bs_bs_banner
Asia-Pacific Journal of Clinical Oncology 2014; 10: 340–345
doi: 10.1111/ajco.12177
ORIGINAL ARTICLE
EGFR gene mutations in patients with adenosquamous lung carcinoma Tomasz POWRÓZEK,1 Paweł KRAWCZYK,1 Rodryg RAMLAU,4,6 Sylwia SURA,1 Kamila WOJAS-KRAWCZYK,1 Tomasz KUCHARCZYK,1,8 Beata WALCZYNA,2 Justyna SZUMIŁO,2 Katarzyna SZYSZKA-BARTH,6 Piotr MILECKI,5,7 Aleksander BARINOW-WOJEWÓDZKI6 and Janusz MILANOWSKI1,3 1
Pneumonology, Oncology and Allergology Department, 2Clinical Pathomorphology Department, Medical University, and Institute of Agricultural Medicine, Lublin, 4Thoracic Surgery Department, Chair of Cardio-Thoracic Surgery, 5Electroradiology Department, University of Medical Sciences, and 6Wielkopolskie Center of Pulmonology and Thoracosurgery of Eugenia and Janusz Zeyland, and 7Radiotherapy Department, Wielkopolskie Cancer Center, Poznan, and 8Postgraduate School of Molecular Medicine, Medical University, Warsaw, Poland 3
Abstract Aim: Adenosquamous (ADSQ) carcinoma accounts for 1–4% of non-small cell lung cancer (NSCLC). The origin of ADSQ carcinoma and its genetic background is not fully understood. Most studies concerning epidermal growth factor receptor (EGFR) mutation status are performed in adenocarcinoma, while there is limited information about the prevalence of this mutation in ADSQ-bearing Caucasian patients and the efficacy of EGFR tyrosine kinase inhibitors. Methods: EGFR gene status has been examined in 1000 non-squamous NSCLC patients of Polish origin. Polymerase chain reaction (PCR) followed by DNA fragment length analysis and allele-specific PCR as well as real-time PCR technique were used to estimate EGFR gene status. Complete clinical data were obtained for all examined patients. Results: In the group of 1000 non-squamous NSCLC patients, ADSQ was diagnosed in 14 (1.4%) cases. Activating mutations of EGFR were observed in 28.6% (four out of 14) of ADSQ-bearing patients and included deletions of 15 base-pairs in exon 19 in three cases (one man and two women) and substitution of L861Q with coexistence of G719X mutation in one non-smoking male patient. Deletions were diagnosed in two non-smoking patients and one current-smoking female patient (50 pack-years). One non-smoking man with deletion in exon 19 of EGFR gene was successfully treated with gefitinib in first-line therapy. Conclusions: EGFR gene mutations in ADSQ carcinoma patients may be more common than previously thought. EGFR mutation testing is appropriate in ADSQ-bearing patients, in which response for molecularbased therapy is predictable. Key words: EGFR mutation, gefitinib, lung adenosquamous carcinoma.
INTRODUCTION Correspondence: Professor Paweł Krawczyk MD PhD, Pneumonology, Oncology and Allergology Department, Medical University of Lublin, Jaczewskiego 8, 20-954 Lublin, Poland. Email: [email protected] Conflict of interest: none Accepted for publication 29 December 2013.
© 2014 Wiley Publishing Asia Pty Ltd
Squamous cell carcinoma, adenocarcinoma and largecell carcinoma accounted for 95 percent of non-small cell lung cancer (NSCLC) cases and the differences in etiology and biology between these types are frequently observed. Adenocarcinoma is most often diagnosed in
341
EGFR mutations in ADSQ cell carcinoma
non-smoking women; moreover, epidermal growth factor receptor (EGFR) gene mutations are reported far more frequent in this type of cancer than in other diagnosed NSCLC tumors. Squamous cell carcinoma occurs mainly in long-term tobacco smoking patients, very rarely in non-smokers, and EGFR gene mutations are diagnosed much less frequent than in adenocarcinoma.1 According to the 2009 World Health Organization classification, tumor containing of both squamous and adenocarcinoma component of at least 10 percent of each, is defined as adenosquamous (ADSQ) carcinoma. ADSQ carcinoma is a rare (1–4 percent of NSCLC cases), but very aggressive lung cancer with 5-year overall survival lower of 20 percent when compared with adenocarcinoma and squamous cell carcinoma patients.2 The etiology of ADSQ is similar to adenocarcinoma in terms of age of patients, smoking status and ethnicity.3 However, there is limited information about genetic alterations and tumor biology of ADSQ carcinoma. Therefore, the tumor morphology and the scarcity of cytological materials could not be sufficient enough for pathomorphological diagnosis of ADSQ. In many cases, tumor samples of ADSQ histology had not been considered for EGFR mutation testing. In these cases, immunohistochemistry (IHC) staining and molecular techniques could be an additional tool for diagnostic. The recent studies showed that EGFRactivating mutations are observed in 15–40 percent of ADSQ carcinoma-bearing patients which are likely to benefit from molecularly targeted drugs.2 The bad prognosis of ADSQ carcinoma-bearing patients has encouraged searching for more effective treatments. Some expectations are associated with the EGFR tyrosine kinase inhibitors (EGFR-TKI), so the evaluation of EGFR gene mutation status could be worthwhile. The increased incidence of ADSQ carcinoma in Caucasian population and the deficient in information about their genetic status has led us to intensive histological and molecular analysis to qualify the patients to more precise treatment.2,4
METHODS EGFR gene status has been examined in 1000 nonsquamous NSCLC patients of Polish origin. DNA was isolated from different types of materials (including paraffin-embedded tissue, cell blocks, fresh human tissue) using the QIAamp DNA FFPE Tissue Kit (Qiagen, Hilden, Germany) in accordance with the manufacturer’s instructions. Polymerase chain reaction (PCR) followed by DNA fragment length analysis and
Asia-Pac J Clin Oncol 2014; 10: 340–345
allele-specific PCR techniques with Cy5 fluorescentlabeled primers were applied. Analysis was performed using an ALF Express II sequencer and ALFWin Fragment Analysis software (Amersham Pharmacia, Biosciences, Amersham Place, Little Chalfont, UK). Moreover, real-time PCR technique (m2000rt, Abbott Molecular, Abbott Park, IL, USA) with fluorescent-labeled probes (Entrogen, Tarzana, Los Angeles, CA, USA) were also used. The sensitivity of the techniques was sufficient to perform EGFR mutation analysis in all examined samples. Objective response to EGFR-TKI treatment was evaluated with Response Evaluation Criteria in Solid Tumors criteria.
