Prevalence and Patterns of EGFR Mutations in Non-Small Cell Lung Cancer in the Middle East and North Africa: A Systematic Review

We conducted a systematic review of literature published on EGFR mutation prevalence and its association with geographic region/country and clinic-pathological features in NSCLC patients in MENA region. We carried out a literature search of original articles published in six databases (PubMed, Science Direct, Web of science, Embase, Scopus, and Google scholar) from the time of inception until April 2021. Included articles have been published in English in Abstract To summarize current evidence and estimate the prevalence of epidermal growth factor receptor (EGFR) mutation frequency and its association with ethnicity and clinic-pathological features in non-small cell lung cancer (NSCLC) patients in the Middle East (ME) and North Africa (NA), a systematic literature review was undertaken. We conducted a literature search of original articles published in six databases (PubMed, Science Direct, Web of Science, Embase, Scopus, and Google scholar) from the time of inception until April 2021. Search terms included “lung cancer”, “NSCLC”, “EGFR mutation”, “Middle East”, “North Africa”, and specic country names belonging to the considered region. The included studies had to meet the following criteria: the study must relate to the role of the EGFR gene in NSCLC, analyze mutations in exon 18, 19, 20, and 21 or select exons of the EGFR gene, and provide sucient information on the clinic-pathological characteristics of the included NSCLC patients. A total of 24 eligible studies were included [(66.6%) in the ME and (34.4%) in NA]. Overall, 6544 patients with NSCLC were analyzed for EGFR mutations [(55.1%) in the ME and (44.8%) in NA]. The overall prevalence of EGFR mutations was 17.9%. In the ME, the reported frequency was 17.3%, whereas in NA, the prevalence of EGFR mutations was 18.5%. The most frequently encountered mutations were the exon 19 deletions (45.2%) and exon 21 substitutions (30.9%). Exon 20 alterations were detected in 11.2%, of which, the T790M resistance mutation was the most prevalent (45.5%). Exon 18 mutations were reported in 3.8%. In the ME, 50.5% of NSCLC patients were positive for exon 19 deletions versus 48.3% in NA. Exon 21 mutations were slightly more commonly detected in the ME (36.3%) than NA (31.3%). There was 1.2% of patients that had concurrent EGFR mutations. Overall, EGFR mutations prevalence was higher in females, non-smokers, and patients with adenocarcinoma. Our systematic literature review concurs that EGFR mutation prevalence among MENA populations is slightly higher than that seen in NSCLC patients of Caucasian ethnicity but is lower than that identied in Asian NSCLC patients. The distribution of these mutations varies signicantly throughout the MENA region.


Introduction
Lung cancer occurred in approximately 2.2 million patients, representing 22.7% of the global cancer burden. In 2020, 1.8 million patients died of lung cancer [1]. It is the leading cause of cancer morbidity and mortality in men, whereas in women, it is the third most common cancer, behind breast and colorectal cancers, and the second leading cause of female cancer death. Incidence and mortality rates are roughly 2 times higher in men than in women, and the maleto-female ratio varies widely across regions, ranging from 1.2 in Northern America to 5.6 in Northern Africa [1]. The incidence and mortality estimates for lung cancer are 3 to 4 times higher in countries with a high Human Development Index (HDI) than in countries with a low HDI; this pattern may well change as the tobacco epidemic evolves given that 80% of smokers aged ≥15 years resided in low-income and middle-income countries in 2016 [2]. Worldwide, the lung cancer mortality rate is foreseen to increase up to 3 million by 2035. The gures are set to double for both genders (men: from 1.1 million in 2012 to 2.1 million by 2035; and women: from 0.5 million in 2012 to 0.9 million by 2035) and the existing gender gap is expected to persist. Most prominent increases are expected in Africa and the Eastern Mediterranean region [3].
