Incidence and mortality trend of liver cancer in Iran from 1990 to 2021: an epidemiological investigation using the Global Burden of Disease study

Abstract

Background:

The burden of liver cancer across the world is one of the most concerning. Iran, as a developing country, lacks a comprehensive review of liver cancer burden in recent years. This study aimed to evaluate the changing burden of liver cancer in Iran from 1990 to 2021.

Methods:

Data were extracted from the Global Burden of Disease study. Age-standardized incidence and mortality rates (ASIR, ASMR) were the main metrics used in this study. Average annual percent change (AAPC) was calculated for Iran and each province.

Results:

Liver cancer incidence rate in Iran increased to 2.70 [2.39 to 3.04] per 100 000 population in 2021 with an AAPC of 0.85 [0.61 to 1.08]. Mortality of liver cancer in 2021 increased to 2.87 [2.55 to 3.23], representing an AAPC of 0.83 [0.32 to 1.35]. Men faced a higher burden than women. The burden of liver cancer became more prominent with increasing age. The Incidence of NASH-related liver cancer shifted remarkably during the studied time period, surpassing trends of other risk factors (AAPC: 1.89 [1.75 to 2.03]). Ardebil showed the most significant increasing trends in incidence and mortality rates compared to other provinces. Incidence of liver cancer remained low in lower SDI provinces but gradually increased when moving towards high SDI ones, suggesting a positive correlation.

Conclusion:

This study highlights the concern for liver cancer rate and related death in Iran due to increasing shifts in mortality rates and deaths. Iranian policymakers must implement feasible preventive approaches to control liver cancer's major risk factors, such as NASH and Alcohol use.

Introduction

Liver cancer is among the three cancers with the highest mortality rate and has the second-highest death rate among the most prevalent cancers^[1–4]. Based on estimations of GLOBACAN, in 2020, nearly one million individuals were diagnosed with liver cancer and 830 200 died of it^[1,3]. While liver cancer incidence trends vary across regions, global cases have been rising^[2]. Chronic infection with hepatitis B or C virus is a risk factor for liver cancer, along with high alcohol consumption, and conditions like non-alcoholic steatohepatitis (NASH)^[5]. In recent decades, dramatic changes in lifestyle and diet have contributed to a staggering rise in NASH prevalence^[6]. Between 2010 and 2015, NASH prevalence increased by 10%^[7]. Also, based on reports worldwide, people diagnosed with NASH doubled in the same period^[7]. Notably, 154 700 cases of liver cancer in 2020 were attributable to alcohol consumption^[8]. In aggregate, evaluating the epidemiological burden of liver cancer in regions with the highest prevalence of these risk factors is necessary.

The global burden of cancer has undergone a striking shift in recent years, with the most profound changes seen in developing regions^[9–11]. Liver cancer prevalence has surged 5 to 10-fold in developing countries, mostly across Asia^[12]. Iran has not been an exception to this trend. The burden of cancer in Iran has been rising considerably along with substantial economic costs^[13,14]. Gastrointestinal cancers are among the most prevalent cancers with the highest rates of mortality and morbidity in Iran^[15,16]. It is also estimated that their rate will increase in the near future^[15]. This concerning trend in Iran is derived from unhealthy habits like smoking, sedentary lifestyles, high-calorie diets, and exposure to dietary aflatoxins^[9,16]. Moreover, key liver cancer risk factors are on the rise in Iran, with increasing rates of hepatitis B and C infections, NASH, and alcohol consumption contributing to the growing burden^[17].

Researchers have been keeping a close watch on the problem due to worrying changes in liver cancer around the world. It has continuously been investigated and studied on a global and regional level. Despite the significance of this issue, no research has been done on it in Iran as of yet. This study uses data from the Global Burden of Diseases study to examine the incidence and mortality of liver cancer in Iran between 1990 and 2021.

