This article is mentioned in:

Introduction

Lung cancer is a devastating disease and the leading cause of cancer-associated mortality worldwide (1). The most frequent type of lung cancer is non-small-cell lung cancer (NSCLC), which accounts for ~80% of lung cancer cases (2). The short survival time of lung cancer patients is mainly attributed to poor outcomes from conventional chemotherapeutic treatments (3). However, progress in defining the molecular mechanism of carcinogenesis has led to a notable improvement in the response to chemotherapy (4). In 2004, it was revealed that epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs), including gefitinib and erlotinib, are only effective in patients that harbor tumorigenic EGFR mutations that cause aberrant tyrosine kinase activity (5,6). Thus, identification of oncogenic driver mutations in cancer patients has become key for the identification of an effective treatment for NSCLC (7). One of the previously identified oncogenic driver mutations is the fusion of anaplastic lymphoma kinase (ALK) with echinoderm microtubule-associated protein-like 4 (EML4) on chromosome 2p, which was identified in 2007 in a subpopulation of Japanese patients with NSCLC (8). In this study, 6.7% (5/75) of the enrolled Japanese patients with NSCLC possessed EML4-ALK fusion transcripts, resulting from ALK translocation within chromosome 2p (8). Other ALK-fusion genes, including KIF5B-ALK, have also been identified within chromosome 2p (9–12). Patients possessing ALK fusions are usually resistant to EGFR-TKIs (13), but respond to the ALK-TKI crizotinib (14). Therefore, screening for oncogenic driver mutations, including tumorigenic EGFR mutations and ALK fusions, has become a crucial step in disease diagnosis and designing an effective personalized or tailored therapy plan.

Since the identification of the EML4-ALK fusion gene, numerous studies have been performed to determine the frequency of occurrence in patients with NSCLC (8,12,15–24). However, these numbers varied significantly between studies (7), ranging between 1.6% in a cohort of Japanese patients (21) and 11.7% in a cohort of Chinese patients (22). This is likely to reflect the differences in detection techniques, sample size and patient selection criteria. Although the EML4-ALK translocation was first identified in a NSCLC patient with a history of smoking (8), subsequent studies have suggested that the translocation is more frequently detected in never-smokers (13,16,21,22). A never-smoker is defined as an individual that has smoked <100 cigarettes per lifetime, according to the US Center for Disease Control (25). Although inconclusive, studies have also suggested that the frequency of the incidence is likely to be increased in female patients compared with male patients (24). Thus, it is possible that the frequency of the EML4-ALK translocation may be markedly higher in female never-smokers. A previous study reported that the incidence was as high as 15.2% (5/33) in a small cohort of female patients with adenocarcinoma (24).

To determine the frequency of EML4-ALK fusion more precisely in female never-smokers, in the present study a large cohort of patients with NSCLC was assembled. In total, 280 female patients that were never-smokers were enrolled and the presence of mutations were detected by Multiplex one-step reverse transcription-polymerase chain reaction (RT-PCR) in the tumor specimens collected from these patients. The clinical characteristics that are associated with these mutations were also analyzed. The present study aimed to increase the understanding of the EML4-ALK fusion in NSCLC and provide information for improving the diagnosis procedure and designing personalized treatment plans.

Materials and methods

Patients and sample collection

The present study was approved by the Institutional Ethics Committee of Henan Cancer Hospital (Zhengzhou, China). In total, 280 never-smoking female patients with NSCLC were recruited (Table I). These patients were enrolled between 2012 and 2013 at Henan Cancer Hospital. Carcinoma tissue samples were collected from these patients and preserved as formalin-fixed paraffin-embedded (FFPE) tissue blocks. The FFPE tissue blocks were used as the only tissue sources for the experiments performed in the present study, including the detection of EML4-ALK fusions and measurement of the expression level of the ALK tyrosine kinase (ALK TK) mRNA and protein. As a standardized procedure during diagnosis, tumor subtypes and pathological characteristics were determined independently by two pathologists. The opinion of a third pathologist was required when there was a discrepancy. The tumors were graded based on the abnormality of the appearance of the tumor cells and tissues compared to the surrounding normal cells and normal tissues. Tumors that appeared close to normal were graded as well-differentiated. Tumors that appeared intermediately and highly abnormal were classified as moderately- and poorly-differentiated, respectively. The tumors that appeared to be abnormal were suspected to be fast-growing and malignant, collectively described as undifferentiated or less-differentiated tumors in the present study.

