The incidence of oral cavity cancer in Iran: A systematic review and meta‐analysis

Abstract Background and Aims Oral cancer is now a top priority for non‐communicable illnesses and universal health care plans, according to the WHO. There is no general estimate of the incidence of oral cavity cancer in Iran, despite multiple investigations. The purpose of this study is to evaluate the age‐standardized incidence rate (ASR) of oral cavity cancers in Iran. Method In accordance with the MOOSE (Meta‐analyses of Observational Studies in Epidemiology) Checklist recommendations, this systematic review was conducted. PubMed/MEDLINE, Web of Science, ScienceDirect, Embase, Scopus, ProQuest, and Google Scholar were used as the international databases for the systematic literature search, while SID (Scientific Information Database), Magiran and element were used as the Iranian databases. The heterogeneity of the research will be evaluated by means of the inverse variance and Cochran Q tests, along with random‐effect models. It was determined what caused the heterogeneity using a meta‐regression model. By eliminating experiments one at a time, sensitivity analysis was used. The meta‐analysis was corrected utilizing the Trim‐and‐fill method due to the identification of noteworthy publication bias via the Egger's test and asymmetry of the funnel plot. Results This research incorporated a total of 22 journal articles. The pooled ASR of oral cavity cancer for males and females was estimated at 1.96 (95% CI: 1.65–2.26) (Q statistic = 1118.09, df = 25, p < .0001, I 2 = 97.8%), and 1.46 (95% CI: 1.14–1.77) (Q statistic = 2576.99, df = 26, p < .0001, I 2 = 99.0%), respectively. According to the funnel plots and Egger's test, there is no evidence of publication bias in studies reporting on males (bias = 1.3220, 95% CI: −3.9571, 6.6012, p = .610), but for ASR in females, Egger's test was significant (bias = −7.6366, 95% CI: 2.2141, 13.05904, p = .008). Based on Trim‐and‐fill methods, overall ASR corrected in females was estimated to be 1.36 (95% CI: 1.05%–1.66%). Conclusion Iran's oral cavity cancer incidence was lower than the global average, but owing to variables including an aging population, a rise in life expectancy, and exposure to risk factors like smoking, we anticipate an increasing trend.


| INTRODUCTION
Cancer is one of the leading causes of mortality globally. Globally, it was predicted that 10.3 million people will die from cancer in 2020 and 19.3 million people would get the disease. Oral malignancies account for around 3.2% of all cancer-related fatalities. 1 Oral cancer is now a top priority for non-communicable diseases and universal health coverage programs, according to the World Health Organization (WHO). 2 Oral cancer incidence and fatality rates vary considerably around the globe. Developing nations, particularly those in South Asia, including India, Pakistan, and Bangladesh, have the highest incidence of oral cancer. 3,4 The primary risk factors for oral cavity cancer include age, male gender, tobacco and alcohol consumption, exposure to UV radiation, immunosuppression (such as from HIV infection or organ transplantation), and human papillomavirus (HPV) infection. [5][6][7][8] In Iran, cancer is currently the second leading cause of death, resulting in over 70 000 deaths each year. 9 According to GLOBOCAN 2020, there are 6 oral and 2.3 lip cavities per 100 000 individuals in the globe for men and women, respectively. 1 However, in Iranian men and women, 2.2 and 1.8 per 100 000 people per year were computed, respectively. 10 In general, in the Iranian population, lip and oral cancer with 10 139 new cases and 454 deaths are ranked 20th and 22nd for morbidity, and mortality among all cancers. 9 The availability of a system for tracking and giving precise information on the incidence of different cancers is one of the crucial elements of the cancer control program. The establishment of the Iran Cancer Registry system dates back to 1984, with the aim of creating a cancer registry at the national level, which would collect and analyze information from various hospitals and health facilities across the country. 11 Cancer control policies and cancer epidemiological research in developed countries owe much to cancer registries and accurate cancer statistics of incidence and mortality. The lack of coverage and quality of cancer registration programs in the country resulted in various conflicting reports from pathology centers, and cancer registration centers regarding the incidence and prevalence of various cancers.
For more efficient planning and management of cancer prevention initiatives, it is vital to be informed of the existing status and trends in cancer incidence. There have been some comprehensive investigations of the incidence of different malignancies in Iran. As far as we know, there has not been a systematic review or meta-analysis conducted in Iran to evaluate the incidence and trend of oral cavity malignancies. Therefore, the objective of this study is to conduct a thorough review and metaanalysis of the age-standardized incidence rate of oral cavity malignancies in Iran.