Methods

Data source

Data from a variety of sources, such as hospital records, population-based surveys, censuses, vital registration systems, systematic reviews, and disease-specific registries, are compiled via the Global Health Data Exchange (GHDx) platform. The Global Burden of Disease (GBD) 2021 study, which assesses the burden of 371 illnesses, 88 risk factors, and numerous injuries across 204 countries and territories between 1990 and 2021, provided the data used in this analysis^[18,19]. The GBD study uses sophisticated statistical modeling and established procedures to guarantee uniformity in disease classification and prevalence estimation. Estimates were produced by integrating data from several demographics (age, sex, and location) and managing missing or insufficient data using DisMod-MR 2.1, a Bayesian meta-regression program. GHDx, a freely accessible database created by GBD study collaborators and run by the Institute for Health Metrics and Evaluation (IHME), is where the data for liver cancer in Iran were taken from. Additionally, stringent data cleaning and validation protocols were put in place to improve accuracy and preserve consistency across various data sources. The Global Burden of Disease (GBD) study provides comprehensive estimates by modeling data from various sources, including surveys, registries, and literature. However, these sources often vary in quality, completeness, and availability across countries and time periods. For diseases like liver disease, which have diverse etiologies and context-specific risk factors, modeled estimates may not accurately reflect local epidemiological trends. Additionally, the reliance on data imputation and statistical assumptions introduces uncertainty and potential bias. Therefore, while GBD data enable valuable comparisons, their limitations must be acknowledged when interpreting results. We obtained estimates of liver cancer incidence, mortality, prevalence, and disability-adjusted life years (DALYs) in Iran from 1990 to 2021 using the Global Burden of Disease (GBD) 2021 study data, accessible through the Institute for Health Metrics and Evaluation (IHME) online tools (GBD Results Tool and GBD Compare). These estimates are not raw data but modeled outcomes generated by IHME using standardized methods that synthesize multiple data sources (vital registration, cancer registries, verbal autopsies, and published studies). Missing or incomplete data within country-specific registries are addressed in the GBD framework through statistical modeling approaches such as the Cause of Death Ensemble model (CODEm) and spatiotemporal Gaussian process regression. Thus, all values used in our analysis were adjusted estimates provided by IHME, already corrected for underreporting, misclassification, and missing data, rather than being imputed by the authors. For our study, we extracted age- and sex-specific rates as well as absolute numbers, reported with 95% uncertainty intervals, exactly as presented in the GBD database. No further adjustments or imputations were made by us.

HIGHLIGHTS

• Despite the significant shifts in liver cancer around the world, this matter has never been studied in a detailed way in Iran. This study focused on dissecting the changes liver cancer has undergone in Iran from 1990 to 2021.

• Average annual percentage change of liver cancer incidence and mortality rate per 100 000 population in Iran from 1990 to 2021 was 0.85 and 0.83, respectively, surpassing both the world and MENA.

• Hepatitis B and C had the highest incidence and mortality rate per 100 000 population back in 1990 and 2021. However, NASH-related liver cancer increased significantly compared to other types.

• Males and older populations in Iran have always shown significantly higher burdens compared to other groups of sex and age. Higher SDI was linked to higher liver cancer rates.

• Liver cancer incidence and mortality rates in Iran are higher than the world and MENA rates. Furthermore, this study also identifies particularly vulnerable groups.

Major indices

In this study, we collected data on Iran’s liver cancer incidence and mortality rates at the national and provincial levels for every age group, sex, and year from 1990 to 2021. All 31 of Iran’s provinces were covered by the data that was extracted at the national and subnational levels. The age-standardized incidence rate (ASIR) and age-standardized mortality rate (ASMR) per 100 000 population was the main indicator used to assess the burden of liver cancer. Comparisons between various geographical areas, demographic groups, and historical periods were made possible by these indices^[20,21]. Additionally, we used the Socio-Demographic Index (SDI), a composite measure with values ranging from 0 to 1, to investigate the link between socioeconomic development and the burden of liver cancer. Higher values indicate greater development. The SDI is calculated using the mean number of years of schooling (for people 15 and older), the fertility rate (for people under 25), and the per capita income^[22]. Regional differences in Iran’s liver cancer incidence and mortality trends were evaluated with the aid of the SDI categorization.

Statistical analysis

Incidence and mortality rates per 100 000 people were computed using 95% UI in accordance with the GBD study protocol. The age-standardized mortality rate (ASMR) and age-standardized incidence rate (ASIR) were the main metrics presented in this study. We used the Joinpoint Regression Model, which finds joinpoints (timepoints where notable trend changes occur) within the study period, to do a time-trend analysis in order to examine patterns over time. We determined the Annual Percent Change (APC) and its 95% confidence interval (CI) for each segment that these joinpoints defined. Furthermore, we weighted the APC values based on the length of each segment to calculate the average annual percent change (AAPC). The AAPC represented the general trend for the duration of the study if no joinpoints were found. The Grid Search Modeling Method was used to do a joinpoint regression analysis. By default, the method requires a minimum of two data points between and adjacent to any joinpoints. Using a maximum of five joinpoints, we used the Monte Carlo Permutation approach to identify statistically significant trend alterations during the course of the study. With a p-value threshold of 0.05, a weighted Bayesian Information Criterion (BIC) test was used to test for significance for both APCs and AAPCs. A thorough trend analysis of the incidence and mortality rates of liver cancer in Iran was carried out using the Joinpoint Regression Program (Version 4.9.1.0.23) after the initial data preparation and visualization in Excel. We used Joinpoint regression analysis to assess temporal trends in liver cancer burden indicators from 1990 to 2021. Joinpoint regression was chosen because it allows for the identification of statistically significant changes in trend over time (joinpoints), providing a flexible yet parsimonious approach compared to simple linear or polynomial regression. This method yields interpretable measures of the annual percent change (APC) and average annual percent change (AAPC), which are standard metrics in cancer epidemiology and burden-of-disease research. The joinpoint regression analysis is the commonly used analysis in reporting and identifying the trends and changes throughout the years when handling GBD data. It utilizes independence of errors, linearity within segments, constant variance of errors, and normality of residuals. The mentioned assumptions are major and key parts of ensuring the validity of its results. However, they also pose potential drawbacks. Take, for example, linearity within segments; this assumption simplifies complex trends within each year, or independence of errors can lead to biased standard errors and falsely significant trends. Knowing these, we made sure of using a proper dataset and took the right measures to ensure the validity of results, and throughout the text, we use AAPC to explain probable causes as a complementary analysis, and not just for describing patterns.

