitosis in pulmonary neuroendocrine carcinomas. Am. J. Clin. Pathol. 136, 252–259 (2011) 12. A. Warth, L. Fink, A. Fisseler-Eckhoff, D. Jonigk, M. Keller, G. Ott, R.J. Rieker, P. Sinn, S. So¨der, A. Soltermann, K. Willenbrock, W. Weichert, Pulmonary Pathology Working Group of the German Society of Pathology: interobserver agreement of proliferation index (Ki-67) outperforms mitotic count in pulmonary carcinoids. Virchows Arch. 462, 507–513 (2013) 13. P.L. Filosso, A. Oliaro, E. Ruffini, G. Bora, P. Lyberis, S. Asioli, L. Delsedime, A. Sandri, F. Guerrera, Outcome and prognostic factors in bronchial carcinoids: a single-center experience. J. Thorac. Oncol. 8, 1282–1288 (2013) 14. E. Aydin, U. Yazici, M. Gulgosteren, Y. Agackiran, S. Kaya, E. Gulhan, I. Tastepe, N. Karaoglanoglu, Long-term outcomes and prognostic factors of patients with surgically treated pulmonary carcinoid: our institutional experience with 104 patients. Eur. J. Cardiothorac. Surg. 39, 549–554 (2011) 15. C. Cao, T.D. Yan, C. Kennedy, N. Hendel, P.G. Bannon, B.C. McCaughan, Bronchopulmonary carcinoid tumors: longterm outcomes after resection. Ann. Thorac. Surg. 91, 339–343 (2011) 16. M.A. Can˜izares, J.M. Matilla, A. Cueto, J. Algar, I. Muguruza, N. Moreno-Mata, R. Moreno-Balsalobre, R. Guijarro, R. Arrabal, E. Garcia-Fontan, A. Gonzalez-Pin˜eiro, M. Garcia-Yuste, EMETNE-SEPAR Members: Atypical carcinoid tumours of the lung: prognostic factors and patterns of recurrence. Thorax 69, 648–653 (2014) 17. N. Daddi, M. Schiavon, P.L. Filosso, G. Cardillo, M.C. Ambrogi, A. De Palma, L. Luzzi, A. Bandiera, C. Casali, A. Ruffato, V. De Angelis, L.G. Andriolo, F. Guerrera, F. Carleo, F. Davini, M. Urbani, S. Mattioli, U. Morandi, P. Zannini, G. Gotti, M. Loizzi, F. Puma, A. Mussi, A. Ricci, A. Oliaro, F. Rea, Multi-Institutional Italian Pathology Group: Prognostic factors in a multicentre study of 247 atypical pulmonary carcinoids. Eur. J. Cardiothorac. Surg. 45, 677–686 (2014) 18. R. Johnson, S. Trocha, M. McLawhorn, M. Worley, G. Wheeler, L. Thompson, N. Schisler, D. Schammel, C. Schammel, J. Stephenson, W. Bolton, Histology, not lymph node involvement, predicts long-term survival in bronchopulmonary carcinoids. Am. Surg. 77, 1669–1674 (2011) 19. G. Pelosi, G. Rindi, W.D. Travis, M. Papotti, Ki-67 antigen in lung neuroendocrine tumors: unraveling a role in clinical practice. J. Thorac. Oncol. 9, 273–284 (2014) 20. T. Zahel, S. Krysa, E. Herpel, A. Stenzinger, B. Goeppert, P. Schirmacher, H. Hoffmann, P.A. Schnabel, A. Warth, Phenotyping of pulmonary carcinoids and a Ki-67-based grading approach. Virchows Arch. 460, 299–308 (2012) 21. G. Rindi, C. Klersy, F. Inzani, G. Fellegara, L. Ampollini, A. Ardizzoni, N. Campanini, P. Carbognani, T.M. De Pas, D. Galetta, P.L. Granone, L. Righi, M. Rusca, L. Spaggiari, M. Tiseo, G. Viale, M. Volante, M. Papotti, G. Pelosi, Grading the neuroendocrine tumors of the lung: an evidence-based proposal. Endocr. Relat. Cancer 21, 1–16 (2014) 22. F. Grimaldi, D. Muser, C.A. Beltrami, P. Machin, A. Morelli, S. Pizzolitto, G. Talmassons, F. Marciello, A.A. Colao, R. Monaco, G. Monaco, A. Faggiano, Partitioning of bronchopulmonary carcinoids in two different prognostic categories by Ki-67 score. Front Endocrinol. (Lausanne) 2, 20 (2011) 23. A.E. Walts, D. Ines, A.M. Marchevsky, Limited role of Ki-67 proliferative index in predicting overall short-term survival in patients with typical and atipica pulmonary carcinoid tumors. Mod. Pathol. 25, 1258–1264 (2012) 24. K. Tsuta, M.G. Raso, N. Kalhor, D.D. Liu, I.I. Wistuba, C.A. Moran, Histologic features of low- and intermediate-grade
