Abstract(resumo):

Environmental exposure to heavy metals has been associated with cancer. Nonetheless, the effects of exposure to combined heavy metals on cancer mortality of urinary and gastrointestinal systems have not been examined. This study assessed the long-term association between environmental exposure to multiple heavy metals and cancer mortality rates in the digestive and urinary tracts. The death rates/100,000 inhabitants over 41 years were collected from Santo Amaro, the exposed town with a former lead smelter, and Ribeira do Pombal, the reference town, in Bahia, Brazil. The exposure duration, age and gender were considered predictors. Data were analyzed using random forest models. The overall accuracy of the models ranged from 84 to 97%. For the population above 80 years, the probability of total mortality went from 20% in the first 6 years of exposure to over 70% after 20 years. The same trend was observed for the urinary system. For the digestive system, the mortality in the elders started after 15 years of exposure and peaked with a 30% probability after 35 years. In conclusion, environmental exposure to multiple heavy metals was linked to cancer-related deaths in the gastrointestinal and urinary tracts of the elderly population.

Keywords(palavra-chave):

Heavy metal; Gastrointestinal cancer; Urinary Bladder Cancer; Environmental pollutants

Content(conteúdo):

Introduction
Excessive exposure to heavy metals is considered detrimental to human health. Lead (Pb), cadmium (Cd), and mercury (Hg) are among the most toxic, even at low concentrations, and are among the top 10 chemicals of primary global public health concern 1,2. Heavy metal pollution is mainly due to anthropogenic activities, including sewage, industrial activities, and traffic 3. These heavy metals may contaminate the air, water, and soil 4. Human contamination occurs through exposure to polluted air and ingestion of water, fish, grains such as rice, and medicinal plants 5,6. Given the direct exposure of the digestive and urinary systems to these contaminants through ingestion, studying cancers within these systems is crucial. The digestive system, the primary site for ingesting and processing contaminated food and water, may be particularly vulnerable to the carcinogenic effects of heavy metals.
There are many mechanisms of heavy metal toxicity, including enzyme inhibition, chromosomal damage, production of oxidative stress, and the release of stress proteins 7, which can cause inflammatory reactions, cancer, hypersensitivity, and allergic and autoimmune diseases 8. Exposure to a single heavy metal alone can affect major organ systems and has been associated with cancer development. For example, Pb contamination has increased cancer mortality 9,10 and Cd exposure has been significantly related to breast, lung, and non-specified cancer 11,12. Exposure to multiple metals exhibited enhanced toxicity compared to single metals, even at low doses. In humans exposed occupationally or environmentally, molecular biomarkers that identify disturbances in cellular metabolic pathways that can trigger cancer revealed an additive interaction among Pb, Cd, and arsenic (As) 13. Nonetheless, the combined potential effects of environmental exposure to heavy metals on cancer mortality associated with the urinary and digestive systems have not been examined.
In Santo Amaro, a town in Bahia, Brazil, the lithosphere, hydrosphere and anthroposphere have all been contaminated by heavy metals, including Pb, Cd, Hg, copper (Cu), nickel (Ni), arsenic (As), antimony (Sb), zinc (Zn), by a multinational lead company that operated metal smelters14,15,16,17,18. The company was founded in 1960 and, after a few years, signs of contamination emerged, including otherwise unexplained deaths of animals on nearby farms 18. The lead smelter operated until 1993, leaving approximately 500,000 tons of industrial dross containing toxic elements and contaminating the company grounds and urban environment 18. Contaminated slag was used to pave public and private areas 16. The Pb and Cd levels exceeded limits established by the World Health Organization by a factor of 46 and 17 in soil, respectively, and 260 and 84 in water 16,19. Even 26 years after the company ceased operations, children continued to be born with high blood Pb concentrations, and chromosomal alterations were detected in cattle and adult women 14. Peripheral lymphocytes of 30 women who lived 1,000 meters from the polluter plant showed high cytogenetic aberrations, including chromosome alterations, chromosome instability, and altered mitotic index and cytogenetics, compared to non-exposed women 15. However, to our knowledge, cancer mortality has not been evaluated in the population.
This study is essential as it explores the link between prolonged environmental exposure to various heavy metals and cancer mortality. By focusing on two distinct municipalities- Santo Amaro, contaminated by heavy metals, and Ribeira do Pombal, uncontaminated- the research enhances our understanding of industrial activities' potential health impacts. The investigation aimed to determine whether prolonged exposure to multiple heavy metals in Santo Amaro increased cancer mortality rates, particularly for digestive and urinary system cancers. Additionally, water samples from Santo Amaro were tested for heavy metals. Since the entire population relies on water from the same source, water analysis provides direct evidence that the population remains exposed. The hypothesis is that more prolonged exposure to multiple heavy metals correlates positively with higher cancer mortality.

