Exposure to Heavy Metal Elements May Signicantly Increase Serum Prostate-Specic Antigen Levels With Overdosed Dietary Zinc

Background: Serum prostate-specic antigen (PSA) is a primary metric for diagnosis and prognosis of prostate cancer (PCa). Exposure to heavy metals, such as lead, cadmium, mercury, and zinc can impact PSA levels in PCa patients. However, it is unclear whether this effect also occurs in men without PCa, which may lead to the overdiagnosis of PCa. Method: Data on a total of 5,089 American men who had never been diagnosed with PCa were obtained from the National Health and Nutrition Examination Survey performed from 2003–2010. The relationship between serum PSA levels (dependent variable) and concentrations of lead (μmol/L), cadmium (nmol/L), and mercury (μmol/L) were investigated with dietary zinc intake being used as a potential modier or covariate in a weighted linear regression model and a generalized additive model. A series of bootstrapping analyses were performed to evaluate sensitivity and specicity using these models. Results: Regression analyses suggested that, in general, lead, cadmium, or mercury did not show an association with PSA levels, which was consistent with the results of the bootstrapping analyses. However, in a subgroup of participants with a high level of dietary zinc intake ( ≥ 14.12 mg/day), a signicant positive association between cadmium and serum PSA was identied (1.06, 95% CI, P=0.0268, P for interaction=0.0249). Conclusions: With high-level zinc intake, serum PSA levels may rise in PCa-free men as the exposure to cadmium increases, leading to a potential risk of an overdiagnosis of PCa and unnecessary treatment. Therefore, environmental variables should be factored in the current diagnostic model for PCa that is solely based on PSA measurements. Different criteria for PSA screening are necessary based on geographical variables. Further investigations are needed to uncover the biological and biochemical relationship between zinc, cadmium, and serum PSA levels PCa.


Introduction
The latest data from the United States Cancer Statistics (USCS) program at the US Centers for Disease Control and Prevention (UCDCP) indicate that prostate cancer (PCa) is the most common cancer among men in the US [1]. There were a total of 3,650,030 PCa cases in 2019, which is more than threefold that of colorectal cancer (994,210 cases), the second most common cancer in men [3]. In 2030, the estimated number of PCa patients will be greater than that for female breast cancer patients. In each subsequent year, there will be about 164,960 new PCa cases, among which 29,430 patients will die from the disease, which has a high mortality rate of 17.84% [2]. A global study found that the incidence and mortality of PCa are higher in the US than in other regions and countries [4][5]. Interestingly, data show that the incidence of PCa in US male adults or in male Japanese immigrants are signi cantly higher than in males in Japan or Korea [6][7]. This suggests that the environment might play a signi cant role in causing PCa. cadmium, or zinc, and PSA levels in blood [38], while a study on 1,320 male participants over 40 years old using data from NHANES came to the same conclusion [39]. To date, data that evaluate the relationship between mercury and PSA levels have not been generated.
To reduce the potential bias from studies based on small sample sizes, we performed the current study using data on 5,089 participants from NHANES to investigate the association between lead, cadmium, mercury, or zinc intake levels (recorded on the rst day on the survey record) with PSA levels in American males over the age of 40. The results indicated that given a high level of dietary zinc, serum PSA levels may rise in PCa-free males as exposure to cadmium increases.

Materials And Methods
Data source NHANES is a program designed to assess the health and nutritional status of adults and children in the US. NHANES includes data on demographic, socioeconomic, dietary, and health-related issues. NHANES includes medical, dental, and physiological measurements including height, weight, and blood pressure, all based on national standards, with all laboratory tests conducted by trained medical personnel. Data and information acquired by the program are being used in epidemiological and health science research to (1) assess nutritional status and its relationship with health promotion and disease prevention, and (2) determine the prevalence and risk factors of major diseases. The results of the research will help to develop public health policies and also help guide designing health plans and services. More detailed information about this data resource may be found at http://www.cdc.gov/nchs/nhanes/nhanes_questionnairees.htm.