Results

Burden of liver cancer in Iran

In Iran, the age-standardized rate of liver cancer (ASIR) increased from 2.10 [1.76 to 2.64] per 100 000 in 1990 to 2.70 [2.39 to 3.04] per 100 000 in 2021, with an AAPC of 0.85 [0.61 to 1.08]. Concurrently, the age-standardized mortality rate (ASMR) increased from 2.26 [1.89 to 2.86] per 100 000 to 2.87 [2.55 to 3.23] per 100 000, representing an AAPC of 0.83 [0.32 to 1.35]. Compared to the MENA region, Iran had an increased rate of ASIR (Iran: 0.85 vs. MENA: 0.17 [0.15 to 0.19]) and ASMR (Iran: 0.83 vs. MENA: 0.09 [−0.24 to 0.42]). Globally, there was an increased rate of ASIR (Global: 0.12 [0.06 to 0.18]), whereas there was a decline in ASMR (Global: −0.11 [−0.23 to 0.01]). As seen, Iran faced a higher increase compared to both MENA and the whole world (Fig. 1).

Figure 1.

Comparing age-standardized incidence (up) and mortality (down) rate of liver cancer in Iran, Middle East and North Africa, and the world.

Liver cancer among various sexes and age groups

Throughout the course of the study, males had a continuously higher incidence and mortality rate of liver cancer than females. Liver cancer burden remained higher in males than females throughout the study period (Fig. 2). In 1990, the ASIR for males was 2.11 [1.75 to 2.77] per 100 000, compared to 2.06 [1.72 to 2.63] per 100 000 in females. By 2021, ASIR rose to 3.06 [2.73 to 3.44] per 100 000 in males and 2.35 [1.91 to 2.83] per 100 000 in females. Similarly, the ASMR was 2.26 [1.87 to 2.98] per 100 000 in males and 2.22 [1.85 to 2.88] per 100 000 in females in 1990, increasing to 3.23 [2.88 to 3.63] per 100 000 and 2.53 [2.06 to 3.04] per 100 000, respectively, in 2021. The incidence and mortality rates of liver cancer rose with age, with older persons bearing the greatest burden. In 1990, the incidence was 0.70 per 100 000 for children under five. By 2021, it had increased significantly in senior groups, surpassing 30 per 100 000 for those aged 80 and above (Fig. 3). In a similar vein, mortality rates rose with age, reaching 40 per 100 000 in 2021 for those aged 80 and over, after being low in younger populations (0.46 per 100 000 in <5 years).

Figure 2.

Sex-based comparison of age-standardized incidence (up) and mortality (down) rate of liver cancer in Iran.

Figure 3.

Incidence (up) and mortality (down) rate of liver cancer across various age groups in Iran in 1990 and 2021.

Provincial burden of liver cancer in Iran

In 2021, Sistan and Baluchistan (ASIR = 4.65 per 100 000), Ilam (ASIR = 3.94), and Ardebil (ASIR = 3.79) recorded the highest ASIR for liver cancer in Iran (Table 1). The provinces with the lowest ASIR values included Hormozgan (ASIR = 1.16), Alborz (ASIR = 1.59), and Qazvin (ASIR = 1.77) (Fig. 4). Over the period 1990–2021, the highest AAPC in ASIR was observed in Ardebil (AAPC = 3.99), Mazandaran (AAPC = 3.68), and Kohgiluyeh and Boyer-Ahmad (AAPC = 3.44), while Chahar Mahaal and Bakhtiari (AAPC = −0.81), Markazi (AAPC = −0.84), and Kerman (AAPC = −0.30) exhibited declining or relatively stable trends (Fig. 5).

Table 1.