RESULTS In the group of 1000 non-squamous NSCLC patients, according to IHC staining (Fig. 1) ADSQ was diagnosed in 14 (1.4 percent) cases with a similar incidence in women and men. Activating mutations of EGFR gene were observed in 28.6 percent (four out of 14) of ADSQbearing patients and included deletions of 15 base-pairs in exon 19 in three cases (one men and two women) and substitution of L861Q with coexistence of G719X mutation in one non-smoking male patient. Deletions were diagnosed in two non-smoking patients and one current-smoking female patient (50 pack-years). The analysis of EGFR gene mutations status in different histological types of NSCLC is presented in Table 1. Clinical characteristics of ADSQ carcinoma-bearing patients with defined EGFR status are presented in Table 2. The EGFR mutations are observed in similar frequency in female as in male. The female patient W.M., with EGFR gene mutation diagnosed in post-surgical material, did not receive EGFR-TKI therapy, as she was qualified to adjuvant cisplatin/gemcitabine-based therapy. The patient received two courses of the treatment and she had been excluded from further systemic therapy because of deterioration of kidney parameters, decrease of overall performance status and disease progression. The female patient I.W., did not require further systemic treatment after surgical resection. The male patient J.P., in stage IV of ADSQ carcinoma and with the distant metastases observed in the central nervous system, in the bone and in the liver, had been previously treated with central nervous system and spine radiotherapy. He received gefitinib (250 mg daily) in first-line therapy and during 5 months of treatment, the disease remission and the rash of III grade has been observed. Complete remission of brain metastases (Fig. 2a,b) and
© 2014 Wiley Publishing Asia Pty Ltd
342
(a)
(b)
Figure 1 (a) The representative immunohistochemistry staining of adenosquamous carcinoma tumor sample. Positive reaction with thyroid transcription factor 1 (TTF-1) antibody is visible in nuclei of adenocarcinoma cell component (tumor cells are present in vessel lumen). Internal control is visible as a positive reaction with TTF-1 antibody in pneumocytes. Negative nuclear reaction with TTF-1 antibody is observed in squamous carcinoma pattern. Background is slightly colored. (b) Positive reaction with p63 antibody is visible in nuclei of squamous carcinoma cell component.
partial remission of lung tumor (Fig. 3a,b) as well as liver (Fig. 4a,b) and bones metastases were observed. After 9 months of treatment, the patient remains in good condition. The male patient K.L. has been diagnosed recently and he is being qualified to erlotinib therapy.
DISCUSSION The EGFR gene mutations are rarely detected in tumors of histology other than adenocarcinoma, especially in
© 2014 Wiley Publishing Asia Pty Ltd
T Powrózek et al.
