Stage I non-small cell lung cancer: the presence of the lymphocyte-specific protein tyrosin kinase in the tumour infiltrate is associated with a better long-term prognosis

Abstract

We studied the expression in the tumour infiltrate of a T-cell activation marker, the lymphocyte-specific protein tyrosin kinase (LCK), to assess if it could be associated with a better prognostic outcome in early stage non-small cell lung cancer (NSCLC) patients. This retrospective study included 25 patients undergoing lobectomy with systematic hilo-mediastinal lymphadenectomy for pathological stage I NSCLC between July 2003 and June 2005. The presence of LCK was detected in the tumour infiltrate by immunohistochemistry on the specimens of all patients. No patient received adjuvant therapy. Twelve patients resulted LCK-positive and 13 LCK-negative. The distribution of patients according to the T-stage was similar between the LCK-positive group (1 T1a, 5 T1b, 6 T2a) and the LCK-negative group (1 T1a, 5 T1b, 7 T2a). Median overall survival (OS) time was not reached in the LCK-positive group and 30 months in the LCK-negative group (P = 0.01). OS was longer than 40 months in 75% of the LCK-positive patients and in 31% of the LCK-negative patients (P = 0.01). Median time to relapse (TTR) was significantly longer in LCK-positive patients than in LCK-negative patients (not reached vs. 25 months; P < 0.001). In conclusion, LCK-positive tumour infiltrate has been found to be associated with a significantly longer OS and TTR in patients with radically resected stage I NSCLC.

INTRODUCTION

Radical resection and early stage represent the main factors influencing survival in patients with lung cancer. However, as many as 30–35% of patients undergoing radical surgery for stage I non-small cell lung cancer (NSCLC) experience recurrence within 5 years from the operation [1]. This high rate of tumour relapse suggests that lung cancer is a heterogeneous disease with variable biological behaviour, and that other factors besides the pathological stage may influence prognosis of these patients.

Increasing evidence reported in the literature show that the immune system plays an important role in tumour surveillance, although its impact on prognosis has not been completely defined. In particular, the prognostic value of the tumour-infiltrating immune cells is controversial. Since cancer cells often suppress surrounding lymphocytes, it is challenging to distinguish whether an infiltrate is active or merely inactivated.

For this reason, we have studied the expression in the tumour infiltrate of a CD8-positive T-cell activation marker, the lymphocyte-specific protein tyrosine kinase (LCK), in order to assess if it could be associated with a better prognostic outcome in early stage NSCLC patients.

A noteworthy evidence of the correlation between this biological marker and a favourable prognosis in early lung cancer has been recently provided by Chen et al. [2]. These authors have identified five genes that are closely associated with relapse free and overall survival in patients with radically resected NSCLC. Among these genes, that for LCK was found to be the one with the highest protective value for cancer-related death. No other study has been published in the literature investigating the role of LCK in the long-term outcome of NSCLC.

We have restricted our analysis to pathological stage I patients, since at this stage the lack of adjuvant therapy avoids possible bias determined by the potential benefit of the chemotherapic treatment on tumour recurrence risk.

MATERIALS AND METHODS

This retrospective study included 25 patients with pathological stage I NSCLC undergoing radical lobectomy with systematic hilo-mediastinal lymphadenectomy between July 2003 and June 2005 at the Sant'Andrea Hospital, University ‘La Sapienza’ in Rome. This was a part of a larger investigation that included an immunohistochemistry analysis on paraffin-embedded tumour sections (whose results are reported in the present article), and a parallel gene expression study performed on frozen fragments of tumour tissue (whose results are not included in this article). For this reason, only patients with available frozen specimens allowing gene-expression profiling were included in this study. Patients operated in the same period with no frozen specimen available were not included. Twenty-three of the included patients were males and 2 females with a median age of 67 ± 10 years. The pathological stage was IA in 12 patients and IB in 13 patients. The tumour histotype was adenocarcinoma in 17 patients and squamous cell carcinoma in 8.

Fragments of tumour tissue were snap-frozen in liquid nitrogen, and stored at −80°C until sectioning; the remaining tissue was formalin-fixed and paraffin-embedded for conventional histology.