Age-standardized burden of liver cancer in Iran

Locations	1990		2021		1990–2021
	ASIR [95% UI]	ASMR [95% UI]	ASIR [95% UI]	ASMR [95% UI]	Incidence AAPC [95% CI]	Mortality AAPC [95% CI]
Global	5.90 [5.43 to 6.48]	5.86 [5.38 to 6.46]	6.15 [5.58 to 6.90]	5.65 [5.13 to 6.30]	0.12 [0.06 to 0.18]	−0.11 [−0.23 to 0.01]
MENA	4.77 [3.73 to 6.72]	5.10 [3.96 to 7.22]	5.03 [4.29 to 5.82]	5.29 [4.51 to 6.11]	0.17 [0.15 to 0.19]	0.09 [−0.24 to 0.42]
Iran	2.10 [1.76 to 2.64]	2.26 [1.89 to 2.86]	2.70 [2.39 to 3.04]	2.87 [2.55 to 3.23]	0.85 [0.61 to 1.08]	0.83 [0.32 to 1.35]
Alborz	0.66 [0.49 to 0.91]	0.73 [0.54 to 1.02]	1.59 [1.25 to 2.01]	1.71 [1.34 to 2.17]	2.91 [2.78 to 3.03]	2.82 [2.59 to 3.05]
Ardebil	1.14 [0.85 to 1.55]	1.21 [0.89 to 1.66]	3.79 [2.96 to 4.82]	4.04 [3.18 to 5.14]	3.99 [3.82 to 4.16]	4.02 [3.61 to 4.43]
Bushehr	2.30 [1.70 to 3.17]	2.48 [1.83 to 3.43]	2.31 [1.82 to 2.86]	2.46 [1.94 to 3.06]	0.11 [−0.27 to 0.50]	0.02 [−0.81 to 0.86]
Chahar Mahaal and Bakhtiari	3.48 [2.66 to 4.42]	3.73 [2.86 to 4.69]	2.68 [2.10 to 3.39]	2.80 [2.21 to 3.54]	−0.81 [−1.04 to −0.57]	−0.82 [−2.52 to 0.91]
East Azerbaijan	2.43 [1.86 to 3.17]	2.64 [2.02 to 3.47]	2.46 [1.98 to 3.08]	2.70 [2.17 to 3.37]	0.11 [−0.22 to 0.44]	0.17 [−0.63 to 0.97]
Fars	1.87 [1.35 to 2.58]	2.01 [1.46 to 2.78]	2.09 [1.63 to 2.66]	2.22 [1.73 to 2.84]	0.36 [0.10 to 0.63]	0.42 [−0.06 to 0.90]
Gilan	1.45 [1.09 to 2.05]	1.56 [1.17 to 2.22]	2.08 [1.66 to 2.61]	2.22 [1.77 to 2.78]	1.23 [0.90 to 1.56]	1.24 [0.80 to 1.68]
Golestan	2.13 [1.60 to 2.85]	2.28 [1.72 to 3.10]	3.54 [2.86 to 4.29]	3.76 [3.04 to 4.59]	1.72 [1.14 to 2.30]	1.80 [1.24 to 2.35]
Hamadan	1.62 [1.22 to 2.20]	1.72 [1.29 to 2.35]	2.41 [1.88 to 3.01]	2.55 [1.98 to 3.19]	1.34 [0.98 to 1.70]	1.42 [0.80 to 2.03]
Hormozgan	0.72 [0.55 to 0.97]	0.77 [0.59 to 1.04]	1.16 [0.93 to 1.45]	1.25 [1.00 to 1.56]	1.53 [1.37 to 1.70]	1.64 [0.75 to 2.53]
Ilam	3.46 [2.64 to 4.58]	3.70 [2.81 to 4.94]	3.94 [3.16 to 4.84]	4.19 [3.33 to 5.12]	0.43 [0.12 to 0.75]	0.37 [−0.55 to 1.30]
Isfahan	1.58 [1.17 to 2.18]	1.71 [1.27 to 2.38]	2.38 [1.91 to 2.98]	2.54 [2.04 to 3.19]	1.37 [1.08 to 1.66]	1.41 [0.73 to 2.11]
Kerman	2.51 [1.93 to 3.28]	2.70 [2.07 to 3.60]	2.28 [1.78 to 2.88]	2.47 [1.95 to 3.09]	−0.30 [−0.52 to −0.08]	−0.26 [−0.88 to 0.37]