Materials and Methods
This study employed an ecological design based on secondary data sources to evaluate the association between environmental heavy metal exposure and cancer mortality at the municipal level. Therefore, ethical approval was not required for this study as the data were obtained from a public database.

Exposure Assessment
Two municipalities were included: 1) an area contaminated with heavy metals (Santo Amaro, State of Bahia, Brazil); 2) an area not contaminated with heavy metals (Ribeira do Pombal, State of Bahia, Brazil- Supplementary table). The reference town was selected based on demographic, cultural, economic, geographic, and climate similarities21,22,23. Santo Amaro and Ribeira do Pombal are in Northeast Brazil.
Fourteen samples of drinking water were collected in 2023 (on the same day) from points of treated water for human supply in different regions of Santo Amaro, Bahia, Brazil, to analyze the presence of heavy metals. The water samples were collected 48 hours after a light rainfall during the spring. Sampling points were strategically selected throughout the municipality to represent the entire water supply consumed by the city accurately. Two water samples were collected at each sampling point using 250 mL plastic bottles containing 1.0 mL of nitric acid. The bottles were not rinsed prior to sampling and were not filled to the brim, leaving approximately 2 cm of headspace for sample homogenization. The samples were then stored in an insulated container with sufficient reusable ice packs to maintain a temperature between 3°C and 10°C during transport, ensuring sample integrity. The metals were quantified using an inductively coupled plasma mass spectrometer (ICP-MS, model 820-MS, Bruker, Madson, WI). The water samples from Santo Amaro exhibited varying concentrations of Pb, Cd, Ni, Fe, Sb, and As (Table 2).
Table 1 supports the classification of Ribeira do Pombal as an uncontaminated reference area.

Cancer mortality data
The death rate per 100,000 inhabitants was obtained from the Brazilian National Cancer Institute database (https://mortalidade.inca.gov.br/MortalidadeWeb/). The institute gathers information on mortality through the systematic collection of data present in death certificates in Brazil. Information about the deceased is organized by gender, age group, municipality, state, geographic region, and cause of death. Malignant neoplasm is classified according to the 10th International Classification of Diseases Revision20. Population estimates from demographic censuses are used as denominators for calculating mortality rates. The specific rate per 100,000 inhabitants is calculated as the quotient between the total number of events for a particular neoplasm (by age group, gender, and period) and the aimed population (also stratified by age group, gender, and year).
Data were collected between March 2021 and May 2024 by two researchers. Cancer mortality rate/100,000 inhabitants for 41 years (from 1979 to 2020) on neoplasms on the digestive (including lip, tongue, gum, the floor of the mouth, palate, other parts of the mouth, salivary glands, pharynx, esophagus, stomach, intestine, pancreas and liver) and urinary systems (including kidney and bladder) were obtained. Mortality rates were acquired according to gender and age group (0-9, 10-19, 20-29, 30-39, 40-49, 50-59, 60-69, 70-79, and 80 years and over). The data were organized in a database using Microsoft Excel®.