Table I. Clinical features of non-small cell lung cancer tumors in female never-smokers.

Table I.

Clinical features of non-small cell lung cancer tumors in female never-smokers.

Characteristics	Total, n (%)
Histology
Adenocarcinoma	274 (97.86)
Squamous cell carcinoma	5 (1.79)
Others	1 (0.36)
Age
<40 years	4 (1.43)
40–49 years	57 (20.36)
50–59 years	105 (37.50)
≥60 years	114 (40.71)
Differentiation
Poorly-differentiated	91 (32.50)
Moderately-differentiated	95 (33.93)
Well-differentiated	94 (33.57)

Detection of EML4-ALK fusion genes

The FFPE tissue blocks were sliced to a width of 3 µm, and the tumor regions were identified and collected for RNA extraction. Total RNA was extracted using RNeasy FFPE kit (Qiagen, Valencia, CA, USA), according to the manufacturer's instructions, and treated with DNase I (DNA-free; Ambion Life Technologies, Carlsbad, CA, USA) to remove any DNA contamination. The RNA samples were then subjected to Multiplex One-step RT-PCR with fluorescent RT-PCR to detect EML4-ALK fusion transcripts using the human Lung Cancer Related Fusion Gene Detection kit (Yuanqi Bio-Pharmaceutical Co., Ltd., Shanghai, China), according to the manufacturer's instructions. Briefly, the mixture of each reaction contained 3 µl total RNA, 20 µl Multiplex RT-PCR buffer and 2 µl Multiplex Enzyme Mix in a total volume of 25 µl. The primers included in the reaction for the detection of EML4-ALK fusion subtype variants were as follows: V1 and V6 forward, 5′-ATT TGT GCA GTG TTT AGC ATT C-3′; V2 forward, 5′-CGG GAG ACT ATG AAA TAT TGT ACT-3′; V3a and V3b forward, 5′-AGT CAC ATA ATT CTT GGG AA-3′; V4b and V7 forward, 5′-GGG AAA GGA CCT AAA GGT G-3′; V4a forward, 5′-GTA GCA GAA GGA AAG GCA GAT C-3′; V5a and V5b forward, 5′-GCT AAA GGC GGC TTT GGC TG-3′; and E17-A20 and V9 forward, 5′-CGC TAC TCA ATA GAT GGT ACC T-3′; and the common reverse primer ALK-E20, 5′-CAT GAT GGT CGA GGT GCG C-3′ (Sangon Biotech Co., Ltd., Shanghai, China). RT-PCR was performed using the 7300 Real Time PCR System (Applied Biosystems Life Technologies, Foster City, CA, USA) under the following conditions: 42°C for 30 min; 94°C for 5 min; 40 cycles at 94°C for 15 sec; and 60°C for 1 min. The PCR products were subjected to DNA sequencing (sequencing primer, 5′-TTG CTC AGC TTG TAC TCA GGG CTC TG-3′; Sangon Biotech Co., Ltd.) to determine the variant types of the EML4-ALK fusion transcripts. All patients expressing the EML4-ALK fusion, as determined by Multiplex RT-PCR and direct DNA sequencing, were further confirmed by Vysis ALK Break Apart fluorescence in situ hybridization (FISH) analysis.

ALK break apart FISH analysis

FISH experiments were performed using FFPE tissue sections in order to identify ALK rearrangements. The Vysis ALK Break Apart FISH Probe kit (Abbott Molecular Inc., Des Plaines, IL, USA) was used according to the manufacturer's instructions. Briefly, two DNA probes that targeted sequences prior to and following the ALK breaking point were labeled with green and red florescent dye, respectively, and used for FISH. In normal nuclei without ALK rearrangement, the red and green fluorescent signals from the two probes colocalize to form a yellowish signal, or the two signals are less than two signal diameters apart. However, when ALK rearrangement occurs, the signals from the two probes are located more than two signal diameters apart in a single nucleus, which indicates a cell with ALK rearrangement. A tumor sample was considered to lack ALK rearrangement if <5 cells out of 50 (<5/50 or <10%) appeared positive for ALK rearrangement. If >25 cells out of 50 (>25/50 or >50%) were positive for ALK rearrangement, the tumor was considered positive, but the tumor is equivocal if 5–25 cells (10 to 50%) are positive for ALK rearrangement. When the tumor was equivocal, additional evaluation was required, which was performed according to the manufacturer's instructions.