cancer of squamous histology. This type of cancer, in contrast to adenocarcinoma, is strongly associated with smoking history. However, in NSCLC patients of mixed histology of tumor, EGFR mutations had been detected in tumor cells of both adeno- and squamous pattern. Moreover, tumor cells of ADSQ carcinoma histology could be morphologically different from “pure” adenocarcinoma or squamous carcinoma cells.2,5 Therefore, the point to discussion is polyclonality or monoclonality of ADSQ carcinoma. These observations allow us to conclude that EGFR gene mutations could be observed even in progenitor cells. Under specific effect of these mutations, the progenitor cells could develop into both types of tumor cells.5 The presence of identical mutations in both tumor cells’ components could explain the monoclonality of carcinogenesis’ pathway. However, it should be noted that exogenous carcinogenic factors could act on bronchial epithelial cells. The combined effect of these factors resulted in other genetic abnormalities in both types of tumor cells component. Moreover, the protein expression profile is changing what could lead to the differentiation of progenitor cells into two types of phenotypical different lung cancers. Furthermore, the “collision” and the interaction of these cells in later multistage carcinogenesis results in secondary polyclonality of ADSQ cancer cells.2,5 Even though ADSQ carcinoma appears to be a rare subtype of lung cancer, some cases might have been undiagnosed or misdiagnosed as a different histological subtype. It is still not precisely described if squamous tumors had a component of adenocarcinoma cells and what is the ratio of this component to the entire tumor cells. It was mainly by unappreciated immunohistochemical staining and abusing of only hematoxilin and eosin staining for the identification of NSCLC types. The representative image of IHC staining used for ADSQ carcinoma diagnosis in our patients is showed on Figure 1. Therefore, many tissues samples could have been omitted or rejected from further EGFR gene mutations analysis, and the patients were not qualified to EGFR-TKI therapy. This could also result from the etiology of ADSQ carcinoma, which is a type of lung cancer connected with tobacco smoking, while EGFR gene mutations are more often observed in non-smokers. Currently, for precise identification of NSCLC subtypes by IHC staining, the specific monoclonal antibodies are used, for example anti-thyroid transcription factor 1 (TTF-1, for adenocarcinoma) and anti-p63 (for squamous cell carcinoma). Nevertheless, the expression
Asia-Pac J Clin Oncol 2014; 10: 340–345
343
EGFR mutations in ADSQ cell carcinoma
Table 1 Clinical characteristics of patients according to the pathological diagnosis and epidermal growth factor receptor (EGFR) mutation status EGFR gene status EGFR gene mutations (n = 90; 9%)
Pathomorphological diagnosis (n = 1000) Adenocarcinoma (n = 806; 80.6%) Large-cell carcinoma (n = 43; 4.3%) Adenosquamous carcinoma (n = 14; 1.4%) Not otherwise specified (n = 137; 13.7%)
Table 2
Wild-type EGFR gene (n = 910, 91%
Exon 19 (deletion)
Exon 21 (L858R)
726 (90.07%) 41 (95.34%) 10 (71.43%) 133 (97.08%)
44 (5.46%) 1 (2.33%) 3 (21.43%) 2 (1.46%)
33 (4.09%) 1 (2.33%) – 2 (1.46%)
Exon 20 (insertion)
Exon 18 (S768I and G719X)
Exon 18 and 21 (G719X and L861Q)
2 (0.25%) – – –
1 (0.13%) – – –
– – 1 (7.14%) –
Clinical and molecular characteristics of adenosquamous carcinoma patients
Patient
Sex
Age
Smoking status
Stage of disease/procedures for material collection
EGFR gene mutation status
W.M. R.S. S.C. J.K. I.W. J.P. E.S. T.G. J.C. L.K. A.S. J.F. M.P. K.L.
Female Male Female Male Female Male Male Male Female Male Male Male Male Male
73 71 77 74 68 58 62 64 77 46 68 48 69 51
Current (50 pack-years) Current (60 pack-years) NA Non-smoker Non-smoker Non-smoker Current (40 pack-years) Current (50 pack-years) NA Current (60 pack-years) NA Non-smoker Ex-smoker (35 pack-years) Non-smoker
IIIB/wedge resection of the tumor† IIIB/endobronchial biopsy IIIB/lobectomy† IV/resection of brain metastase IIA/lobectomy IV/FNA IIIB/FNA IIIB/endobronchial biopsy IIB/wedge resection of the tumor IIIA/wedge resection of tumor IIB/wedge resection of tumor IIIA/lobectomy IIIA/endobronchial biopsy IV/endobronchial biopsy
Ex 19 deletion Wild type Wild type Wild type Ex 19 deletion Ex 19 deletion Wild type Wild type Wild type Wild type Wild type Wild type Wild type L861Q and G719X substitution
†
Stage of the disease was estimated during qualification for epidermal growth factor receptor- tyrosine kinase inhibitors treatment; surgical material was obtained in earlier stages than mentioned in table. FNA, fine needle aspiration; NA, data not available.
of TTF-1 antigen is sometimes observed in squamous cell carcinoma and p63 antigen expression is weakly or locally observed in adenocarcinoma. Hence, the expanding of IHC panel of additional antibodies (e.g. anti-CK5/6 and anti-CK34βE12 for squamous cell carcinoma, anti-CK7, anti-SPA, anti-SPB [surfactant proteins A and B] and anti-napsin A for adenocarcinoma, especially with a negative TTF-1 staining), which allow precise determination of histological type of NSCLC should be necessarily performed. According to the problems described above, little is known about the frequency of EGFR gene mutations in ADSQ carcinoma patients of different ethnicity, especially of Caucasian origin. The results of the latest
Asia-Pac J Clin Oncol 2014; 10: 340–345
studies regard mainly patients of Asian origin and concluded that EGFR mutations among ADSQ carcinoma patients are much more common than it was previously thought. Kang et al.6 had found 11 (44 percent) EGFRactivating mutations in the group of 25 ADSQ carcinoma-bearing patients of Korean origin (2.5 percent of NSCLC patients). Toyooka et al.1 had found three (27.3 percent) EGFR mutations in 11 patients of Japanese origin with ADSQ carcinoma diagnosis (2.8 percent of NSCLC patients). These results suggest that the differences in the incidence of EGFR mutations could be seen in the same ethnic group, although it could also be related to a small number of studied population. In a meta-analysis by Tochigi et al.4 concerning
© 2014 Wiley Publishing Asia Pty Ltd
344
T Powrózek et al.
(a)
(b)
Figure 2 Central nervous system computed tomography showed adenosquamous carcinoma metastasis (a) and its complete remission after 5 months of gefitinib treatment (b).
(b)
Figure 3 Chest computed tomography showed primary tumor in the right lung (a) and its partial remission after 5 months of gefitinib treatment (b).