Statistical Analysis
A descriptive analysis was initially conducted, considering the average mortality rate over the 41 years of analysis, followed by exploratory analyses using annual mortality rates. Data was analyzed using random forest machine learning models24. For the models, the mortality rate per 100,000 inhabitants was first categorized into three groups: 0 - when the rate was zero; 1- when the rate was below the median (lower mortality); 2- when the rate was above the median (higher mortality). As predictors of the death rate, the municipality, exposure duration, gender and age were considered. Random forest models were conducted for each cancer group (digestive and urinary) and the total.
The models were trained with 80% of the data and validated with the remaining 20%; this split was done in a balanced way among the three death rate levels. No missing data were present in the dataset, so no imputation or data handling techniques were required. Hyperparameters were set to optimize model stability and accuracy. All the models were parameterized to perform 100 trees, 5 repeated cross-validations, and 5 repetitions to prevent overfitting and ensure robustness. The models were trained with a bootstrapping procedure, where each tree was built using a randomly selected subset (with replacement) of the training data, while approximately one-fifth of the data (out-of-bag samples) was used to estimate model performance and variable importance internally. The number of variables randomly selected at each split (mtry) was done using the combination of 2, 3, and 4 variables. Additional parameters remained in the software's default configuration.
Models’ performance was analyzed using the overall accuracy, kappa index, and confusion matrix25. For each model, the feature importance analysis was performed to select the significant variables (p<0.05), and the partial dependence analysis to evaluate the interaction among the significant ones. The partial dependence analyses were performed using 99 interactions and a grid resolution of 15. The simulated grids were further used to visualize the probabilities of each death rate level among the predictors. The analyses were performed in iMESC, an open R-language app25.

Results
General characteristics
Table 1 shows the mean (± standard deviation) for cancer mortality rates from 1979 to 2020 per 100,000 inhabitants, considering all age groups, computed for each municipality's oral, digestive and urinary groups. There were higher mean death rates in the exposed town compared to the reference town across multiple cancer sites. For the digestive system, notable differences included the esophagus (9.62 ± 35.92 vs. 0.79 ± 8.12) and stomach (22.53 ± 69.42 vs. 5.83 ± 35.50). Similarly, for the urinary system, the bladder (4.32 ± 30.24 vs. 1.02 ± 11.50) and kidney (1.83 ± 17.34 vs. 0.30 ± 5.87) showed higher mean death rates in the exposed town. From the whole dataset, 91% of the deaths were associated with the digestive system, while 9% were associated with the urinary. During the study period, Santo Amaro showed 4.15 times more deaths related to cancer than Ribera do Pombal. In the urinary system, the total mortality was more associated with the kidney (r = 0.52); in the digestive system, the total mortality was more associated with the stomach (r = 0.69) (Figure 1).

Results of Random Forest Machine Learning Models
The random forest analysis yielded an overall accuracy of 84 and 97% and a kappa between 18 and 62% in the training part of the data (Table 3). The highest kappa was associated with urinary cancers. Specifically for the higher mortality group (group 2), the accuracies of the model for the total rate of deaths were 70% and 67% for the training and test part of the data, respectively (Table 3). Yet, for the urinary cancers, the model correctly classified 79% of the higher mortality group in the training data and 55% in the test. For the digestive, it was 100% and 50%, respectively.
For all the models, the feature importance analysis significantly selected only the variables age and exposure duration (Figure 2- A1, B1, C1). The partial dependence analysis showed that these two variables interacted with each other. Modeling results demonstrated that during the whole exposure period, the population under 50 years of age showed zero probability of dying from any evaluated cancer (Figure 2- A2, B2, C2). Nonetheless, for the population above 80 years, the probability of mortality (total) went from 20% in the first 6 years of exposure to over 70% after 20 years. It remained stable until the end of the exposure period (Figure 2- C3). The same trend was observed for the urinary (Figure 2- B3) and digestive systems (Figure 2- A3).