Determining the level of ALK mRNA

One-step RT-PCR was performed on the total RNA extracted from FFPE tissue blocks to determine the level of ALK RNA in the tissues. Primers specific to the ALK TK domain were designed for the ALK mRNA expression analysis. The expression level of the reference gene Abelson murine leukemia viral oncogene homolog (ABL) was also determined as a control. The primers used were as follows: ALK forward, 5′-AGA AAC TGC CTC TT GAC CTG-3′ and reverse, 5′-GGG CAT CCA CTT AAC TGG C-3′; and ABL forward, 5′-TAC CTG AGG GAG TGC AAC C-3′ and reverse, 5′-TTT TCT TCT CCA GGT ACT CCA-3′. In addition, the DNA sequencing primers were as follows: ALK sequencing primer, 5′-CCC TTT CTA TAG TAG CTC GCC CTG TAG AT-3′; and ABL sequencing primer, 5′-CCA TGT ACA GCA GCA CCA CGG CGT-3′. RT-PCR was performed as aforementioned.

Immunohistochemistry (IHC)

IHC was used to evaluate the expression of ALK TK protein, as previously described (26). Briefly, the experiments were performed on FFPE tissue sections using a Ventana ALK (D5F3) Cdx Assay (rabbit monoclonal antibody against ALK; cat no. 790–4796; Ventana Medical Systems, Inc., Tucson, AZ, USA), which detects the endogenous levels of total ALK protein and ALK fusion proteins, according to the manufacturer's instructions. The tissue sections were deparaffinized and incubated with 3% H2O2 (Ventana Medical Systems, Inc.) to quench endogenous peroxidase activity, which was followed by heat-induced antigen retrieval for 30–60 min (Ventana Medical Systems, Inc.). Subsequent to blocking with 10% normal goat serum (Ventana Medical Systems, Inc.), the aforementioned primary antibody were applied, followed by goat anti-rabbit secondary antibody conjugated with horseradish peroxidase (ultraView Universal DAB Detection kit; cat no. 760–500; Ventana Medical Systems, Inc.), according to the manufacturer's instructions. The immunosignals were visualized using diaminobenzidine (UltraView; Ventana Medical Systems, Inc.) and counterstained with hematoxylin (Ventana Medical Systems, Inc.). Images were captured to assess the intensity of staining, as follows: 0, no staining; 1+, light staining; 2+, moderate staining; and 3+, strong staining. The distribution of ALK immunostaining was also determined. Any tissue specimens exhibiting IHC intensity >0 were defined as IHC positive.

Statistical analysis

The differences between data were analyzed using χ2 test, Mann-Whitney U test or Student's t-test, as appropriate. P<0.05 was considered to indicate a statistically significant difference.

Results

Clinical features of participating patients

To avoid the variation caused by a small sample size and to more precisely determine the frequency of EML4-ALK translocation, a total of 280 patients with NSCLC were recruited for the present study. All patients were female and were defined as never-smokers (25). Based on the information collected from these patients, the majority of the patients were diagnosed with adenocarcinoma (97.86%; Table I). The median age of this cohort of patients was 56.69 years, ranging between 23 and 76 years. Out of these patients, 40.71% of patients were aged ≥60 years, while 37.50, 20.36 and 1.43% of patients were aged 50–59, 40–49 and <40 years, respectively (Table I). The number of patients with poorly, moderately or well-differentiated tumor cells was similar between the groups, with 91 (32.50%), 95 (33.93%) and 94 (33.57%) patients in each group, respectively (Table I).

