SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

p53 and epidermal growth factor receptor (EGFR) are common genes involved in the pathogenesis of lung cancer, but their roles in lymphoepithelioma-like carcinomas (LELC) are unclear. In this study, we investigate the roles of p53 and EGFR in LELC carcinogenesis. Forty-six pulmonary LELCs were identified to evaluate p53 and EGFR aberrations. p53 mutations were identified in three patients, which all occurred in exon 8. EGFR mutations were detected in 8 of 46 cases with a majority of exon 21 mutations but without L858R. The other cases harbored mutations in exons 20 and 18. Only one case gained a deletion in exon 19. Notably, EGFR mutation was more commonly observed in patients with tumor size ≤3 cm (P = 0.014). In addition, there was a trend of more common EGFR overexpression in female (22/30) than in male patients (7/16, P = 0.061). However, there was no correlation between p53/EGFR mutations and protein expressions, suggesting the presence of complex mechanisms. p53 and EGFR mutations are uncommon in LELCs, indicating that these genes are not the important events in carcinogenesis for this tumor subtype. The EGFR mutation in 35% patients with LELC tumors <3 cm in size suggests the potential benefits to EGFR tyrosine kinase inhibitors of inoperable LELCs. (Cancer Sci 2011; 102: 282–287)

The majority of tumors have evidence of mutational inactivation of tumor suppressor genes and activation of oncogenes. Mutation of one p53 allele and loss of the normal p53 allele occur in many tumors including lung cancers. These alterations apparently contribute to the development of cancer by interfering with the tumor suppressor activity of p53. In addition, overexpression of mutant p53 may cause cellular transformation, whereas transfection and expression of a normal p53 gene in cells having a mutated p53 can suppress cellular transformation and inhibit cell proliferation.(1) It has been reported that somatic mutations and increased expression of p53 are frequently found in approximately 23% and 65% of non-small-cell lung cancer (NSCLC), respectively.(2–4)

Epidermal growth factor receptor (EGFR), a transmembrane glycoprotein, is involved in promoting cell division, migration and angiogenesis, and inhibiting apoptosis.(5) Loss of control of EGFR because of deregulation, amplification, or mutations may result in malignant change of cells.(6) Gefitinib (Iressa, ZD1839; AstraZeneca, Wilmington, DE, USA), an inhibitor of EGFR tyrosine kinase, has a novel antitumor effect on NSCLC. The better treatment outcomes of tyrosine kinase inhibitor in NSCLC patients are strongly associated with mutations at the EGFR tyrosine kinase domain (exons 18–21) in EGFR of tumor tissues.(7,8) Depending on the tumor subtypes and different ethnicity, Caucasian NSCLC patients have an EGFR mutation frequency of <10% compared with a mutation rate of at least 30% in East Asian patients.(2,9,10)

Lymphoepithelioma-like carcinoma of the lung (LELC) was first reported in 1987.(11) It is a rare form of NSCLC predominantly affecting young non-smoking Asians(12–14) and has been described as being closely associated with Epstein–Barr virus (EBV) infection.(12,13) Lymphoepithelioma-like carcinoma of the lung shows a poorly differentiated epithelial component admixed with lymphocytic infiltrate(15) identical to undifferentiated nasopharyngeal carcinoma. It is also associated with a lower incidence of nodal disease or distant metastases and a better prognosis than other NSCLCs.(12,16,17) Due to its rarity, the treatment of advanced LELC is not only empirical, but controversial.(12,14)

In our previous study, we were unable to establish a significant role of p53 or bcl-2 in tumorigenesis of LELCs at the molecular level.(12) The aim of this study is to elucidate the mechanisms responsible for the development of LELCs by p53 and EGFR mutations, as well as p53 and EGFR expressions. Furthermore, the results of these gene expression statuses are also characterized for correlation with clinicopathologic features, including sex, smoking status, stage, tumor size, lymph node metastasis, and survival.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Patients.  We investigated a cohort of 46 individuals with pulmonary LELCs, including 18 cases from our previous report,(12) managed by our team at National Taiwan University Hospital (Taipei, Taiwan) from November 1, 1995 to March 31, 2009. Pulmonary LELCs were diagnosed according to the criteria set by the World Health Organization.(18) Undifferentiated carcinomas without dense lymphoid infiltrates and EBER-1 staining were excluded in this study.