Discussion
This study confirmed the hypothesis of a positive association between urinary and gastrointestinal system cancer mortality rates and the time to exposure to multiple heavy metals. The concentrations of Pb, Ni, and Fe, and zinc exceeding the limits stipulated by regulations, detected in the exposed group have also been reported by previous studies in the area where the present exposed group lives 14-18. According to our results and previous studies, environmental contamination by heavy metals was associated with a negative impact on human health 14,17,26. In the present study, the negative effects manifested primarily among adults above 80 years old. Given that the disposal started in the 60’s and the cancer mortality began to increase in 1985, the latency period was about 15 years of exposure. Among the elders, the death rates kept increasing till 1999 and remained high till recently. It is known that the incidence of cancer related to exposure to heavy metal pollution has a latency period, which can be over 10 years 27.
Nonetheless, the latency period between exposure and the manifestation of cancer may differ for different tissues 28, which can be observed in our study. In the present study, the increase in death rates started to be registered earlier for the urinary system than for the digestive. Cancer in different regions may respond differently to environmental exposures, reflecting the complex and multifactorial nature of carcinogenesis. Factors such as the varying susceptibility of tissues to heavy metal toxicity, the diverse molecular mechanisms, and the interplay of genetic and environmental factors underlie cancer pathogenesis29. Particularly in the present study, the total mortality in the urinary system was more associated with the kidney, while in the digestive system, with the stomach. The association with esophagus and stomach cancer agrees with previous reports on exposure to Cd and Pb 12,13,30, the same heavy metals found in the present exposed group. It is crucial to consider that the mouth, esophagus, and stomach have direct and prolonged contact with food and water contaminated by heavy metals.
The present analysis reveals an increase in mortality rates over the duration of exposure, highlighting a concerning trend. Even with the introduction of new regulations aimed at reducing future environmental pollution, insufficient efforts have been made to remediate existing contamination. While no new contamination cases have been reported in the region of Santo Amaro, the legacy of prior pollution continues to impact mortality rates. Therefore, comprehensive interventions addressing past contamination and ongoing health risks are required to ensure that public health measures effectively tackle the long-term consequences of environmental exposures.
As expected, age was associated with cancer mortality. Several biological changes linked to the aging process can explain this fact, such as the accumulation of oxidative stress and DNA damage over the years caused by lifelong exposure to free radicals, UV irradiation, and foods, as well as a progressive decay of immune function 31. Within this context, the combined effects of aging and heavy metals heightened cancer mortality in this study.
Heavy metals may contribute to carcinogenesis at the molecular and cellular levels, involving many pathways. They can generate reactive oxygen species within cells, leading to oxidative stress and damaging proteins, lipids, and DNA 32. Cd, Pb, Ni, Fe and As, found in the tap water of Santo Amaro, can interact with DNA, causing strand breaks, cross-links, and the formation of DNA adducts, and can also interfere with DNA repair mechanisms, which may lead to the accumulation of genetic mutations 33, 34, 35. Moreover, heavy metals have been linked to the development of chronic inflammation, a key driver of carcinogenesis. They can activate signaling pathways, leading to prolonged release of proinflammatory cytokines and chemokines and creating a microenvironment conducive to tumor growth 36. Heavy metals can also induce alterations in the epigenome, modifying gene expression patterns without changing the underlying DNA sequence. These modifications include DNA methylation, histone modifications, and non-coding RNA expression. Dysregulation of epigenetic processes can lead to the silencing of tumor-suppressor genes or activation of oncogenes, contributing to carcinogenesis 37.
The limitations of this study are related to the nature of the dataset, which did not permit a control for other cancer risk factors such as level of exposure, occupational and other exposures, diet, habits, genetic/family history, lifestyle, socioeconomic status, smoking and alcohol consumption, and infection with Helicobacter pylori; factors that are also known to affect cancer mortality rates 12,27,38,39. Although these confounders could not be controlled, the exposed and non-exposed groups had similar environmental, cultural, dietary, and geographic characteristics, minimizing the risk of bias. Moreover, at the beginning of the time series, the death rates by cancer were nearly 3.5 times lower than after 20 years of exposure, pointing to the long-term exposure effect.
In summary, measures to mitigate the effects of environmental contamination with heavy metals and reduce the risk of gastrointestinal and urinary cancers associated with heavy metal contamination are required for this population. Although the contamination ceased more than 30 years ago when the polluting industry was shut down, the environment remains contaminated with heavy metals, and the population continues to be exposed to these contaminants. Water contamination varies according to temperature and rainfall; however, since temperature and rainfall remain constant throughout the year in Santo Amaro, water contamination is also expected to be consistent year-round 40. The mitigating measures require minimizing exposure, enhancing detoxification, and implementing early detection strategies and regulation. As mitigating measures, ion exchange, precipitation chemistry, membrane filtration, flocculation, solvent extraction, and adsorption using bentonite clays eliminate Cd, Pb, and zinc from contaminated water and effluents41. Water filtration methods such as reverse osmosis and activated carbon effectively remove As, Cd, Pb, and Hg from household drinking water, while dietary modifications—such as avoiding crops grown in contaminated soil and seafood with high metal accumulation—help limit ingestion 42,43. Furthermore, electroremediation of the soil can be suggested, as the population is also exposed to soil contamination and to food contaminated 18. Detoxification strategies include increasing antioxidant intake, consuming soluble fiber to bind metals in the gut, and using probiotics to aid excretion 44. Regular screenings, including colonoscopy and urine cytology, support early cancer detection. Additionally, stricter industrial regulations and cleaner production technologies are crucial to reducing environmental contamination and its carcinogenic effects.

