Immunohistochemical detection of neuroendocrine differentiation in non-small-cell lung cancer and its clinical implications

Abstract

Purpose

The purpose of this study was to prospectively assess the clinical implications of neuroendocrine (NE) differentiation in non-small-cell lung cancer (NSCLC) tumors.

Methods

This study accrued subjects suspected to have lung cancer who underwent diagnostic bronchoscopy. Bronchoscopically-biopsied specimens were subjected to routine pathologic examination, and immunohistochemical studies were then performed if lung cancer was diagnosed. Chromogranin-A, synaptophysin, neural cell adhesion molecule, and Leu7 were used to demonstrate NE differentiation.

Results

A total of 280 subjects were accrued to this study over a period of 2 years. Among them, 149 subjects were assessable for this study, and 130 were diagnosed as having NSCLC tumors (55 adenocarcinomas, 50 squamous cell carcinomas, 24 NSCLCs not otherwise specified, and 1 typical carcinoid). Large cell NE carcinoma was not observed in this study. Immunohistochemically, NE differentiation was detected in 16% of NSCLC tumors excluding typical carcinoid. By status of NE differentiation of NSCLC tumors, progression-free survivals were similar in 73 patients undergoing non-surgical treatment (positive, n = 10; negative, n = 63) and 43 patients undergoing surgical resection (positive, n = 8; negative, n = 35), respectively. Overall survival of patients with NE-positive tumors appeared to be favorable both for those undergoing non-surgical treatment and those undergoing surgical resection, though the differences in survival were not significant (P = 0.11 and 0.35, respectively).

Conclusions

NE differentiation was detected in 16% of NSCLC tumors in our study. However, the prognostic implications of the presence of this feature could not be clearly determined in this study.

Introduction

Lung cancer became a leading cause of cancer death in industrialized countries in the twentieth century (Smith and Khuri 2004), and its mortality has continued to increase in Japan (Center for Cancer Control and Information Services 2007). In the recent revision of the WHO classification of lung and pleural tumors (Travis et al. 1999), the terminology for pulmonary neuroendocrine (NE) lesions includes two distinct categories, as follows: (1) a series of common neoplasms with NE light-microscopic appearance, which include low-grade typical carcinoid (TC), intermediate-grade atypical carcinoid (AC), and high-grade large-cell NE carcinoma (LCNEC) and small-cell carcinoma (SCLC), and (2) non-small-cell carcinoma (NSCLC) with NE differentiation detected by immunohistochemical or ultrastructual studies. NCSLC with NE differentiation is known to account for 10–20% of histologically ordinary NSCLCs (Travis et al. 2000). However, controversy persists regarding whether NSCLC with NE differentiation is a favorable or unfavorable prognostic factor (Berendsen et al. 1989; Graziano et al. 1989; Kibbelaar et al. 1991; Skov et al. 1991; Linnoila, et al. 1994; Schleusener et al. 1996). In addition, studies reporting findings related to this issue included biases in subject selection, since only one was prospective (Berendsen et al. 1989). We report here the results of our prospective study assessing the clinical implications of NE differentiation in NSCLC tumors obtained by diagnostic bronchoscopy.

Materials and methods

Patients

This prospective study consecutively accrued subjects suspected of having lung cancer who underwent diagnostic bronchoscopy in our institution. Recruitment of subjects was performed between April 2004 and March 2006 after written informed consent had been obtained in accordance with our institutional guidelines. Eligibility criteria for this study included only a histological diagnosis of lung cancer using bronchoscopically-biopsied specimens.

Each eligible patient gave a complete medical history and underwent physical, laboratory, and staging work-up examinations. Laboratory examinations included complete blood cell counts with differentials, serum chemistry and tumor marker analyses, arterial blood gas analysis, electrocardiography, and pulmonary function tests. Staging work-up examination consisted of plain radiographs of the chest, computerized tomography of the chest and abdomen, magnetic resonance imaging of the brain, and radionucleotide bone scan. Disease stage was determined based on the international staging system (Mountain 1997).

Treatment

Treatment was performed on a practical basis, based on NSCLC disease stage as well as medical characteristics of patients such as performance status, preservation of organ function, and comorbidities. Surgical approaches included usual thoracotomy with mediastinal lymph node dissection and video-assisted thoracoscopic surgery. Radiotherapy was performed alone or in combination with induction or concurrent chemotherapy. The treatment volume consisted principally of original and boost volumes sequentially administered (Segawa et al. 1997). The radiation dose for the original volume was usually set at 40 Gy, and an additional 20–30 Gy was administered as the boost volume. Chemotherapy consisted of a wide range of regimens, which included monotherapy with third-generation agents and platinum-based doublet chemotherapy. Postoperative adjuvant chemotherapy was also performed.