(b)
Figure 4 Abdominal computed tomography showed adenosquamous carcinoma metastases (a) and their partial remission after 5 months of gefitinib treatment (b).
(a)
(a)
molecular profile of ADSQ carcinoma patients, nine articles and 78 patients with ADSQ carcinoma of Asian origin had been analyzed. Mutations in EGFR gene were detected in 23 of patients (32.1 percent of cases)
© 2014 Wiley Publishing Asia Pty Ltd
and mostly included 17 cases of deletions in exon 19, two L858R substitutions in exon 21 and two mutations in exon 20. Taking into consideration the results of the cited articles and the results of our study, it can be
Asia-Pac J Clin Oncol 2014; 10: 340–345
345
EGFR mutations in ADSQ cell carcinoma
assumed that the deletion in exon 19 of EGFR gene is more common for ADSQ carcinoma than the substitution L858R in exon 21 (in adenocarcinoma patients both mutations occur with a similar frequency of about 45 percent). Till date, the biggest study concerning molecular profile of 6990 Chinese NSCLC patients was carried out by Jia et al.3 The percentage of ADSQ carcinoma-bearing patients was up to 9.6 percent, although only 55 patients were qualified for molecular diagnosis. EGFR gene mutations were found in 21 patients (38.2 percent) and included 10 deletions in exon 19, six substitutions L858R in exon 21 and a number of a rare mutations. The presented studies were conducted in small group of patients of different Asian origin, we could expect that the percentage of EGFR mutations in ADSQ carcinoma-bearing Asian patients is similar to the percentage of the mutations in adenocarcinoma patients (even up to 40 percent). This paper shows that EGFR gene mutations could be detected in Caucasian ADSQ carcinoma patients in an even higher percentage than in adenocarcinoma cases. Unfortunately, systematic studies have not been carried out in Caucasian ADSQ carcinoma patients. The biggest studies provided by Rosell et al., Bell et al. and Lynch et al. (over 4500 NSCLC cases altogether)7–9 are mainly concerned on the detection of EGFR mutation in adenocarcinoma patients with an objective response after molecularly targeted therapy. Moreover, the authors did not found EGFR-activating mutations in ADSQ carcinoma patients. This confirms that a little attention was given to other histological types of NSCLC. Only one study by Tochigi et al.4 had presented the EGFR mutations status in Caucasian ADSQ carcinoma patients. Activating EGFR mutations were observed in 13 percent (three out of 23) of patients and included only deletions in exon 19. In Caucasian population, the percentage of EGFR mutations detected in ADSQ carcinoma patients was comparable with the percentage of EGFR mutations reported in adenocarcinoma cases. In conclusion, the high percentage of EGFR mutations reported in our paper resulted from a small number of studied patients. However, similar to the patients of Asian origin, the percentage of EGFR mutations may fluctuate between the Caucasian patients from different geographical regions. It should be mentioned that this paper is probably the second one concerning
Asia-Pac J Clin Oncol 2014; 10: 340–345
EGFR gene mutation status in ADSQ carcinoma patients of Caucasian origin, and probably the first one in Europe. These observations suggest merit in more scrupulous molecular studies on larger groups of ADSQ carcinoma patients, which should enable proper qualification for EGFR-TKI therapy.
REFERENCES 1 Toyooka S, Yatabe Y, Tokumo M et al. Mutations of epidermal growth factor receptor and K-ras genes in adenosquamous carcinoma of the lung. Int J Cancer 2006; 118: 1588–90. 2 Iwanaga K, Sueoka-Aragne N, Nakamura T, Mori D, Kimura S. The long-term survival of a patient with adenosquamous lung carcinoma harboring EGFR-activating mutations who was treated with gefitinib. Intern Med 2012; 51: 2771–4. 3 Jia X-L, Chen G. EGFR and KRAS mutations in Chinese patients with adenosquamous carcinoma of the lung. Lung Cancer 2011; 74: 396–400. 4 Tochigi N, Dacic S, Nikiforova M, Cieply KM, Yousem SA. Adenosquamous carcinoma of the lung: a microdissection study of KRAS and EGFR mutational and amplification status in a western patient population. Am J Clin Pathol 2011; 135: 783–9. 5 Ohtsuka K, Ohnishi H, Fujiwara M et al. Abnormalities of epidermal growth factor receptor in lung squamous cell carcinoma, adenosquamous carcinoma and large cell carcinoma: tyrosine kinase domain mutations are not rare in tumors with a adenocarcinoma component. Cancer 2007; 109: 741–50. 6 Kang SM, Kang HJ, Shin JH et al. Identical epidermal growth factor receptor mutations in adenocarcinomatous and squamous cell carcinomatous components of adenosquamous carcinoma of the lung. Cancer 2007; 109: 581–7. 7 Rosell R, Moran T, Queralt C et al. Screening for epidermal growth factor receptor mutations in lung cancer. N Engl J Med 2009; 361: 958–67. 8 Bell DW, Lynch TJ, Haserlat SM et al. Epidermal growth factor receptor mutations and gene amplification in nonsmall-cell lung cancer: molecular analysis of the IDEAL/ INTACT gefitinib trials. J Clin Oncol 2005; 23: 8081– 92. 9 Lynch TJ, Bell DW, Sordella R et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 2129–39.