Conclusion
The improper disposal of heavy metals since the 1960s has been linked to increased cancer mortality in the digestive and urinary tracts among older adults. Higher and rising rates were observed in older age groups and with longer exposure durations, highlighting an interaction between age and exposure time. Identifying environmental risk factors for cancer is crucial for planning preventive measures and allocating resources to reduce risks and improve health. Additionally, mitigation efforts are necessary for contaminated areas.

Acknowledgments: This work was supported by the CNPQ (Conselho Nacional de Desenvolvimento Científico e Tecnológico) under Grants (number 404088/2023-6,
403344/2022-0 and 306780/2022-40; CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior); and FAPESB (Fundação de Amparo à Pesquisa do Estado da Bahia).

Declaração de Disponibilidade de Dados
Os dados de pesquisa estão disponíveis mediante solicitação ao autor de correspondência.


References
1) Al-Saleh I, Shinwari N, Mashhour A. Heavy metal concentrations in the breast milk of Saudi women. Biol Trace Elem Res 2003; 96(1-3): 21-37.
2) Tchounwou PB, Yedjou CG, Patlolla AK, Sutton DJ. Heavy metal toxicity and the environment. In: Luch A, ed. Molecular, Clinical and Environmental Toxicology. Experientia Supplementum, vol 101. Basel: Springer; 2012.
3) Sljivic Husejnovic M, Bergant M, Jankovic S, Zizek S, Smajlovic A, Softic A, Music O, Antonijevic B. Assessment of Pb, Cd and Hg soil contamination and its potential to cause cytotoxic and genotoxic effects in human cell lines (CaCo-2 and HaCaT). Environ Geochem Health 2018 Aug;40(4):1557-1572.
4) Järup L. Hazards of heavy metal contamination. Br Med Bull 2003;68:167-82.
5) Asgari Lajayer B, Ghorbanpour M, Nikabadi S. Heavy metals in contaminated environment: Destiny of secondary metabolite biosynthesis, oxidative status and phytoextraction in medicinal plants. Ecotoxicol Environ Saf 2017 Nov;145:377-390.
6) Glava? NK, Djogo S, Raži? S, Kreft S, Veber M. Accumulation of heavy metals from soil in medicinal plants. Arh Hig Rada Toksikol 2017 Sep 26;68(3):236-244.
7) Jomova K, Valko M. Advances in metal-induced oxidative stress and human disease. Toxicology 2011 May 10;283(2-3):65-87.
8) Lehmann I, Sack U, Lehmann J. Metal ions affecting the immune system. Met Ions Life Sci 2011;8:157-85.
9) Wani AL, Ara A, Usmani JA. Lead toxicity: a review. Interdiscip Toxicol 2015 Jun;8(2):55-64.
10) Guo X, Su W, Li N, Song Q, Wang H, Liang Q, Li Y, Lowe S, Bentley R, Zhou Z, Song EJ, Cheng C, Zhou Q, Sun C. Association of urinary or blood heavy metals and mortality from all causes, cardiovascular disease, and cancer in the general population: a systematic review and meta-analysis of cohort studies. Environ Sci Pollut Res Int 2022 Sep;29(45):67483-67503.
11) Song M, Huang X, Wei X, Tang X, Rao Z, Hu Z, Yang H. Spatial patterns and the associated factors for breast cancer hospitalization in the rural population of Fujian Province, China. BMC Womens Health.2023 May 9;23(1):247.