Detection and characterization of EML4-ALK translocation

In the present study, the frequency of EML4-ALK translocation in female never-smokers with NSCLC was determined. Therefore, Multiplex one-step RT-PCR was performed on the total RNA extracted from the FFPE tissue samples prepared from the resected tumors of the enrolled patients. Out of the 280 carcinoma specimens, EML4-ALK fusion gene transcripts were detected in 21 tissue specimens, accounting for 7.5% of the total patients. The representative PCR products for each detected EML4-ALK fusion variant are shown in Fig. 1A. The presence of ALK rearrangements were confirmed using the Vysis ALK Break Apart FISH Probe kit, which detects ALK rearrangements using two fluorescence-labeled probes that target sequences flanking the ALK breaking point. Representative images of FISH are shown in Fig. 1B. ALK rearrangements, indicated by red and green fluorescent signals located at least two signal diameters apart in a single nucleus, were detected in carcinoma samples harboring EML4-ALK translocation, but not in carcinoma samples without EML4-ALK fusion. The identity of these RT-PCR products was confirmed by DNA sequencing. All RT-PCR products from 21 tissue samples harboring EML4-ALK translocation were subsequently sequenced. The DNA sequencing traces and exon organization for each representative variant are presented in Fig. 2A. In total, 5 different variants were detected, including variant 1 in 12 samples and variant 3a/3b in 6 samples. In addition, variants 4, 5a and E17-A20 were detected in one sample each (Fig. 2B). Variant 2 was not detected in any of the present tissue samples. Therefore, it is possible that EML4-ALK variant 1, which accounted for 57.1% of all EML4-ALK translocations, is the most common form of the fusion products in female never-smokers with NSCLC. By contrast, the variant 2 is less likely to be detected in the same group of patients.

Figure 1. Screening for EML4-ALK fusion transcripts in the tissues obtained from female never-smokers with non-small cell lung cancer. (A) Representative DNA gel images revealing the expression of the EML4-ALK fusion transcript variants V1, V3a, V3b, V4a, V5a and E17-20, and ABL controls, as determined by Multiplex one-step reverse transcription-polymerase chain reaction. (B) ALK Break Apart fluorescence in situ hybridization analysis revealed ALK inversion in carcinoma tissues. Two probes targeting the sequences prior to and following the breaking point were labeled by red and green fluorescent dye, respectively. Representative images of EML4-ALK-negative (left) and -positive (right) carcinomas are shown. The ALK inversion in EML4-ALK-positive carcinomas is indicated by two separated red and green dots (right) located at least two signal diameters apart. NC, negative case; M, DNA ladder; EML4-ALK, echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase; ABL, Abelson murine leukemia viral oncogene homolog.

Figure 2. EML4-ALK fusion transcript variants in female never-smokers with non-small cell lung cancer. (A) Representative DNA sequencing chromatographs and schematic drawings revealing EML4-ALK fusion junctions. Five representative EML4-ALK fusion transcript variants were detected by Multiplex reverse transcription-polymerase chain reaction. The sequences and variant types were determined by direct DNA sequencing. The 5 detected EML4-ALK fusion variants consisted of V1, V3a/3b, V4a, V5a and E17-20. Variant 2 was not detected in the present cohort of patients. Schematic drawings are exhibited below the corresponding DNA sequences to indicate the organization of fusion junctions. (B) Pie graph showing the frequency of EML4-ALK fusion variants. A total of 5 different EML4-ALK fusion variants were detected, consisting of V1, V3a/3b, V4a, V5a, and E17-20. The number and the percent of each variant are indicated in the graph. Variant 2 was not detected in these carcinoma tissues. EML4-ALK, echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase.

Aberrant expression of ALK mRNA is associated with EML4-ALK translocation

The expression level of ALK mRNA was measured by one-step RT-PCR. The primers were specific to the ALK TK domain that was translocated and fused with EML4 in EML4-ALK fusion genes, so the mRNAs transcribed from the intact and translocated ALK genes would be detected. This method revealed that ALK TK mRNA was highly expressed in tissues harboring EML4-ALK fusion, but not in tissues without the EML4-ALK fusion (Fig. 3A). Despite expressing different variants of the EML4-ALK fusion gene, all carcinoma specimens harboring the fusion expressed ALK TK RNA aberrantly. By contrast, ALK TK mRNA was not detectable in the majority of the samples not harboring EML4-ALK translocation. IHC also revealed that the ALK protein is highly expressed in specimens harboring EML4-ALK fusion, but the protein was not expressed or was expressed at an extremely low level in carcinomas without EML4-ALK fusion (Fig. 3B).

