Incidence and Mortality of Eye Cancer in Shandong, China: A Population-based Study During 2013-2017

Background: Eye cancer is relatively rare. Current prevalence and disease burden of eye cancer are unlcear.The purpose of this study was to understand the epidemiology in the incidence and mortality of eye cancer in Shandong Province, and to provide reference for the prevention and control of eye cancer. Methods: Population-based cancer incidence and death data from cancer registries in 2013-2017 was collected by Shandong Center for Disease Control and Prevention (SDCDC). Extracted data were rstly assessed for data quality and then were aggregated by area (urban/rural), gender, and age group [0, 1-4, 5-9, 10-14, …, 85+]. Crude and age-adjusted incidence and mortality rates were calculated. Results: The reported numbers of eye cancer incident cases and deaths in Shandong cancer registry from 2013 to 2017 were 169 and 43, respectively. The crude incidence rate, age-standardized rate of incidence by Chinese population (ASRIC, 2000) and world population (ASRIW) were 11.12/10 7 , 8.92/10 7 and 12.44/10 7 , respectively. The crude mortality rate, age-standardized rate of mortality by Chinese population (ASRMC, 2000) and world population (ASRMW) were 2.83/10 7 , 1.89/10 7 and 2.58/10 7 , respectively. There were no marked differences in the incidence and mortality rates between male and female. The incidence rates were similar between rural and urban areas. The mortality rate in rural areas was higher than that in urban areas. The mortality-to-incidence (M/I) ratio in rural areas (0.30) was considerably higher than that in urban areas (0.18). The highest incidence was observed in children aged 0-4 years old, which was true for both urban and rural areas, and for both genders, Conclusion: Eye cancer is not a common malignant tumor in this population. Prevention and control measures should be tailored according to the epidemiological characteristics and risk factors of eye cancer.


Background
The mortality and incidence rates of cancer were increasing rapidly worldwide and in China [1,2]. Since 2010, cancer had become the leading cause of death and a central public health problem in China [3]. Unlike most other ophthalmic diseases, eye cancer does not only cause vision impairment, but also affect the integrity of the eyeball, and even is lifethreatening [4].
Cancer or malignant tumor of the eye is likely the only ocular disease which directly lead to life loss. Death usually occurs when cancer cells invade or metastasize to other organs such as the brain, lung and liver, and only can be prevented by early detection and treatment [5]. Eye cancer is relatively rare, and little is known about the determinants and distribution of this disease [5]. According to the latest global cancer incidence and death data released by the IARC (International Agency for Research on Cancer) under the WHO (World Health Organization) in 2018, eye cancer was classi ed as one of other tumors and was not listed separately. Current prevalence and disease burden of eye cancer are unlcear [1,6]. As the second most populous province in China, Shandong has a population of nearly 100 million, which almost doubles that of South Korea [7]. There have been no reports on the mortality and incidence of eye cancer in Shandong. This study aimed to evaluate the disease burden of eye cancer among people in Shandong Province and to provide scienti c evidence to inform the prevention and control of eye cancer.

Data Sources
Shandong Center for Disease Control and Prevention (SDCDC) is responsible for the management of the population-based cancer registration system in Shandong Province. Data provided by 30 cancer registries were collected from local hospitals, medical insurance institutions and health service centers of local communities [7]. All cancer cases were classi ed according to the ICD-10 (the 10th edition of the International Classi cation of Diseases). Cases of eye cancer with ICD-10 encoded as C69 were extracted. Population data were estimated based on the statistical yearbook of Shandong Province in 2013-2017 and the data from the sixth census.

Quality control
According to the "Guideline for Chinese Cancer Registration [8]" and the relevant requirements for quality of registration in the 9th volume of Cancer Incidence on Five Continents issued by the International Cancer Research Center and the International Association for the Advancement of Cancer Registry [9], data review and evaluation were reported to the registries. The completeness, reliability, validity and timeliness of the data were assessed based on the main indicators such as: the percentage of morphologically veri ed cases (MV%), the proportion of death certi cate-only Cases (DCO%), the ratio of mortality-to-incidence (M/I) and the proportion of unknown diagnosis basis(UB%). The nal registry data included in the analysis meet the following criteria (average of all cancers): MV%>66%, 0.6%<M/I < 0.8%, DCO%<15%,UB%<0.5%. In total, 25 registries met the inclusion criteria, covering a total population (2013-2017) of 151,922,468 (of which the urban population was 59,738,017 and the rural population was 92,184,451). This sample accounted for 30.93% of the total population of the province. The IARC veri cation software was used to determine whether the variables were valid and complete [10,11].