Endocrine neuroendocrine carcinoma (typical and atipica carcinoid tumors) of the lung. Lung Cancer 71, 34–41 (2011) 25. P. Perkins, B.L. Kemp, J.B. Putnam Jr, J.D. Cox, Pretreatment characteristics of carcinoid tumors of the lung which predict aggressive behavior. Am. J. Clin. Oncol. 20, 285–288 (1997)
26. S.Y. Ha, J.J. Lee, J. Cho, J. Hyeon, J. Han, H.K. Kim, Lung parenchymal invasion in pulmonary carcinoid tumor: an important histologic feature suggesting the diagnosis of atypical carcinoid and poor prognosis. Lung Cancer 80, 146–152 (2013)
123
VIEWPOINT
Classification of lung neuroendocrine tumors: lights and shadows Marco Volante1 • Gaia Gatti1 • Mauro Papotti1
Received: 27 November 2014 / Accepted: 11 March 2015 Ó Springer Science+Business Media New York 2015
Abstract Neuroendocrine tumors of the lung are classified into low-grade typical and intermediate-grade atypical carcinoids, and high-grade poorly differentiated neuroendocrine carcinomas of the large and small cell types. This scheme is strongly predictive of patients’ prognosis but relies on few and scarcely reproducible pathological parameters (namely mitotic count and assessment of the presence of necrosis), which have been demonstrated to affect the inter-observer agreement of the classification. Moreover, tumor and nodal staging schemes are not specific for lung carcinoids, at variance with neuroendocrine tumors of the gastro-entero-pancreatic system, despite these tumors have specific features that strongly differ from conventional lung cancer. Finally, there is no grading for lung neuroendocrine neoplasms and prognostication, as well as the definition of treatment modalities and clinical strategies, which are based on tumor histotypes, only. However, literature data indicate that the evaluation of Ki-67 proliferation index may be a reliable and useful tool to determine the biological and clinical behavior of neuroendocrine tumors, with special reference to carcinoids, both in pre-operative and surgical samples. Keywords Lung Carcinoid Neuroendocrine Carcinoma Classification
& Marco Volante [email protected] 1
Department of Oncology, University of Turin at San Luigi Hospital, Regione Gonzole 10, Orbassano, 10043 Turin, Italy
Lung neuroendocrine tumor diagnosis and classification: what is ‘‘under the sun’’ Neuroendocrine tumors of the lung are a heterogeneous group of neoplasms which includes four histological types representative of a spectrum of tumor differentiation and having considerably different clinical and biological behavior. On the one side of this spectrum, lung carcinoids are separated into low-grade typical and intermediate-grade atypical carcinoids, while high-grade poorly differentiated lung neuroendocrine neoplasms include both large and small cell neuroendocrine carcinomas, on the other. All four histological types are characterized by different cytological and architectural findings. However, according to the most recent World Health Organization classification, year 2004 [1], the differential diagnosis among them relies on the assessment of two parameters, only, namely the presence/absence of