© 2014 Wiley Publishing Asia Pty Ltd
Asia-Pacific Journal of Clinical Oncology 2014; 10: 340–345
doi: 10.1111/ajco.12177
ORIGINAL ARTICLE
EGFR gene mutations in patients with adenosquamous lung carcinoma Tomasz POWRÓZEK,1 Paweł KRAWCZYK,1 Rodryg RAMLAU,4,6 Sylwia SURA,1 Kamila WOJAS-KRAWCZYK,1 Tomasz KUCHARCZYK,1,8 Beata WALCZYNA,2 Justyna SZUMIŁO,2 Katarzyna SZYSZKA-BARTH,6 Piotr MILECKI,5,7 Aleksander BARINOW-WOJEWÓDZKI6 and Janusz MILANOWSKI1,3 1
Pneumonology, Oncology and Allergology Department, 2Clinical Pathomorphology Department, Medical University, and Institute of Agricultural Medicine, Lublin, 4Thoracic Surgery Department, Chair of Cardio-Thoracic Surgery, 5Electroradiology Department, University of Medical Sciences, and 6Wielkopolskie Center of Pulmonology and Thoracosurgery of Eugenia and Janusz Zeyland, and 7Radiotherapy Department, Wielkopolskie Cancer Center, Poznan, and 8Postgraduate School of Molecular Medicine, Medical University, Warsaw, Poland 3
Abstract Aim: Adenosquamous (ADSQ) carcinoma accounts for 1–4% of non-small cell lung cancer (NSCLC). The origin of ADSQ carcinoma and its genetic background is not fully understood. Most studies concerning epidermal growth factor receptor (EGFR) mutation status are performed in adenocarcinoma, while there is limited information about the prevalence of this mutation in ADSQ-bearing Caucasian patients and the efficacy of EGFR tyrosine kinase inhibitors. Methods: EGFR gene status has been examined in 1000 non-squamous NSCLC patients of Polish origin. Polymerase chain reaction (PCR) followed by DNA fragment length analysis and allele-specific PCR as well as real-time PCR technique were used to estimate EGFR gene status. Complete clinical data were obtained for all examined patients. Results: In the group of 1000 non-squamous NSCLC patients, ADSQ was diagnosed in 14 (1.4%) cases. Activating mutations of EGFR were observed in 28.6% (four out of 14) of ADSQ-bearing patients and included deletions of 15 base-pairs in exon 19 in three cases (one man and two women) and substitution of L861Q with coexistence of G719X mutation in one non-smoking male patient. Deletions were diagnosed in two non-smoking patients and one current-smoking female patient (50 pack-years). One non-smoking man with deletion in exon 19 of EGFR gene was successfully treated with gefitinib in first-line therapy. Conclusions: EGFR gene mutations in ADSQ carcinoma patients may be more common than previously thought. EGFR mutation testing is appropriate in ADSQ-bearing patients, in which response for molecularbased therapy is predictable. Key words: EGFR mutation, gefitinib, lung adenosquamous carcinoma.
INTRODUCTION Correspondence: Professor Paweł Krawczyk MD PhD, Pneumonology, Oncology and Allergology Department, Medical University of Lublin, Jaczewskiego 8, 20-954 Lublin, Poland. Email: [email protected] Conflict of interest: none Accepted for publication 29 December 2013.
© 2014 Wiley Publishing Asia Pty Ltd
Squamous cell carcinoma, adenocarcinoma and largecell carcinoma accounted for 95 percent of non-small cell lung cancer (NSCLC) cases and the differences in etiology and biology between these types are frequently observed. Adenocarcinoma is most often diagnosed in
341
EGFR mutations in ADSQ cell carcinoma
non-smoking women; moreover, epidermal growth factor receptor (EGFR) gene mutations are reported far more frequent in this type of cancer than in other diagnosed NSCLC tumors. Squamous cell carcinoma occurs mainly in long-term tobacco smoking patients, very rarely in non-smokers, and EGFR gene mutations are diagnosed much less frequent than in adenocarcinoma.1 According to the 2009 World Health Organization classification, tumor containing of both squamous and adenocarcinoma component of at least 10 percent of each, is defined as adenosquamous (ADSQ) carcinoma. ADSQ carcinoma is a rare (1–4 percent of NSCLC cases), but very aggressive lung cancer with 5-year overall survival lower of 20 percent when compared with adenocarcinoma and squamous cell carcinoma patients.2 The etiology of ADSQ is similar to adenocarcinoma in terms of age of patients, smoking status and ethnicity.3 However, there is limited information about genetic alterations and tumor biology of ADSQ carcinoma. Therefore, the tumor morphology and the scarcity of cytological materials could not be sufficient enough for pathomorphological diagnosis of ADSQ. In many cases, tumor samples of ADSQ histology had not been considered for EGFR mutation testing. In these cases, immunohistochemistry (IHC) staining and molecular techniques could be an additional tool for diagnostic. The recent studies showed that EGFRactivating mutations are observed in 15–40 percent of ADSQ carcinoma-bearing patients which are likely to benefit from molecularly targeted drugs.2 The bad prognosis of ADSQ carcinoma-bearing patients has encouraged searching for more effective treatments. Some expectations are associated with the EGFR tyrosine kinase inhibitors (EGFR-TKI), so the evaluation of EGFR gene mutation status could be worthwhile. The increased incidence of ADSQ carcinoma in Caucasian population and the deficient in information about their genetic status has led us to intensive histological and molecular analysis to qualify the patients to more precise treatment.2,4
METHODS EGFR gene status has been examined in 1000 nonsquamous NSCLC patients of Polish origin. DNA was isolated from different types of materials (including paraffin-embedded tissue, cell blocks, fresh human tissue) using the QIAamp DNA FFPE Tissue Kit (Qiagen, Hilden, Germany) in accordance with the manufacturer’s instructions. Polymerase chain reaction (PCR) followed by DNA fragment length analysis and
Asia-Pac J Clin Oncol 2014; 10: 340–345
allele-specific PCR techniques with Cy5 fluorescentlabeled primers were applied. Analysis was performed using an ALF Express II sequencer and ALFWin Fragment Analysis software (Amersham Pharmacia, Biosciences, Amersham Place, Little Chalfont, UK). Moreover, real-time PCR technique (m2000rt, Abbott Molecular, Abbott Park, IL, USA) with fluorescent-labeled probes (Entrogen, Tarzana, Los Angeles, CA, USA) were also used. The sensitivity of the techniques was sufficient to perform EGFR mutation analysis in all examined samples. Objective response to EGFR-TKI treatment was evaluated with Response Evaluation Criteria in Solid Tumors criteria.