Statistical Analysis
The mortality and incidence rates were calculated by area, gender, and age group. The Segi's world population and Chinese census population in 2000 were used for calculating age standardized rates [12]. The cumulative risk of eye cancer (developing or dying from eye cancer) before the age of 75 years old was calculated. The IARCcrgTools2.13 software (Cours Albert Thomas, Lyon, France) and MS-Excel (Microsoft, Redmond, WA, USA) software were used for data veri cation and evaluation [11]. SAS software (SAS Institute, Cary, NC, USA) was employed to calculate the incidence, mortality rates and cumulative risk of eye cancer.

Data Quality
The MV%, DCO% and M/I of eye cancer data from the 25 cancer registries were 65.14%, 0.49% and 0.25 respectively. The indicators were 70.17%, 0%, 0.22 in urban areas, and 61.15%, 0.95%, 0.30 in rural areas ( Table 1). UB% of eye cancer in urban and rural areas were all zero.

Incidence of Eye Cancer
There were 169 incident cases of eye reported during 2013-2017. The crude incidence, ASRIC and ASRIW of eye cancer were 11.12/10 7 , 8.92/10 7 and 12.44/10 7 , respectively. The cumulative incidence rate (aged 0-74 years old) was 0.82/10 4 . The crude incidence rate, ASRIC and ASRIW of eye cancer in males were 11.34/10 7 , 9.35/10 7 and 12.25/10 7 , respectively. The crude incidence of female eye cancer, ASRIC and ASRIW were 10.90/10 7 , 8.54/10 7 and 12.75/10 7 , respectively ( Table 2). The crude incidence and ASRIC of eye cancer in males were higher than that of in females. While the ASIRW in males was lower than in females. Age-speci c Incidence In both urban and rural areas, regardless of gender, 0-4 years old had the highest incidence. The lowest incidence was in the 10-15 age group. Subsequently, the incidence increased with age. The age distribution of eye cancer is different from that of common malignant tumors such as lung cancer, esophageal cancer, gastric cancer and liver cancer. There were two incidence peaks in the youngest age group and the oldest age group ( Fig. 1 and Table 3). The number of new eye cancer cases in rural and urban areas in 2013-2017 were 104 and 65, respectively.
The ASRIW, ASRIC and crude incidence of eye cancer in urban areas were 11.27/10 7 , 8.57/10 7 and 10.88/10 7 , respectively, which were all slightly lower than that in rural areas.  Table 4). The ASRMC and ASRMW of eye cancer in males were lower than that in females. While crude mortality in males was higher than in females. Age-speci c Mortality There were two peaks in age-speci c mortality rates. One was in 0-9 years old, and the other in those aged 60-85 years old. The mortality rate of eye cancer was extremely low in the other age groups (10-59) ( Fig. 2 and Table 5).

Differences in Mortality between Areas
The number of eye cancer deaths in rural and urban areas in 2013-2017 were 31 and 12, respectively. The ASRIC, ASRIW and crude incidence in rural areas were 2.39/10 7 , 3.44/10 7 and 3.36/10 7 , respectively, which were all higher than that in urban areas.