Figure 3. Aberrant expression of ALK mRNA and protein. (A) Scatter graph demonstrating that the expression of ALK mRNA is significantly higher (P<0.0001) in carcinoma tissues harboring EML4-ALK fusion compared with tissues without the fusion gene in female never-smokers with non-small cell lung cancer. The values are expressed as the mean ± standard error of the mean. (B) Representative images exhibiting the immunohistochemical staining for ALK in carcinoma tissues. The ALK protein was aberrantly expressed in carcinoma tissues harboring the EML4-ALK fusion gene (EML4-ALK+), but ALK protein expression was absent in tissues without the EML4-ALK fusion gene (EML4-ALK-). EML4-ALK, echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase.

EML4-ALK translocation is more frequently detected in younger patients and undifferentiated carcinomas

All patients harboring the EML4-ALK translocation in the present cohort of patients were diagnosed with adenocarcinoma (Table II). However, it is possible that the number of other types of carcinoma was too low for EML4-ALK fusion genes to be detected (6 out of 280 patients; 2.14%). Therefore, the frequency of EML4-ALK translocation in other types of carcinomas was unclear. In total, 50, 12.28, 7.62 and 3.51% of the patients aged <40, 40–49, 50–59 and ≥60 years, respectively, harbored EML4-ALK fusion genes (Table II). The distribution of the patients with EML4-ALK fusion in the four different age groups was significantly different from the distribution of patients without EML4-ALK fusion (P<0.0019; Table II). Furthermore, patients harboring EML4-ALK translocation (median age, 50.95±2.29 years) were significantly younger than patients without EML4-ALK translocation (median age, 57.15±0.56; P<0.01) and all patients combined (median age, 56.69±0.54; P<0.05) (Fig. 4). Histological examination revealed that all carcinomas harboring EML4-ALK fusion were undifferentiated, either poorly- or moderately-differentiated, but never well-differentiated (Table II). The difference in the differentiation level between carcinomas harboring EML4-ALK fusion and those without the fusion gene is significant (P<0.0014). Overall, these results indicate that the EML4-ALK fusion is more frequent in younger patients and in undifferentiated or less-differentiated carcinomas.

Figure 4. Summary of the median age of ALK+ and ALK- patients and the two combined. The median age in ALK+ was significantly lower compared with the median age in ALK- patients or in the two groups combined. The values are expressed as the mean ± standard error of the mean. *P<0.05 and **P<0.01. EML4-ALK, echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase; ALK+, patients harboring EML4-ALK fusion; ALK-, patients without EML4-ALK fusion.

Table II. Clinical features of tissues harboring the EML4-ALK fusion gene in female never-smokers with non-small cell lung cancer.

Table II.

Clinical features of tissues harboring the EML4-ALK fusion gene in female never-smokers with non-small cell lung cancer.

	EML4-ALK fusion gene
Characteristics	Present, n (%)	Absent, n (%)	Total, n (%)	P-value
Histology
Adenocarcinoma	21 (7.66)	253 (92.34)	274 (100)
Squamous cell carcinoma	0 (0.00)	5 (100.00)	5 (100)
Other	0 (0.00)	1 (100.00)	1 (100)
Patient age
<40 years	2 (50.00)	2 (50.00)	4 (100)	0.0019
40–49 years	7 (12.28)	50 (87.72)	57 (100)
50–59 years	8 (7.62)	97 (92.38)	105 (100)
≥60 years	4 (3.51)	110 (96.49)	114 (100)
Differentiation
Poorly-differentiated	8 (8.79)	83 (91.21)	91 (100)	0.0014
Moderately-differentiated	13 (13.68)	82 (86.32)	95 (100)
Well-differentiated	0 (0.00)	94 (100.00)	94 (100)

[i] EML4-ALK, echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase.