Immunohistochemical detection of NE differentiation

Formalin-fixed paraffin-embedded bronchoscopically-biopsied samples were processed by routine hematoxylin and eosine staining, and immunohistochemical studies were performed if lung cancer was diagnosed. The classification criteria were based on the revised WHO classification of lung carcinoma (Travis et al. 1999). For demonstration of NE phenotype, chromogranin-A (DakoCytomation, Denmark), synaptophysin (DakoCytomation, Denmark), neural cell adhesion molecule (CD56; Novocastra, UK), and Leu7 (CD57; Novocastra, UK) were used. Staining was performed according to the protocols supplied by the manufacturers. Immunohistochemically, tumor was considered positive if tumor cells exhibited focal, patchy, or diffuse staining intracellularly for at least one of the antigens. Evaluation of immunohistochemical staining was performed by mutual agreement of two expert pathologists (R.N. and N.T.).

Statistical analysis

Statistical analysis was performed using SPSS software (version 11.1J; SPSS Inc., Chicago, IL). For comparison of proportions for categorical variables, the χ² test was used. The significance of differences between two independent groups was determined by the Mann–Whitney U test. Survival time was defined as the period from the initiation of treatment to death or last follow-up evaluation. In addition, progression-free survival was defined as the period from initiation of treatment to disease progression. Survival curves were calculated using the method of Kaplan and Meier, and differences in survival distribution between two categorized groups were assessed using the log-rank test. To estimate the prognostic significance of covariates for survival, a Cox regression model was employed using backward step-wise technique. P values less than 0.05 on two-tailed analyses were considered significant.

Results

This prospective study accrued 280 subjects over a period of 2 years. Among them, 95 with lung cancer diagnosed cytologically or using surgically-resected specimens, 27 with benign lung disease, and eight with metastatic lung tumor were excluded from this study. One patient with lung cancer was also excluded, since the tumor specimen for this patient was too small to appropriately determine tumor NE differentiation.

Incidence of lung cancer with NE differentiation

Of the 149 tumors assessable, a total of 130 were diagnosed as NSCLC tumors, whereas SCLC histology was confirmed in 19 tumors. NSCLC tumors consisted of 55 adenocarcinomas, 50 squamous cell carcinomas, 24 NSCLCs not otherwise specified, and one TC. No LCNEC or AC tumors were observed in this study. Immunohistochemically, NE differentiation was detected in 15 (27%) adenocarcinomas, four (8%) squamous cell carcinomas, and two (8%) NSCLCs not otherwise specified, with an overall incidence of 16% among these subtypes (Table 1; representative positive staining for each antigen is shown in Fig. 1). In addition, all of the SCLC tumors as well as the TC tumor exhibited NE differentiation.

Table 1.

Results of immunohistochemical examination

Ad adenocarcinoma, CGA chromogranin-A, NCAM neural cell adhesion molecule, NSCLC-NOS non-small-cell lung carcinoma not otherwise specified, SCLC small-cell lung carcinoma, SP synaptophysin, Sq squamous cell carcinoma, TC typical carcinoid

Fig. 1.

Positive immunostaining of non-small-cell lung cancer tumors for chromogranin-A (a), synaptophysin (b), neural cell adhesion molecule (c), and Leu7 (d)

Prognosis by NE differentiation in NSCLC tumors

Clinical background and treatment outcome were examined according to NE differentiation in NSCLC tumors. Thirteen patients undergoing best supportive care alone and one with TC were excluded from this analysis.

Among 73 patients undergoing non-surgical treatment (chemotherapy, n = 37; radiotherapy ± chemotherapy, n = 36), ten patients exhibited NE differentiation in tumors. There were no significant differences in clinical background factors between the NE-positive and -negative groups, although patients with advanced-stage disease were more frequent in the NE-negative group (Table 2). At the median follow-up time of 25.0 months (range, 0.5–44.4 months), median progression-free survival was similar in the NE-positive (7.7 months) and -negative (6.1 months) groups (P = 0.49) (Fig. 2a). Overall survival appeared to be better in the NE-positive group (median value, not reached) than in the NE-negative group (15.3 months) (Fig. 2b), though the difference between these groups was not significant (P = 0.11).

Table 2.