Study population
The NHANES database contains PSA data collected from 2003-2010. We integrated this data from the four biennial NHANES survey cycles (2003-2004, 2005-2006, 2007-2008, and 2009-2010) and conducted a secondary data analysis. We only considered males age 40 and over in the NHANES database. Men with PCa, prostatitis or recent prostate surgery (i.e., rectal examination within a week, and prostate biopsy, surgery, or cystoscopy within a month) were not included in the study. We also excluded men who used 5-alpha-reductase inhibitors or other hormone therapies (such as testosterone replacement or medical castration), or whose clinical or social demographic data were incomplete. A total of 5,089 men out of 41,156 participants in NHANES were nally included in our study (See ow chart in Fig. 1).
Categorical variables included hypertension history (no/yes), diabetes history (no/yes), DMDEDUC2 (lower than high school, high school, higher than high school), DMDMARTL (married, living with partner versus not, single), coronary heart disease (no/yes), stroke (yes/no), race/ethnicity (Mexican American, other Hispanic American, non-Hispanic white, non-Hispanic black, other race), smoked at least 100 cigarettes during lifetime (no/yes), enlarged prostate (no/yes), and average level of physical activity each day (PAQ180) (daily exercise: none, slight, normal, vigorous). All variables, which were derived from the participants' self-reports and used in the analysis, are summarized in Table 1.

Data analysis
The skewed distribution of blood lead, cadmium, mercury exposure, zinc intake, and PSA levels were log2 transformed prior to the analysis. Four subgroups were formed according to PSA levels: Q1 (0.07-0.577), Q2 (0.58-0.96), Q3 (0.97-1.88), and Q4 (1.89-74.02). Continuous variables were expressed as β (95% CI), and categorized variables were expressed as percentages. Data analysis was based on the following steps: (1) Univariate analysis, (2) multivariate analysis, (3) fully adjusted model, and (4) generalized additive model (GAM). The variables modi ed in the fully adjusted model included hypertension history, body mass index (kg/m 2 ), diabetes history, DMDEDUC2.NEW NEW, DMDMARTL.NEW NEW, alcohol (gm) rst day, high-density lipoprotein (HDL), poverty income ratio, enlarged prostate, LBDLDL, C-reactive protein (mg/dL), glycohemoglobin (%), triglycerides (mg/dL), coronary heart disease, stroke, PAQ180, age, race/ethnicity, smoked at least 100 cigarettes during lifetime, vitamin D (VITD), and zinc (mg) rst day. The GAM was used to identify the non-linear relationship for the non-equidistant variation in PSA levels after adjustments for hypertension history, body mass index (kg/m 2 ) (smooth), diabetes history, DMDEDUC, DMDMARTL, alcohol (gm) rst day (smooth), HDL (smooth), poverty income ratio, enlarged prostate, LBDLDL (smooth), C-reactive protein (mg/dL) (smooth), glycohemoglobin (%) (smooth), triglycerides (mg/dL) (smooth), coronary heart disease, stroke, PAQ180, age (year) (smooth), race/ethnicity, smoked at least 100 cigarettes during lifetime, VITD (smooth), and zinc (mg) rst day (smooth). For each of the four metals, if the non-linear correlation model was observed, a two-piecewise linear regression model was performed to calculate the in ection point (or threshold effect) of the curve using a recursive method based on a maximum likelihood model [40]. Finally, a subgroup analysis was performed using strati ed linear regression models with different zinc intake dose levels (low, medium, and high). The modi cations and interactions of each subgroup, including lead, cadmium, and mercury with PSA levels, were tested using the likelihood ratio test. Data analysis was performed based on the R package. A P value less than 0.05 was considered statistically signi cant.
The analyses indicated that lead exposure, body mass index, DMDMART (single), enlarged prostate, and age were signi cantly associated with elevated serum PSA levels, with only lead exposure having a 10% or more effect size. The results also suggested that cadmium, mercury exposure, or zinc intake was irrelevant to changes in serum PSA levels.

Adjusted model
A non-adjusted model and a fully adjusted model were attempted to analyze the associations between serum PSA levels with lead, mercury, cadmium exposure, and zinc intake (Table 3). After adjustment of the variables, including hypertension history, body mass index (kg/m 2 ), diabetes history, DMDEDUC, DMDMARTL, alcohol (gm) rst day, HDL, poverty income ratio, enlarged prostate, LBDLDL, C-reactive protein (mg/dL), glycohemoglobin (%), triglycerides (mg/dL), coronary heart disease, stroke, PAQ180, age (year), race/ethnicity, smoked at least 100 cigarettes during lifetime, VITD, and zinc (mg) rst day, the effect size for lead exposure was reduced by nearly 60% from 0.196 (multivariate analysis) to 0.079 (P = 0.00752). The other three variables (mercury exposure, cadmium exposure, and zinc intake) were not signi cantly associated with changes in serum PSA levels.

