Prognostic Significance of Synaptophysin in Stage I of Squamous Carcinoma and Adenocarcinoma of the Lung

Transcription

1 1776 Prognostic Significance of Synaptophysin in Stage I of Squamous Carcinoma and Adenocarcinoma of the Lung Federico González-Aragoneses, MD, PhD 1,2 Nicolás Moreno-Mata, MD, PhD 1,2 María Cebollero-Presmanes, MD, PhD 3 Mariano García-Yuste, MD, PhD 4 Miguel Ángel Ca~nizares-Carretero, MD 5 Laureano Molins-López-Rodó, MD, PhD 6,7 Santiago Quevedo-Losada, MD, PhD 8 Juan Torres-Lanzas, MD, PhD 9,10 Emilio Álvarez-Fernández, MD, PhD 3,11 and the Spanish Multicenter Study of Neuroendocrine Tumours of the Lung of the Spanish Society of Pneumonology and Thoracic Surgery (EMETNE-SEPAR)* 1 Thoracic Surgery, Complutense University, Madrid, Spain. 2 Department of Thoracic Surgery, Gregorio Maranon University Hospital, Madrid, Spain. 3 Department of Pathology, Gregorio Maranon University Hospital, Madrid, Spain. 4 Department of Thoracic Surgery, Valladolid University Hospital, Valladolid, Spain. 5 Department of Thoracic Surgery, Xeral Hospital, Vigo, Spain. 6 Thoracic Surgery, Barcelona University, Barcelona, Spain. 7 Department of Thoracic Surgery, Sagrado Corazon Hospital, Barcelona, Spain. BACKGROUND. The prognostic significance of the presence of a neuroendocrine marker (synaptophysin, SY) was analyzed in stage I of squamous carcinoma and adenocarcinoma of the lung. METHODS. A multicentric retrospective study was conducted with immunohistochemical staining in a single center of 318 patients resected for squamous carcinoma or adenocarcinoma in pathologic stage I. RESULTS. In all, 162 cases of squamous carcinoma and 156 cases of adenocarcinoma were identified, which included 105 patients in stage IA (50 patients with squamous carcinoma and 55 patients with adenocarcinoma) and 213 in stage IB (112 with squamous carcinoma and 101 with adenocarcinoma). Eighty-six tumors showed a presence of SY1 (27%). Univariate analysis showed lower survival rates at 5 years for those patients older than 70 years of age compared with those patients younger than 70 years of age (60.35% vs 70.57%; P 5.007) and for those patients with SY1 compared with those with SY2 (52.48% vs 72.68%; P ). Patients with SY1 tumors showed a higher rate of recurrence than patients with SY2 tumors (50% vs 33.6%; P 5.008). Multivariate analysis showed that those patients greater that 70 years of age (hazard ratio [HR], 1.74; 95% confidence interval [CI], ) and the presence of SY (HR, 2.15; 95% CI, ) were significant independent prognostic factors associated with a poor outcome. CONCLUSIONS. Stage I of squamous carcinoma and adenocarcinoma of the lung with SY1 has a poor prognosis, with a higher frequency of recurrence and lower survival rates. Cancer 2007;110: Ó 2007 American Cancer Society. KEYWORDS: squamous carcinoma and adenocarcinoma of the lung, neuroendocrine differentiation, immunohistochemistry, survival. It is estimated that surgical resection for nonsmall-cell lung cancer (NSCLC) in stage I (T1N0M0 and T2N0M0) has a survival rate of approximately 75% 1,2 at 5 years. One of the possible causes contributing to such a significant percentage of the cases (25%) having a negative 8 Department of Thoracic Surgery, Insular Hospital, Las Palmas, Gran Canaria, Spain. 9 Thoracic Surgery, Murcia University, Murcia, Spain. 10 Department of Thoracic Surgery, Virgen de la Arrixaca Hospital, Murcia, Spain. 11 Pathology, Complutense University, Madrid, Spain. Partly financed by FIS grant PI *Unit of Investigation, University Hospital, Valladolid: Ana Almaraz, MD, and María F. Mu~noz. EMETNE-SEPAR Invited Foreign Members: William D. Travis, MD (Sloan Kettering Cancer Center, New York, NY); Richard Battafarano, MD (Washington University, St. Louis, MO); Pierre Fuentes, MD (University Hospitals of Marseille, France). The authors thank Mark and Marga Fronmuller for assistance in the English translation. Address for reprints: Federico González-Aragoneses, MD, PhD, Dept. of Thoracic Surgery, Gregorio Maranon University Hospital, Madrid, Spain; Fax: (911) ; Received December 18, 2006; revision received May 8, 2007; accepted July 18, ª 2007 American Cancer Society DOI /cncr Published online 27 August 2007 in Wiley InterScience (

