Abstract

Background: This study aims to investigate the frequency, distribution, and morphological/immunohistochemical features of epidermal growth factor receptor mutations and to examine the possible relationship between the material type and technical success of mutation analysis in Turkish population with non-small cell lung cancer.

Methods: Between September 2012 and December 2015, a total of 499 consecutive, treatment-naïve patients (437 males, 163 females; mean age 61 years; range, 30 to 84 years) with primary or metastatic non-small cell lung cancer who underwent epidermal growth factor receptor mutation testing using Sanger sequencing method were retrospectively analyzed. Archival records and hematoxylin-eosine and immunohistochemically stained sections were re-examined. The thyroid transcription factor-1 and napsin A immunohistochemical stains were performed on tissue array blocks.

Results: Seventy-five mutations were detected in 70 patients (14%). The success rate of testing and intact deoxyribonucleic acid fragment length were significantly higher in the cytological material, compared to tissue specimens (p<0.001). The mutation rate in adenocarcinomas was 33.9% for women and 9.4% for men. The most common mutation was L746-E750del in exon 19 (29.3%), followed by the L858R mutation in exon 21 (28%). The mutation rate was the highest in micropapillary (40%) and lowest in solid (5.4%) adenocarcinomas. All epidermal growth factor receptor mutations, except for one, were positive for the thyroid transcription factor-1. The single nucleotide polymorphism Q787Q in exon 20 was observed in 79.6% of patients.

Conclusion: The frequency and distribution of epidermal growth factor receptor mutations in the Turkish patients with non-small cell lung cancer are similar to the European populations. These results also demonstrate that cytological materials are highly reliable for epidermal growth factor receptor mutation testing, and the probability of detection of wild-type epidermal growth factor receptor is low in cases of thyroid transcription factor-1 negativity.

Lung carcinoma is the leading cause of cancer deaths worldwide.[1,2] The majority of lung carcinomas are non-small cell lung carcinomas (NSCLC), and adenocarcinoma is the most frequent type.[3,4] An important proportion of NSCLC cases is diagnosed at advanced stage and a candidate for chemotherapy and/or personalized therapies.[3,5] The epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) are important tools in the treatment of NSCLC patients whose tumors show mutations in the EGFR gene.[1,5] Therefore, molecular testing for EGFR mutation has become an essential part of pathological examination in the last decade. The EGFR mutation is more frequent in female and non-smoker patients.[3,6] The mutation rate also shows variation between different ethnic groups, being higher in Asians[1,2,7] and lower in Europeans.[3,5,8] Although there are some reports regarding EGFR mutations from Turkey, most of these are smaller series, and targeted methods are used in larger series.[4,9-18] As a result, little is known about the exact profile of EGFR mutations in the Turkish NSCLC patients.

Thyroid transcription factor-1 (TTF-1), a transcription factor involved in normal thyroid and lung development, plays an active role in sustaining lung cancer and has been proved to be a highly specific and sensitive immunohistochemical marker for lung adenocarcinomas.[19] The TTF-1 expression has been shown to be associated with female gender, never-smoking status, and presence of EGFR mutations.[19,20] Napsin A is an aspartic protease presenting in the epithelial cells of the lung and is expressed in type 2 pneumocytes and also in alveolar macrophages.[19] Napsin A s taining is associated with TTF-1 expression and presence of EGFR mutations.[19,20]

In the present study, we aimed to investigate the frequency and distribution of EGFR mutations in the Turkish population with NSCLC and to identify relationship between the TTF-1 and napsin A staining and EGFR mutations.

Methods

This single-center, retrospective study was conducted at Ankara University Faculty of Medicine between September 2012 and December 2015. A total of 600 consecutive, treatment-naïve patients (437 males, 163 females; mean age 61 years; range, 30 to 84 years) with primary or metastatic NSCLC who underwent EGFR mutation testing using the Sanger sequencing method in our pathology lab, irrespective of the tumor stage, were retrospectively analyzed. The specimens were obtained by resection (n=212), biopsy (n=196), and cytology (n=192). Of the tumors, 397 were primary and 203 were metastatic tumors. The primary source of the material was one of the two hospitals of our university in 409 patients and specimens were consultation materials sent for EGFR mutation analysis in the remaining 191 patients.

