Serum levels of Vitamin D3, Zinc, and Parathyroid Hormone in HCV-induced Hepatocellular Carcinoma.

Background and Aim The role of the metabolic syndrome in hepatocellular carcinoma (HCC) has been previously reported. This study aims to investigate the possible role of vitamin D3, Zinc, Parathyroid hormone (PTH), calcium and phosphorus serum levels as non-traditional metabolic risk factors in HCV-related HCC. Method This cross-sectional observational study recruited HCV infected patients with and without HCC. All patients were subjected to demographic, biochemical, and hematological assessment. Serum levels of vitamin D3, Zinc, PTH, calcium, and phosphorus were determined in all the study participants. This study includes 50 patients with HCV-related HCC compared to 40 patients with HCV-related liver cirrhosis and 30 patients with HCV chronic hepatitis C (CHC) without HCC. Our results show signicantly higher age, male sex, aspartate transaminase (AST), PTH and corrected serum calcium levels in the HCC patients compared to values in the other two groups, (p < 0.001); while signicant lower vitamin D3 and zinc levels were detected among the HCC patients compared to patients with non-HCC liver cirrhosis and CHC, (p < 0.001).Vitamin D3 deciency was detected in 96% of the HCC patients, while it was detected in only 22.5% of the cirrhotic patients and in none of the CHC patients, (p < 0.001). However, on multiple stepwise regression analysis, only the age, AST, PTH, and corrected calcium levels were the independent predictors for HCC when studied in relation to chronic liver disease. study indicates however, a relationship


Background
Hepatocellular carcinoma (HCC) is the fth most common cancer and third cause of death due to cancer universal. 1,2 A variety of risk factors for the development of HCC have been clearly identi ed. Hepatitis B and C viral infections are the main HCC risk factors. 3 In addition, dietary a atoxin exposure, alcoholic and non-alcoholic cirrhosis are considered risk factors of HCC. 4 Metabolic syndrome (MS) associated with insulin resistance is recognized as a potential risk factor for the development of hepatocellular carcinoma. 5 The liver plays an important role in the metabolism of vitamin D which is hydroxylated in the liver into 25hydroxyvitamin D (Vitamin D3) and then transported to the kidney to undergo a second hydroxylation, to form 1, 25(OH) Vitamin D. 6 Vitamin D3 is particularly known for its function in calcium homeostasis and was measured using kits supplied by Abcam ( ab102507, USA). Serum phosphorus level was measured by kit supplied by Abcam (ab65622, USA). The patients were classi ed according to serum Zinc levels in patients with zinc de ciency (< 70 μg/dl), patients with normal zinc levels ≥ 70 μg/dl). 15 Finally both Parathyroid hormone (PTH) and Serum 25-OH vitamin D3 levels were measured by mini Vida an automation using the ELFA (Enzyme Linked Fluorescent Assay) technique (Mini Vidas, bioMerieux, France). According to serum 25-OH vitamin D3 levels, all patients were classi ed into patients with vitamin D3 de ciency (< 20 ng/dl), patients with vitamin D3 insu ciency (20-30 ng/dl), and patients with vitamin D3 su ciency (> 30 ng/ml). 16

Statistical analysis
The collected data were coded, tabulated, and statistically analyzed using SPSS program software version 24. Test of normality (Kolmogorov-Smirnov) was done to determine the distribution of the quantitative data. Descriptive statistics were done for parametric quantitative data by mean, standard deviation, and minimum & maximum of the range; while the nonparametric quantitative data was presented by the range and median. The categorical data was presented as number and percentage. An independent sample T test was used for analyzing the parametric quantitative data of the two groups, and Mann Whitney test was used for the non-parametric quantitative data. Chi-Square test was used for analyzing the qualitative data. Simple regression analysis was done to determine the possible predictive factors for HCC. Multiple logistic and stepwise regressions were done to detect the independent predictors for HCC. ROC curve study was done to determine the cutoff point, AUC, sensitivity, speci city, PPV, NPV and accuracy for the signi cant independent predictors for HCC. The level of signi cance was considered when P value was < 0.05.