Kermanshah	3.04 [2.35 to 3.97]	3.27 [2.51 to 4.25]	3.18 [2.50 to 3.95]	3.38 [2.68 to 4.15]	0.19 [−0.12 to 0.50]	0.24 [−0.30 to 0.78]
Khorasan-e-Razavi	3.12 [2.41 to 4.23]	3.34 [2.57 to 4.57]	2.91 [2.27 to 3.67]	3.15 [2.46 to 3.98]	−0.14 [−0.53 to 0.26]	−0.16 [−0.96 to 0.65]
Khuzestan	0.84 [0.65 to 1.12]	0.89 [0.69 to 1.19]	2.33 [1.90 to 2.96]	2.47 [2.02 to 3.13]	3.32 [3.08 to 3.57]	3.35 [2.99 to 3.70]
Kohgiluyeh and Boyer-Ahmad	0.67 [0.50 to 0.91]	0.71 [0.54 to 0.97]	1.90 [1.51 to 2.43]	1.98 [1.56 to 2.52]	3.44 [3.26 to 3.62]	3.37 [2.95 to 3.80]
Kurdistan	2.49 [1.94 to 3.29]	2.66 [2.05 to 3.53]	2.40 [1.91 to 2.95]	2.55 [2.05 to 3.13]	−0.04 [−0.38 to 0.29]	−0.02 [−0.87 to 0.83]
Lorestan	1.10 [0.82 to 1.50]	1.18 [0.88 to 1.64]	3.13 [2.47 to 3.93]	3.35 [2.66 to 4.24]	3.43 [3.31 to 3.55]	3.48 [3.15 to 3.80]
Markazi	2.92 [2.29 to 3.69]	3.15 [2.47 to 3.99]	2.23 [1.78 to 2.76]	2.37 [1.87 to 2.90]	−0.84 [−1.08 to −0.61]	−0.83 [−1.28 to −0.37]
Mazandaran	0.71 [0.51 to 1.03]	0.77 [0.55 to 1.12]	2.19 [1.67 to 2.80]	2.30 [1.76 to 2.96]	3.68 [3.57 to 3.80]	3.66 [3.39 to 3.93]
North Khorasan	1.11 [0.85 to 1.42]	1.18 [0.89 to 1.52]	3.10 [2.37 to 3.85]	3.32 [2.54 to 4.10]	3.42 [3.26 to 3.58]	3.43 [3.07 to 3.79]
Qazvin	0.99 [0.75 to 1.33]	1.06 [0.80 to 1.45]	1.77 [1.40 to 2.19]	1.89 [1.50 to 2.34]	1.94 [1.76 to 2.11]	1.97 [1.45 to 2.49]
Qom	0.97 [0.75 to 1.31]	1.05 [0.82 to 1.42]	2.06 [1.62 to 2.65]	2.21 [1.73 to 2.83]	2.47 [2.35 to 2.60]	2.44 [2.02 to 2.85]
Semnan	2.20 [1.65 to 2.98]	2.36 [1.77 to 3.22]	2.18 [1.70 to 2.74]	2.33 [1.81 to 2.93]	0.05 [−0.29 to 0.39]	0.00 [−0.86 to 0.87]
Sistan and Baluchistan	3.03 [2.41 to 3.87]	3.20 [2.55 to 4.04]	4.65 [3.69 to 5.89]	4.94 [3.92 to 6.23]	1.37 [1.14 to 1.59]	1.40 [1.04 to 1.75]
South Khorasan	0.85 [0.66 to 1.11]	0.91 [0.70 to 1.18]	1.92 [1.51 to 2.44]	2.06 [1.61 to 2.61]	2.68 [2.53 to 2.84]	2.69 [2.39 to 2.99]
Tehran	2.97 [2.14 to 4.32]	3.22 [2.30 to 4.70]	3.59 [2.77 to 4.58]	3.73 [2.86 to 4.81]	0.60 [0.41 to 0.79]	0.47 [0.18 to 0.76]
West Azerbaijan	3.03 [2.29 to 4.07]	3.28 [2.48 to 4.43]	2.93 [2.39 to 3.61]	3.22 [2.62 to 3.95]	−0.04 [−0.40 to 0.31]	0.02 [−0.96 to 1.01]
Yazd	2.18 [1.62 to 2.83]	2.37 [1.75 to 3.08]	2.09 [1.61 to 2.67]	2.27 [1.74 to 2.90]	−0.09 [−0.35 to 0.16]	−0.12 [−0.80 to 0.57]
Zanjan	2.25 [1.76 to 2.96]	2.39 [1.87 to 3.15]	2.19 [1.80 to 2.63]	2.34 [1.93 to 2.80]	0.04 [−0.39 to 0.47]	0.02 [−0.79 to 0.84]