All hematoxylin and eosin (H-E) stained slides of 600 patients were retrieved from archives and reclassified according to the 2015 World Health Organization (WHO) classification of lung tumors. Carcinomas with adequate tumor tissues were reevaluated for the tumor patterns (i.e., acinar, lepidic, papillary, micropapillary or solid) using a cut-off value of 5%.

A written informed consent was obtained from each patient. The study protocol was approved by the Ankara University Medical School Human Research Ethics Committee (20^th April 2 015). The study was conducted in accordance with the principles of the Declaration of Helsinki.

Immunohistochemistry
The TTF-1 and/or napsin A stained slides were reviewed, when available. Tissue array blocks were prepared for 57 cases, which were not evaluated for TTF-1 and/or napsin A at the time of the initial diagnosis. Four-micrometer-thick sections were stained for TTF-1 (8G7G3/1, Thermo Fisher Scientific Inc., Waltham, MA, USA) and napsin A (TMU-AD2, Biocare Medical, Pacheco, CA, USA) by streptavidinbiotin peroxidase method using the UltraView Universal DAB Detection Kit on a Ventana (BenchMark XT, Woonsocket, RI, USA) stainer and evaluated on the light microscope (Olympus BX50, Olympus Inc., Tokyo, Japan). For stained cases for TTF-1 and/or napsin A at the time of the initial diagnosis (TTF-1 n=320; napsin A n=293), immunohistochemically stained sections were retrieved and reevaluated with the same criteria. In consultation cases, whose slides and paraffin blocks were not available for tissue array preparation, pathology reports from the external institutions were examined and the TTF-1 and/or napsin A staining results were recorded, if available (TTF-1 n=100; napsin A n=53).

Sequencing
The EGFR mutations in exons 18, 19, 20, and 21 were detected by polymerase chain reaction-based direct Sanger sequencing. Deoxyribonucleic acid (DNA) was extracted from the formalin-fixed, paraffin-embedded (FFPE) tissue obtained from resection or biopsy specimens. For cytological materials, air dried and May Grünwald Giemsa (MGG) stained slides were preferred, when cellularity was adequate; however, sections from the FFPE cell blocks were also used, where applicable. The DNA extraction was performed using a commercial kit (QIAamp DNA FFPE tissue kit, Qiagen, Hilden, Germany) in accordance with the manufacturer"s instructions. Sequencing was performed on an automated single-capillary genetic analyzer (ABI 310; Applied Biosystems, Foster City, CA, USA) with forward and reverse primers (Table 1) separately, and final nucleotide changes were detected by comparing the sequence with the National Center for Biotechnology Information database (reference sequence: NM_005228.2).

Table 1: Sequences of the primers for EGFR exon 18, 19, 20, and 21

Statistical analysis
Statistical analysis was performed using the SPSS for Windows version 15.0 software (SPSS Inc., Chicago, IL, USA). Descriptive data were expressed in mean ± standard deviation (SD), median (min-max), or number and frequency. The EGFR status and clinicopathologic features were analyzed using the chi-square or Fisher's exact test. Pearson correlation analysis was performed to analyze possible correlations between variables. A p value of <0.05 was considered statistically significant.

Results

EGFR mutations
Of a total of 600 patients included in the study, the EGFR mutation testing was successful in 499 patients. Clinical characteristics and the outcome of EGFR mutation analysis of successful cases are shown in Table 2. In 495 cases, analyses of all four exons were successful. In additional four cases, an EGFR mutation was detected in one exon, while the analysis of one or more of the remaining exons were unsuccessful. The overall mutation rate was 14% in this series, and of 141 women, 43 (30.5%) showed EGFR mutation, which was more frequent than men (27/358, 7.5%) (p<0.001). The EGFR mutations were more frequent in adenocarcinomas (33.9% in women, 9.4% in men), compared to NSCLC-not otherwise specified cases (p=0.001). In 67 cases, either DNA extraction was unsuccessful or we were unable to obtain any amplicon. Mutation analysis was able to be performed for only one, two, and three exons; in one, seven, and 30 cases, respectively (n=4 among 38 cases who showed EGFR mutation in one exon and were considered successful per protocol).