Clinical data were tabulated from the patients’ medical records and analyzed in conjunction with gross and microscopic findings. Pathologic staging in 27 lobectomy, four bilobectomy, and five pneumonectomy cases, and clinical staging in six wedge resection, two lymph node dissection, and two biopsy cases were determined according to the International Union Against Cancer (UICC) TNM classification of malignant tumors.(19) Date of diagnosis, chemotherapy, radiotherapy, and combined chemoradiotherapy received were recorded. Written informed consent for the use of tissues in analysis was acquired from patients at the time of procurement of tumor specimens. The research ethics committee of the hospital approved this study.

Mutational analysis of p53 and EGFR.  Five-micrometer-thick paraffin sections were used for genomic DNA isolation. After deparaffinization with xylene, tissue sections were stained with hematoxylin, and areas were carefully microdissected with a laser microdissection system (LMD 7000; Leica, Wetzlar, Germany). An estimated 1000 microdissected tumor cells were digested in 50 μL buffer consisting of 20 mmol/L Tris–HCl (pH 8.0), 1 mmol/L EDTA (pH 8.0), 1% Tween 20, and 1 mg/mL proteinase K for 24 h at 56°C. The protease-treated DNA mixture was heat inactivated after incubation for 10 min at 95°C, as described by Sugio et al.(20)p53 exons 5, 6, 7, 8 and EGFR tyrosine kinase exons 18, 19, 20, 21 were amplified by nested PCR, using specific primers. The primers and amplification conditions have been described previously.(8) The resulting PCR amplicons were purified with a Gel/PCR DNA Fragments Extraction kit and sequenced using the BigDye Terminator kit (Applied Biosystems, Foster City, CA, USA) and ABI Prism 3700 DNA Analyzer (Applied Biosystems) according to the manufacturer’s instructions. All sequencing reactions were carried out in both forward and reverse directions, using tracings from at least two independent PCRs. Mutations were also checked against the corresponding sequences from an adjacent non-neoplastic lung tissue DNA and single nucleotide polymorphism database.

Immunohistochemical analysis of p53 and EGFR expre-ssion.  The immunohistochemical analysis of p53 and EGFR protein in the tumors was carried out as previously described.(2) Immunostaining was classified into two groups according to both intensity and extent: (i) negative, when no staining or positive staining was detected in ≤50% of the cells; or (ii) positive, when immunostaining was present in >50% of the cells. Two independent pathologists (Y.-L.C. and C.-T.W.) were involved in the assessment of the expression.

In situ hybridization of EBER-1.  The presence of EBV was examined by in situ hybridization using digoxigenin-labeled 30-base oligonucleotide probes targeted to EBER-1, as previously described.(12)

Statistical analyses.  The correlation between various clinicopathologic variables and the somatic aberrations of p53 and EGFR were analyzed by Fisher’s exact test. The effect of p53 and EGFR on survival was tested using Kaplan–Meier survival plots and analyzed using the log-rank test. All tests were two-sided, and P < 0.05 was considered significant. SPSS version 10.0 software (SPSS, Chicago, IL, USA) was used for all analyses.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Patient demographics.  Sixteen (34.8%) of the patients were men, and 30 (65.2%) patients were female. Six (13.0%) patients were smokers. The mean age at diagnosis was 57 years (range, 40–85 years). All patients had normal nasopharyngoscopic findings. The tumors ranged in size from 1.0 to 11.0 cm, including 20 tumors of 3 cm or less and 26 tumors of more than 3 cm in size (Table 1). All tumors were indistinguishable from lymphoepithelioma of the nasopharynx and were characterized by a seemingly syncytial pattern of growth, large vesicular nuclei, prominent nucleoli, and a heavy lymphocytic infiltration (Fig. 1A,B). In the tumor cells of all patients, the in situ hybridization for EBER-1 showed positive signals in the nuclei of the large undifferentiated neoplastic cells. The accompanying lymphocytes and plasma cells were negative.