Discussion
In order to control and prevent cancer, it is crucial to collect and generate cancer statistics on a populationbased cancer registry system [13]. Cancer registration data is widely used in cancer control and decisionmaking as well as scienti c research [14]. Public health and cancer control greatly rely on a good understanding of the disease burden in the population. Reliable estimates of the disease burden provides a comprehensive picture of how the impact of cancer varies with geographic regions and populations [15]. Population-based cancer registries have been operating in China for about 60 years. Their roles have expanded to include the development and evaluation of national cancer control programs and interventions to improve the care for cancer patients [5].
This study provided the mortality and incidence of eye cancer based on the data from Shandong Cancer Registries. The age-standardized mortality and incidence rates with the world standard population were 2.58/10 7 and 12.44/10 7 , respectively. The ASRIW in males and females were 12.25/10 7 and 12.75/10 7 , respectively. They are lower than those in Singapore (male 18.9/10 7 and female 18.1/10 7 ) [17]. Earlier studies showed that the incidence rate of eye cancer varied in different countries and regions. These rates of eye cancer in Asia were generally lower than those in the West [17,18]. For example, the rates in Canadian residents were 8.1/10 6 and 6.0/10 6 for males and females, respectively in 1970-1982 [19]. And the rates of eye cancer in New York State of American were 5.4/10 6 in males and 4.9/10 6 in females between 1975 and 1986 [20]. The rates were much higher than those in Singapore (male 1.89/10 6 and female 1.81/10 6 ,1968-1995) and Taiwan (male 2.57/10 6 and female 2.33/10 6 ,1979-1996) [17]. Unlike most other malignant tumors, there are no signi cant differences between male and female in the incidence and mortality of eye cancer. While in other common malignant tumors, such as lung cancer, esophageal cancer, liver cancer and colorectal cancer, gastric cancer, the incidence and mortality in males are much higher than that of females, or even several times that of females [1,21,22]. In addition, according to the incidence and mortality curves in different age groups, the incidence and mortality of eye cancer have two peaks in the youngest age group and an oldest age group, while the other common malignant tumors (such as respiratory system and digestive system cancers) have only one peak in the old age group [23,24].
Eye cancer is a malignant tumor that can occur in all parts of the eye. Retinoblastoma (RB) is the most common intraocular malignant tumor in children, accounting for 3% of all pediatric malignant tumors [25,26]. Research showed that the median age for diagnosis of RB was 17.0 months [27]. This is also the reason why there is a peak of eye cancer incidence and mortality in the low age group (0-10 years old) in our study. There is a signi cant regional difference in Rb mortality worldwide, with higher than 40% (40-70%) in Africa and Asia, and only 3-5% in the USA, Canada and Europe [28]. Some factors, such as delayed diagnosis, poor access to eye cancer experts, ophthalmologic pathology, gene detection and socio-economic factors, are the factors leading to RB related life loss [27]. Uveal melanoma (UM) is the most common intraocular malignant tumor in adults, accounting for about 5% of all melanomas [29]. UM can affect any part of the uveal tract including the ciliary body, iris and choroid [30]. Most of the UM patients are between 50 and 70 years old.
Rarely occur before adulthood [31]. The risk factors of UM include light-coloured eyes, fair skin, melanocytoma, congenital ocular melanocytosis and BAP1-tumour predisposition syndrome [32]. Nearly 50% of the UM patients occur metastatic tumor, which usually involves the liver, and usually die within one year [30].
Eye cancer not only seriously affects patient's quality of life, but also directly threatens their lives [33]. Early detection and treatment strategies are considered to be the best countermeasures to prevent and treat eye cancer, which may improve the cure rate of eye cancer, reduce public health expenditure, and exert the least impact on the patient's quality of life.

Conclusion
The age-standardized rate of eye cancer incidence using world population was 12.25/10 7 and 12.75/10 7 in males and females, respectively. These rates are lower than those in Taiwan, Singapore, and some Western countries (Canada and USA). Unlike most other malignant tumors, there is no signi cant difference between males and females in the incidence and mortality of eye cancer. In both urban or rural areas and in both males and females, 0-4 years old has the highest incidence. Therefore, adequate attention should be paid to eye cancer prevention and control in children. Availability of data and materials The datasets generated and analyzed during the current study are not publicly available due to the con dentiality policy of the China Central Cancer Registry but are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
This study was approved by the Ethics Committee of Preventive Medicine in SDCDC (Permission No: 2013020). In the whole study, only data from the annual cancer report was used. In Shandong Province, cancer registration data were collected and managed by SDCDC. All the data used in this study were approved by SDCDC.The consent to participate was not applicable. No information is included to identify individuals. According to the regulations of China's health administrative department, Cancer registration information collection is the responsibility of all kinds of medical and health institutions at all levels. Relevant management documents can be found on the o cial website of the Chinese government (http://www.gov.cn/gongbao/content/2015/content_2868879.htm).

Consent for publication
Not applicable.
Competing interests Figure 1 Age-speci c incidence of eye cancer in Shandong Province in 2013-2017.

Figure 2
Age-speci c mortality of eye cancer in Shandong Province in 2013-2017.