Table 2: Clinical characteristics and results of EGFR mutation testing (n=499)

Regarding the specimen type, cytological specimens showed the best DNA quality, median DNA fragment length being 600 bp, 400 bp, and 400 bp in cytological, resection and biopsy specimens, respectively (p<0.001). Analytical success was higher in cytological specimens (p<0.001); 94.8% of cytological specimens, 77.8% of resectional specimens, and 75.5% of biopsy specimens were examined for all four exons. However, analytical success was poor in decalcified specimens (p<0.001) and in consultation cases (p=0.02).

In this series, a total of 75 mutations were detected in 70 of 499 cases (14%); 5 mutations in exon 18 (6.7%), 40 mutations in exon 19 (53.3%), seven mutations in exon 20 (9.3%), and 23 mutations in exon 21 (30.7%). Of 75 mutations including double mutations in five patients are documented in Table 3 and Figure 1. In addition, 79.6% of our cases showed heterozygous or homozygous single nucleotide polymorphism (SNP) Q787Q in exon 20.

Table 3: Genomic alterations in tyrosine kinase domain (exons 18-21) of EGFR gene

Figure 1: DNA sequencing electropherogram of a doublet mutation (p.H870R+p.L858R) in exon 21 of EGFR gene.
DNA: Deoxyribonucleic acid; EGFR: Epidermal growth factor receptor.

In 179 cases, a predominant adenocarcinoma pattern was evaluated. The most frequent primary pattern was acinar pattern (51.4%), followed by solid (22.9%), lepidic (17.9%), papillary (4.5%), and micropapillary (3.4%) patterns. The EGFR mutations were the most frequently detected in micropapillary adenocarcinomas and rare in solid adenocarcinomas, indicating a significant correlation between the EGFR mutation status and dominant pattern (p=0.029). In 11 of 14 (78.6%) EGFR-mutant acinar adenocarcinomas, mutation was located on exon 19; however, it did not reach statistical significance.

TTF-1 and napsin A
The TTF-1 and napsin A expressions were detected in 477 and 403 cases, respectively. In 75.7% and 72.5% of the cases, TTF-1 and napsin A were positive, respectively. Among 57 EGFR-mutant cases, whose TTF-1 expression was also detected, 56 (98.2%) were positive and only one case was negative for TTF-1. Among 48 EGFR-mutant cases, whose napsin A expression was detected, 46 (95.8%) were positive and only two cases were negative for napsin A. The TTF1 positivity and napsin A staining were strongly correlated with the presence of EGFR mutations (p<0.001).

Discussion

The role of EGFR in carcinogenesis clarified in the 1980s and the first EGFR-TKIs were synthesized in the 1990s. The EGFR somatic mutation frequencies differ among ethnic groups and geographic regions, and Asians show the highest EGFR mutation prevalence worldwide ranging from 33.7 to 59%,[1,2,7] followed by Latin Americans (26%),[21] and it is the lowest in Europeans (5.4 to 15%).[3,5,8]