Table 1.   Clinicopathologic characteristics of lymphoepithelioma-like carcinoma patients with p53 and epidermal growth factor receptor (EGFR) mutations
VariablesTotalp53 mutationEGFR mutation
(+)(−)P-value(+)(−)P-value
Patient no.46343 838 
Sex
 Male160160.5421150.230
 Female30327723
Smoking status
 Smokers6150.349151.000
 Non-smokers40238733
Tumor size (cm)
 ≤3202180.5727130.014
 >326125125
Lymph node metastasis
 Positive210210.2392190.260
 Negative25322619
Stage
 I and II263230.2465211.000
 III and IV20020317
p53 expression
 Positive9180.488270.645
 Negative37235631
EGFR expression
 Positive292271.0004250.443
 Negative17116413
image

Figure 1.  (A) Large tumor cells containing vesicular nuclei and prominent nucleoli intermixed with a lymphoid infiltrate (H&E). (B) The typical characters of cancer cells shown in high power.

Download figure to PowerPoint

The stages of LELC were 19 (41.3%) in stage I, 7 (15.2%) in stage II, 15 (32.6%) in stage III, and 5 (10.9%) in stage IV. Forty-two patients underwent surgery, nine of them with neoadjuvant chemotherapy (a combination of gemcitabine and cisplatin). Five patients received adjuvant concurrent chemoradiotherapy, two with adjuvant chemotherapy and one with adjuvant radiotherapy. For the two inoperable cases, concurrent chemoradiotherapy was taken. In addition, palliative therapy including chemotherapy, radiotherapy, and concurrent chemoradiotherapy was undertaken in four, two, and three patients, respectively. Most of the patients achieved partial response. Only one patient of stage IIIA LELC with wild-type EGFR received gefitinib treatment following the first-line chemotherapy with gemcitabine plus cisplatin and second-line chemotherapy with docetaxel due to progressive disease. However, the patient died of multiple distant metastases.

p53 and EGFR mutation analyses.  There were three p53 gene mutations in 46 (6.5%) pulmonary LELCs, which all occurred in exon 8. EGFR mutations were identified in eight tumors (17.4%). We did not find any mutations in the corresponding DNA from normal lung tissues, confirming that these mutations were somatic in origin. The type and location of the mutations and corresponding amino acid sequence alterations in tumors are shown in Table 2. Four of the patients (8.7%) had mutations in exon 21: one patient with A839T (with coexistent p53 mutation in exon 8 with 892G>T E298stop, Fig. 2A–C); one with G874S; one with a triple-mutation of E829K, R836C, plus H870Y; and one with E872G (who also had a simultaneous D800N mutation in exon 20). Three of the 46 (6.5%) LELCs harbored mutations in exon 20: one patient with V765M; one with D800N (as shown above with a simultaneous mutation in exon 21); and one with T785I. Another case had a double-mutation of N700Y plus E711K that was detected in exon 18. Only one case (2.2%) gained a deletion (E746–A750) in exon 19.

Table 2.   Mutations detected in p53 and epidermal growth factor receptor (EGFR) genes in patients with lymphoepithelioma-like carcinoma
CaseExonMutation inExonMutation in
p53 genep53 proteinEGFR geneEGFR protein
  1. Del, deletion; NA not applicable.

 1NANAWild182098A>T 2131G>AN700Y E711K
 2NANAWild19Del 2235–2249Del E746–A750
 3NANAWild202293G>AV765M
 4NANAWild202354C>TT785I
 5NANAWild20 212398G>A 2615A>GD800N E872G
 6NANAWild212620G>AG874S
 7NANAWild212485G>A 2506C>T 2608C>TE829K R836C H870Y
 88892G>TE298stop212515G>AA839T
 98817C>TR273CNANAWild
108835G>AG279RNANAWild
image

Figure 2.  (A) Hematoxylin–eosin histology of lympho-epithelioma-like carcinoma from case 8. Sequencing chromatograms of the same case show base pair changes. (B) p53 mutation. Sequencing of exon 8 shows a G>T base pair change (arrow). (C) Epider-mal growth factor receptor (EGFR) mutation. Sequencing of exon 21 shows a G>A base pair change (arrow).

Download figure to PowerPoint

Correlations of p53 and EGFR mutations with clinicopathologic features and survival.  We examined the relationship between the presence of p53 and EGFR mutations and several important parameters. EGFR mutations were observed more frequently (7 out of 20; 35.0%) in tumors of 3 cm or less than in tumors of more than 3 cm (1 out of 26; 3.8%; P = 0.014). There was no statistically significant correlation between EGFR mutation and other clinicopathologic features of sex, smoking status, lymph node metastasis, stage (Table 1), or survival (P = 0.079). p53 mutations were not associated with any clinicopathologic parameters or survival (P = 0.316).