12) Lee NW, Wang HY, Du CL, Yuan TH, Chen CY, Yu CJ, Chan CC. Air-polluted environmental heavy metal exposure increase lung cancer incidence and mortality: A population-based longitudinal cohort study. Sci Total Environ.2022 Mar 1;810:152186.
13) Wang M, Song H, Chen WQ, Lu C, Hu Q, Ren Z, Yang Y, Xu Y, Zhong A, Ling W. Cancer mortality in a Chinese population surrounding a multi-metal sulphide mine in Guangdong province: an ecologic study. BMC Public Health.2011 May 16;11:319.
14) Carvalho FM, Silvany Neto AM, Tavares TM, Costa AC, Chaves Cd, Nascimento LD, Reis Mde A. Chumbo no sangue de crianças e passivo ambiental de uma fundição de chumbo no Brasil [Blood lead levels in children and environmental legacy of a lead foundry in Brazil]. Rev Panam Salud Publica 2003 Jan;13(1):19-23.
15) Rodrigues CV. Efeito mutagênico ambiental de metais pesados em mulheres de Santo Amaro da Purificação - Bahia. [dissertação]. Salvador: Universidade Federal da Bahia; Centro de Pesquisas Gonçalo Moniz, Fundação Oswaldo Cruz; 2002.
16) Machado S, Ribeiro LD, Kiperstok A, Botelho M, Carvalho MF. Diagnóstico da contaminação por metais pesados em Santo Amaro-BA. Eng Sanit Ambient 2004;9:140-155.
17) Hatje V, Barros F, Figueiredo DG, Santos VL, Peso-Aguiar MC. Trace metal contamination and benthic assemblages in Subaé estuarine system, Brazil. Mar Pollut Bull 2006 Aug;52(8):982-7.
18) de Andrade Lima LRP, Bernardez LA. Characterization of the soil contamination around the former primary lead smelter at Santo Amaro, Bahia, Brazil. Environ Earth Sci 2017;76:470.
19) Anjos JÂSA dos, Sánchez LE. Plano de gestão ambiental para sítios contaminados por resíduos industriais-o caso da Plumbum em Santo Amaro da Purificação, BA [Internet]. Bahia Análise & Dados 2001;10(4):306-309.
20) Center of Disease Control and Prevention (CDC). International Classification of Diseases, Tenth Revision (ICD-10); 2022.
21) InfoSanBAS. Ribeira do Pombal: Perfil Municipal. Brasil. 2024.
22) Instituto Brasileiro de Geografia e Estatística (IBGE). Cidades e Estados. Brasil; 2024.
23) Ministério da Saúde, Gabinete do Ministro. Ordenação GM/MS No. 888. GM/MS No. 5. 28 de Setembro, 2017. Consolidação das normas sobre controle da qualidade da água para consumo humano. Brasília, DF: Diário Oficial da União (2021).
24) Breiman L. Random Forests. Machine Learning 2001;45:5-32.
25) Vieira DC, Fonseca G. iMESc: An Interactive Machine Learning App for Environmental Science. Zenodo; 2023.
26) Wang G, Fowler BA. Roles of biomarkers in evaluating interactions among mixtures of lead, cadmium and arsenic. Toxicol Appl Pharmacol 2008 Nov 15;233(1):92-99.
27) Su CC, Tsai KY, Hsu YY, Lin YY, Lian IeB. Chronic exposure to heavy metals and risk of oral cancer in Taiwanese males. Oral Oncol.2010 Aug;46(8):586-90. Epub 2010 Jul 8.