| METHODS
As advised by the MOOSE (Meta-analyses of Observational Studies in Epidemiology) Checklist, 12

| Search strategy
The researchers searched for relevant articles without a time limit till December 2022 in Iranian databases SID (Scientific Information Database), Magiran and element as well as international databases PubMed/MEDLINE, Web of Science, ScienceDirect, Embase, Scopus, ProQuest, and Google Scholar. Additionally, the reference list of the discovered papers was manually reviewed to boost the search's sensitivity. Using ICD-0-3 codes, the team determined that the oral cavity consists of the lip (Vermillion, mucosal surfaces, both commissures), tongue, mouth floor, gum, hard palate, mouth, buccal mucosa, and vestibule. The search method was performed using MeSH terms in combination or separately using "AND," and "OR" functions (Supplementary Table 1). The search process and the process of selecting related articles are shown in the PRISMA flowchart ( Figure 1).

| Inclusion and exclusion criteria
The review included studies that provided clear descriptions of the age-standardized rate (ASR) of oral cavity cancer and contained reports from populations in Iran. Studies with limited sample numbers and those that reported prevalence were excluded. The final metaanalysis did not take into account review studies or publications presented as conference posters.

| Data extraction
The reviewers were aware of the name of the publication and the authors to gather information from relevant papers. The age-standardized incidence rate (ASR), which was taken from the publications and put into this form, the sample size, sex, and the time and place of the research were all included in the data collecting form, which was an Excel sheet that was constructed based on the study's goals. The authors were contacted for access to additional data, and the evaluation of the articles for which the full text was not available.
All articles were assessed for duplicates. They were then chosen based on the titles and abstracts of linked papers. To evaluate the quality of the studies, two impartial reviewers assessed the entire text based on the inclusion and exclusion criteria. The cases of disagreement between two referees were independently evaluated, and resolved by the third referee (Table 1).

| Quality assessment (risk of bias)
The risk of bias (internal validity) of the included studies was evaluated using the modified Newcastle Ottawa scale for observational study checklist, 14,15 which assigns a score ranging from 0 to 10. Based on the scores obtained for each study using the checklist, the risk of bias was classified into three categories: low risk (scores of 6-10), medium risk (scores of 3-5), and high risk (scores of 0-3). 15,16 Records removed before screening: Duplicate records removed (n =174)

IdenƟficaƟon of studies via databases
Reports assessed for eligibility (n = 54) 32 documents did not meet inclusion criteria in the full text. Review, letter, and case report (n = 9) The incidence of other cancers (n = 16) Report the prevalence of oral cancer (n = 7) Studies included in systematic review and meta-analysis (n = 22) Reports sought for retrieval (n = 106) Reports not retrieved (n = 52)

| Statistical analysis
The meta-analysis was performed using Stata version 16 software.
The researchers assessed the heterogeneity using statistical techniques, such as the I 2 statistic and Cochran's Q test. When no heterogeneity was detected, a fixed-effect model was used. However, in cases of heterogeneity, a random-effect model was utilized. 17,18 The random-effects model was used to extend the study's findings beyond the included studies by assuming that the selected studies are random samples from a larger population. 19 The researchers employed multivariable meta-regression analysis to estimate the effects of potential factors in heterogeneity. 18 The sensitivity analysis approach, which involved one-by-one elimination, was used to investigate the influence of each study on the pooled incidence estimate. The researchers assessed the robustness of each model and ultimately selected the best mode. 14,20 The forest plot was used to display the overall age-standardized rate (ASR) and its corresponding 95% confidence intervals (CI). To assess publication bias, Egger's and Begg's tests were performed.
Additionally, the trim-and-fill approach was employed to determine the number of missing studies and adjust the total estimate. 21,22 3 | RESULTS