Our study represents the largest, single-center experience on EGFR mutation status of the Turkish NSCLC patients as assessed by a screening method. We found EGFR mutation in 14% of our patients, which is consistent with the results of other studies from Europe.[8] The results of studies on the Turkish NSCLC patients show a great variation and ranges from 4 to 48.1%.[4,9-18] The number of cases in the series reporting the highest (48.1%, 37.5%, and 44%, respectively) and lowest (4%, 7.1%, 8%, respectively) frequency of EGFR mutations varies between 40 and 52.[9,10,15] Larger series from Turkey including 122, 218, 300, and 959 cases showed an EGFR mutation rate of 14.39%, 28.9%, 15%, and 16.7%, respectively.[4,13,17,18] This discrepancy may be related with small sample size and different methods used in these studies. Even in some of these studies, different methods were used for EGFR mutation testing within the same study,[4,17] and all of four exons were not examined in some others.[9,14,15] In our study, the Sanger sequencing method, which has a low sensitivity, was used, and tumor concentration increased by dissection of tumor from slides. As a result, we believe that 14% EGFR mutation frequency represents the Turkish NSCLC patients successfully. We also believe that using a screening method, which investigate both known and unknown mutations, is more useful than the targeted methods to detect all mutations present in a newly studied population similar to ours. However, this rate can be expected to increase with a more sensitive analysis method. Similar to previous studies, the most common EGFR mutations were short, in-frame deletions in exon 19, and L858R point mutation in exon 21.[1,3,7] The mutations observed in exon 19 are ranked first in terms of the frequency of EGFR mutations and are reported to be associated with a good response to EGFR-TKIs.[1,7,8] Among all EGFR mutations, the exon 19 mutation rate was found to be 31.55 to 67.8% in European[3,5,8] and 39.4 to 48% in Asian populations.[1,2,7] In this study, EGFR mutations in exon 19 represented 53.3% of all EGFR mutations, and 55% of exon 19 mutations were p.E746_A750del, similar to a previous study.[3] In our study, we detected very rare mutational events in exon 19 (p.746_753insATAT and p.T751_A755insAT), which has not been reported in the literature to date.

Exon 21 mutations are the second most common type of mutations among EGFR mutant cases, p.L858R being the most frequent.[1,5,18] Accordingly, the frequency of exon 21 mutations was 30.7% in this study, while the exon 21 mutation rate was 27.04 to 46.4% in European[3,5,8] and 43 to 49.8% in Asian cohorts.[1,2,7]

Mutations observed in exon 20 are mainly observed as insertion or point mutations and associated with de novo resistance to EGFR-TKIs.[1] The exon 20 mutation rate was reported to be 1.29 to 10.3% in European[3,5,8] and 3.4 to 9.3% in Asian cohorts.[1,2,7] In our study, we observed that EGFR mutations in exon 20 constituted 9.3% of all EGFR mutations.

Mutations in exon 18 are relatively rare and most are point mutations.[7] Frequency of mutations in exon 18 are reported to be 1.29 to 3.2% in European[3,5,8] and 3 to 6.5% in Asian populations.[1,2,7] In our study, EGFR mutations in exon 18 represented 6.7% of all EGFR mutations, and we identified five patients (7.1%) with complex mutations (p.T790M+p.L858R, p.L858R+p.H870R, p.E709A+p.G719A, p.L858R+p. V834L, p.G719C+p.S768I).

In previous studies investigating the relationship between the dominant adenocarcinoma pattern and EGFR mutations, solid adenocarcinomas were mostly associated with the wild-type EGFR.[6,22] Similarly, the lowest EGFR mutation rate was observed in solid adenocarcinomas in the present study. In addition, micropapillary adenocarcinomas showed the highest incidence of EGFR mutations consistent with some previous reports.[6,22]

On the other hand, there is a limited number of studies evaluating the relationship between the mutation types and dominant pattern. As shown in a previous study, reporting exon 19 mutations were more common in acinar adenocarcinomas, and 11 of 14 cases (78.6%) showed mutations in exon 19 with an acinar pattern.[22] This rate is of relevance, when compared to 53.3% overall frequency of exon 19 mutations in this series. Despite this, it did not reach statistical significance (p>0.05). However, further studies with larger series may be useful to reveal a significant relationship.

In previous studies, TTF-1 positivity was observed in 89.4 to 99% of EGFR-mutant tumors, whereas TTF-1 was negative in only 1 to 10.6% of cases.[4,20] Our study results showed a significant correlation between the EGFR mutational status and TTF-1 protein expression, indicating that patients with TTF-1-negative adenocarcinomas had at least a 99% chance of being wild-type EGFR. Although TTF-1 status should not be used to include or exclude cases for EGFR testing, it may be used for prioritizing molecular tests in resource-poor settings or when the available tissue is not adequate for all molecular tests.

Lee et al.[19] and Jie-Liu et al.[20] reported that napsin A positivity rate was 94.6%, and 72.3%, respectively in EGFR mutant cases. Our study results demonstrated a significant association between napsin A expression and EGFR mutation, consistent with the results of Lee et al[19] and Jie et al.[20]

Furthermore, in our study we observed a high frequency (79.6%) of SNP Q787Q in the EGFR gene which is more common in Caucasians (23 to 28.3%)[23] than in Asians (70.6 to 82.7%).[24,25] This SNP frequency was 79% in the EGFR-mutant cases and 80% in wildtype cases, probably due to ethnic origin, rather than a disease-related phenomenon.