necrosis and mitotic index per 2 mm2, with a cut-off of \2 to segregate typical and atypical carcinoids and of C10 for high-grade poorly differentiated neuroendocrine carcinomas. Cell size, nuclear morphology, and architecture are additional characteristics useful to distinguish large from small cell neuroendocrine carcinomas. In 2015, the 7th edition of the WHO classification of Tumors of the Lung, Pleura, Thymus, and Heart is expected, based on the proposal of the International Association for the Study of Lung Cancer. A major change in the new classification will concern large cell neuroendocrine carcinoma, which will be moved into the group of neuroendocrine neoplasms and not considered, as in the current classification, a variant of large cell carcinomas. However, no other changes—with special reference to carcinoid
123
Endocrine
classification—will be present, based on the proven highly prognostic performance of the currently employed system. The use of immunohistochemistry is restricted to general neuroendocrine markers, such as chromogranin A and synaptophysin, which are required, especially in small biopsies, to prove the neuroendocrine nature of these neoplasms. At variance with the gastro-entero-pancreatic system, Ki-67 labeling index is not mandatory in the diagnostic flow of lung neuroendocrine tumors, but may be useful in distinguishing carcinoids from high-grade carcinomas, especially in small biopsies [2]. There is no specific staging for lung neuroendocrine neoplasms, but they are staged according to the AJCC/ IASLC TNM staging system following the same recommendations formulated for non-neuroendocrine lung cancer [3]. Based on the aforementioned, the classification system and the diagnostic workflow of lung neuroendocrine neoplasms are apparently easy and well settled. However, several gray zones and unresolved issues still exist in lung tumor classification and pathological characterization, which will be briefly outlined here below (Table 1).
Unclear clinical significance of pre-neoplastic lesions Pre-neoplastic neuroendocrine lesions include neuroendocrine cell hyperplasia and tumorlets. Neuroendocrine cell hyperplasia is defined by a linear or nodular proliferation of neuroendocrine cells confined within the bronchial basal membrane, whereas tumorlets may show signs of stromal invasion. In both cases, however, the largest dimension should not exceed 5 mm, which is the unique criterion to differentiate these lesions from carcinoids. Tumorlets and neuroendocrine cell hyperplasia may be not infrequently observed in lung parenchyma at the periphery
of carcinoids [4] or in patients affected by chronic diseases leading to respiratory failure. Moreover, they may be the hallmark of a rare idiopathic condition, the so-called diffuse idiopathic pulmonary neuroendocrine cell hyperplasia (DIPNECH) [5]. Irrespective of the clinical situation, the pathogenetic link between neuroendocrine cell hyperplasia/tumorlets and carcinoids, their distinction (based currently only on size of the lesion) and the real clinical significance as pre-invasive lesions of the former, is questioned. In fact, no molecular nor clinical data are available to demonstrate or disprove a true progression from neuroendocrine cell hyperplasia/tumorlets to carcinoids, and rare cases of tumorlets with nodal metastases have been described [6].