Correlations of p53 and EGFR protein overexpression with clinicopathologic features and survival.  Overexpression of p53 protein (Fig. 3) was detected in 9 out of 46 cases (19.6%). There was no statistically significant correlation between p53 overexpression and sex, smoking status, tumor size, lymph node metastasis, stage (Table 3), or survival (= 0.563). However, positive EGFR expression (Fig. 4) was shown in 29 out of 46 cases (63.0%). There was a tendency for more frequent EGFR overexpression in female (22/30, 73.3%) than in male patients (7/16, 43.8%) (P = 0.061), whereas overexpression was not associated with other clinicopathologic parameters or survival (P = 0.789). Furthermore, our results showed that there was no correlation of p53/EGFR mutations with protein expression in LELCs (Table 1; Fig. 5A,B).

image

Figure 3.  p53 immunostain of lymphoepithelioma-like carcinoma cells showing diffuse positivity with strong intensity (avidin biotin complex method).

Download figure to PowerPoint

Table 3.   Clinicopathologic characteristics of lymphoepithelioma-like carcinoma patients with p53 and epidermal growth factor receptor (EGFR) protein expression
VariablesTotalp53 expressionEGFR expression
(+)(−)P(+)(−)P
Patient no.46937 2917 
Sex
 Male162140.463790.061
 Female30723228
Smoking status
 Smokers6060.327240.174
 Non-smokers409312713
Tumor size (cm)
 ≤3205150.4721190.369
 >326422188
Lymph node metastasis
 Positive215160.7111290.545
 Negative25421178
Stage
 I and II265211.00016101.000
 III and IV20416137
image

Figure 4.  Epidermal growth factor receptor immunostain of lympho-epithelioma-like carcinoma cells showing strong membranous staining (avidin biotin complex method).

Download figure to PowerPoint

image

Figure 5.  (A) p53 immunostain and (B) epidermal growth factor receptor immunostain of lympho-epithelioma-like carcinoma cells (case 8) are negative (ABC method).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

Primary LELC of the lung is a rare and clinicopathologically distinctive neoplasm. In the past two decades, there have been just over 150 cases reported in published work.(12,21,22) In our study, 16 patients were male and 30 patients were female. The male:female ratio was different from other published reports of 2.2:1(23) and 6:5(16); the male:female ratio of NSCLC was 2.2:1 in our previous study.(24) The patients ranged in age from 40 to 85 years (mean, 57 years), similar to previous reports.(23,25) The mean age of the patients with LELC was 10 years younger than that of other histological types of NSCLCs.(12)

The etiologic and molecular events responsible for the occurrence of LELC are almost entirely unknown. The presence of EBV was indicated by in situ hybridization for EBV DNA and RNA in all of our cases, suggesting a relationship between pulmonary LELC and EBV. It has been well documented that specific genetic alterations and aberrant gene expressions in tumor cells may influence the biologic behaviors of a tumor. p53 is a tumor suppressor gene involved in cell cycle control and preservation of genomic integrity. Mutation in the p53 gene is relatively frequent in lung cancer,(26) and it is mutated in approximately 50% of NSCLCs.(2,27) To the best of our knowledge, this is the first time a study has revealed that a very low frequency of somatic aberration of p53, suggesting different somatic or inherited genetic mechanisms, could underlie cancer development in these patients. Our results also indicate that rare p53 mutations in these cases were closely related to favorable survival in pulmonary LELC.(12) This observation is different from ordinary NSCLC results.(24,27)

The low frequency of association with cigarette smoking in LELC of the lung (13% in the current series and 40% in cases in published reports),(22,28) in contrast to other lung cancers (smoking history usually more than 60%),(29) suggests that smoking probably does not play an etiologic role, further mandating separation of this tumor from other lung cancers.(21,22)