2 Significance of Synaptophysin/González-Aragoneses et al evolution is that these were cases of an NSCLC with neuroendocrine (NE) differentiation and a higher degree of aggressiveness. 3 According to the WHO 3,4 10% to 20% of NSCLC cases show NE differentiation. This NE differentiation has been found in all histologic types, although some studies have found that it is more frequent in adenocarcinoma and large cell carcinoma. 5,6 It has been suggested that NE differentiation in NSCLC is accompanied by a lower survival rate and higher chemosensitivity, but other studies have not been able to show any correlation. 7,8 The objective of this study was to evaluate the prevalence and prognostic significance of NE differentiation in a series of patients with squamous carcinoma and adenocarcinoma in pathologic stage I in whom a complete resection was carried out using synaptophysin (SY) as a marker. MATERIALS AND METHODS Criteria for inclusion in the study included cases of squamous carcinoma and adenocarcinoma in pathologic stage IA (pt1n0m0) and IB (pt2n0m0) who underwent radical surgery, without adjuvant or neoadjuvant therapy, surviving a minimum of 30 days and undergoing a complete clinical follow-up. In all, 360 patients operated for stage Ip NSCLC (T1-T2N0M0) in 6 hospitals of the Spanish Multicenter Study of Neuroendocrine Tumors of the Lung (EMETNE-SEPAR) with histology of squamous carcinoma and adenocarcinoma were studied. The tumors were classified according to the WHO classification system. All tissue blocks were sent for central histological typing and immunohistochemical (IHC) staining. All tumors from the institution in which the central review and IHC techniques were performed (162 patients) were embedded in toto in paraffin after fixation in 4% formaldehyde and Carnoy fixative. Tissue samples were cut up as large as possible so that histologic sections of at least 2 3 2mm(4mm 2 for staining and study) could be obtained. The rest of the specimen was fixed by transbronchial instillation of buffered formol and left in fixative for 24 hours. All paraffin blocks were sectioned in a rotary microtome and 3 to 4-lm sections were obtained mounted in silane-coated slides and left to dry overnight in an oven set at 308C. They were deparaffined with xylene (2 3 5 minutes) and rehydrated using decreasing concentrations of alcohol (100%, 96%, and 70%) for 5 minutes and later washed for 5 minutes in distilled water. Lastly, they were dyed with hematoxylin-eosin. For IHC staining the rehydrated sections were pretreated with protease K for 10 minutes. To demonstrate SY expression a polyclonal antisynaptophysin antibody 223M (Biomeda, Foster City, Calif) obtained from a synthetic peptide of human SY at a working dilution of 1 of 8 was used. The techniques were evaluated taking into account the intensity of positivity and the percentage of positive cells. A presence of 10% or greater of neoplastic cells was considered positive. The evaluation of the histochemical techniques was performed by 2 separate pathologists (M.C.P., E.A.F.) in 2 separate rounds so as to obtain a consensus diagnosis. All the patients studied were included in a protocol of EMETNE-SEPAR using information derived from clinical follow-up, as is carried out with all NE tumors. 