Low reproducibility of lung neuroendocrine neoplasm classification As already stated above, two parameters are fundamental for lung neuroendocrine neoplasm classification, namely mitotic index and necrosis. However, the assessment of these two parameters may be affected by high subjectivity. In a study on 110 lung neuroendocrine tumors, up to 44 % were re-classified after revision by a panel of pathologist experienced in the field of lung cancer [7]. More recently, carcinoid classification was demonstrated to have a fair inter-observer agreement (Kappa: 0.32), even between experienced pathologists [8]. The low number of mitoses in a rather large microscopic area, the fact that mitoses are not homogeneously distributed over the tissue section and that they may be confused with apoptotic cells, crushed cells, and granulocytes are proposed to explain this low reproducibility. In addition, punctate necrosis may be difficult to identify. The distinction between large and small cell highgrade neuroendocrine carcinomas also shows a low reproducibility, due to problems related to sampling procedure and the resulting presence of artifacts, and to the fact that
Table 1 Lights and shadows in lung neuroendocrine tumor classification Cons
Item
Pros
Histological classification is not assessable in small biopsies/cytological samples
Current four-ties classification scheme
Worldwide accepted
Lung cancer overall staging is prognostic also for carcinoids A carcinoid-specific staging might be a confounding factor
Carcinoid-specific staging
Mean carcinoid size is lower than non-neuroendocrine lung cancer (inappropriate size cut-offs)
Histological classification is the most relevant prognostic factor at multivariate analysis in most studies
Carcinoid-specific grading
Not fully reproducible (low inter-observer agreement for mitotic index and necrosis)
Familiar to clinicians Relevant for prognosis
pTN stage is not prognostic for typical carcinoids
Ki-67 cut-off values are not uniform nor consolidated
123
Proliferation index is more easy to evaluate in small biopsies Homogenization with neuroendocrine tumors of the GEP area is desirable
Endocrine
some tumors present transitional cell characteristics between small and large cell neuroendocrine carcinomas [9, 10]. The use of mitosis-specific antibodies, such as phosphohistone H3 [11], has been proposed to improve the mitotic index determination indistinguishing typical from atypical carcinoids. Moreover, Ki-67 proliferation index has been shown to outperform mitotic count in terms of reproducibility of carcinoid classification [12].
Lack of a carcinoid-specific staging Although lung carcinoids are staged as non-neuroendocrine lung cancer and staging overall has been shown to be prognostic in lung neuroendocrine neoplasms [3], some carcinoid-specific features suggest that this classification may be not completely representative of the real clinical situation. In fact, TNM staging cut-off values of 3, 5, and 7 cm are poorly representative of carcinoids, since both typical and atypical carcinoids are often less than 3 cm in largest diameter. This is also clear from the original paper by Travis and co-workers [3] which showed that the number of carcinoid cases in the SEER database having a diameter below/equal to or above five cm was 1223 and 41, respectively. Moreover, literature data—which confirmed a prognostic significance of the TNM classification—also showed an overlapping survival of carcinoid patients in stages I and II [13, 14]. In fact, in both papers, either in the complete group of carcinoids [13] or in typical and atypical ones considered separately [14], a significantly different prognosis was observed only for patients in stage III (and IV) as compared to stages I and II, but not between stages I and II. Moreover, in two other papers, the prognostic impact of the TNM staging was lost at multivariate analysis which retains histological type as the major prognostic indicator [4, 15]. The impact of nodal involvement as a prognostic factor in carcinoids is also controversial. In fact, the presence of lymph node metastases has been demonstrated to be and independent negative prognostic factor by some Authors [13] but not by others [4, 18]. Moreover, it is a strong negative prognostic indicator in atypical carcinoids [16, 17] but no data have been generated specifically in the group of typical ones, which indeed in some series present a survival rate significantly higher that atypical ones despite a similar rate of cases with positive nodal status [4, 18].