RESULTS In the group of 1000 non-squamous NSCLC patients, according to IHC staining (Fig. 1) ADSQ was diagnosed in 14 (1.4 percent) cases with a similar incidence in women and men. Activating mutations of EGFR gene were observed in 28.6 percent (four out of 14) of ADSQbearing patients and included deletions of 15 base-pairs in exon 19 in three cases (one men and two women) and substitution of L861Q with coexistence of G719X mutation in one non-smoking male patient. Deletions were diagnosed in two non-smoking patients and one current-smoking female patient (50 pack-years). The analysis of EGFR gene mutations status in different histological types of NSCLC is presented in Table 1. Clinical characteristics of ADSQ carcinoma-bearing patients with defined EGFR status are presented in Table 2. The EGFR mutations are observed in similar frequency in female as in male. The female patient W.M., with EGFR gene mutation diagnosed in post-surgical material, did not receive EGFR-TKI therapy, as she was qualified to adjuvant cisplatin/gemcitabine-based therapy. The patient received two courses of the treatment and she had been excluded from further systemic therapy because of deterioration of kidney parameters, decrease of overall performance status and disease progression. The female patient I.W., did not require further systemic treatment after surgical resection. The male patient J.P., in stage IV of ADSQ carcinoma and with the distant metastases observed in the central nervous system, in the bone and in the liver, had been previously treated with central nervous system and spine radiotherapy. He received gefitinib (250 mg daily) in first-line therapy and during 5 months of treatment, the disease remission and the rash of III grade has been observed. Complete remission of brain metastases (Fig. 2a,b) and
© 2014 Wiley Publishing Asia Pty Ltd
342
(a)
(b)
Figure 1 (a) The representative immunohistochemistry staining of adenosquamous carcinoma tumor sample. Positive reaction with thyroid transcription factor 1 (TTF-1) antibody is visible in nuclei of adenocarcinoma cell component (tumor cells are present in vessel lumen). Internal control is visible as a positive reaction with TTF-1 antibody in pneumocytes. Negative nuclear reaction with TTF-1 antibody is observed in squamous carcinoma pattern. Background is slightly colored. (b) Positive reaction with p63 antibody is visible in nuclei of squamous carcinoma cell component.
partial remission of lung tumor (Fig. 3a,b) as well as liver (Fig. 4a,b) and bones metastases were observed. After 9 months of treatment, the patient remains in good condition. The male patient K.L. has been diagnosed recently and he is being qualified to erlotinib therapy.
DISCUSSION The EGFR gene mutations are rarely detected in tumors of histology other than adenocarcinoma, especially in
© 2014 Wiley Publishing Asia Pty Ltd
T Powrózek et al.