The Middle East (ME) and North Africa (NA) countries have witnessed a steady increase in the incidence rates of lung cancer [4]. In 2018, an estimated new 79887 lung cancer cases were registered in the MENA region versus 470000 new diagnoses in Europe. The age-standardized incidence rate of lung cancer in the MENA region is less than international rates, with gures varying from lowest in Yemen (4.2 per 100,000) to highest in Lebanon (23 per 100,000) [5]. Lung cancer incidence rates increases are more eminent among older age groups in the MENA area [6]. Despite recent breakthroughs in lung cancer management, the 5-year relative survival rate in the region doesn't surpass 8%. This is largely due to late diagnoses. In the MENA countries, the highest mortality rates were reported in Morocco and Tunisia, whereas the lowest were in Yemen and Egypt [5].
Lung carcinomas are categorized by the size and appearance of the malignant cells and are divided into two broad categories of small cell lung cancers (SCLC) and non-small cell lung cancers (NSCLC). SCLC comprises about 10%-15% of all lung cancers. NSCLC is the most common type of lung cancer and accounts for 80-90% of all lung tumors. SCLC, commonly centrally located in the major airway, tends to grow and spread faster than NSCLC. It is estimated that 70% of SCLC patients present with locally advanced or distant metastatic disease at the time of diagnosis [7]. NSCLC is a highly heterogeneous disease and is mainly divided into three major histological subtypes: adenocarcinoma, squamous cell carcinoma, and large cell carcinoma; it harbors various genetic alterations within each subtype. The identi cation of mutations in certain histological subtypes has led to molecular sub-classi cation of NSCLC and also opened therapeutic opportunities for personalized medicine based on targeted drugs [8,9]. Several mutations in NSCLC are considered actionable with available or promising targeted therapies. Some of the most common mutations for NSCLC occur in epidermal growth factor receptor (EGFR) and favor cell survival, proliferation and migration, and metastasis development by increasing the activity of EGFR tyrosine kinase [10]. EGFR tyrosine kinase inhibitors (TKIs) are established effective therapies in patients who have mutations in exons 18, 19, 20, and 21 of EGFR, leading to longer progression-free survival intervals with fewer or at least different side-effects than chemotherapy [11,12].
Previous studies have established marked variations in EGFR mutation rates depending on different geographic locations and race/ethnicity backgrounds. It occurs at the rate of 10-15% in North Americans and Europeans, 15-20% in African-Americans, 20-30% in various East Asian series (Chinese, Koreans, Japanese), and 20-25% in patients from the Indian subcontinent [13][14][15][16][17]. In ME and African populations, the EGFR mutation frequency is higher than that shown in white populations but still lower than the frequency reported in Asian populations [18]. In the MENA, the frequency of EGFR mutations is considered among the lowest. To summarize current evidence and estimate the prevalence of EGFR mutations and its association with geographic region/country and clinic-pathological features of EGFR mutation-positive NSCLC patients in the Middle East (ME) and North Africa (NA), A systematic literature review was undertaken in Bahrain, Egypt, Iran, Iraq, Jordan, Kingdom of Saudi Arabia, Kuwait, Lebanon, Oman, Palestine, Qatar, Syria, Turkey, United Arab Emirates, Yemen, Algeria, Egypt, Morocco, and Tunisia. indexed and peer-reviewed journals. Search terms included lung cancer, or lung tumor, or lung adenocarcinoma, or NSCLC, or EGFR, or EGFR mutation, or EGFR oncogene mutations, or EGFR oncogenic driver mutation, or EGFR activating mutation, or EGFR prevalence, or EGFR rate, or EGFR incidence or EGFR frequency. An additional literature search was also conducted using Middle East, Middle Eastern, North Africa, North African and speci c country names belonging to the considered region and any other variant names for any of the MENA countries (ex: Maghreb, Levant, Gulf, Arab). We manually checked reference lists of the included studies and relevant review articles to identify additional studies. We also searched relevant abstracts reported in the most important multi-disciplinary societies of medical oncology such as the American Society of Clinical Oncology (ASCO) meetings to identify unpublished studies.