9 A retrospective analysis of different clinical and pathologic variables was carried out: sex, age, pathologic stage, histology, and SY. Of the 360 patients, 42 were excluded because of a different histology, unknown cause of death, or lack of follow-up. Clinical Follow-up and Statistical Analysis Survival data were obtained from the successive reviews of the patients in each hospital. In all the patients the date and location of the tumoral recurrence, the date of death or the last living date, and the cause of death for those who died during followup were recorded. Only cancer-related mortality was considered; other causes of mortality were censored. Descriptive statistics were used to describe the patients characteristics and outcomes. Normally distributed continuous data are expressed as mean and standard deviation (SD). Categorical data are expressed as counts and proportions. The v 2 or Fisher exact tests were used to analyze the categorical data. All other data were compared using the analysis of variance test. Survival was defined as the interval between the date of surgery and the date of death related to lung cancer or the date of the most recent contact for censored cases. Cumulative survival rates were calculated by the Kaplan-Meier method. Survival comparison between groups of patients was completed using the Mantel-Haenszel log-rank test. The Cox Multivariate Proportional Hazards Regression Model was used to identify independent risk factors for death after surgical resection for stage I squamous carcinoma and adenocarcinoma. A probability value (P) less than or equal to.05 was considered significant. RESULTS The 318 tumors included 162 (51%) cases of squamous carcinoma and 156 (49%) cases of adenocarcinoma. There were 282 (89%) men and 36 women

3 1778 CANCER October 15, 2007 / Volume 110 / Number 8 TABLE 1 Distribution of the Series by Sex, Age, Histology, and Stage Sex Male n % Female n % Age Male Average 64.5 y SD 8.59 Female Average 63.3 y SD <70 y n ,2% >70 y n ,8% Histology Squamous carcinoma n % Adenocarcinoma n % Stage IA n % IB n % SD indicates standard deviation. TABLE 3 Relation Between SY and DFI DFI, mo SD, mo 95% CI Sig (P) SY SY SY indicates synaptophysin; DFI, disease-free interval. TABLE 4 Univariate Analysis. Survival Related to Age, Histology, Stage, and SY % Survival at 5 y 95% CI Sig (P) TABLE 2 Relation Between Histology, and Stage, and SY Stage IA Stage IB SY1 SY2 No. Squamous carcinoma Adenocarcinoma Total no. (%) 105 (33) 213 (67) 86 (27) 232 (73) 318 (100) SY indicates synaptophysin. with an average age of 64.5 years (standard deviation [SD], 8.59) in men and 63.3 years (SD, 10.12) in women (Table 1). A complete clinical follow-up for the survival analysis was available from all patients, with an average of months (SD, 32.44). The study included 105 patients in stage IA (50 patients with squamous carcinoma and 55 patients with adenocarcinoma) and 213 patients in stage IB (112 with squamous carcinoma and 101 with adenocarcinoma). A total of 86 (27%) tumors showed SY1, 34 (21%) in squamous carcinoma and 52 (33.3%) in adenocarcinoma (Table 2). Recurrent disease was observed in 121 (38%) patients. Among the 140 deceased patients, 53% died of distant metastasis, 12% of local recurrence, and 35% of causes not related to the tumor. In patients with SY1 there was 50% recurrence as opposed to 33.6% in SY2 (P 5.008), this difference being statistically significant. Average disease-free interval (DFI) in SY1 patients was months (SD, months; 95% confidence interval [CI]; ), whereas in SY2 patients, the DFI was months (SD months; 95% CI, ), this difference being statistically significant (P 5.01) (Table 3). The global survival rate of the series at 5 years was 67.40%. In the univariate analysis the 220 (69%) patients younger than 70 years of age had a survival Global Age <70 y Age >70 y Squamous carcinoma Adenocarcinoma Stage IA Stage IB SY SY Squamous Ca SY Squamous Ca SY Adenocarcinoma SY Adenocarcinoma SY SY indicates synaptophysin. rate at 5 years of 70.57% (95% CI, ) compared with 60.53% of the 98 patients older than 70 years of age (95% CI, ), this difference being significant (P 5.007). By histology, the survival rate at 5 years of the patients with squamous carcinoma tumors was 65.45% (95% CI, ) and the survival rate of the patients with adenocarcinoma tumors was 69.46% (95% CI, ), a difference that is not significant (P 5.35). By stages, the survival rate at 5 years of stage IA was 74.19% (95% CI, ) and of stage IB 63.98% (95% CI, ) with a difference close to being statistically significant (P 5.08) (Table 4). Survival at 5 years in patients with SY1 was 52.48% (95% CI, ) and in patients with SY % (95% CI, ), the difference being significant (P 5.015) (Fig. 1). By histology, the survival rate at 5 years of patients with squamous tumors with SY1 was 40.14% (95% CI, ) and with SY % (95% CI, ), the difference being significant (P ). Survival at 5 years in patients with adenocarcinoma with SY1 was 60.44% (95% CI, ) and with SY %