Preoperative evaluation of patients included lung, brain and abdominal computed tomography (CT) and bronchoscopy. No patient presented enlarged hilar and/or mediastinal lymph nodes (diameter >1 cm) at preoperative CT scan. Preoperative cyto/histologic diagnosis was achieved by means of fine needle aspiration biopsy in 16 (66%) of these patients. In the nine patients with no preoperative cyto/histologic diagnosis, intraoperative frozen section analysis on the lung tumour was performed before radical resection. Positron emission tomography was performed before the operation in nine patients (36%) showing positivity at the level of the lung lesion in all of them. Preoperative cytologic or histologic assessment of mediastinal lymph node status by transbronchial needle aspiration or mediastinoscopy was performed in none of these patients, since we consider it indicated only in case of clinical suspicion of metastatic nodal involvement (lymph node diameter >1 cm at CT scan).

No patient received preoperative or postoperative chemo- and/or radiotherapy. No patient had the history of other cancers. No patient experienced major complications after surgery. The presence of preoperative major comorbidity including cardiac disease, diabetes, chronic obstructive pulmonary disease and renal insufficiency was registered.

Oncological follow-up included brain, chest and abdomen CT scans with contrast medium associated with physical examination performed every 6 months for the first 5 years and once a year for the following time. Additional checks were performed during the first 2 years by chest X-ray and physical examination performed at 3, 9, 15 and 21 months from the operation. Flexible bronchoscopy was performed when evidence of recurrence involving the lung was found at radiological examination.

The study was performed according to the informed consensus law of Italy and was approved by the local Ethical Committee. All patients gave written informed consent for the study.

Immunohistochemistry

The study was performed on paraffin sections of NSCLC tissue using immunohistochemistry and mouse mAbs against LCK (clone Ab18896; 1:50 dilution; UCS Diagnostics, Italy). Paraffin-embedded sections were deparaffinized and treated for antigen retrieval (0.01 M citrate buffer pH 6.0), followed by preincubation with 3% hydrogen peroxide and with the DAKO protein blocking solution to prevent non-specific binding (DAKO; Dakopatts, Copenhagen, Denmark). The slides were then incubated with an optimal dilution of the primary antibody for 60 min at room temperature; the reaction product was developed using DAKO LSAB Kit-peroxidase and DAB Substrate Chromogen (DAKO; Dakopatts, Copenhagen, Denmark) and counterstained with haematoxylin.

The presence of LCK was evaluated on both tumour tissue and lymphocytic infiltrate by two expert pathologists, independently [Fig. 1]. Only patients showing lymphocytes in the tumour infiltrate with immunoreactivity for LCK, according to the evaluation of both pathologists, were considered LCK positive.

Figure 1:

Tumour infiltrating lymphocytes immunostained for LCK. Several cells with lymphocyte morphology show imunoreactivity (brown areas) (ABC-peroxidase, ×400). Some groups of cells showing immunoreactivity are pointed by arrows.

Statistical analysis

Survival was defined as the time from surgery to death from any cause. Time to relapse was defined as the time from surgery to documented disease progression or cancer-specific death. We used a two-sided unstratified log-rank test to compare survival rates in the two study groups. We estimated event–time distributions using the Kaplan–Meier method. All reported P-values are two-sided, and confidence intervals are at the 95% level. Descriptive statistics are also used.

RESULTS

Immunohistochemistry analysis showed the presence in the peri-tumour infiltrate of lymphocytes presenting immunoreactivity for LCK in 12 of 25 patients. In the other 13 patients, no expression of the LCK by the tumour-infiltrating lymphocytes was found. Patients have been divided into two groups according to the presence (LCK positive group) or the absence (LCK negative group) of LCK in the peri-tumour infiltrate.

Mean age of patients was 61.4 years (range 48–74) in the LCK-positive group and 63.2 years (range 50–75) in the LCK-negative group.

The distribution of patients according to major comorbidity, T stage (T1a, T1b or T2a) and histology (adenocarcinoma or squamous cell carcinoma) was similar between the two groups [Table 1].