| Search results and eligibility studies
Three local databases (n = 24 from Magiran, n = 22 from SID, and n = 14 from Elmnet) and seven international databases (n = 84 from T A B L E 1 The article met the eligibility criteria of this systematic review and meta-analysis. Moreover, 32 papers were disqualified from meeting the inclusion criteria, encompassing reviews, letters, case reports, prevalence of other cancers, and accounts of oral cancer prevalence in the complete text. Figure 1 shows that a total of 22 publications were incorporated in the meta-analysis and systematic review (Figure 1).

| Characteristics of the eligible studies
The study comprised 22 journal papers in total. The NOS criteria were used to evaluate the risk of bias, and the results showed that six of the studies had a moderate risk of bias, while 16 of the studies had a low risk of bias (Table 1). Regarding the study location, three studies were conducted in Kerman, [23][24][25] two studies in Fars, 26,27 two studies in East Azerbaijan, 28,29 two studies in Ardabil, 30,31 one study in Tehran, 32 Semnan, 33 Gilan, 34 and Khuzestan, 35 and nine studies were conducted in Iran [36][37][38][39][40][41][42][43][44] (Table 1).

| The results of individual studies
Based on the findings of included studies, the ASR ratio for oral cancer of men to women for oral cavity cancer was 1.20. Khanali (2) (2016) Rabiei (5)

| The pooled ASR of oral cavity cancer in male
We estimated pooled ASR of oral cavity cancer 1.96 (95% CI: 1.65-2.26) (Q statistic = 1118.09, df = 25, p < .0001, I 2 = 97.8%), with random effect model after sensitivity analysis using one-by-one elimination of studies and the best robustness model was chosen among the males (Figure 2).
Because the Q statistic and I 2 = 97.8% were significant, indicating high heterogeneity among the studies, univariate and multivariable meta-regression analyses were utilized to determine the source of the heterogeneity.
Furthermore, meta-regression analysis demonstrated that geography, bias risk, research year, and sample size are not likely sources of heterogeneity ( p > .05) ( Table 2).

| The pooled ASR of oral cavity cancer in female
After performing sensitivity analysis using one-by-one elimination of studies, a random effect model was used to estimate the pooled agestandardized incidence rate (ASR) of oral cavity cancer for females, which was determined to be 1.46 (95% CI: 1.14-1.77) (Q statistic = 2576.99, df = 26, p < .0001, I 2 = 99.0%) (Figure 3).
The high heterogeneity among the studies was observed based on the significant Q statistic and I 2 = 99%. To identify the source of heterogeneity, univariate and multivariable meta-regression analyses were employed.
Through a univariate meta-regression analysis, it was found that sample size studies ( p = .044) and the study period ( p = .042) were potentially responsible for the observed heterogeneity among the studies. A multivariable meta-regression also suggested that the study period may have contributed to the data's heterogeneity (B-coefficient = 0.04517, p = .009). As a result, altering the study period can increase the incidence by 0.04517. In a multivariable analysis, the location, sample size, or bias risk of the studies did not show a significant difference between them (with p value <.05).

| Publication bias
To evaluate publication bias, Egger's regression test and funnel plots were utilized. In the case of verified publication bias, the trim-and-fill approach was applied to estimate the number of missing studies and adjust the pooled estimate from the meta-analysis. 14,22 According to the results of the funnel plots and Egger's test, there is no evidence of publication bias in studies reporting on oral cavity cancer in males (bias = 1.3220, 95% CI: À3.9571, 6.6012, p = .610) ( Figure 4A).
Based on Trim-and-fill analysis using non-parametric methods, the anticipated values of three censored studies were calculated, and overall ASR corrected by the random effects model was calculated to be 1.36 (95% CI: 1.05%-1.66%) in females.