A total of 24 studies met the inclusion criteria. In most studies, materials were formalin-xed para n-embedded (FFPE) tissues and included small biopsies such as trans-bronchial biopsy or tru-cut biopsy and also resection materials. DNA extraction was applied on tissue samples using kits that extracted DNA from para n blocks. Mutations in exon 18 (codon 719), exon 19 deletions, exon 20 (codons 768 and 790), and exon 21 (codons 858 and 861) were assessed in 79.1% (19/24) of the studies. A wide variety of detection methods were used to identify recognized mutations of the EGFR kinase domain, from exon 18 to 21. Direct sequencing was broadly used, as it was used in the most of the studies [19-21, 23, 24, 26, 33, 34, 36, 37, 40]. RT-PCR-based assays, namely scorpions-ampli cation refractory mutation system (ARMS/Scorpion) methodology, was also widely used [27, 29, 31-33, 37, 40]. EGFR mutation analysis was carried out with quantitative PCR analysis in the study from Gulf region [41]. The INFINITI system using BioFilmChip-based microarray assay was used in one study from Turkey [22]. Details of the study methods and population characteristics are summarized in Table 1.  Table 2.

Discussion
The identi cation of EGFR mutations in tumors of NSCLC patients has led to personalized molecular therapies and to a paradigm shift for patients with lung cancer candidates for targeted therapy. Furthermore, it has been established that EFGR mutations are key diagnostic biomarkers in NSCLC, therefore NSCLC patients genotyping for these alterations should be a standard of care right along standard clinical examination, pathology and imaging studies [43]. Practice guidelines outlined by the National Comprehensive Cancer Network (NCCN) and the European Society for Medical Oncology (ESMO) now include EGFR genotyping to guide therapy selection [44,45].  [50][51][52][53]. In our systematic review, an overall prevalence of 17.8% was identi ed in patients with EGFR mutation-positive NSCLC across the MENA region. The reported prevalence was slightly higher than those observed among Western populations but still lower than frequencies reported in Latino and Asian populations. In the Levant countries, a region anked by the ME and Europe, and the Gulf region (also known as Arabian Gulf), the reported EGFR mutation frequency was 15.6% [34] and 28.7 [28], respectively. The lowest mutation frequencies were seen in Lebanon (8.8 to 12,7%) [31][32][33]. Among the Turkish population, an EGFR mutation frequency of 42.6% in NSCLC patients was identi ed in western Turkey [25], when Tezel et al., showed that the mutations rate in Turkish patients with EGFR mutation-positive NSCLC was 16.7% [25]. Regional distribution of genetic mutations of lung cancer in Turkey, as reported in the (REDIGMA) study, including 25 centers, showed that mutation tests were found to be positive in 18.9% of these patients. The mutations were 69.9% EGFR, 26.3% ALK, 1.6%ROS and 2.2% PDL [26]. Our systematic review also highlights a wide range in EGFR mutation frequencies in NA populations. The overall EGFR mutation rate of NSCLC patients varied from 15.9% [37] to 26.8% [35] in Morocco. Additionally, one Moroccan study showed similar EGFR incidence rates (21%) as in patients of Caucasian descent [36]. An overall rate of 39.6% was found in EGFR mutation-positive NSCLC patients in Algeria [42]. In Tunisia, while rst reports account for an EGFR mutation frequency as low as 5.5% [38], other reports show discrepant data of 11.5% and 44% [40,39].
The mechanism behind the differences of EGFR mutation rates across geographic regions and the race/ethnicity is still unclear. A persistent nding in the literature is the substantial variation in EGFR mutation prevalence across different geographic areas and among various race/ethnicity backgrounds [54]. Although such mutations are over-represented in more than 40% of EGFR mutation-positive NSCLC in Japan and China, they are detected in roughly 15% of EGFR mutation-positive NSCLC patients in France and Italy [48]. It has been demonstrated that ethnic genetic variation may explain these differences [55,56].