Considerable evidence indicates that the EGFR pathway also plays an important role in both the pathogenesis and the progression of lung cancer.(30) A detailed study of mutation patterns within the EGFR gene and their associations with clinicopathologic features of lung cancers suggested that lung cancer patients whose tumors have EGFR mutations respond better than those without such mutations to drugs that target the EGFR tyrosine kinase domain.(7,8,31) However, there are no data addressing the EFGR mutation status and EGFR tyrosine kinase inhibitor treatment response in LELC of the lung. In this series, we tested a larger number of primary pulmonary LELCs for EGFR mutations and explored the relationship between EGFR mutation status and multiple clinicopathologic parameters. In contrast to previous reports that 40–80% of patients of East Asian ethnicity have NSCLCs with EGFR mutations,(2,10,32) we found that only 17.4% of patients with LELCs carried EGFR mutations. This relatively low EGFR mutation frequency existing in LELCs, concurring with the results reported by Tam et al.,(33) is a proof that it is a distinct entity of special interest, especially in the Asia-Pacific region.

It has been previously indicated that mutations of a subset of NSCLCs are limited to the first four exons (exons 18–21) of the tyrosine kinase. Three types of mutations constitute 94% of the total: deletions in exon 19; duplications and/or insertions in exon 20; and a single-point mutation in exon 21.(10) Other investigators have also noted that the predominant mutations are single classical mutations (del-19 or L858R),(32,34) which all flank the ATP-binding pocket that is important for tyrosine kinase activity.(7,8) Among the patients with single classical mutations, a better response to gefitinib than those without the classical mutations has been noted.(35,36) In our study, we identified four mutations in exon 21, all of them non-classical mutations. Surprisingly, a relatively high prevalence of exon 20 mutations, without duplications or insertions, was detected in three cases. One of them coexisted with an exon 21 mutation. According to the published reports, there is an average 34% gefitinib response rate for the exon 20 mutation, which is much lower than that for classical mutations in exons 19 and 21.(37–39) These results suggest that EGFR target therapy is not an encouraging treatment for patients with inoperable LELCs. Our experience with the chemotherapy regimen, a combination of gemcitabine and cisplatin, showed a good partial response without accompanying adverse reactions. Based on the better prognosis, we would recommend future use of combination chemotherapy or neoadjuvant chemotherapy with additional surgical resection in advanced LELC of the lung.

It has also been noticed that EGFR mutations are statistically significantly more frequent in females and non-smokers in adenocarcinoma.(7,8,10,32,39) However, these characteristics are not present in patients with LELC. The meaning of this discrepancy is not clear.

As in the reports mentioned before, EGFR mutations are not associated with stage or patient survival, suggesting that EGFR mutations occur relatively early in the clinical course and are associated with pathogenesis of adenocarcinoma rather than progression.(2,10,32) Interestingly, the correlation between EGFR mutations and various clinicopathologic factors of LELCs reveals that EGFR mutations are tightly associated with tumor sizes of 3 cm or less. This observation suggests that EGFR mutations may be responsible for the relatively different clinical and biologic behaviors.

In this study, we did not find a concordance between p53/EGFR mutations and p53/EGFR protein expressions in LELCs, suggesting the presence of complex mechanisms. There was a tendency for more common EGFR overexpression in female patients. Notably, the frequency of the expression of p53 protein (19.6%) was much higher than the results in our previous study.(12) This was because the authors used bifunctional SkipDewax pretreatment solution (Insitus Biotechnologies, Albuquerque, NM, USA) as the antigen retrieval technique in this study instead of citrate buffer, as previously used. This is a product that combines the three pretreatment steps, deparaffinization, rehydration, and unmasking in immunohistochemistry stains, which generally results in a higher proportion of positive cells, greater staining intensity, and lower backgrounds than citrate solution, thereby producing more consistent, reliable results.(40)

In conclusion, this study shows for the first time that p53 and EGFR mutations are infrequent in LELC of the lung. The absence of association between classical EGFR mutations and female non-smokers, and the significant correlation between EGFR mutation and a tumor size of 3 cm or less imply that LELC is clearly a distinct entity of NSCLCs. The 35%EGFR mutation rate in tumors <3 cm suggests the potential benefits of EGFR tyrosine kinase inhibitors of inoperable LELCs.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

The authors would like to thank Ms. Chih-Hsin Chen for her skillful technical support. This work was supported by the National Taiwan University Hospital, (NTUH. 99-S1290), and the National Science Council, Taiwan, (NSC 97-2314-B-002-044-MY3).

Disclosure Statement

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References

All the authors declare no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure Statement
  8. References