| DISCUSSION
In Iran, cancer ranks third as a cause of mortality. 45 According to Globocan 2020, lip and oral cancer are ranked 20th and 22nd for morbidity, and mortality among cancers in the Iranian population. 9 The variation in the incidence rate of cancer between Globocan's report and Iran's report may be due to differences in estimates and sources of information in Iran. Globocan data, which provides estimates of cancer incidence, mortality, and prevalence at the global, regional, and national levels, is calculated using a combination of data sources and mathematical modeling. To date, there has been no systematic review T A B L E 2 Univariate and multivariable meta-regression to find possible causes of heterogeneity between studies included in the metaanalysis.  50 On the other hand, the population profile in Iran will have an increasing trend in individuals more than 65 years old. 40 This means that as time goes on, oral cavity cancer may become an even more concerning issue than ever. It is estimated that the projection of cancer incidence by 2030 would be doubled in comparison to 2012. 51 Similar to the majority of other reports, the ASR ratio for oral cancer showed dominance in men. The reason for the higher incidence of oral cancer in men compared to women could be due to their increased exposure to risk factors, including smoking, alcohol consumption, tobacco use, and outdoor activities that increase exposure to solar radiation. Our result showed a 1.2 age-standardized incidence rate (ASR) ratio for oral cancer of men to women, however, various numbers are reported in other studies for instance in Taiwan men were predominantly 15 times more than women. 52 In some particular high-risk countries in South Asia like Indonesia 53 and Singapore, 54 women were at the same risk as men.

ASR
Some justifications for this diversity might be in terms of differences in risk factors exposure, behavioral, cultural, dietary habits, occupational, and environmental factors.
In both genders, East-Azerbaijan province stands among the top 2 with the highest ASR rate. Moreover, Ardebil as the neighboring  According to the reports, the age-standardized incidence rate (ASR) of all cancers has been found to be higher in the northern countries of the West Asia region, including the north of Iran, when compared to the southern countries of the region, which includes the south of Iran. 43 The same trend was observed, as the least amount of ASR rate is reported in Fars and Khuzestan provinces. Furthermore, accessibility to medical facilities varies in different provinces which leads to this verity in ASR incidence rate.
Unlike most of the world 58 Alcohol consumption cannot be taken as the main risk factor in Iran, 23 because it is forbidden due to local regulations. However, tobacco specifically abuse in the form of a water pipe can be considered one of the most notable risk factors. 59 Our study had several limitations. First, Although the study shows that the ASR incidence rate for oral cavity cancer in Iran is lower than in the world, taking into the fact that it is reported that all cancers in Iran are underestimated, 39  Funnel plot with pseudo 95% confidence limits F I G U R E 4 Funnel plot with pseudo 95% confidence limits for the detection of publication bias among included studies.

| CONCLUSION
In Iran, the incidence of oral cavity cancer was lower than the world average. However, we expect an upward trend, as a result of an upward trend in life expectancy, the number of elderly, and various exposures to risk factors, such as smoking. To detect cancer at an early stage, a cancer registry system, and screening programs should be developed.

CONFLICT OF INTEREST STATEMENT
The authors have stated explicitly that there are no conflicts of interest in connection with this article.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

ETHICS STATEMENT
In this study, we adhered to all ethical principles in the systematic review and meta-analysis studies. The Research Ethics Committee of Jahrom University of Medical Sciences approved the study protocol (ID IR.JUMS.REC.1401.112.).

TRANSPARENCY STATEMENT
The lead author (Vahid Rahmanian) affirms that the manuscript presents a comprehensive, truthful, and clear account of the study. The study plan was adhered to, and any deviations were elucidated, and if necessary, documented.