Paik, Jin Ho MD, PhD*; Choe, Gheeyoung MD, PhD*; Kim, Hyojin MD*; Choe, Ji-Young MD*; Lee, Hyun Ju MD*; Lee, Choon-Taek MD, PhD†; Lee, Jong Seok MD, PhD†; Jheon, Sanghoon MD, PhD‡; Chung, Jin-Haeng MD, PhD*
Lung cancer is the leading cause of cancer-related deaths in the world, despite improvements in detection methods and treatments.1,2 Recent studies on personalized treatment by selecting patients who are likely to respond to a particular therapeutic agent may allow improved treatment efficacy.3–5 Most notably, a subset of patients with non-small cell lung cancer (NSCLC) with activating mutations in the epidermal growth factor receptor (EGFR) gene show dramatic responses to drugs that inhibit EGFR kinase activity, resulting in prolonged survival.3,5 It is expected that the use of targeted therapies for NSCLC in biomarker-selected patients will increase.
Recently, activation of the anaplastic lymphoma kinase (ALK) gene in lung cancer by fusion to echinoderm microtubule-associated protein-like 4 or other gene partners was reported.6 ALK is a receptor tyrosine kinase described in anaplastic large cell lymphoma with t(2;5)(p23;q35) resulting in a nucleophosmin-ALK fusion protein. The ALK-signaling pathway includes several biologically important pathways involving cell proliferation, differentiation, and antiapoptosis.4 In NSCLC, the ALK gene is most commonly fused with echinoderm microtubule-associated protein-like 4, and the frequency of the fusion gene is approximately 3 to 6.7%.4,7–10 A novel ALK tyrosine kinase inhibitor named crizotinib (PF-02341066) is currently in phase II and phase III clinical trials for advanced NSCLC with ALK rearrangement, and early studies have demonstrated dramatic clinical responses in patients with ALK-rearranged NSCLC assessed using fluorescence in situ hybridization (FISH).11,12
The introduction of an effective and widely applicable screening method using immunohistochemistry (IHC) to detect NSCLC with ALK rearrangement is vital because ALK rearrangement status has been determined in few patients at the time of excision of the primary lesion. Nonetheless, ALK IHC testing in lung cancer remains challenging because of the relatively low level of ALK transcription, few data about the sensitivity or specificity of the test, and the absence of a universally accepted and evidence-based IHC scoring algorithm to predict ALK rearrangement using FISH.4,7–10,13
FISH is standard method to detect ALK rearrangement, but it is not readily available to as a routine method in the pathology practice. As IHC is a commonly used technique in the pathology laboratories, several strategies have been tried to improve the accuracy of the ALK IHC assays in lung cancer including amplification of the signal with a tyramide cascade,14 intercalation of an antibody-enhanced polymer,15 or development of novel antibodies.13
In this study, we semiquantitatively evaluated ALK rearrangement using IHC and correlated the results with results obtained using FISH. The aim of this study was to acquire the baseline ALK rearrangement status of Korean patients with NSCLC and to determine a sensitive screening method using IHC to predict FISH results.
PATIENTS AND METHODS
Patients and Samples
This study included 465 consecutive patients with NSCLC who underwent surgical resection at the Seoul National University Bundang Hospital in Korea from May 2003 to May 2008 as a test set using tissue microarray (TMA). The cohort consisted of 269 cases of adenocarcinoma (ADC), 169 cases of squamous cell carcinoma, 10 cases of adenosquamous carcinoma, five cases of pleomorphic carcinoma, two cases of large cell carcinoma, eight cases of large cell neuroendocrine carcinoma, one case of carcinosarcoma, and one case of lymphoepithelioma-like carcinoma. The mean age of the 465 patients was 63.8 years and ranged from 21 to 84. To validate our results, we examined whole sections of surgically resected or biopsied samples from 187 patients with ADC at the Seoul National University Bundang Hospital in Korea from September 2009 to May 2010 using IHC. This study was approved by the institutional review boards at the Seoul National University Bundang Hospital.
Construction of the TMA
One representative core tissue (2 mm in diameter) was taken from a paraffin block of each case and arranged in new TMA blocks using a trephine apparatus (Superbiochips Laboratories, Seoul, Korea), as described previously.16 Two serial sections were cut and examined using IHC and FISH.
Formalin-fixed and paraffin-embedded (FFPE) tissues were sectioned at a thickness of 4 μm and stained using a Ventana automated immunostainer (Ventana Medical Systems, Tucson, AZ) according to the manufacturer's protocol. Briefly, the slides were dried at 60°C for 1 hour and deparaffinized using EZ Prep (Ventana Medical Systems) at 75°C for 4 minutes. The cells were conditioned (heat pretreatment) using a CC1 solution containing Tris/borate/ethylenediaminetetraacetic acid at 100°C for 20 minutes. The antibody for ALK (mouse monoclonal, clone 5A4, Novocastra, Newcastle, United Kingdom) was diluted to 1:30, treated, and incubated at 42°C for 2 hours. Signals were detected using an i-view detection kit (Ventana Medical Systems) based on the labeled streptavidin-biotin method. The steps used with the kit included treatment with an inhibitor (1% H2O2, 4 minutes), biotinylated immunoglobulin (8 minutes), streptavidin-horseradish peroxidase (8 minutes), diaminobenzidine (chromogen + substrate) (8 minutes), and copper (4 minutes) at 37°C. Counterstaining was performed with Mayer's hematoxylin (ScyTek, Logan, UT) for 2 minutes at room temperature.
Fluorescence In Situ Hybridization
FISH was performed on FFPE tumor tissues using a break-apart probe specific to the ALK locus (Vysis LSI ALK dual-color, break-apart rearrangement probe; Abbott Molecular, Abbott Park, IL) according to the manufacturer's instructions. FISH-positive cases were defined as those presenting more than 15% split signals or an isolated red signal (IRS) in tumor cells, as described previously.17,18 Briefly, 3-μm-thick sections from FFPE tissue blocks were deparaffinized, dehydrated, immersed in 0.2 N HCl, and washed. The sections were immersed in 0.01 M citrate buffer, boiled in a microwave for 5 minutes, treated with pretreatment reagent (Abbott Molecular) at 80°C for 30 minutes, and reacted with protease mixed with a protease buffer. Dual-probe hybridization was performed using the LSI ALK dual-color probe, which hybridizes to the 2p23 band with SpectrumOrange (red) and SpectrumGreen on either side of the ALK gene breakpoint (Abbott Molecular). After applying the probe mixture, they were treated with protease, incubated in a humidified atmosphere with HybriteTM (Abbott Molecular) at 75°C for 5 minutes to denature the probe and target the DNA, and incubated at 37°C for 16 hours to allow hybridization. They were then immersed in 0.3% NP-40 (Abbott Molecular)/2× saline sodium citrate for washing. For the nuclei counterstaining, 4,6-diamidino-2-phenylindole II and an antifade compound (p-phenylenediamine) were applied. Signals for each probe were evaluated under a microscope equipped with a triple-pass filter (diamidino-2-phenylindole/Green/Orange; Abbott Molecular) and an oil immersion objective lens. FISH tests were performed without knowledge of the IHC results for ALK.
EGFR and K-ras Mutation Study
All the cases that were determined to be ALK FISH- positive and some of the cases that were determined to be ALK FISH-negative were analyzed for EGFR mutations at exons 18 to 21 and K-ras mutations at codons 12, 13, and 61 using the polymerase chain reaction and a direct DNA sequencing method, as described previously.19 EGFR mutations in some patients included in this study had been identified previously.19
SPSS version 12.0 (Systat, Chicago, IL) was used for the statistical analysis, which included Pearson's χ2 test, Pearson's R test, and Fisher's exact test. Statistical significance was defined as p < 0.05.
ALK Immunoreactivity by IHC in the Test Set
ALK protein expression was observed in tumor cells with a predominantly cytoplasmic staining pattern. Nonneoplastic bronchial epithelium, alveolar type I and type II pneumocytes, mesenchymal tissue, and inflammatory cells in the adjacent lung tissue were negative for ALK. A few alveolar macrophages adjacent to positively stained tumor cells showed faint granular cytoplasmic staining. Nevertheless, alveolar macrophages were not stained in the NSCLC cases in the absence of ALK protein expression. The mucin-rich cytoplasm of the tumor cells showed faint, granular immunoreactivity for ALK. The expression of ALK was assessed semiquantitatively by three pathologists (J.H.P., G.C., and J.-H.C.) without knowledge of the results obtained using FISH or the clinicopathologic information. Semiquantitative assessment was performed by estimating the staining intensity and the percentage of tumor cells using positive cytoplasmic staining. Each cell was first scored as 0, 1, 2, or 3, which corresponded to negative, weak, moderate, and strong staining intensities, respectively. Next, we evaluated the percentage of positively stained cells. Increases in the staining intensity were associated with increases in the number of positively stained cells observed: a staining intensity of 1 resulted in an average of 14.7% positively stained cells, a staining intensity of 2 resulted in 58.2% positively stained cells, and a staining intensity of 3 resulted in 97.3% positively stained cells. ALK IHC scores were assigned as follows: 0, no stained cells; 1, faint or weak staining intensity with more than 5% tumor cells or any staining intensity with ≤5% tumor cells; 2, moderate staining intensity with more than 5% tumor cells; and 3, strong staining intensity with more than 5% tumor cells.
ALK protein expression was detected in 8.6% (40/465) of the NSCLC cases, which included scores of 0 (n = 425), 1 (n = 14), 2 (n = 10), and 3 (n = 16) (Figures 1A–D). The interobserver agreement between pathologists was excellent (κ value, 0.92).
ALK Rearrangement Assessed by FISH in the Test Set
ALK rearrangement was assessed using FISH in 453 patients (97.1%), and ALK rearrangement was identified in 19 patients (4.2%). The majority of the ALK FISH-negative cases showed two fusion signals or close proximity of the red and green signals. Nevertheless, a few cells demonstrated an isolated green signal (IGS) with loss of the corresponding red signal. As the tyrosine kinase domain resides in the 5′ end probe of ALK (the green signal), an IGS was considered to correspond to no ALK rearrangement. The clinical significance of IGS is unknown.
The ALK FISH-positive NSCLC cases showed two major patterns using the LSI ALK dual-color break-apart probe as follows: (1) the break-apart (split) pattern was observed in 73.7% of cases (14/19) (Figure 1E) and (2) IRS pattern: the isolated (single) red signal was predominant in 26.3% of cases (5/19) (Figure 1F). The interobserver agreement between pathologists was excellent (κ value, 0.94). Regarding discordant cases (usually IHC 2+ cases), three pathologists repeatedly examined, scored the cases, and discussed together, which led to the consensus scoring.
Correlation between ALK Protein Expression as Determined by IHC and ALK Rearrangement as Determined by FISH
To determine the best criteria for IHC assessment to predict gene rearrangements, we compared the ALK results using IHC and FISH in the test set (Table 1). The average percentage of cells positively stained using IHC was higher in the FISH-positive tumors than in the FISH-negative tumors (92.4% versus 19.3%). All the cases with IHC scores of 3 were FISH-positive, and all the cases with IHC scores of 0 or 1 were FISH-negative. For cases with scores of 2, 30% (3/10) were FISH-positive and 70% (7/10) were FISH-negative. The overall sensitivity of IHC was 100%, and the specificity was 95.2% in the test set. There were 21 false-positive cases (14 cases of score 1 and seven cases of score 2) and no false-negative cases using IHC. Assuming that ALK IHC scores of 0 and 1 were ALK rearrangement-negative, an ALK IHC score of 3 was ALK rearrangement-positive, and an ALK IHC score of 2 was equivocal, the results of the ALK IHC assay using the tiered scoring method and FISH were strongly correlated (p < 0.001). After excluding cases with ALK IHC scores of 2 (equivocal), the ALK IHC scores (0 and 1 versus 3) and ALK FISH (negative versus positive) were completely correlated. These results indicate that an ALK IHC score of 3 could strongly predict FISH-positivity and an IHC score of 0 or 1 could predict FISH-negativity. A diagnostic algorithm was derived from these results (Figure 2).
Validation of the IHC Interpretation Criteria to Detect ALK FISH-Positive NSCLC
As a validation set, we examined surgically resected or biopsied specimens from 187 patients with ADC and applied the IHC interpretation algorithm. Among these specimens, 14 were positively stained for ALK, including two cases of ALK IHC score of 1, six with a score of 2, and six with a score of 3 (Table 2). The FISH results were as follows: all the patients with an IHC score of 0 or 1 were FISH-negative, and six of the patients with an IHC score of 3 were FISH-positive. The cases with scores of 2 were 50% (3/6) FISH-positive (Table 2). Nine FISH-positive cases included seven break-apart (split) patterns and two IRS patterns. There were five false positives (two cases with a score of 1 and three cases with a score of 2) and no false-negative cases. The diagnostic algorithm derived from the test set was effective in the validation set.
Table 3 lists the correlation of the results using ALK IHC and FISH in all 640 cases, including the test and validation sets.
Analysis of Clinicopathologic Characteristics of Lung Cancer with ALK Rearrangement
Clinicopathologic factors associated with ALK rearrangement in the surgically resected NSCLC (test set)
Nineteen ALK FISH-positive patients (4.2%), composed of 11 men and eight women, presented ADC histology (except for one case of adenosquamous carcinoma). Neither EGFR nor K-ras mutations were concurrent with ALK rearrangement. The patients with ALK-rearranged NSCLC in the surgically resected group exhibited ADC histology (p < 0.001) and an absence of EGFR or K-ras mutations (p < 0.001). Nevertheless, sex, age, smoking habit, tumor size, and pathologic stage were not significantly different between ALK-positive and -negative groups (Table 4).
Clinicopathologic factors associated with ALK rearrangement in the consecutive ADCs (validation set)
ALK rearrangement was identified in 9 of 187 patients (4.8%) having ADC between September 2009 and May 2010 (Table 5). The mean age of these ALK FISH-positive patients was 48.7 years (range: 28–71). In contrast to the test set, ALK FISH-positive patients were younger than ALK FISH-negative patients (mean 62.0 years; range: 27–91) in the ADC-only validation set (p = 0.014). There were six women and three men constituting the ALK FISH-positive patients. Neither EGFR nor K-ras mutations were identified. In addition to the younger age, the patients with ALK rearrangement demonstrated an absence of EGFR or K-ras mutations (p < 0.001). Nevertheless, sex, smoking habit, tumor size, and pathologic stage were not significantly different between ALK FISH-positive and ALK FISH-negative groups. Even though nonsmokers tended to exhibit ALK rearrangement more frequently (6/104, 5.8%) than smokers (3/83, 3.6%), this was not statistically significant (Table 5, p = 0.733).
The clinicopathologic features of all 28 ALK FISH-positive patients are summarized in Table 6.
FISH analysis using the ALK break-apart probe was the standard test for enrollment in a clinical trial with crizotinib.12 Because the incidence of ALK rearrangement in NSCLC is relatively low and the ALK rearrangement status is infrequently determined at the time of excision of primary lesions, it is difficult to use FISH on all the biopsied or archived NSCLC samples. Therefore, the development of a screening method to identify ALK-rearranged tumors is imperative, and the diagnostic test needs to be applicable to archived, formalin-fixed tissues that have been removed up to several years earlier. Because IHC is readily available in pathology laboratories, it is important to optimize the condition of ALK IHC assay as a screening method and to establish an interpretation guideline.
In this study, we observed a good correlation between results obtained using IHC and FISH in a large-scale, single-institution study using a semiquantitative IHC scoring assessment. We also presented a diagnostic algorithm to screen for NSCLC with ALK rearrangements using IHC (Figure 2).
In the test set, good correlations were observed between cases assigned ALK IHC scores of 3 and FISH-positivity, as well as ALK IHC scores of 0 and 1 and FISH-negativity. Nevertheless, cases assigned ALK IHC scores of 2 showed variable FISH results. The proposed diagnostic algorithm was also effective in the validation set, which included 187 consecutive ADCs. Because there were no false-negative ALK IHC results using the 5A4 antibody (Novocastra) in 640 patients with NSCLC (453 in the test set and 187 in the validation set), ALK IHC would be a suitable method to screen for ALK rearrangement.
For FISH, we used break-apart probes specific to the 5′ and 3′ ends of ALK (Abbott Molecular) for identifying ALK rearrangement. Although this method is sensitive for detecting the breakage of the ALK, it does not identify the 5′ fusion partner or the exact breakpoint. To overcome these problems, several new methods including novel break-apart probes designed for the detection of both ALK and fusion partner genes are being explored or developed.20
In this study, there were two major patterns of ALK rearrangement, i.e., break apart (75%, 21/28) and IRS (25%, 7/28). The IRS pattern appears to be unique in NSCLC, unlike anaplastic large cell lymphoma. Although some reports have described this pattern previously in NSCLC, its relatively high frequency and correlations of immunoreactivity have not been reported.18,20 Significant differences in clinicopathologic factors were not identified between the two patterns of ALK rearrangement (data not shown). Although the mechanism of the IRS pattern is not yet clear, it might be associated with various breakpoints of the ALK gene. On the other hand, a few patients showed an IGS, where the tyrosine kinase domain resides. Loss of the red-labeled probe might be associated with the loss of the probe on 2p outside the area encoding for the ALK kinase domain. The clinical significances of IRS and IGS need to be clarified. In this study, it was impossible to assess the specimens using FISH in 12 cases of TMA due to signal loss or background autofluorescence. Inappropriate fixation in formalin before tissue processing may result in background autofluorescence or more aggressive tissue permeabilization, possibly leading to loss of signal and poor tissue preservation. The standardization of all preanalytical variables, including tissue handling, fixation, and processing, is important when using FISH.21
Patients exhibiting ALK rearrangement in the test set demonstrated ADC histology and the absence of EGFR or K-ras mutations. Sex, age, smoking habit, tumor size, and pathologic stage were not significant factors in the surgically resected NSCLC specimens. When the test and validation sets were integrated, the cohort of screened patients harboring ALK rearrangement was enriched in young patients, patients with ADC histology, and the absence of EGFR or K-ras mutations (Table 6). Sex, smoking habit, pathologic stage, and tumor size were not associated with ALK-rearranged NSCLC. These features were not fully consistent with previous reports in which the patients most likely to harbor ALK-rearranged NSCLC were men, young, never/light smokers with ADC histology.17 In our study, sex and smoking history were not significant clinical features, whereas young age was significantly associated with ALK rearrangement only in the consecutive ADC group (the validation set). These trends might be due to the low incidence of ALK rearrangement, which might affect the statistics by infrequently including patients with ALK-rearranged NSCLC. Therefore, a large-scale meta-analysis would be needed to delineate the specific features corresponding to ALK-rearranged NSCLC.
In summary, we present a diagnostic algorithm using ALK IHC to predict ALK gene rearrangement. ALK FISH results were completely predicted in patients with IHC scores of 0, 1, and 3, whereas patients with IHC scores of 2 demonstrated variable FISH results. This algorithmic approach might be useful as a screening method for ALK gene rearrangement. In addition, patients with ALK rearrangement in the test set demonstrated ADC histology and an absence of EGFR or K-ras mutations. Nevertheless, sex, age, and smoking status were not solid variables associated with ALK-rearranged NSCLC. Our results suggest that more careful interpretation of the clinicopathologic characteristics of ALK-rearranged NSCLC is necessary.
Supported by a grant-in-aid from the Korea Institute of Science & Technology (2E21496-09-300) and the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2010-0022451).
The authors are indebted to J. Patrick Barron, Professor and Chairman of the Department of International Medical Communication, Tokyo Medical University, for his pro bono review of this manuscript.
1. Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 2006;24:2137–2150.
2. Jemal A, Clegg LX, Ward E, et al. Annual report to the nation on the status of cancer, 1975–2001, with a special feature regarding survival. Cancer 2004;101:3–27.
3. Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med 2009;361:947–957.
4. Palmer RH, Vernersson E, Grabbe C, et al. Anaplastic lymphoma kinase: signalling in development and disease. Biochem J 2009;420:345–361.
5. Pao W, Miller VA. Epidermal growth factor receptor mutations, small-molecule kinase inhibitors, and non-small-cell lung cancer: current knowledge and future directions. J Clin Oncol 2005;23:2556–2568.
6. Soda M, Choi YL, Enomoto M, et al. Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature 2007;448:561–566.
7. Wong DW, Leung EL, So KK, et al. The EML4-ALK fusion gene is involved in various histologic types of lung cancers from nonsmokers with wild-type EGFR and KRAS. Cancer 2009;115:1723–1733.
8. Inamura K, Takeuchi K, Togashi Y, et al. EML4-ALK lung cancers are characterized by rare other mutations, a TTF-1 cell lineage, an acinar histology, and young onset. Mod Pathol 2009;22:508–515.
9. Boland JM, Erdogan S, Vasmatzis G, et al. Anaplastic lymphoma kinase immunoreactivity correlates with ALK gene rearrangement and transcriptional up-regulation in non-small cell lung carcinomas. Hum Pathol 2009;40:1152–1158.
10. Takeuchi K, Choi YL, Soda M, et al. Multiplex reverse transcription-PCR screening for EML4-ALK fusion transcripts. Clin Cancer Res 2008;14:6618–6624.
11. Camidge DR, Christensen J, Bang YJ. Addressing right drug-right target-right patient in phase I studies to accelerate bench to clinical benefit time: ALK gene rearrangements and the development of PF- 02341066 in NSCLC. J Thorac Oncol 2010;5:S233.
12. Bang Y, Kwak EL, Shaw AT, et al. Clinical activity of the oral ALK inhibitor PF-02341066 in ALK-positive patients with non-small cell lung cancer (NSCLC). J Clin Oncol 2010;28:18s.
13. Mino-Kenudson M, Chirieac LR, Law K, et al. A novel, highly sensitive antibody allows for the routine detection of ALK-rearranged lung adenocarcinomas by standard immunohistochemistry. Clin Cancer Res 2010;16:1561–1571.
14. Rodig SJ, Mino-Kenudson M, Dacic S, et al. Unique clinicopathologic features characterize ALK-rearranged lung adenocarcinoma in the western population. Clin Cancer Res 2009;15:5216–5223.
15. Takeuchi K, Choi YL, Togashi Y, et al. KIF5B-ALK, a novel fusion oncokinase identified by an immunohistochemistry-based diagnostic system for ALK-positive lung cancer. Clin Cancer Res 2009;15:3143–3149.
16. Lee HJ, Xu X, Choe G, et al. Protein overexpression and gene amplification of epidermal growth factor receptor in nonsmall cell lung carcinomas: comparison of four commercially available antibodies by immunohistochemistry and fluorescence in situ hybridization study. Lung Cancer 2010;68:375–382.
17. Shaw AT, Yeap BY, Mino-Kenudson M, et al. Clinical features and outcome of patients with non-small-cell lung cancer who harbor EML4-ALK. J Clin Oncol 2009;27:4247–4253.
18. Martelli MP, Sozzi G, Hernandez L, et al. EML4-ALK rearrangement in non-small cell lung cancer and non-tumor lung tissues. Am J Pathol 2009;174:661–670.
19. Chung JH, Choe G, Jheon S, et al. Epidermal growth factor receptor mutation and pathologic-radiologic correlation between multiple lung nodules with ground-glass opacity differentiates multicentric origin from intrapulmonary spread. J Thorac Oncol 2009;4:1490–1495.
20. Perner S, Wagner PL, Demichelis F, et al. EML4-ALK fusion lung cancer: a rare acquired event. Neoplasia 2008;10:298–302.
21. Hicks DG, Tubbs RR. Assessment of the HER2 status in breast cancer by fluorescence in situ hybridization: a technical review with interpretive guidelines. Hum Pathol 2005;36:250–261.