In the present study, we also showed that cytological materials were highly reliable for EGFR mutation testing. Cytological specimens showed a higher DNA fragment length, compared to surgical and biopsy specimens, showing a higher DNA quality, which can be attributed to the lack of damaging effect of formalin fixation on DNA. Analytical success rate was also higher in cytological specimens.

The main limitation of the study is the low sensitivity of Sanger sequencing. Variants with low allele frequency were unable to be detected in the FFPE blocks or cytological specimens which had limited tumor cells (>20%).

In conclusion, the frequency and distribution of epidermal growth factor receptor mutations in the Turkish patients with non-small cell lung cancer a re similar to the European populations. These results also demonstrate that cytological materials are highly reliable for epidermal growth factor receptor mutation testing, and thyroid transcription factor-1 negativity appears to be a good predictor of wildtype mutations. Further large-scale, prospective studies are needed to gain a better understanding of this issue.

Declaration of conflicting interests
The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding
The authors received no financial support for the research and/or authorship of this article.

References

1) Lai Y, Zhang Z, Li J, Sun D, Zhou Y, Jiang T, et al. EGFR mutations in surgically resected fresh specimens from 697 consecutive Chinese patients with non-small cell lung cancer and their relationships with clinical features. Int J Mol Sci 2013;14:24549-59.

2) Sun PL, Seol H, Lee HJ, Yoo SB, Kim H, Xu X, et al. High incidence of EGFR mutations in Korean men smokers with no intratumoral heterogeneity of lung adenocarcinomas: correlation with histologic subtypes, EGFR/TTF-1 expressions, and clinical features. J Thorac Oncol 2012;7:323-30.

3) Papadopoulou E, Tsoulos N, Tsirigoti A, Apessos A, Agiannitopoulos K, Metaxa-Mariatou V, et al. Determination of EGFR and KRAS mutational status in Greek non-smallcell lung cancer patients. Oncol Lett 2015;10:2176-84.

4) Ozluk Y, Firat P, Yegen G, Hocaoglu J, Tas S, Yilmazbayhan D. EGFR mutation testing using archivalstained smears in non-small cell lung carcinoma. Cytopathology 2017;28:35-45.

5) Skov BG, Høgdall E, Clementsen P, Krasnik M, Larsen KR, Sørensen JB, et al. The prevalence of EGFR mutations in non-small cell lung cancer in an unselected Caucasian population. APMIS 2015;123:108-15.

6) Chen Z, Liu X, Zhao J, Yang H, Teng X. Correlation of EGFR mutation and histological subtype according to the IASLC/ ATS/ERS classification of lung adenocarcinoma. Int J Clin Exp Pathol 2014;7:8039-45.

7) Ninomiya H, Hiramatsu M, Inamura K, Nomura K, Okui M, Miyoshi T, et al. Correlation between morphology and EGFR mutations in lung adenocarcinomas Significance of the micropapillary pattern and the hobnail cell type. Lung Cancer 2009;63:235-40.

8) Szumera-Cie?kiewicz A, Olszewski WT, Tysarowski A, Kowalski DM, G?ogowski M, Krzakowski M, et al. EGFR mutation testing on cytological and histological samples in non-small cell lung cancer: a Polish, single institution study and systematic review of European incidence. Int J Clin Exp Pathol 2013;6:2800-12.

9) Cetin Z, Ozbilim G, Erdogan A, Luleci G, Karauzum SB. Evaluation of PTEN and Mcl-1 expressions in NSCLC expressing wild-type or mutated EGFR. Med Oncol 2010;27:853-60.

10) Baykara O, Tansarikaya M, Demirkaya A, Kaynak K, Tanju S, Toker A, et al. Association of epidermal growth factor receptor and K-Ras mutations with smoking history in non-small cell lung cancer patients. Exp Ther Med 2013;5:495-8.