cancer of squamous histology. This type of cancer, in contrast to adenocarcinoma, is strongly associated with smoking history. However, in NSCLC patients of mixed histology of tumor, EGFR mutations had been detected in tumor cells of both adeno- and squamous pattern. Moreover, tumor cells of ADSQ carcinoma histology could be morphologically different from “pure” adenocarcinoma or squamous carcinoma cells.2,5 Therefore, the point to discussion is polyclonality or monoclonality of ADSQ carcinoma. These observations allow us to conclude that EGFR gene mutations could be observed even in progenitor cells. Under specific effect of these mutations, the progenitor cells could develop into both types of tumor cells.5 The presence of identical mutations in both tumor cells’ components could explain the monoclonality of carcinogenesis’ pathway. However, it should be noted that exogenous carcinogenic factors could act on bronchial epithelial cells. The combined effect of these factors resulted in other genetic abnormalities in both types of tumor cells component. Moreover, the protein expression profile is changing what could lead to the differentiation of progenitor cells into two types of phenotypical different lung cancers. Furthermore, the “collision” and the interaction of these cells in later multistage carcinogenesis results in secondary polyclonality of ADSQ cancer cells.2,5 Even though ADSQ carcinoma appears to be a rare subtype of lung cancer, some cases might have been undiagnosed or misdiagnosed as a different histological subtype. It is still not precisely described if squamous tumors had a component of adenocarcinoma cells and what is the ratio of this component to the entire tumor cells. It was mainly by unappreciated immunohistochemical staining and abusing of only hematoxilin and eosin staining for the identification of NSCLC types. The representative image of IHC staining used for ADSQ carcinoma diagnosis in our patients is showed on Figure 1. Therefore, many tissues samples could have been omitted or rejected from further EGFR gene mutations analysis, and the patients were not qualified to EGFR-TKI therapy. This could also result from the etiology of ADSQ carcinoma, which is a type of lung cancer connected with tobacco smoking, while EGFR gene mutations are more often observed in non-smokers. Currently, for precise identification of NSCLC subtypes by IHC staining, the specific monoclonal antibodies are used, for example anti-thyroid transcription factor 1 (TTF-1, for adenocarcinoma) and anti-p63 (for squamous cell carcinoma). Nevertheless, the expression
Asia-Pac J Clin Oncol 2014; 10: 340–345
343
EGFR mutations in ADSQ cell carcinoma
Table 1 Clinical characteristics of patients according to the pathological diagnosis and epidermal growth factor receptor (EGFR) mutation status EGFR gene status EGFR gene mutations (n = 90; 9%)
Pathomorphological diagnosis (n = 1000) Adenocarcinoma (n = 806; 80.6%) Large-cell carcinoma (n = 43; 4.3%) Adenosquamous carcinoma (n = 14; 1.4%) Not otherwise specified (n = 137; 13.7%)
Table 2
Wild-type EGFR gene (n = 910, 91%
Exon 19 (deletion)
Exon 21 (L858R)
726 (90.07%) 41 (95.34%) 10 (71.43%) 133 (97.08%)
44 (5.46%) 1 (2.33%) 3 (21.43%) 2 (1.46%)
33 (4.09%) 1 (2.33%) – 2 (1.46%)
Exon 20 (insertion)
Exon 18 (S768I and G719X)
Exon 18 and 21 (G719X and L861Q)
2 (0.25%) – – –
1 (0.13%) – – –
– – 1 (7.14%) –
Clinical and molecular characteristics of adenosquamous carcinoma patients
Patient
Sex
Age
Smoking status
Stage of disease/procedures for material collection
EGFR gene mutation status
W.M. R.S. S.C. J.K. I.W. J.P. E.S. T.G. J.C. L.K. A.S. J.F. M.P. K.L.
Female Male Female Male Female Male Male Male Female Male Male Male Male Male
73 71 77 74 68 58 62 64 77 46 68 48 69 51
Current (50 pack-years) Current (60 pack-years) NA Non-smoker Non-smoker Non-smoker Current (40 pack-years) Current (50 pack-years) NA Current (60 pack-years) NA Non-smoker Ex-smoker (35 pack-years) Non-smoker
IIIB/wedge resection of the tumor† IIIB/endobronchial biopsy IIIB/lobectomy† IV/resection of brain metastase IIA/lobectomy IV/FNA IIIB/FNA IIIB/endobronchial biopsy IIB/wedge resection of the tumor IIIA/wedge resection of tumor IIB/wedge resection of tumor IIIA/lobectomy IIIA/endobronchial biopsy IV/endobronchial biopsy
Ex 19 deletion Wild type Wild type Wild type Ex 19 deletion Ex 19 deletion Wild type Wild type Wild type Wild type Wild type Wild type Wild type L861Q and G719X substitution
†
Stage of the disease was estimated during qualification for epidermal growth factor receptor- tyrosine kinase inhibitors treatment; surgical material was obtained in earlier stages than mentioned in table. FNA, fine needle aspiration; NA, data not available.
of TTF-1 antigen is sometimes observed in squamous cell carcinoma and p63 antigen expression is weakly or locally observed in adenocarcinoma. Hence, the expanding of IHC panel of additional antibodies (e.g. anti-CK5/6 and anti-CK34βE12 for squamous cell carcinoma, anti-CK7, anti-SPA, anti-SPB [surfactant proteins A and B] and anti-napsin A for adenocarcinoma, especially with a negative TTF-1 staining), which allow precise determination of histological type of NSCLC should be necessarily performed. According to the problems described above, little is known about the frequency of EGFR gene mutations in ADSQ carcinoma patients of different ethnicity, especially of Caucasian origin. The results of the latest
Asia-Pac J Clin Oncol 2014; 10: 340–345
studies regard mainly patients of Asian origin and concluded that EGFR mutations among ADSQ carcinoma patients are much more common than it was previously thought. Kang et al.6 had found 11 (44 percent) EGFRactivating mutations in the group of 25 ADSQ carcinoma-bearing patients of Korean origin (2.5 percent of NSCLC patients). Toyooka et al.1 had found three (27.3 percent) EGFR mutations in 11 patients of Japanese origin with ADSQ carcinoma diagnosis (2.8 percent of NSCLC patients). These results suggest that the differences in the incidence of EGFR mutations could be seen in the same ethnic group, although it could also be related to a small number of studied population. In a meta-analysis by Tochigi et al.4 concerning
© 2014 Wiley Publishing Asia Pty Ltd
344
T Powrózek et al.
(a)
(b)
Figure 2 Central nervous system computed tomography showed adenosquamous carcinoma metastasis (a) and its complete remission after 5 months of gefitinib treatment (b).
(b)
Figure 3 Chest computed tomography showed primary tumor in the right lung (a) and its partial remission after 5 months of gefitinib treatment (b).