AAPC, average annual percent change; ASIR, age-standardized incidence rate; ASMR, age-standardized mortality rate; MENA, Middle East and North Africa.

Note that provinces with the most significant AAPC are in bold type, making them easier to spot.

Rates were reported per 100 000 population.

Figure 4.

Age-standardized incidence (up) and mortality (down) rate of liver cancer in Iran across various provinces in 1990 and 2021.

Figure 5.

Average annual percent change of incidence (up) and mortality (down) rate of liver cancer in Iran across various provinces from 1990 to 2021.

For ASMR in 2021, the highest values were recorded in Sistan and Baluchistan (ASMR = 4.94 per 100 000), Ilam (ASMR = 4.19), and Ardebil (ASMR = 4.04), while Hormozgan (ASMR = 1.25), Alborz (ASMR = 1.71), and Qazvin (ASMR = 1.89) had the lowest rates (Fig. 4). The AAPC in ASMR from 1990 to 2021 was highest in Ardebil (AAPC = 4.02), Mazandaran (AAPC = 3.66), and Lorestan (AAPC = 3.48), whereas Markazi (AAPC = −0.83), Chahar Mahaal and Bakhtiari (AAPC = −0.82), and Kerman (AAPC = −0.26) exhibited the most significant reductions (Fig. 5).

Burden of various liver cancer causes in Iran

In 1990, liver cancer due to hepatitis B had the highest incidence (0.99 per 100 000), followed by hepatitis C (0.46 per 100 000) and NASH (0.25 per 100 000) (Fig. 6). By 2021, the incidence of NASH-related liver cancer increased significantly (AAPC: 1.89 [1.75 to 2.03]), while hepatitis B and C showed moderate growth (AAPC: 0.73 [0.53 to 0.92] and 0.85 [0.56 to 1.14], respectively). Conversely, hepatoblastoma incidence declined (AAPC: −2.22 [−3.03 to −1.41]). A similar trend was observed in mortality rates. Hepatitis B-related liver cancer had the highest mortality in 1990 (1.04 per 100 000), followed by hepatitis C (0.53 per 100 000) and NASH (0.28 per 100 000). Over time, NASH-related mortality rose (AAPC: 1.83 [1.31 to 2.35]), while alcohol-related liver cancer showed a lower increase (AAPC: 1.10 [0.43 to 1.76]). In contrast, hepatoblastoma mortality declined significantly (AAPC: −3.05 [−3.70 to −2.40]).

Figure 6.

Age-standardized incidence (up) and mortality (down) rate of various causes of liver cancer in Iran.

Socio-Demographic Index (SDI) and incidence of liver cancer in Iran

A positive correlation was observed between the Socio-Demographic Index (SDI) and ASIR of liver cancer in Iran. As the SDI increased, the ASIR showed a corresponding rise, suggesting that higher levels of socio-economic development were associated with a greater burden of liver cancer incidence (Fig. 7). In regions with lower SDI values (~0.45–0.50), the ASIR remained relatively low (~2.10–2.18 per 100 000). However, as the SDI approached higher levels (>0.50), the ASIR progressively increased, reaching its peak values.

Figure 7.

The association between the Sociodemographic Index (SDI) and age-standardized incidence of liver cancer in Iran.

Discussion

This study aimed to investigate the incidence and mortality rates of liver cancer in Iran using GBD studies between 1990 and 2021. Iran’s ASIR of liver cancer increased significantly, as well as ASMR. Men had a significantly higher burden of liver cancer than women. While mortality and incidence rates increased with age, they remained mostly unchanged in the younger population. Two main underlying risk factors for liver cancer continued to be hepatitis B and chronic infections. Even so, the burden of NASH-related and alcohol-related liver cancer increased substantially. The most rapidly growing trend was observed in Ardabil province, and conversely, the most declining in Markazi. This study also demonstrated a positive relationship between SDI and ASIR.

Previous studies conducted in Iran have reported an increase of up to fourfold in mortality rate between 1990 and 2015^[12]. Many attempts have been made to investigate this issue at the provincial level and also to identify the effects of risk factors on the burden of liver cancer^[23–27]. Notably, the burden of liver cancer in Iran was found to be substantially higher than the globe and the Middle East and North Africa (MENA) region. The most frequent cause of hepatocellular carcinoma in Iran is reported to be the viral agents; precisely hepatitis B and C virus^[16]. Chronic infection with viral agents in the past^[28,29] has probably led to the accumulative burden of liver cancer over time. High aflatoxin level in Iranian food might be also a considerable risk factor^[30]. Besides, life expectancy in Iran has increased and it could have caused a surge in overall cancer rates and deaths^[31]. Another reason behind the increased burden of liver cancer in Iran could be the improvement of cancer screening programs^[32].

Numerous studies indicate a greater global burden of liver cancer in men compared to women^[33,34]. Men have a higher tendency to carry out precarious and detrimental behaviors, including excessive alcohol intake, binge drinking, risky sexual practices, and substance abuse^[35–38]. Female sex hormones play a protective role against malignancies. For instance, it is shown that IL-6 was suppressed by estrogen by Naugler et al^[39]. This chemokine is shown to be responsible for hepatic response to inflammation and is linked to hepatocarcinogenesis^[40]. Moreover, a positive relationship between testosterone and liver cancer was observed^[41]. However, the role of androgens in the development of liver cancer and suppressing them and their receptor as a cure is still not definitive^[42]. Moreover, higher incidence and mortality rates seen in older populations are consistent with previous findings conducted in Iran^[12,43,44]. It indicates an accumulative effect of exposure to multiple risk factors, decreased immune responsiveness, and troubles in accessing health care systems due to aging. Obesity, low physical activity, and metabolic syndromes in general are major risk factors for cancer, mostly present in the elderly population^[45–47].