In the ME, the frequency of EGFR mutations was reported to range between 16.6% and 44% in the Turkish population [22][23][24][25][26]. This disproportion is a result of the genetic heterogeneity and the ethnic diversity that characterize Turkey, a country endowed with a distinguished geographic location that is between Europe and Asia and near the ME. In NA, the EGFR mutation frequency in Tunisians range from 5.5% and 44% [38][39][40]. This disparity in frequencies is mainly attributable to the ethnicities that have succeeded in Tunisia, contributing to this country's ethnic diversity and therefore genetic heterogeneity [57].
Some studies showed that difference in EGFR mutations frequencies might be caused by exposing to indoor and/or outdoor air pollution [58]. The unique EGFR mutation spectrum in southwestern China might be related to the exposure of air pollution from local smoky coal and can re ect a speci c environmental exposure [59].In Europe, a positive association between various indicators of indoor air pollution and lung cancer risk has also been reported Several studies have reported a higher incidence of EGFR mutations among women in comparison to men, with gures up to 69.7%. In effect, up to 42% of females versus only 14% of males with NSCLC are expected to harbor an EGFR TK domain mutation [46,47,66]. In our review, EGFR mutation prevalence was higher in females (females versus males: 33.4% versus 17%).This is similar to data from Europe, Spain and other Asian studies which concluded that EGFR mutations were more common in women [67][68][69]. A systematic review covering 151 worldwide studies published in 2014 observed that the EGFR mutationpositive proportions were 60% and 37% in women and in men, respectively [48]. Previous studies showed that women can be more exposed to domestic radon which poses a risk for lung cancer at exposure levels approaching those for underground miners [70]. Others studies reported that domestic radon is associated with a low excess risk for lung cancer [71,72]. Generally, women tend to be non-smokers or light smokers compared to men, but their domestic lifestyle may expose them to certain indoor mutagen. If the occurrence of EGFR mutations is associated with potential indoor mutagens, women would have a higher mutation rate than men [73]. Furthermore, female endocrine factors such as progesterone receptor and aromatase expression could also play a role in the prevalence of the EGFR mutations [74]. Further studies are needed to investigate the role of hormones in EGFR mutation-positive NSCLC.
In our review article, the prevalence of EGFR mutations was more than two folds higher in non-smokers than current smokers (31.1% versus 11.1%). In European or American studies, EGFR mutations rate in non-smokers ranged from 10-30% [75]. EGFR mutations are the most common driver gene found in never-smoker adenocarcinoma from East Asia, constituting 60-78% of this subgroup [76][77][78]. While some studies found that non-smokers were associated with a signi cantly higher EGFR mutations prevalence [79]. Others have reported an association between EGFR mutations and the amount and duration of cigarette smoking, with a higher incidence of mutations than that seen in never smokers [80]. Furthermore, clinical studies have suggested that the pathogenesis, clinical manifestation, and prognosis of non-smokers and smokers are different in lung cancer tumors [81-83]. Genetic differences have been also found in the tumors of non-smokers versus smokers [84,85]. The proportion of non-smokers with NSCLC is increasing. Multiple environmental factors are implicated in lung carcinogenesis including exposure to secondhand tobacco smoke, pre-existing lung diseases, and family history of cancer. Exposure to industrial substances such as toxin (ex: arsenic, nickel, chromium, tar, soot), some organic chemicals (ex: radon, asbestos), radiation exposure, air pollution, tuberculosis and environmental tobacco smoke in non-smokers also increases the risk of developing lung cancer. More thorough investigations are needed to pinpoint causal mutagens and determine the amplitude of their potential mutagenic capability.