Lack of grading for lung neuroendocrine neoplasms In pulmonary neuroendocrine neoplasms, no formal grading system has been created, as tumor grade is actually part of the histological classification. Morphology is undoubtedly able
to predict the clinical behavior of tumors. However, grading systems have been proposed in the literature to integrate histotype and to implement the prediction of patients’ outcome. The potential impact of such an approach is limited to carcinoids, since both small cell and large cell neuroendocrine carcinomas typically die quickly and they are considered ‘‘high grade’’ tumors by definition. Similar to grading of neuroendocrine neoplasms of the gastro-enteropancreatic tract, a major role for these systems is played by Ki-67 proliferation index, which has been demonstrated to be a strong prognostic indicator in lung neuroendocrine neoplasms [12, 19]. Moreover, the homogenization of a lung carcinoid grading approach with that of neuroendocrine tumors of the gastro-entero-pancreatic system would be also a benefit for designing novel clinical trials, based on the fact that lung carcinoids are usually randomly included in trials recruiting other types of neuroendocrine cancers and therefore not fully comparable. However, a definite grading for lung carcinoids is still a matter of debate. In fact, different prognostic cut-offs of Ki-67 have been proposed to segregate G1 and G2 tumors, from 3 % (as in the case of neuroendocrine tumors of the gastrointestinal tract) [20], to 4 % [21, 22], to 5 % [8]. Moreover, Ki-67 proliferation index has been proposed either alone or integrated with mitotic count and necrosis, as recently proposed by the group of Rindi and coworkers [21]. In addition, although not extensively investigated in lung neuroendocrine neoplasms so far, the assessment of Ki-67 is affected by either technical issues or poor reproducibility of scoring [19]. Finally, the real prognostic role of Ki-67 proliferation index, independently from carcinoid histotypes, has been called into question by some other Authors [23].
Doubtful clinical meaning of additional pathologic parameters Apart from morphological (or phenotypical) parameters included in classification or proposed grading systems, some other pathological characteristics have been analyzed as potential predictors of biological and clinical behavior in lung neuroendocrine neoplasms, with special reference to carcinoids. In a study by Tsuta and co-workers, several pathological parameters were associated to either the presence of lymph node metastases (blood and lymphovascular invasion, nuclear features) or survival (pleural invasion and active fibroblast proliferation) [24]. Vascular invasion has been associated to worse prognosis in another single study [25]. Moreover, invasion of the lung parenchyma has been also associated to atypical carcinoid histotypes and poor prognosis [26]. However, most of the features above have been investigated in single studies lacking multivariate statistical analysis, and are associated
123
Endocrine
to the atypical carcinoid histotypes; further studies are therefore needed to validate these findings and possibly insert these parameters in future classifications/diagnostic grids for pathology reporting of lung neuroendocrine neoplasms. Acknowledgments This study was partially supported by a grant from the Italian Association for Cancer Research (AIRC) (grant nr. IG 13567 to MV). Conflict of interest All Authors declare the absence of any potential conflict of interest.