Table 1:

Patients' characteristics

	LCK+ group (12 patients)	LCK− group (13 patients)
Stage
T1a	1	1
T1b	5	5
T2a	6	7
Histology
Adeno	8	9
Squamous	4	4
Comorbidity
Yes	7	8
No	5	5

Median follow-up time of the whole cohort of patients was 48 months (range 6–73). Median follow-up time of surviving patients was 56 months (range 48–73). At the time of last follow-up (31 January 2011), 9 of 12 patients of the LCK-positive group were alive with no evidence of recurrence, while 3 patients in this group were deceased due to tumour recurrence. In the LCK-negative group, four patients were alive with no evident tumour recurrence, while nine patients were dead due to the tumour relapse. Tumour recurrence sites in the LCK-positive group included: brain in one patient, lung in one and pleura in one. Tumour recurrence sites in the LCK-negative group included: brain in two patients, bone in one, liver in one, lung in two and multiple metastatic sites in three. Cytologic or histologic confirmation of tumour recurrence was obtained in all patients, except for those experiencing only brain metastases.

Survival longer than 40 months was registered in 75% of patients in the LCK-positive group and in 31% of patients in the LCK-negative group (P = 0.01; Fisher exact test).

Median survival time was not reached in the LCK-positive group and 30 months in the LCK-negative group. This difference was statistically significant (P = 0.01; log-rank test). Similarly, median time to relapse resulted significantly longer in the LCK-positive group if compared with that of the LCK-negative group (not reached vs. 25 months; P < 0.001—log-rank test).

DISCUSSION

The current staging system for lung cancer shows evident limitations since even patients with similar clinical and pathological features of the tumour may present variable biological behaviour of the disease and different oncological outcome.

Additional information on the biological characteristics of the tumour is therefore required, in order to increase the prognostic value of the pathological staging and to plan the most appropriate treatment strategy for each patient.

There are accumulating evidences that many developing tumours produce a response by the immune system, although this is usually insufficient to achieve a complete eradication of the disease.

The spontaneous tumour regression observed in patients with active autoimmune phenomena indicates a high potential efficacy of the immune system in the tumour rejection. These evidences are supported further by the finding of increased rate of tumours in immunosuppressed patients and animals. A significantly increased developing of tumours has been observed in mice deficient in immune function [3].

Several studies in the literature have shown a correlation between the presence of peritumoral and intratumoral T cells and an improved clinical outcome in different types of tumours, including ovarian, colorectal and oesophageal carcinoma [4–6], although the mechanisms of this anti-cancer activity have still not been completely defined. Moreover, there are some conflicting data suggesting that this favourable prognostic impact may be absent when the tumour infiltrating T cells are not activated. Based on these evidences, we have tried to evaluate the effect of the activated T cell on the biologic behaviour of early stage NSCLC by studying the intratumoral expression of an important CD8-positive T cell activation marker: the LCK.

The immune-editing hypothesis postulates that tumour-associated antigens are recognized by CD8-positive lymphocytes and a number of spontaneous tumours are destroyed in such a way [7].

LCK belongs to Src family tyrosine kinases and modulates a variety of cellular functions. Of note, LCK represents a critical effector of T cell receptor signalling by binding either CD4- or CD8-positive lymphocyte coreceptors [8]. In other words, LCK is a marker of immunosurveillance. Masaki et al. [9] showed LCK overexpression in some patients' lung adenocarcinoma, but its impact on the long-term prognosis has not been studied. Differently, LCK overexpression has been observed in other cancers (i.e. gastric cancer) by other authors [10], and it has been found related to a better prognosis.

Although our study has been conducted on a small statistical sample, we have found a significantly better long-term prognosis in patients showing the presence of LCK in the tumour infiltrate.

Since the study by Chen et al. which reported the first evidence of a significantly better outcome associated to the expression of the gene for LCK in NSCLC, no other investigation has been published assessing the role of LCK in early stage NSCLC prognosis.

The main limitations of this study include the lack of validation of the immunohistochemical findings by a gene expression analysis, and the bias in patient selection due to the enrolment in this retrospective analysis only of patients with available frozen specimens. Therefore, although the results observed in the present study provide some original information, data reported in this paper need to be confirmed in further larger studies with the inclusion of gene-expression analysis data and a prospective enrolment of patients. If the here-reported evidences concerning the role of LCK in NSCLC can be adequately validated, it could be hypothesized for the future a possible role of this biological marker in the clinical practice, since it would allow the identification of patients with a potentially more aggressive behaviour of the tumour, suggesting the indication for adjuvant therapy.