Significant rise of NASH-related liver cancer in Iran, surpassing that of viral types, is an interesting trend to note. Recent studies have pointed out the increasing prevalence of non-alcoholic fatty liver disease (NASH) and NASH in Iran^[48]. These trends around the world have received much attention from clinicians, pathologists, and commoners as well^[49]. Persistent adipose expansion observed in obesity over time leads to adipose tissue necrosis due to insufficient vascularization^[50,51]. Released free fatty acids cause lipotoxicity in the liver and set the stage ready for steatosis^[52]. Kupffer cells then become triggered and release cytokines, attracting monocytes^[53]. This unremitting cycle of inflammation and necrosis eventually causes scarring, then fibrosis and cirrhosis^[52]. To this extent, NASH and related complications, as well as cirrhosis, are major risk factors of liver cancer^[54]. Improved socioeconomic status in Iran during recent decades is to blame for increased cirrhosis and chronic liver diseases^[55]. Obesity and metabolic syndromes have been reported to be on the rise in Iran^[56–59]. Metabolic syndrome (MetS) in Iran is reportedly higher in older age groups and in urban areas^[59]. This trend can be explained by lower physical activity in the senior population and the shift towards Western dietary pattern. Persian cuisine is more similar to Mediterranean cuisine, but in recent years, urban areas have embraced Western diet more^[60]. It’s shown that in Iran, NASH patients have a higher consumption of red meat, sugar, and fats, whereas the consumption of vegetables, fruits, and whole grains is low. This change can explain the higher prevalence and incidence of MetS and NASH. The rising prevalence of NASH in Iran runs parallel to the global trends^[61]. NASH has become the leading cause of chronic liver disease in the US and Europe and is rampaging freely worldwide as well^[7]. The underlying reasons of this event are common around the globe. With modernization, people are attending jobs that require less physical activity^[62]. We consume more meals than we need, and they’re often saturated with sugars and fats. Combined with lower physical activity and high stress, the population around the world, specifically in urban areas, has become overweight or obese^[63,64]. Abdominal obesity, high blood sugar levels, and increased blood pressure all result in MetS^[65]. MetS is one of the most important contributing causes of NASH and consequent NASH^[66]. We’ve seen these trends playing a part in Iranian society^[67] as well, and they’ve precisely happened in urban parts of the country more than any other regions. The burden of alcohol-related liver cancer has also increased in recent years, even though Sharia law forbids drinking for any Muslim. Unlike what is believed, studies suggest that alcohol consumption is not a rare habit in Iran^[68,69]. The fact that alcohol consumption is taboo and something that warrants punishment might make reports unreliable and the predictive models less accurate. Policymakers and health care professionals must carefully plan to disrupt increasing trends of obesity and metabolic syndromes.

Data lead us to conclude that despite the overall global decline in incidence of acute viral hepatitis (AVH), Iran has gone through a much different path. The cumulative burden of AVH shows an increasing shift from 1990 to 2021. In a study performed in Fars province, researchers found that a significant portion of Hepatitis B subjects were immigrants from Afghanistan^[70]. Afghan immigrants in Iran have a poor social condition, and their integration into the society usually never happens. This almost always leads to disparities when health care and education are needed. Most of them usually do manual labor and don’t get paid enough^[71]. Another reason could be the insufficient education of people on the possible transmission routes of viral hepatitis^[48,72].

The burden of liver cancer was observed to be more prominent in high SDI provinces. Observed differences in the burden of liver cancer demonstrate a complex interaction between socioeconomic development and geographical factors^[73,74]. These shifts are explainable by dietary choices^[60,75–77], geographical and population distribution^[44,78], increased mean age^[79,80], and improved data registries^[81,82]. Higher SDI regions often enjoy better access to health care facilities, preventive programs, and education, resulting in more registries of liver cancer^[83]. Affluent cities of Iran have better life standards and are abundant in food supply chains that can hypothetically lead the population towards high-calorie diets and more sedentary lifestyles^[73]. In contrast, low SDI regions are probably still suffering from a lack of proper education, fastidious vaccination routines, implemented screening methods, precise diagnostics, frail infrastructure, low-income, and rigorous health care^[84]. Among provinces with the highest burden, Ardebil has experienced the most pronounced shift in incidence and mortality of liver cancer. Ardebil was also listed amongst provinces with substantial prevalence of liver cancer in recent years^[44]. In a study done by Nemati M. et al^[85], all of the milk samples from Ardabil were contaminated with aflatoxin M1 (AFM1) metabolite. A considerable number of these samples contained AFM1 levels exceeding higher values than the determined maximum residue limit (MRL) by the European Union. This province is one of the key producers of milk in the country as well, and can potentially contaminate the whole milk industry of Iran. Markazi province recorded the lowest burden. It was shown previously that some provinces, like Markazi, had considerable amounts of incorrect classifications in their cancer registry facilities^[86,87].

