Cancer epidemiology · 2025
Identifying environmental factors and biological metrics associated with cancer prevalence and mortality: An environment-wide association study.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Li Zongyuan, Yu Cheng, Hao Jianqi, Shu Yueli, Zhang Jian, Zhao Kejia, Pu Qiang, Liu Lunxu
Аннотация
BACKGROUND: Present knowledge about determinants of oncogenesis and cancer mortality remains incomplete, inconsistent, and controversial. We aimed to conduct an environment-wide association study (EWAS) to systematically investigate and tentatively validate correlations of environmental factors and biological metrics with prevalence and mortality of cancer.
METHODS: All eligible participants were selected from the US National Health and Nutrition Examination Survey (NHANES) and randomly split into training and testing sets by survey years. Environmental and biological exposures were assessed through either physical examinations or laboratory tests. We conducted survey-weighted logistic regression and COX proportional hazards regression models to investigate the relationships of 398 factors with cancer prevalence and 380 factors with cancer mortality, respectively. To adjust for multiple comparisons, positive findings in the training set (false discovery rate [FDR] < 5 %) were tentatively validated in the testing set (P value < 0.05). Random forest models were further fitted to evaluate the importance and diagnostic value of identified factors in relation to cancer prevalence.
RESULTS: Overall, 55,021 general participants and 5163 cancer survivors were included in the study of cancer prevalence and mortality, respectively. After adjusting potential confounders, we identified 7 environmental or biological factors (e.g. total bilirubin, testosterone, and beta-cryptoxanthin) associated with cancer prevalence in the general population, as well as 21, 8, and 6 indicators associated with all-cause (e.g. C-reactive protein), cancer-specific (e.g. blood selenium), and noncancer mortality (e.g. albumin) among individuals with cancer, respectively. EWAS-identified factors contributed to better performance of random forest models in predicting cancer prevalence.
CONCLUSIONS: Employing an EWAS approach, this study provided novel insights into potential targets for prevention and control of cancer.
METHODS: All eligible participants were selected from the US National Health and Nutrition Examination Survey (NHANES) and randomly split into training and testing sets by survey years. Environmental and biological exposures were assessed through either physical examinations or laboratory tests. We conducted survey-weighted logistic regression and COX proportional hazards regression models to investigate the relationships of 398 factors with cancer prevalence and 380 factors with cancer mortality, respectively. To adjust for multiple comparisons, positive findings in the training set (false discovery rate [FDR] < 5 %) were tentatively validated in the testing set (P value < 0.05). Random forest models were further fitted to evaluate the importance and diagnostic value of identified factors in relation to cancer prevalence.
RESULTS: Overall, 55,021 general participants and 5163 cancer survivors were included in the study of cancer prevalence and mortality, respectively. After adjusting potential confounders, we identified 7 environmental or biological factors (e.g. total bilirubin, testosterone, and beta-cryptoxanthin) associated with cancer prevalence in the general population, as well as 21, 8, and 6 indicators associated with all-cause (e.g. C-reactive protein), cancer-specific (e.g. blood selenium), and noncancer mortality (e.g. albumin) among individuals with cancer, respectively. EWAS-identified factors contributed to better performance of random forest models in predicting cancer prevalence.
CONCLUSIONS: Employing an EWAS approach, this study provided novel insights into potential targets for prevention and control of cancer.