4 Significance of Synaptophysin/González-Aragoneses et al TABLE 5 Multivariate Cox Regression Analysis Beta HR 95% CI Sig (P) Age Sex Histology Stage SY Beta indicates regression coefficient; HR, hazard ratio: relative risk of more than 1.0 show higher risk of death (95% CI); CI, confidence interval. Sig (P): significant if P <.05. (95% CI, ) with a difference that was not significant (P 5.12) (Fig. 1) (Table 4). In the multivariate analysis, no interactions between variables (age, sex, stage, histology, and SY) were identified. Significant differences were found for less probability of survival if the age was above 70 years (P 5.009) and with the presence of SY (hazard ratio [HR], 2.15; 95% CI, ; P ), whereas the stage was near statistical significance (P 5.07). No significant differences were found regarding sex or histology (Table 5). FIGURE 1. (Top) Survival rate in patients with SY1 and SY2, P (Middle) Survival rate in squamous carcinoma with SY1 and SY2, P (Bottom) Survival rate in adenocarcinoma with SY1 and SY2, P DISCUSSION In the histologic classifications of the WHO, cases of NSCLC that presented IHC or ultrastructural evidence of NE differentiation were included as a specific group. It was mentioned in the comments that evidence of NE differentiation appeared in 10% to 20% of the cases of NSCLC, admitting a controversy about its prognostic significance and a different response to chemotherapy. It was suggested that further study was required to define them as an independent group. 3,4 Previous reports concerning the incidence of NE differentiation and its relation to the histologic type and prognostic significance are mostly retrospective and have offered variable and contradictory results As far as this incidence is concerned, the first problem is the identification of these tumors because there is no gold standard that defines NE differentiation. 8 These tumors are identified by electron microscopy or IHC using antibodies related to the neurosecretion system. Some authors 10,11,12 propose a series of markers to configure a routine panel and consider NE differentiation as being those tumors with 2 or more markers. But the results published vary substantially depending on the type of antibody, the laboratory that carried out the process, the type of material used, and the criteria used when inter-

5 1780 CANCER October 15, 2007 / Volume 110 / Number 8 TABLE 6 Global Incidence of SY1 in NSCLC Author No. of cases Stages Histology % SY1 Carles et al II III IV NSCLC: all types 22.6 Graziano et al I II NSCLC: all types 11.2 Schleusener et al III IV NSCLC: all types 24.3 Lyda et al NI NSCLC: all types 3 Ionescu et al NI NSCLC: all types 7.5 Gonzalez-Aragoneses et al. 318 I II NSCLC: Squam-Adeno 27 SY indicates synaptophysin; NSCLC, nonsmall-cell lung cancer; NI, not indicated; Squam, squamous; Adeno, adenocarcinoma. preting the samples. Another difficulty is that these tumors are very heterogeneous and have a small amount of neurosecretor granules, which means that if inadequate material is used, like the 1 used by endoscopy, 13 the results can be erroneous. In this study we tried to reduce to a minimum these biases, carrying out the study in a single laboratory. If we analyze only the SY marker, we can observe that in 560 cases published in 4 articles, which include all the subtypes of NSCLC and all the stages, the positivity varies from 3% 12 to 24.3%. 