References 1. W.D. Travis, E. Brambilla, H.K. Muller-Hermelink, C.C. Harris, Pathology and genetics of tumours of the lung, pleura, thymus and heart (IARC Press, Lyon, 2004) 2. G. Pelosi, J. Rodriguez, G. Viale, J. Rosai, Typical and atypical pulmonary carcinoid tumor overdiagnosed as small-cell carcinoma on biopsy specimens: a major pitfall in the management of lung cancer patients. Am. J. Surg. Pathol. 29, 179–187 (2005) 3. W.D. Travis, D.J. Giroux, K. Chansky, J. Crowley, H. Asamura, E. Brambilla, J. Jett, C. Kennedy, R. Rami-Porta, V.W. Rusch, P. Goldstraw, International Staging Committee and Participating Institutions: the IASLC Lung Cancer Staging Project: proposals for the inclusion of broncho-pulmonary carcinoid tumors in the forthcoming (seventh) edition of the TNM Classification for Lung Cancer. J. Thorac. Oncol. 3, 1213–1223 (2008) 4. P. Ferolla, N. Daddi, M. Urbani, A. Semeraro, R. Ribacchi, P. Giovenali, S. Ascani, V. De Angelis, L. Crino`, F. Puma, G. Daddi, Regional multidisciplinary group for the diagnosis and treatment of neuroendocrine tumors, CRO, Umbria region cancer network, Italy: tumorlets, multicentric carcinoids, lymph-nodal metastases, and long-term behavior in bronchial carcinoids. J. Thorac. Oncol. 4, 383–387 (2009) 5. A.A. Nassar, D.E. Jaroszewski, R.A. Helmers, T.V. Colby, B.M. Patel, F. Mookadam, Diffuse idiopathic pulmonary neuroendocrine cell hyperplasia: a systematic overview. Am. J. Respir. Crit. Care Med. 184, 8–16 (2011) 6. S.M. Liu, H.H. Wu, C.J. Wu, C.L. Kuo, L.R. Mo, Adrenocorticotropin-producing pulmonary tumorlets with lymph node metastasis. Pathol. Int. 53, 883–886 (2003) 7. B.G. Skov, M. Krasnik, S. Lantuejoul, T. Skov, E. Brambilla, Reclassification of neuroendocrine tumors improves the separation of carcinoids and the prediction of survival. J. Thorac. Oncol. 3, 1410–1415 (2008) 8. D.R. Swarts, R.J. van Suylen, M.A. den Bakker, M.F. van Oosterhout, F.B. Thunnissen, M. Volante, A.M. Dingemans, M.R. Scheltinga, G.P. Bootsma, H.M. Pouwels, B.E. van den Borne, F.C. Ramaekers, E.J. Speel, Interobserver variability for the WHO classification of pulmonary carcinoids. Am. J. Surg. Pathol. 38, 1429–1436 (2014) 9. M.A. den Bakker, S. Willemsen, K. Gru¨nberg, L.A. Noorduijn, M.F. van Oosterhout, R.J. van Suylen, W. Timens, B. Vrugt, A. Wiersma-van Tilburg, F.B. Thunnissen, Small cell carcinoma of the lung and large cell neuroendocrine carcinoma interobserver variability. Histopathology 56, 356–363 (2010) 10. A.M. Marchevsky, A.A. Gal, S. Shah, M.N. Koss, Morphometry confirms the presence of considerable nuclear size overlap between ‘‘small cells’’ and ‘‘large cells’’ in high-grade pulmonary neuroendocrine neoplasms. Am. J. Clin. Pathol. 116, 466–472 (2001)
123
11. K. Tsuta, D.C. Liu, N. Kalhor, I.I. Wistuba, C.A. Moran, Using the mitosis-specific marker anti-phosphohistone H3 to assess mitosis in pulmonary neuroendocrine carcinomas. Am. J. Clin. Pathol. 136, 252–259 (2011) 12. A. Warth, L. Fink, A. Fisseler-Eckhoff, D. Jonigk, M. Keller, G. Ott, R.J. Rieker, P. Sinn, S. So¨der, A. Soltermann, K. Willenbrock, W. Weichert, Pulmonary Pathology Working Group of the German Society of Pathology: interobserver agreement of proliferation index (Ki-67) outperforms mitotic count in pulmonary carcinoids. Virchows Arch. 462, 507–513 (2013) 13. P.L. Filosso, A. Oliaro, E. Ruffini, G. Bora, P. Lyberis, S. Asioli, L. Delsedime, A. Sandri, F. Guerrera, Outcome and prognostic factors in bronchial carcinoids: a single-center experience. J. Thorac. Oncol. 