When comparing our findings to other neighboring countries and countries in the MENA region, we see a similar trend in common. The overall burden of liver cancer in the MENA region has had an increasing trend since 1990^[88]. Iran doesn’t stand out from other countries in the region and has followed a similar track. The identified reason behind this orchestration seems to be present in all of the countries. NASH, alcohol consumption, and viral hepatitis are the major drivers of the trajectory^[89]. All of the mentioned causes are also active in Iran, as it was discussed before.

The burden of liver cancer in Iran has increased during the timeline that was studied. This is absolutely worrisome, and proper, planned, and targeted solutions must be recruited in order to control it. Participation rates in screenings in Iran still have not reached the ideal levels proposed by the World Health Organization (WHO)^[90]. Addiction is on the rise in Iran^[91], and the ever-growing young Iranian community lacks a creditworthy source of education^[92]. Improvements in education and awareness of liver diseases, screenings, and transmission routes would surely and greatly aid the Iranians to better prepare themselves when running up against these concerns. Distributing needles to addicts alongside opioid substitutes can also increasingly and demonstrably reduce the rate of viral hepatitis transmission^[93].

The Global Burden of Disease (GBD) studies offer invaluable insights and information about a wide list of diseases. However, its drawbacks must be considered when we are presented with this data. Data and health registries across the globe do not possess a unified standard. Registered information in low and middle-low income countries typically lacks rigorous and cohesive quality. An undeniable and significant misclassification of the incidence of HCC in neighboring provinces was noted previously by researchers. The average estimated misclassification rate between neighboring provinces was as high as 47% between Khorasan Razavi Province and Sistan and Baluchistan Province, for instance. Khorasan Razavi has a considerable concentration of medical facilities compared to Sistan and Baluchistan, and has an even higher SDI^[87]. Some of the data gathered worldwide were recorded in a periodic order, and some countries even prohibited sharing for some time. Statistical models used by GBD lack certainty due to built-in assumptions about disease risk factors and progression. Some of the societal trends in countries may get overlooked in this conventional method of IHME data reporting. Namely, the potential growing trend of alcohol use that is considered a crime in Muslim countries, and even reported data may not be reliable^[69]. Moreover, health policies implemented take time to take effect, and to see the results requires more time. These points can lead toan underestimation of the true burden of liver cancer in Iran during the studied time-line, and when interpreting results of the study should be taken into account. These issues are inherently forsaken in this way of forecasting disease burden. Estimates often focus on national levels, and in large or diverse countries, significant subnational information is ignored completely. The quality level of data gathered is still inevitably heterogeneous which results in data granularity. Despite these limitations, GBD is still the most vital instrument when performing global and epidemiological studies to guide researchers, policymakers, and alert others about where and when to look after what.

Conclusion

Overall, the burden of liver cancer seemed to be increasing during the last three decades in both men and women, and especially in older individuals. Among the probable causes, NASH and alcohol abuse indicated the most notable increasing burden. However, most liver cancer cases still emerg due to hepatitis. Provincial differences in ASIR and ASMR were also noted. A higher SDI led to higher mortality and incidence of liver cancer.

Ethical approval

Not applicable.

Consent

Not applicable.

Sources of funding

There is no funding source for the authors to declare.

Author contributions

O.K.: conceptualization, formal analysis, investigation, methodology, project administration, supervision, validation, writing – review & editing; S.H.: conceptualization, data curation, writing – original draft, writing – review & editing; A.G.R.: writing – original draft, writing – review & editing; A.A.: conceptualization, formal analysis, data curation, writing – original draft, methodology; A.S.: conceptualization, investigation, methodology, project administration, supervision, validation, writing – review & editing.

Conflicts of interest disclosure

All of the authors have no relevant financial or non-financial conflict of interest to disclose.

Guarantor

Amirhossein Shirinezhad.

Research registration unique identifying number (UIN)

This paper relies on aggregated secondary data from web sources rather than individual data points. This study only published aggregated epidemiological data, not individual data.

Provenance and peer review

This paper was not invited.

Data availability statement

Data are available in a public, open access repository (https://ghdx.healthdata.org/gbd-results-tool).