15 In our series of 318 cases, limited to squamous carcinoma and adenocarcinoma tumors in stage I, we obtain a percentage of 27% (Table 6), well over the 10% to 20% suggested by WHO. From the histological point of view, the majority of authors find that NE differentiation is more frequent in adenocarcinoma tumors and large-cell carcinoma tumors. 5,6 Our results confirm a higher frequency of SY1 in adenocarcinoma tumors (33.3%) than in squamous carcinoma (21%). SY1 is independent of the histological degree of differentiation of the tumor. No distribution preference of the positive tumoral cells in central or peripheral perivascular areas has been found. When the percentage of positive cells in the tumor is lower than 10%, the cells are distributed randomly, whereas if the percentage is higher they tend to group forming nests of variable size. 16 The existing confusion regarding prognostic significance of the presence of NE differentiation can be attributed to the inclusion of patients in early or advanced clinical stages. In tumors in an advanced stage with SY1 there are series that have shown a better prognosis 17 and there are few studies with comparison of tumors in early stages. Graziano et al. 14 included 260 (193 in stage I and 67 in stage II) cases showing a worse prognosis in the tumors that expressed NE markers, although the difference in survival did not reach statistical significance. Hiroshima et al. 5 determined the degree of NE differentiation in 90 surgically resected specimens of adenocarcinoma in stage IA and IB and found that its presence was a significant adverse factor (P ). Pelosi et al. 18 included 220 patients with tumors in stage IA and IB showing in the cases of adenocarcinoma a worse prognosis (HR, 2.61; 95% CI, ). Our results from the multivariate analysis showed that squamous carcinoma cases and adenocarcinoma cases in stage IA and IB with SY1 had a statistically significant higher recurrence (50% vs 33.6%; P 5.008), lower DFI (40.24 months vs months; P 5.01), and lower survival rate at 5 years (52.48 months vs months; P ). Although in the univariate analysis the presence of SY affected the prognosis of squamous carcinoma tumors and not the adenocarcinoma tumors, contradicting the works of Hiroshima et al. 5 and Pelosi et al., 18 the SY is independent of histology in multivariate analysis (HR, 2.15; P ). Therefore, this group of patients with SY1 has shown a more aggressive evolution and could be comparable to the NE carcinoma of large cells (LCNEC). 18 A recent article by Ionescu et al. 19 states that NE differentiation, as demonstrated by IHC, is of no clinical or prognostic significance. The main differences with our study are the number of cases, the size of tissue samples from each tumor, the wider subset of histologic variants included in the study, and the staging data. We have used routine histologic sections with a minimum surface area of about 4mm 2, whereas the other authors used tissue arrays with circular samples 0.6 mm in diameter. Taking into account the heterogeneity of the reaction it seems reasonable to assume that the larger the section the greater the probability of the presence of areas of positive staining. Conversely, we only included squamous carcinoma cases and adenocarcinoma cases in our series. So although the group of patients is smaller the selection of cases renders our group a very homogeneous one, avoiding possible inconsistencies due to differences in expression by other neoplastic variants as large cell carcinoma. At last the significance of SY expression in our study is restricted to pathologic stages I and II, and in this article no staging data are provided. Therefore, it is probable that at least some advanced cases may be included in the investigation. Although caution is important when interpreting these results and their therapeutic implications, 8 it would appear logical to think that these patients with SY1 could benefit from adjuvant chemotherapy after surgery. Thus, Iyoda et al. 20 published significant improvements in survival in patients with