Discussion

By combining multiple published studies, the frequency of the EML4-ALK translocation in NSCLC worldwide was determined to be ~5% (7). In the present study, the frequency of EML4-ALK fusion determined using Multiplex one-step RT-PCR was 7.5% (21 out of 280 patients) in a large cohort of never-smoking female patients with NSCLC. This result was verified by DNA sequencing, ALK Break Apart FISH analysis, and immunohistochemistry. This frequency is slightly increased compared with all NSCLC patients worldwide, regardless of gender and smoking history, indicating that the combined effect of gender (female) and smoking habit (never-smokers) is minor. Notably, the present results vary from the findings of a previous study that assessed 33 female never-smoker patients with NSCLC and reported that the incidence of EML4-ALK fusion was 15.2% (5 patients) (24). The possible discrepancy may be due to the variance caused by the difference in sample sizes between the two studies, with 33 female non-smoking patients in the previous study and 280 in the present study. In addition, environmental factors, such as local air pollution levels and food choices may vary between the different regions of China and may also be involved in the discrepancy.

Out of the 21 patients with EML4-ALK fusion genes, more than one-half were detected as fusion gene variant 1. Variant 2 was not detected in the present tissue samples. It is therefore possible that variant 1 is the predominant type, while variant 2 is the least likely to be detected in female never-smokers with NSCLC. Regardless of the variant present, the ALK mRNA and protein were aberrantly expressed in all EML4-ALK-positive carcinomas (Fig. 3). This result is consistent with the findings that EML4-ALK fusion genes are carcinoma driver mutations (8) and that overactivation of ALK TK plays a pivotal role in tumor cell proliferation (26–28). Compared with conventional chemotherapy, crizotinib, a selective inhibitor of ALK, has demonstrated a superior ability to improve the treatment outcome and survival rate in patients harboring EML4-ALK fusion mutations in clinical trials (26,29–32). By contrast, these patients demonstrated resistance to EGFR TKIs (13). Therefore, identification of EML4-ALK fusion has a direct impact on disease treatment in NSCLC, particularly with female never-smokers that demonstrate a high frequency of the EML4-ALK fusion.

In the present cohort of patients, EML4-ALK fusions were only detected in undifferentiated or less-differentiated carcinomas, including those graded as poorly- and moderately-differentiated tumors, but not in tumors graded as well-differentiated (Table II). This association between EML4-ALK fusion genes and undifferentiated carcinomas was also reported in a previous study, which was conducted using a cohort consisting of male and female patients with NSCLC (33). Since ALK is aberrantly expressed only in the carcinoma lesions harboring EML4-ALK gene fusion, but not in carcinomas without the fusion gene, it is possible that ALK expression plays an important role in determining the fate of cells in NSCLC and is accountable for decreased differentiation, a feature that is often associated with more rapid-growing and malignant types of carcinomas.

In the present study, the median age of patients with EML4-ALK gene fusion (50.95 years) was >6 years younger than the median age of patients without EML4-ALK gene fusion (57.15 years) in the present cohort of female never-smokers with NSCLC. The difference in age distribution between the patients harboring EML4-ALK gene fusion and those without gene fusion was statistically significant, as determined by χ2 (P<0.05; Table II). In addition, the frequency of EML4-ALK gene fusion in the patients aged <40 years is 50%, while the frequency gradually decreases in the older age groups (Table II). In particular, only 3.5% of patients aged ≥60 years harbored the EML4-ALK fusion gene. These results suggest that the age of disease onset is significantly younger in patients harboring EML4-ALK gene fusion compared with patients that did not harbor gene fusion.

In conclusion, the frequency of the EML4-ALK fusion is 7.5% in a cohort of 280 NSCLC patients that consisted of female never-smokers. Among the identified fusion variants, variant 1 was the most common type, accounting for 57.1% of all EML4-ALK-positive cases. The mRNA and protein of ALK were aberrantly expressed in all EML4-ALK-positive carcinoma lesions, but were absent or expressed at a low level in patients without the fusion gene, suggesting an important role of EML4-ALK translocation in tumorigenesis. The EML4-ALK translocation is detected more frequently in younger patients and in undifferentiated carcinomas. These results may provide useful insights to the knowledge of NSCLC and facilitate the diagnosis and targeted treatment of the disease, leading to an improved treatment outcome and patient life quality.

Acknowledgements

This study was supported by the Henan Province Science and Technology Department (grant no., 132300410448).

Abbreviations:

References