28) Howard J. Minimum Latency & Types or Categories of Cancer. World Trade Center Health Program. Published October 17, 2012.
29) Gupta B, Johnson NW. Emerging and established global life-style risk factors for cancer of the upper aero-digestive tract. Asian Pac J Cancer Prev 2014;15(15):5983-91.
30) Steenland K, Boffetta P. Lead and cancer in humans: where are we now? Am J Ind Med 2000 Sep;38(3):295-299.
31) Finkel T, Serrano M, Blasco MA. The common biology of cancer and ageing. Nature 2007 Aug 16;448(7155):767-774.
32) Pizzino G, Irrera N, Cucinotta M, Pallio G, Mannino F, Arcoraci V, Squadrito F, Altavilla D, Bitto A. Oxidative Stress: Harms and Benefits for Human Health. Oxid Med Cell Longev 2017;2017:8416763. Epub 2017 Jul 27.
33) Chen QY, DesMarais T, Costa M. Metals and Mechanisms of Carcinogenesis. Annu Rev Pharmacol Toxicol 2019 Jan 6;59:537-554.
34) Denkhaus E, Salnikow K. Nickel essentiality, toxicity, and carcinogenicity. Crit Rev Oncol Hematol 2002 Apr;42(1):35-56.
35) Huang X. Iron overload and its association with cancer risk in humans: evidence for iron as a carcinogenic metal. Mutat Res 2003 Dec 10;533(1-2):153-171.
36) Multhoff G, Molls M, Radons J. Chronic inflammation in cancer development. Front Immunol 2012 Jan 12;2:98.
37) Ryu HW, Lee DH, Won HR, Kim KH, Seong YJ, Kwon SH. Influence of toxicologically relevant metals on human epigenetic regulation. Toxicol Res 2015 Mar;31(1):1-9.
38) Zali H, Rezaei-Tavirani M, Azodi M. Gastric cancer: prevention, risk factors and treatment. Gastroenterol Hepatol Bed Bench 2011 Fall;4(4):175-185.
39) Fei X, Wu J, Kong Z, Christakos G. Urban-rural disparity of breast cancer and socioeconomic risk factors in China. PLoS One 2015 Feb 17;10(2):e0117572.
40) Ma R, Zhang B, Zhou X. The effects of climate change and groundwater exploitation on the spatial and temporal variations of heavy metal content in maize in the Luan River catchment of China. Environ Sci Pollut Res Int 2020 Jan;27(1):1035-1052.
41) Chai WS, Cheun JY, Kumar PS, Mubashir M, Majeed Z, Banat F et al. A review on conventional and novel materials towards heavy metal adsorption in wastewater treatment application. J Clea Prod 2021 May 10;296:126589.
42) Carolin CF, Kumar PS, Saravanan A, Joshiba GJ, Naushad, M. Efficient techniques for the removal of toxic heavy metals from aquatic environments: A review. J Environ Chem Eng 2017;5(3): 2782-2799.
43) Islam Eu, Yang XE, He ZL, Mahmood Q. Assessing potential dietary toxicity of heavy metals in selected vegetables and food crops. J Zhejiang Univ Sci B 2007 Jan;8(1):1-13.
44) Zhai Q, Narbad A, Chen W. Dietary strategies for the treatment of cadmium and lead toxicity. Nutrients 2015 Jan 14;7(1):552-571.
45) US Environmental Protection Agency (USEPA). National Primary Drinking Water Regulations. Published (2009).
46) Directive (EU) 2020/2184 of the European Parliament and of the Council of 16 December 2020 on the quality of water intended for human consumption (recast). Off. J Eur Union 2020;63(63).