Characteristics of patients with non-small-cell lung cancer by treatment

	No. of patients treated with
	Non-surgical modality			Surgery
	IHC-positive	IHC-negative	P value	IHC-positive	IHC-negative	P value
Total	10	63		8	35
Median age (range)	71 (51–82)	67 (47–86)	0.76	72 (38–82)	71 (49–86)	0.83
Gender: male/female	9/1	52/11	0.53	6/2	25/10	0.84
ECOG PS: 0–1/2	10/0	59/4	0.27	8/0	35/0	–
Disease stage: I-II/III-IV	3/7	6/57	0.10	8/0	29/6a	0.10
CT/TRT ± CT	4/6	33/30	0.47	0/0	4/0b	–

CT chemotherapy, ECOG PS eastern cooperative oncology group performance status, IHC immunohistochemistry, TRT thoracic radiotherapy

^aPatients with stage IV disease were not included

^bFour patients underwent adjuvant chemotherapy

Fig. 2.

Progression-free survival curves (a) and overall survival curves (b) for 73 patients undergoing non-surgical treatment are depicted by presence of neuroendocrine differentiation in non-small-cell lung cancer tumors

Among 43 patients undergoing surgical resection, eight patients exhibited NE differentiation in tumors. There were no significant differences in clinical background factors between the NE-positive and -negative groups, although, similar to the findings for non-surgical treatment, advanced disease was more frequent in the NE-negative group (Table 2). In addition, four patients in the NE-negative group underwent adjuvant chemotherapy with a combination of carboplatin and paclitaxel. At a median follow-up time of 29.0 months (range, 0.5–45.4 months), progression-free survivals were similar in the two groups (P = 0.60) (Fig. 3a). Overall survival appeared to be better in the NE-positive group (Fig. 3b), though the difference between groups was not significant (P = 0.35).

Fig. 3.

Progression-free survival curves (a) and overall survival curves (b) for 43 patients undergoing surgical resection are depicted by presence of neuroendocrine differentiation in non-small-cell lung cancer tumors

The prognostic implications of NE differentiation in NSCLC tumors were further examined using a Cox regression model. However, prognostic significance of NE differentiation was confirmed for neither of the two treatment groups.

Tumor markers and NE differentiation in NSCLC tumors

Serum carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), and progastrin-releasing peptide (ProGRP) levels were measured in 127, 69, and 97 patients, respectively. The mean CEA values ± standard errors of the mean were 37.0 ± 22.3 ng/mL in the NE-positive group (n = 21) and 36.9 ± 16.1 ng/mL in the NE-negative group (n = 106) (P = 0.74). The respective values were 11.1 ± 1.9 ng/mL (n = 15) and 10.5 ± 1.0 ng/mL (n = 54) for NSE (P = 0.53) and 20.1 ± 3.1 pg/mL (n = 19) and 18.0 ± 1.1 pg/mL (n = 78) for ProGRP (P = 0.57).

Discussion

Given the possibility that NSCLC with NE differentiation might exhibit biological behavior intermediate between NSCLC and SCLC with increased chemosensitivity, the prognostic implications of this feature were extensively studied until the mid 1990’s. In a chemotherapy series, Graziano et al. (1989) reported superior survival with increased response to chemotherapy for NSCLC with NE differentiation. However, subsequent studies reported conflicting results, including lack of superiority in survival with increased response to chemotherapy (Skov et al. 1991) and superior survival without increase in chemotherapy response (Schleusener et al. 1996). In addition, among surgical series, two studies reported inferior survival with NE differentiation (Berendsen et al. 1989; Kibbelaar et al. 1991) though one study reported lack of difference in survival (Linnoila et al. 1994).

We resumed investigation of the clinical implications of NE differentiation in the present prospective study, since a comprehensive project for establishment of standardized diagnosis and treatment for NE lung cancer obtained support from the Japanese government, and since most accumulated findings on NE differentiation were obtained in a series of retrospective studies (Graziano et al. 1989; Kibbelaar et al. 1991; Skov et al. 1991; Linnoila et al. 1994; Schleusener et al. 1996) and appeared to include genuinely conflicting results.

NE differentiation was detected in 16% of NSCLC tumors in our study, an incidence consistent with that noted in a previous report (Travis et al. 2000). However, by status of NE differentiation in tumors, neither differences in progression-free survival nor in overall survival were demonstrated in either the surgical or non-surgical treatment groups, although the overall survival of patients with NE-positive tumors appeared to be favorable in both treatment groups. The lack of significant differences found between groups may have been mainly due to the small proportion of patients with NE-positive tumors as well as the imbalance in distribution of disease stage between the NE-positive and -negative groups. Response to chemotherapy and/or radiotherapy was not included in this study, since treatment was conducted on a practical basis. In addition, expression of NE differentiation in tumors was not associated with elevation of serum NE marker levels such as NSE and ProGRP. It would be useful, though probably not feasible, to conduct a prospective large-scaled randomized study to elucidate the clinical implications of NE differentiation in NSCLC.

In conclusion, NE differentiation was detected in 16% of NSCLC tumors in our prospective bronchoscopy-based study. However, the prognostic implications of the presence of this feature could not be clearly determined in this study.