11) Akca H, Demiray A, Yaren A, Bir F, Koseler A, Iwakawa R, et al. Utility of serum DNA and pyrosequencing for the detection of EGFR mutations in non-small cell lung cancer. Cancer Genet 2013;206:73-80.

12) Unal OU, Oztop I, Calibasi G, Baskin Y, Koca D, Demir N, et al. Relationship between epidermal growth factor receptor gene mutations and clinicopathological features in patients with non-small cell lung cancer in western Turkey. Asian Pac J Cancer Prev 2013;14:3705-9.

13) Sag SO, Gorukmez O, Ture M, Gorukmez O, Deligonul A, Sahinturk S, et al. Spectrum of EGFR gene mutations and ALK rearrangements in lung cancer patients in Turkey. Springerplus 2016;5:482.

14) Bircan S, Baloglu H, Kucukodaci Z, Bircan A. EGFR and KRAS mutations in Turkish non-small cell lung cancer patients: a pilot study. Med Oncol 2014;31:87.

15) Dogan M, Demirkazik A, Tukun A, Sak SD, Ceyhan K, Yalcin B, et al. The Relationship Between Common EGFR, BRAF, KRAS Mutations and Prognosis in Advanced Stage Non-Small Cell Lung Cancer with Response to the Treatment in Turkey. UHOD 2014;24:1-10.

16) Kus T, Aktas G, Sevinc A, Kalender ME, Yilmaz M, Kul S, et al. Prognostic impact of initial maximum standardized uptake value of (18)F-FDG PET/CT on treatment response in patients with metastatic lung adenocarcinoma treated with erlotinib. Onco Targets Ther 2015;8:3749-56.

17) Kanmaz ZD, Aras G, Tuncay E, Bahadır A, Kocatürk C, Yaşar ZA, et al. Contribution of 18Fluorodeoxyglucose positron emission tomography uptake and TTF-1 expression in the evaluation of the EGFR mutation in patients with lung adenocarcinoma. Cancer Biomark 2016;16:489-98.

18) Güler Tezel G, Şener E, Aydın Ç, Önder S. Prevalence of epidermal growth factor receptor mutations in patients with non-small cell lung cancer in turkish population. Balkan Med J 2017;34:567-71.

19) Lee JG, Kim S, Shim HS. Napsin A is an independent prognostic factor in surgically resected adenocarcinoma of the lung. Lung Cancer 2012;77:156-61.

20) Jie-Liu, Li XY, Zhao YQ, Liu RQ, Zhang JB, Ma J, et al. Genotype-phenotype correlation in Chinese patients with pulmonary mixed type adenocarcinoma: Relationship between histologic subtypes, TITF-1/SP-A expressions and EGFR mutations. Pathol Res Pract 2014;210:176-81.

21) Arrieta O, Cardona AF, Martín C, Más-López L, Corrales- Rodríguez L, Bramuglia G, et al. Updated Frequency of EGFR and KRAS Mutations in NonSmall-Cell Lung Cancer in Latin America: The Latin-American Consortium for the Investigation of Lung Cancer (CLICaP). J Thorac Oncol 2015;10:838-43.

22) Lu F, Li S, Dong B, Zhang S, Lv C, Yang Y. Identification of lung adenocarcinoma mutation status based on histologic subtype: Retrospective analysis of 269 patients. Thorac Cancer 2016;7:17-23.

23) Leichsenring J, Volckmar AL, Magios N, Morais de Oliveira CM, Penzel R, Brandt R, et al. Synonymous EGFR variant p.Q787Q is neither prognostic nor predictive in patients with lung adenocarcinoma. Genes Chromosomes Cancer 2017;56:214-20.

24) Lee GH, Choi YM, Kim JM, Shin JJ, Kim JG, Moon SY. Association of epidermal growth factor receptor gene polymorphisms with advanced endometriosis in a Korean population. Eur J Obstet Gynecol Reprod Biol 2012;164:196-9.

25) Sasaki H, Endo K, Takada M, Kawahara M, Tanaka H, Kitahara N, et al. EGFR polymorphism of the kinase domain in Japanese lung cancer. J Surg Res 2008;148:260-3.