6 Significance of Synaptophysin/González-Aragoneses et al large-cell carcinoma with NE differentiation treated with cisplatin and VP-16. In conclusion, the presence of SY in squamous carcinoma tumors and adenocarcinoma tumors is associated with a higher degree of aggressiveness and a worse prognosis that when SY is not present. Future studies should be aimed at confirming these results and testing if the administration of adjuvant chemotherapy increases the survival of these patients. REFERENCES 1. Read RC, Yoder G, Schaeffer RC. Survival after conservative resection for T1N0M0 non-small cell lung cancer. Ann Thorac Surg. 1990;49: Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumours in resected stage I lung cancer. J Thorac Cardiovasc Surg. 1995;109: Travis WD, Colby TV, Corrin B, et al. Histological typing of lung and pleural tumours. WHO International Histological Classification of Tumours. Berlin: Springer; Brambilla E. Large cell carcinoma. Pathology and genetics. In: Travis W, Brambilla E, Muller-Hermelink, Harris CG, eds. Tumours of the Lung. Lyon, France: IARC Press; 2004: Hiroshima K, Iyoda A, Shibuya K, et al. Prognostic significance of neuroendocrine differentiation in adenocarcinoma of the lung. Ann Thorac Surg. 2002;73: Linnoila RI, Mulshine JL, Steinberg SM, et al. Neuroendocrine differentiation in endocrine and nonendocrine lung carcinomas. Am J Clin Pathol. 1988;90: Howe MC, Chapman A, Kerr K, et al. Neuroendocrine differentiation in non-small cell lung cancer and its relation to prognosis and therapy. Histopathology. 2005;46: Carnaghi C, Rimassa L, Garassino I, Santoro A. Clinical significance of neuroendocrine phenotype in non-small-cell lung cancer. Ann Oncol. 2001;12:S Garcia-Yuste M, Matilla JM, Alvarez-Gago T, et al. Prognostic factors in neuroendocrine lung tumours: a Spanish Multicenter Study. Spanish Multicenter Study of Neuroendocrine Tumours of the Lung of the Spanish Society of Pneumonology and Thoracic Surgery (EMETNE-SEPAR). Ann Thorac Surg. 2000;70: Sundaresan V, Reeve JG, Stenning S, et al. Neuroendocrine differentiation and clinical behaviour in non-small cell lung tumours. Br J Cancer. 1991;64: Poola I, Graziano SL. Expression of NSE, Chromogranin A, synaptophysin and Leu-7 in lung cancer cell lines. J Exp Clin Cancer Res. 1998;17: Lyda MH, Weiss LM. Immunoreactivity for epithelial and neuroendocrine antibodies are useful in the differential diagnosis of lung carcinomas. Hum Pathol. 2000;31: Carles J, Rosell R, Ariza A, et al. Neuroendocrine differentiation as a prognostic factor in non-small cell lung cancer. Lung Cancer. 1993;10: Graziano SL, Tatum AH, Newman NB, et al. The prognostic significance of neuroendocrine markers and carcinoembryonic antigen in patients with resected stage I and II nonsmall cell lung cancer. Cancer Res. 1994;54: Schleusener JT, Tazelaar HD, Jung SH, et al. Neuroendocrine differentiation is an independent prognostic factor in chemotherapy-treated nonsmall cell lung carcinoma. Cancer. 1996;77: Cebollero-Presmanes M. Estudio de la historia natural de los carcinomas pulmonares no microciticos con diferenciacion neuroendocrina oculta. Tesis Doctoral. Universidad Complutense de Madrid, Graziano SL, Mazid R, Newman N, et al. The use of neuroendocrine immunoperoxidase markers to predict chemotherapy response in patients with non-small-cell lung cancer. J Clin Oncol. 1989;7: Pelosi G, Pasini F, Sonzogni A, et al. Prognostic implications of neuroendocrine differentiation and hormone production in patients with stage I nonsmall cell lung carcinoma. Cancer. 2003;97: Ionescu DN, Treaba D, Gilks CB, et al. Nonsmall cell lung carcinoma with neuroendocrine differentiation an entity of no clinical prognostic significance. Am J Surg Pathol. 2007;31: Iyoda A, Hiroshima K, Moriya Y, et al. Prospective study of adjuvant chemotherapy for pulmonary large cell neuroendocrine carcinoma. Ann Thorac Surg. 2006;82: