A study of therapy targeted EGFR/ALK mutations in Indian patients with lung adenocarcinoma: A clinical and epidemiological study

a b s t r a c t
Background: Established predictive biomarkers for Non-Small Cell Lung Carcinoma (NSCLC) include sensitizing Epidermal Growth Factor Receptor (EGFR) mutations and Anaplastic Lymphoma Kinase (ALK) fusion oncogene. The primary aim of the study is to ascertain the prevalence of EGFR mutation and ALK gene rearrangement in patients of lung adenocarci- noma in Indian population and the second objective is to impress upon the importance of adequate processing of limited tissue samples.Methods: Histopathologically confirmed cases of lung adenocarcinoma, whose tumour had been tested for both EGFR and ALK gene mutations, were included in this study. The EGFR mutations were analyzed using PCR and Gene Sequencing. ALK fusion oncogene was found by Fluorescence In Situ Hybridization (FISH) technique using kit of Vysis LSI ALK Dual colour Break Apart Rearrangement probe.Results: A total of 152 cases of lung adenocarcinoma were included. Out of which, 92 (60.5%) were male and 60 (39.5%) were female. After exclusion of 17 cases due to unsatisfactory result, EGFR mutations were found positive in 35.5% cases (48/135). ALK gene rearrangement was found in 7.6% (10/131) after excluding 21 cases with unsatisfactory result.Conclusion: EGFR mutations and ALK gene rearrangement was found to be mutually exclu- sive. Incidence of EGFR mutations (35.5%) is much higher in Indian population than in Caucasians (13%) and is close to the incidence in East Asian countries. The 7.6% incidence of ALK fusion oncogene in Indian patients establishes the importance of molecular studies to give maximum benefit of targeted therapy to the patients.

Introduction
Targeted therapies have given new lease of life to cancer patients. Latest to join this field of individualized therapy are patients with lung malignancy. Lung cancers are the leading cause of death all over the world. Depending on their biology, therapy and prognosis, lung cancers have been divided by the WHO in two categories, i.e. Small Cell Lung Carcinoma (SCLC) and Non-Small Cell Lung Carcinoma (NSCLC).1 Several biomarkers have been recognized as prognostic and predictive markers for NSCLC. Established predictive biomarkers are sensitizing Epidermal Growth Factor Receptor (EGFR) muta- tions and Anaplastic Lymphoma Kinase (ALK) fusion onco- gene.2,3 As per the guidelines of the National Comprehensive Cancer Network (NCCN) version 5.2015, testing for EGFR mutations and EML4-ALK is recommended in the NSCLC EGFR mutations as compared to Caucasian.5 There have not been many large studies conducted in Indian population for both EGFR and EML4-ALK mutations. Routine molecular testing of tumour samples is an important paradigm shift in NSCLC therapy and highlights the role of diagnostic faculty in obtaining adequate and representative tissue sample and its proper analysis. The primary objective of the study was to find the prevalence of EGFR mutation and EML4-ALK fusion in patients of lung adenocarcinoma in Indian population and the second objective was to impress upon the importance of proper and adequate processing of critical tissue sample for the total benefit to the patient.All the lung carcinoma patients having Adenocarcinoma histology diagnosed during the period February 2013 to February 2015 and, whose tumour had been tested both for EGFR mutational status and ALK gene rearrangement, were included in this study conducted at our centre. Clinical details of the patients were taken from MDTC registry of our centre and from other treating oncophysician of Pune region. The molecular investigations were outsourced.

The EGFR mutations were tested using PCR and Gene Sequencing. Testing was done on micro-dissected cells from Formalin-Fixed, Paraffin-Embedded (FFPE) tissue blocks. The assay was done on tumour rich region of the FFPE tissue with >40–50% cancer cells. Exon 18–21 of EGFR gene was screened for mutations. Vysis LSI Dual colour Break Apart Rearrange- ment probe kit was used in Fluorescence In Situ Hybridization (FISH) for ALK fusion oncogene detection. It has a spectrum green (green) labelled 300 kb probe at the 50 end and a spectrum orange (red) labelled 250 kb probe at the 30 end of ALK. On evaluation of signals for each probe, two types of ALK rearrangement were identified (Fig. 1). One has split green and red signals with split distance being more than or equal to two signal diameters. The second pattern was loss of green signal because of deletion in the 50 ALK region in association with 2p inversion. Minimum of 50 tumour cells were assessed and cases were termed positive for ALK rearrangement only when more than 15% of the cells revealed single red signals or split signals.3 If no signals were detected even after repeat test, the report of ‘‘test uninterpretable’’ was issued. During the later period of study standardized immunohistochemistry (IHC) instead of FISH was used for screening of ALK gene rearrangement in few patients due to financial constrains. IHC was carried out on the Benchmark XT Autostainer, Roche Diagnostic’s using Rabbit monoclonal antibody D5F3 clone, company Ventana. Optiview DAB IHC detection kit and Optiview amplification kit was used for high sensitivity. The positive and negative tissue controls were run simultaneously. Presence or absence of cytoplasmic granular staining in tumour cells was taken as positive or negative for ALK mutation.Statistical analysis was done by using Pearson’s Chi- squared or Fischer’s exact test, whichever was appropriate for categorical variables. The power of study for EGFR mutation was found to be is 96.4% and for ALK mutation it was 100% at 5% level of significance using one sample formula for percentage. A two sided P < 0.05 was considered significant. Statistical analysis was done using IBM SPSS statistic V22.0 software.

Results
Total of 152 cases of lung adenocarcinoma were included in this study that had undergone molecular studies for EGFR mutations and ALK gene rearrangement status. Out of 152 cases, 92 (60.5%) were male and 60 (39.5%) were female with median age of 57.5 years (range 25–86 years). Baseline patients' characteristics are given in Table 1. Median age and age range in years along with distribution of genders are depicted.The EGFR and ALK gene rearrangement status of cases are shown in Table 2. Forty-eight cases were positive for EGFR mutations out of 152 blocks analyzed. However, in 17 cases (11.2%) result could not be obtained due to scanty or improperly processed tissue. So out of 135 cases with available results the positivity rate of EGFR mutation is 35.5%. Eighty- seven patients had EGFR of wild type. EML4-ALK fusion gene found positive in 10 cases out of 152 however after taking out type of deactivating mutation in exon 20 in which patient does not benefit from TKI therapy.

Discussion
21 cases which yielded unsatisfactory results on molecular analysis the positivity rate for ALK gene rearrangement was 7.6%. One hundred and twenty-one cases were negative for EML4-ALK fusion gene (Fig. 2). Table 3 depicts the association of age and sex of patients with EGFR and EML4-ALK fusiongene mutation. The patients are considered in two groups (≤50 and >50 years) for data correlation and comparison with other studies. There is no significant association seen in cases of EGFR mutation. The EML4-ALK fusion gene mutation shows significant association with age and sex of the patients.Table 4 deals with the type of mutations seen in exon 18–21 amongst the EGFR positive cases and their frequency. Commonest mutation is encountered in exon 19, i.e. de- lE746-A750. Out of 48 positive cases, 34 (70.8%) were positive for exon 19 mutation. Second commonest is exon 21 mutation which is 20.8%. One patient had double EGFR mutation involving exon 20 and 21. In one case we found insertionAccording to NCCN guidelines version 5.2015, Mutational testing for EGFR and EML4-ALK is strongly recommended in all NSCLC favouring adenocarcinoma for targeted therapies. These specific therapies have resulted in the improvement of median survival by more than one year for the patients of advanced stage NSCLC.7Variable incidences of EGFR mutation have been reported by various studies done across different ethnicities in cases of NSCLC. The incidence of EGFR mutations has been found to be much higher in Asian population (47%) as compared to Caucasians (13%).5 The EGFR mutations occurs at the rate of 10–15% in North Americans and Europeans, 19% in African- Americans, 26–30% in various East Asian countries.

Limited literature is available regarding the Indian population, the study involving the largest numbers from India has reported incidence of EGFR mutations to be 26% in cases of lung adenocarcinoma which is comparable to the rates in East Asian countries.12 Other few studies from India reported frequency of EGFR mutation ranging from 29% to 51.8% along with evidence of female dominance.13,14 The incidence of EGFR mutations was found to be 35.5% in our study which is closer to incidence seen in East Asian population. In present study out of total of 54 analyzed female patients, 23 had EGFR mutations (42.6%) as compared to 25 male patients out of total of 81 (28.4%). However this difference was not found to be statistically significant as compared to other studies whichhave proved dominance of female gender in cases of EGFR mutation.12,15 The association of age and sex of patients with regard to EGFR was not found to be significant in our study may be due to the low median age of patients included in study as compared to other studies.12,15 Various studies have shown EGFR mutations to be more common in non-smokers then in smokers.15–17 However, in our study we could not establish the correlation because of lack of details regarding the smoking status of all the patients.Presence of activating or driver mutations in exon 18–21 of EGFR gene has been associated with response to the anti-EGFR therapies.7,8,16,18 The highest response rates to the TKIs were seen in cases particularly with exon 19 deletion, exon 21 L858R and 18 G719X.19

In our study the frequency of mutation was highest in exon 19 (70.8%) followed by exon 21 (20.8%). Both exon 18 and 20 show similar frequency of mutation (2.8%). Deletions (DelE746-A750) are common mutation in exon 19 but it is difficult to distinguish them. Results of our study are similar to study by Noronha et al. with 74% of their patient having in frame deletions in exon 19 and 23% had the L858R point mutation in exon 21 and 2.5% had the G719C point mutation in exon 18.21 In other reported studies in-frame deletion in exon 19 remain the commonest although, frequency varies from 50% to 61% of EGFR mutations, while approximately 36–43% are missense mutations in L858R in exon 21 and 4–9% of mutations are reported in exon 20.12,20,21 These variations in frequencies are probably due to varying sample size in different studies. Exon 21 mutations have been found to be more common in never-smoker females whereas exon 19 mutations are found to be more common in non- smoker males. It has been reported that exon 18 EGFR mutations are detected more frequently in younger patients.12,22 In our study patients with exon 21 mutations were more of female then male and one patient has doublemutation involving exon 20 and 21. Exon 20 insertion mutations in cases of lung carcinoma are known to be resistant to TKIs.23 One of the patients in our study has insertion type of deactivating mutation in exon 20.Predictive biomarkers for NSCLC also include the ALK fusion oncogene; EML4-ALK is the most common known ALK rearrangement in NSCLC.3 The overall incidence of ALK gene rearrangement in NSCLC is found to be approximately 2–7%.24 Some of the studies have reported the incidence ranging from of 5–13% in East Asian population.25 There are very few published reports from India on EML4 ALK mutations with incidence rate approximating 3%.15,26

In present study out of 152 patients of lung adenocarcinoma results regarding ALK gene mutation could not be established in 21 cases due to inadequate tissue or improper processing. Out of remaining 131 analyzed cases the positivity rate for ALK gene rearrange- ment was 7.6%. The ALK fusion gene in our study was found to be positive in younger male patients and this correlation was statistically significant and is comparable with other stud- ies.3,24,25 In present study all the cases were tested for both EGFR and EML4-ALK mutations and they were found to be mutually exclusive.15,27 In light of the approval of targeted crizotinib therapy for ALK mutation and to optimize the use of limited tumour tissue, it is recommended to test ALK simultaneously with EGFR mutations.FISH is the FDA approved method for detection of ALK mutations; other methods include IHC, reverse transcriptase polymerase chain reaction and DNA sequencing.28 FISH is a very expensive screening test and for developing country like ours IHC can turn out to be more affordable alternative. Few studies done in this field has reported a good concordance between the two, with a sensitivity and specificity for FISH after IHC of 35% and 100% respectively.6 In present study few cases were tested by IHC for ALK mutations but positive cases should have been confirmed by FDA approved FISH method. That could be the reason for slightly high frequency seen for ALK mutations in our study. Given the absence of FISH positivity in IHC negative samples, IHC could become primary screening tool for ALK mutations in NSCLC hence reducing the cost.6 However further large studies are required for valida- tion.In the majority of cases a diagnosis of NSCLC is obtained from small tumour biopsies or cytological smears rather than surgical samples.28 Limited IHC panel of TTF-1 and p63 or p40 is enough to differentiate between adenocarcinoma and squamous cell carcinoma of lung. It will help to conserve the critical tissue for molecular studies. In present study, 17cases (11.2%) yielded inconclusive result for EGFR status and 21 cases (13.8%) for ALK gene rearrangement, highlighting the drawbacks of FFPE processing for molecular diagnosis. Formaldehyde tissue fixation not only causes strong cross linking of nucleic acids but also degrades and modifies the nucleic acid.29 Long formalin fixation causes low hybridization efficiency. The small tissue biopsies get fixed faster hence should not be kept in fixative for long time. Eosin stains is known to interfere with fluorescence signals so the biopsy samples should not be marked with eosin during processing. It is essential to balance use of the available tumour tissue for maximum benefit to the patient.

Conclusion
Our study reports data on simultaneous testing of EGFR and ALK mutation in lung adenocarcinoma patients and confirms the exclusivity of these mutations. It establishes the impor- tance of molecular testing in Indian population because of higher prevalence rates. It highlights the importance of optimal processing of biopsy tissue for maximum benefit to the Envonalkib patients.