GAM
In the four-subgroup data (Q1-Q4), we observed a non-equidistant change of regression coe cients when linear regression models were used to analyze the relationship between PSA levels and the four variables of interest, i.e., lead exposure, cadmium exposure, mercury exposure, and zinc intake. Therefore, GAM was used to identify the non-linear relationship between these four variables and PSA levels. The advantage of this model is that it allows other variables to be adjusted using a function, which are then included in a regression model. If a non-linear relationship is identi ed, an in ection point is calculated using a recursive algorithm, followed by an analysis with a weighted linear regression model to integrate data points on both sides of the in ection point. Eventually, a simple linear regression model or a piecewise linear regression model can be established based on the relationship between the logarithmic likelihood ratio with 0.01 selected as a cutoff. In GAM analyses, the variables that had been adjusted included hypertension history, body mass index (kg/m 2 ) (smooth), diabetes history, DMDEDUC, DMDMARTL, alcohol (gm) rst day (smooth), HDL (smooth), poverty income ratio, enlarged prostate, LBDLDL (smooth), C-reactive protein (mg/dL) (smooth), glycohemoglobin (%) (smooth), triglycerides (mg/dL) (smooth), coronary heart disease, stroke, PAQ180, age (year) (smooth), race/ethnicity, smoked at least 100 cigarettes during lifetime, VITD (smooth), and zinc (mg) rst day (smooth). The results showed that only lead exposure was signi cantly associated with changes in serum PSA levels after the adjustment. Univariate analysis showed a signi cant association between zinc intake and serum PSA levels, whereas multivariate analysis did not con rm this association; however, when zinc was adjusted in the GAM analysis, a signi cant association between lead exposure and serum PSA levels was observed.
The sum of the evidence indicated that the zinc in blood in uences serum PSA levels in an indirect manner, which is also supported by the fact that zinc does not directly interfere with PSA pathways.
As shown in Table 3, the logarithmic likelihood ratios for lead, cadmium, and mercury calculated using the GAM model were 0.168, 0.059, and 0.399, respectively, which were all greater than the threshold of 0.01 and suggested a piecewise linear regression model. The analysis with the piecewise linear regression model indicated that when the cadmium concentration was less than 6.06 µmol/L, an increase of cadmium by a unit led to a slight increase in PSA levels by 0.08 units. Such a small effect is equivalent to a 2% change compared with the standard used in clinical practice for diagnosis, i.e., PSA at 4 ng/mL. There was no such association when cadmium concentrations were greater than 6.06 µmol/L. No association was identi ed between serum PSA levels and lead or mercury.  Fully-adjusted model adjust for: Hypertension history, body mass index (kg/m 2 ), diabetes history, DMDEDUC, DMDMARTL, alcohol (gm) rst day, HDL, poverty income ratio, enlarged prostate, LBDLDL, C-reactive protein (mg/dL), glycohemoglobin (%), triglycerides (mg/dL), coronary heart disease, stroke, PAQ180, age (year), race/ethnicity, smoked at least 100 cigarettes during lifetime, VITD, zinc (mg) rst day.

Subgroup analysis
Univariate analyses showed that lead, cadmium, and zinc were all signi cantly associated with changes in serum PSA levels. Multivariate analyses, using a fully adjusted model, and GAM analysis only detected an association between lead and serum PSA. Nevertheless, when zinc was treated as a variable being adjusted in multivariate analysis, the fully adjusted model, or GAM, associations between serum PSA and lead or cadmium could be detected. Thus, we divided the sample into three strata (T1: 0.19-8.69 (low), T2: 8.75-14.03 (median), and T3: 14.12-315.17 (high)) based on zinc intake, and analyzed each of these three strata separately using various models including a non-adjusted model, an adjusted model I, and an adjusted model II ( Table 4). The results showed that only in the T3 (high level of zinc intake) stratum, exposure to cadmium or lead could increase serum PSA levels: an increase of cadmium by 1 unit led to an increase in PSA by 1.06 unit, while an increase in lead by 1 unit yielded an increase in PSA by 1.09 units. Such effects are equivalent to 25% of changes compared with the standard for diagnosis, which is quite signi cant in clinical practice. No such associations were detected in the T1 (low level of zinc intake) or T2 (median level of zinc intake) strata. These results suggest that increased zinc intake may interfere with serum cadmium to increase serum PSA levels in a synergistic manner.  Adjusted model I adjusts for: demography univariate.
Adjusted model II adjusts for: all univariate in Table I.

Discussion
The purpose of this study was to evaluate the effects of lead, cadmium, and mercury exposure, as well as zinc intake on serum PSA levels in PCa-free men. Early studies have shown that exposure to lead, cadmium, and mercury poses a signi cant threat to human health. Zinc is considered to be a useful metal for biological growth and disease recovery, although it does have potential side effects. This study showed that lead exposure increased serum PSA levels in normal men, whereas cadmium or mercury exposure or zinc intake did not affect serum PSA levels. Interestingly, we found that when zinc intake was 14.12 mg/day or higher (high dose), exposure to cadmium signi cantly increased serum PSA levels (β = 1.06, 95% CI), indicating a possible synergistic effect between zinc and cadmium on serum PSA.
Toxicological and physiological studies have shown that lead, cadmium, mercury, and zinc are toxic heavy metals, which can cause various degrees of damage to the human body. Lead may aggregate in blood, liver, and kidney, which can lead to serious toxicity to the nervous system, kidney, and testis [41]. A study has shown that synergism between lead, cadmium, and mercury can cause damage to kidney function [42]. Lead also targets testicular Sertoli cells, which may affect sperm production [43]. Lead toxicity in a children's nervous system will cause typical neurological symptoms [44]. Long time exposure to lead increases the risk of Alzheimer's disease in the elderly [45]. Acute cadmium poisoning usually occurs 12-24 hours after exposure, with severe respiratory symptoms and necrosis of the liver and kidney, while chronic cadmium poisoning is mainly characterized by toxicity in the cardiovascular, cerebrovascular, kidney, and digestive tract [46][47][48][49]. Mercury volatilizes easily and chronic exposure to mercury causes nervous system damage and visual impairment, which poses a high level of risk in certain occupations [50][51][52][53].
It has been a topic of debate as to whether zinc is bene cial in human health. For example, zinc has long been documented to be bene cial in epilepsy, which may be traced back to a study in 1786 [54]. Studies have also shown that zinc may assist in wound healing, and zinc has been applied in the treatment of male infertility [57], impotence [58] and chronic prostatitis [59]. Recent molecular and cellular studies indicated that reduced zinc concentrations were associated with induction or aggravation of ]. An in vitro study also showed that cadmium can induce malignant prostate tumors [71]. However, our results based on previous data did not show such a relationship between cadmium exposure and PSA levels, which might have been because of: (1) a potential bias due to the small sample sizes in each of these studies, and (2) suboptimal timing for data collection in these studies. Interestingly, our results indicated that cadmium exposure increased serum PSA levels when zinc intake exceeded 14.12 mg/day.
Based on NHANES data of 1,320 male cases over 40 years old, Edwin et al. found that dietary zinc intake less than 12.67 mg/day was associated with a protective effect on injured prostate glands, while such an association was not observed in the cohort with a higher zinc intake [72], which was partly consistent with our conclusion. Terrence et al. analyzed NHANES data for 6,488 cases with cadmium exposure and found a signi cant association between zinc intake and serum cadmium [73], which also reiterates what we found. In our study, where the average zinc intake among participants was 13.10 mg/day, we found that 14.12 mg/day was the critical value for zinc intake, beyond which cadmium exposure signi cantly increased serum PSA levels. Of note is that this critical value was lower than the World Health Organization's suggested standard of 15 mg/day but higher than the dose of 12.67 mg/day recommended by Edwin et al. [72]. An interesting question would be whether men living in cadmiumexposed areas should follow the World Health Organization's recommended zinc intake dose (15 mg/day), which may arti cially increase serum PSA levels leading to a false PCa diagnosis. The vast majority of research on the effect of zinc intake suggests that zinc might protect against PCa; nevertheless, a study based on 2,000 Spaniards indicated that high levels of zinc intake will increase the risk of PCa [34]. The sum of the evidence seems to support our ndings that zinc and cadmium might increase serum PSA levels in a synergistic manner, which is similar to the conclusions reached by Mahmoud et al. [75].
This study had a number of limitations. First, there was the potential for selection bias and inaccurate data, which was based on participants' memory, and this may have made it di cult to identify associations between serum PSA levels and the metals studied. Second, only US men were included in the study, which may limit the generalizability of the results. Third, there were no other clinical data for determining whether the increase in PSA was because of physiological or pathological causes. Last, the zinc intake data were completely based on participants' recall of the previous 24 hours, thus preventing long-term and sustainable follow-up.
However, compared with previous studies, our research had the following advantages. First, this study included a total of 5,089 participants from NHANES, which may have reduced the potential biases because of small sample sizes. Second, we used advanced methodologies, including the GAM model, to detect and analyze the potential non-linear associations between serum PSA levels and these metals. In addition, the analysis of subgroups was used to identify the complex relationship between serum PSA levels and the metals with various zinc intake doses.

Conclusions
We found that cadmium exposure increased serum PSA levels given that zinc intake exceeded 14.12 mg/day, leading to a potential overdiagnosis of PCa. Thus, local standards for PSA screening need to be established, and different zinc supplementation strategies could be developed for geographical areas with different levels of cadmium exposure. Our study also indicated that risk factors for PCa should be evaluated with environmental factors taken into account. The underlying biochemical mechanisms  Flow chart of study participants.