8, 1282–1288 (2013) 14. E. Aydin, U. Yazici, M. Gulgosteren, Y. Agackiran, S. Kaya, E. Gulhan, I. Tastepe, N. Karaoglanoglu, Long-term outcomes and prognostic factors of patients with surgically treated pulmonary carcinoid: our institutional experience with 104 patients. Eur. J. Cardiothorac. Surg. 39, 549–554 (2011) 15. C. Cao, T.D. Yan, C. Kennedy, N. Hendel, P.G. Bannon, B.C. McCaughan, Bronchopulmonary carcinoid tumors: longterm outcomes after resection. Ann. Thorac. Surg. 91, 339–343 (2011) 16. M.A. Can˜izares, J.M. Matilla, A. Cueto, J. Algar, I. Muguruza, N. Moreno-Mata, R. Moreno-Balsalobre, R. Guijarro, R. Arrabal, E. Garcia-Fontan, A. Gonzalez-Pin˜eiro, M. Garcia-Yuste, EMETNE-SEPAR Members: Atypical carcinoid tumours of the lung: prognostic factors and patterns of recurrence. Thorax 69, 648–653 (2014) 17. N. Daddi, M. Schiavon, P.L. Filosso, G. Cardillo, M.C. Ambrogi, A. De Palma, L. Luzzi, A. Bandiera, C. Casali, A. Ruffato, V. De Angelis, L.G. Andriolo, F. Guerrera, F. Carleo, F. Davini, M. Urbani, S. Mattioli, U. Morandi, P. Zannini, G. Gotti, M. Loizzi, F. Puma, A. Mussi, A. Ricci, A. Oliaro, F. Rea, Multi-Institutional Italian Pathology Group: Prognostic factors in a multicentre study of 247 atypical pulmonary carcinoids. Eur. J. Cardiothorac. Surg. 45, 677–686 (2014) 18. R. Johnson, S. Trocha, M. McLawhorn, M. Worley, G. Wheeler, L. Thompson, N. Schisler, D. Schammel, C. Schammel, J. Stephenson, W. Bolton, Histology, not lymph node involvement, predicts long-term survival in bronchopulmonary carcinoids. Am. Surg. 77, 1669–1674 (2011) 19. G. Pelosi, G. Rindi, W.D. Travis, M. Papotti, Ki-67 antigen in lung neuroendocrine tumors: unraveling a role in clinical practice. J. Thorac. Oncol. 9, 273–284 (2014) 20. T. Zahel, S. Krysa, E. Herpel, A. Stenzinger, B. Goeppert, P. Schirmacher, H. Hoffmann, P.A. Schnabel, A. Warth, Phenotyping of pulmonary carcinoids and a Ki-67-based grading approach. Virchows Arch. 460, 299–308 (2012) 21. G. Rindi, C. Klersy, F. Inzani, G. Fellegara, L. Ampollini, A. Ardizzoni, N. Campanini, P. Carbognani, T.M. De Pas, D. Galetta, P.L. Granone, L. Righi, M. Rusca, L. Spaggiari, M. Tiseo, G. Viale, M. Volante, M. Papotti, G. Pelosi, Grading the neuroendocrine tumors of the lung: an evidence-based proposal. Endocr. Relat. Cancer 21, 1–16 (2014) 22. F. Grimaldi, D. Muser, C.A. Beltrami, P. Machin, A. Morelli, S. Pizzolitto, G. Talmassons, F. Marciello, A.A. Colao, R. Monaco, G. Monaco, A. Faggiano, Partitioning of bronchopulmonary carcinoids in two different prognostic categories by Ki-67 score. Front Endocrinol. (Lausanne) 2, 20 (2011) 23. A.E. Walts, D. Ines, A.M. Marchevsky, Limited role of Ki-67 proliferative index in predicting overall short-term survival in patients with typical and atipica pulmonary carcinoid tumors. Mod. Pathol. 25, 1258–1264 (2012) 24. K. Tsuta, M.G. Raso, N. Kalhor, D.D. Liu, I.I. Wistuba, C.A. Moran, Histologic features of low- and intermediate-grade
Endocrine neuroendocrine carcinoma (typical and atipica carcinoid tumors) of the lung. Lung Cancer 71, 34–41 (2011) 25. P. Perkins, B.L. Kemp, J.B. Putnam Jr, J.D. Cox, Pretreatment characteristics of carcinoid tumors of the lung which predict aggressive behavior. Am. J. Clin. Oncol. 20, 285–288 (1997)
26. S.Y. Ha, J.J. Lee, J. Cho, J. Hyeon, J. Han, H.K. Kim, Lung parenchymal invasion in pulmonary carcinoid tumor: an important histologic feature suggesting the diagnosis of atypical carcinoid and poor prognosis. Lung Cancer 80, 146–152 (2013)
123