(b)
Figure 4 Abdominal computed tomography showed adenosquamous carcinoma metastases (a) and their partial remission after 5 months of gefitinib treatment (b).
(a)
(a)
molecular profile of ADSQ carcinoma patients, nine articles and 78 patients with ADSQ carcinoma of Asian origin had been analyzed. Mutations in EGFR gene were detected in 23 of patients (32.1 percent of cases)
© 2014 Wiley Publishing Asia Pty Ltd
and mostly included 17 cases of deletions in exon 19, two L858R substitutions in exon 21 and two mutations in exon 20. Taking into consideration the results of the cited articles and the results of our study, it can be
Asia-Pac J Clin Oncol 2014; 10: 340–345
345
EGFR mutations in ADSQ cell carcinoma
assumed that the deletion in exon 19 of EGFR gene is more common for ADSQ carcinoma than the substitution L858R in exon 21 (in adenocarcinoma patients both mutations occur with a similar frequency of about 45 percent). Till date, the biggest study concerning molecular profile of 6990 Chinese NSCLC patients was carried out by Jia et al.3 The percentage of ADSQ carcinoma-bearing patients was up to 9.6 percent, although only 55 patients were qualified for molecular diagnosis. EGFR gene mutations were found in 21 patients (38.2 percent) and included 10 deletions in exon 19, six substitutions L858R in exon 21 and a number of a rare mutations. The presented studies were conducted in small group of patients of different Asian origin, we could expect that the percentage of EGFR mutations in ADSQ carcinoma-bearing Asian patients is similar to the percentage of the mutations in adenocarcinoma patients (even up to 40 percent). This paper shows that EGFR gene mutations could be detected in Caucasian ADSQ carcinoma patients in an even higher percentage than in adenocarcinoma cases. Unfortunately, systematic studies have not been carried out in Caucasian ADSQ carcinoma patients. The biggest studies provided by Rosell et al., Bell et al. and Lynch et al. (over 4500 NSCLC cases altogether)7–9 are mainly concerned on the detection of EGFR mutation in adenocarcinoma patients with an objective response after molecularly targeted therapy. Moreover, the authors did not found EGFR-activating mutations in ADSQ carcinoma patients. This confirms that a little attention was given to other histological types of NSCLC. Only one study by Tochigi et al.4 had presented the EGFR mutations status in Caucasian ADSQ carcinoma patients. Activating EGFR mutations were observed in 13 percent (three out of 23) of patients and included only deletions in exon 19. In Caucasian population, the percentage of EGFR mutations detected in ADSQ carcinoma patients was comparable with the percentage of EGFR mutations reported in adenocarcinoma cases. In conclusion, the high percentage of EGFR mutations reported in our paper resulted from a small number of studied patients. However, similar to the patients of Asian origin, the percentage of EGFR mutations may fluctuate between the Caucasian patients from different geographical regions. It should be mentioned that this paper is probably the second one concerning
Asia-Pac J Clin Oncol 2014; 10: 340–345
EGFR gene mutation status in ADSQ carcinoma patients of Caucasian origin, and probably the first one in Europe. These observations suggest merit in more scrupulous molecular studies on larger groups of ADSQ carcinoma patients, which should enable proper qualification for EGFR-TKI therapy.
REFERENCES 1 Toyooka S, Yatabe Y, Tokumo M et al. Mutations of epidermal growth factor receptor and K-ras genes in adenosquamous carcinoma of the lung. Int J Cancer 2006; 118: 1588–90. 2 Iwanaga K, Sueoka-Aragne N, Nakamura T, Mori D, Kimura S. The long-term survival of a patient with adenosquamous lung carcinoma harboring EGFR-activating mutations who was treated with gefitinib. Intern Med 2012; 51: 2771–4. 3 Jia X-L, Chen G. EGFR and KRAS mutations in Chinese patients with adenosquamous carcinoma of the lung. Lung Cancer 2011; 74: 396–400. 4 Tochigi N, Dacic S, Nikiforova M, Cieply KM, Yousem SA. Adenosquamous carcinoma of the lung: a microdissection study of KRAS and EGFR mutational and amplification status in a western patient population. Am J Clin Pathol 2011; 135: 783–9. 5 Ohtsuka K, Ohnishi H, Fujiwara M et al. Abnormalities of epidermal growth factor receptor in lung squamous cell carcinoma, adenosquamous carcinoma and large cell carcinoma: tyrosine kinase domain mutations are not rare in tumors with a adenocarcinoma component. Cancer 2007; 109: 741–50. 6 Kang SM, Kang HJ, Shin JH et al. Identical epidermal growth factor receptor mutations in adenocarcinomatous and squamous cell carcinomatous components of adenosquamous carcinoma of the lung. Cancer 2007; 109: 581–7. 7 Rosell R, Moran T, Queralt C et al. Screening for epidermal growth factor receptor mutations in lung cancer. N Engl J Med 2009; 361: 958–67. 8 Bell DW, Lynch TJ, Haserlat SM et al. Epidermal growth factor receptor mutations and gene amplification in nonsmall-cell lung cancer: molecular analysis of the IDEAL/ INTACT gefitinib trials. J Clin Oncol 2005; 23: 8081– 92. 9 Lynch TJ, Bell DW, Sordella R et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 2129–39.
© 2014 Wiley Publishing Asia Pty Ltd