Deletions in exon 19 and the single amino acid substitution L858R in exon 21 account for approximately 85%-90% of all EGFR mutations in NSCLC, they are the most common and can predict response to EGFR TKIs and confer sensitivity to EGFR TKIs [86]. Exon 18 and 20 insertion mutations are less common and represent the remaining 10% of EGFR mutants in NSCLC. The exon 20 T790M point mutation, and most EGFR exon 20 mutations, are predictive of treatment resistance to rst-and second-generation EGFR TKI therapies [44,87]. In our article review, the average frequency of the exon 19 and substitutions in exon 21 were 45.2% and 30.9%, respectively, among all EGFR mutations. Together, these two mutations account for up to 76.1% of identi ed EGFR mutations. Our ndings also identi ed potential EGFR TKI-resistant mutations in 11.2% (112/998) among which, the T790M substitution was the most prevalent resistance mutation to rst-generation TKI (45.5%, 51/112). The low frequency of exon 19 del and the point mutation L858R at exon 21 (73.4%) among the MENA population is likely the result of the heterogeneity in screening and targeted methods, potentially engendering inaccuracies in the incidence rates of otherwise common EGFR mutations. Direct sequencing was the most commonly used methodology in MENA studies (45.9%, 11/24). However, Direct sequencing has some critical limitations among which the low mutation detection sensitivity; below a certain threshold of mutant DNA, mutations could not be detected. . The molecular characterization of peripheral blood may provide a strategy for the non-invasive serial monitoring of tumor genotypes during treatment, particularly for the EGFR T790M mutation [90]. The frequency of T790M mutation depends on the types of assays for this mutation [91]. Oxnard et al. found that 31% of NSCLC patients who are negative for T790M on central tumor genotyping have detectable T790M in plasma and recommend that tissue biopsy T790M genotyping would be substituted by liquid biopsy [92]. Other plasma assays have similarly identi ed unexpected false-positives for T790M in the absence of falsepositives for other mutations [93]. T790M mutational analysis in liquid biopsies is currently incorporated in recent guidelines for the management of acquired TKI resistance [94]. Recent studies have con rmed that EGFR mutations from plasma can predict the clinical response to targeted therapy [95,96]. In the MENA region, the T790M mutation, using liquid biopsy, has been conducted only in NSCLC patients from Lebanon [97]. In addition, Next Generation Sequencing (NGS) has the ability to detect the whole exome or genome and is not restricted to speci c target sequences. NGS can simultaneously analyze multiple variations, including uncommon alterations. Uncommon EGFR mutations make up a highly heterogeneous subgroup of NSCLCs that account for approximately 10%-18% of EGFR-mutated patients, and NGS testing can broaden the spectrum of alterations within the uncommon group in NSCLC patients [98]. However, Non-invasive plasma-based detection of EGFR mutations using digital PCR is still the most suitable method in clinical EGFR testing, thanks to its higher sensitivity, easier-to-understand results, low turn-around time and low cost to predicting the e ciency of EGFR-TKI [96].
This report revealed that the molecular epidemiology of EGFR mutations is heavily in uenced by ethnicity and geography; EGFR mutations were found to be more frequent in patients in the MENA region than in patients of caucasian ancestry, in contrast, the rates reported among Asian populations were quite higher. Although results from this study were consistent with ndings in previous reports, they should be considered cautiously due to some limitations. Firstly, a considerable portion of the considered studies have low statistical power as 8 of them included less than 100 patients. This could misrepresent the true prevalence of EGFR mutations in the region. Also, data about the stage of the tumors lacked from the majority of the included studies. Therefore, the correlation of tumor-stage and EGFR mutational status remains unde ned in the region. Furthermore, the majority of the analyzed cases of the studies had adenocarcinomas, consequently, the reported in uence of this particular histological subtype on EGFR mutational status could be inaccurate. Despite these limitations, a major strength of this review is the inclusion of available studies from a wide range of countries in the region. These estimates can serve as a reference for future research or policy making. Since EGFR mutation rates vary depend depending on, inter alia, ethnicity, NSCLC patients genotyping should be a standard of care in the MENA region in order to have more accurate and realistic data on EGFR mutation frequencies.
Declarations Ethics approval and consent to participate: Not applicable. Consent for publication: Not applicable.
Availability of data and materials: The data that support the ndings of this study are available from original articles that have been included in this study.
Data are available from the authors upon reasonable request from the corresponding author.
Competing Interest: The authors declare no competing interest.
Funding: This research received no external funding.