Results
This study includes 50 HCV-related cirrhotic patients with HCC (Group 1), 40 HCV cirrhotic patients without HCC (Group 2) and 30 HCV chronic hepatitis (CHC) patients without cirrhosis (Group 3). Table (1) presents the statistical signi cance of the baseline demographic and clinical characteristics of the studied groups. The data indicates that the cirrhotic groups with and without HCC have signi cantly higher age compared to the CHC patients, (p < 0.001), while there is no signi cant difference in age between the cirrhotic and HCC groups, (p = 0.729). There is statistically higher male distribution in the HCC group (68%) compared to the non-HCC cirrhotic group (45%; p value = 0.028). However, both the CTP and MELD scores are not signi cantly different between the HCC and the non-HCC cirrhotic patients.
Regarding the laboratory data, the results show that the total leucocytes count (TLC), serum albumin, ALT and AST have higher signi cant values in the HCC group compared to the cirrhotic group without HCC, Table (2). Table (3) shows that vitamin D3 has signi cantly lower level in the HCC group than in the non-HCC cirrhotic and CHC groups (values of: 14.7 ± 8.4ng/ml, 28.5 ± 11.3 ng/ml, and 29.3 ± 6.2 ng/ml for the three groups, respectively, with p = < 0.001). However, there is no signi cant difference in vitamin D levels between the non-HCC cirrhotic and CHC groups, (p = < 0.605). According to the guideline classi cation for the cutoff values for de cient and insu cient levels of vitamin D3 16 , vitamin D3 de ciency was detected in 96% of our HCC patients while it was detected in only 22.5% of the cirrhotic patients and in none of the CHC patients, p < 0.001, Table (4). At the same time, zinc level in the HCC group has signi cantly lower level than in the non-HCC cirrhotic (p < 0.02) and in the CHC patients (p < 0.001), Table (3). Zinc levels are strati ed according to the recommended cutoff value (patients with normal zinc levels ≥ 70 µg/dl). 19 Zinc is found to be de cient in 96% of the cases with HCC while this is observed in 75% of the cirrhotic patients without HCC, and in only 16.7% of the CHC patients, p < 0.001, Table (5). PTH has signi cantly higher levels in the HCC group than in the non-HCC cirrhotic group (p < 0.013), and the levels in the non-HCC cirrhotic group are signi cantly higher than in the CHC group (p < 0.001), Table (3). The corrected calcium has signi cantly higher level in the HCC group than the levels in the non-HCC cirrhotic and CHC groups; (p < 0.001), Table (3). The ionized calcium levels present the same statistical signi cance in the studied groups. There is no statistically signi cant difference between the three groups as regards serum phosphorus level, Table (3).
Utilizing multiple stepwise logistic regression analysis for the prediction of HCC when chronic liver disease (CLD) (the cirrhotic and CHC patients) is the reference, the corrected calcium, the PTH, the AST and the age are the independent predictors for HCC. The OR for corrected calcium is 3.18; 95%CI: 1.96-5.14 with p = < 0.001; the OR for PTH is 1.01; 95%CI: 1.004-1.02 with p = 0.002; the OR for AST is 1.03; 95%CI: 1.01-1.05 with p = 0.002 and the OR for age is 1.11; 95%CI: 1.02-1.21 with p = 0.014 in the prediction of HCC, Table (6).  (1) show ROC curve analysis and accuracy indices for the prediction of HCC when chronic liver disease is the reference group. With optimal cut off points of > 100 ng/L for PTH; >9.63 mg/dl for corrected calcium, > 58 years for age, and > 50 U/L for AST; these variables have AUC values of 0.800, 0.790, 0.739 and 0.686, respectively, with P values of < 0.001 for all of them. Corrected calcium has the highest total accuracy of 85.6 %.in the prediction of HCC development on top of chronic liver disease.

Discussion
Our study aims to investigate the status of some non-traditional metabolic variables in relation to HCC in HCV-induced cirrhotic patients. We estimated the serum levels of vitamin D3, Zinc, parathyroid hormone (PTH), calcium, and phosphate in HCV patients with and without HCC in a trial to study their possible predictive or risk abilities in the development of HCC.
In our study, in the HCC patients, the male distribution is twice more frequent than in females (68% vs 32%); with signi cantly higher male distribution in the HCC group than in the cirrhotic one (68% vs 45%, p < 0.028). This re ects the male predominance in HCC as has been reported by Hammad et al. 17 , who stated that HCC was signi cantly more frequent in males than in females (77.7% and 22.3% respectively).
However, in our study, on multiple logistic regression analysis, the sex distribution is not a signi cant predictor for the occurrence of HCC.
The reasons for this gender disparity could be explained by differences in exposure to risk factors.
Furthermore, estrogen is believed to have a protective role against the development of HCC as differences in subtypes of estrogen receptors expressed in males vs. females have been shown to contribute to the progression of HCV related HCC. 18 Our data also show that HCC is more common in older age patients; albeit with no signi cant difference of age between the HCC and non-HCC cirrhotic patients. The age is signi cantly higher in the HCC and the cirrhotic than in the CHC patients, (p < 0.001). On multiple stepwise logistic regression analysis in relation to chronic liver disease rather than to cirrhosis, older age is found to be signi cantly associated with HCC (OR: 1.11; 95% CI; 1.02-1.2; P < 0.014). El Zayadi et al. 19 have reported that HCC in Egypt is signi cantly more prevalent among older age groups than younger age groups and they have suggested that HCV infection in old patients induces a rapid progression to HCC independent of HCV genotype. Omata et al. 20 have reported that older age is a risk factor for HCC, especially in areas where HCV infection is endemic as in Egypt.
Furthermore, our results show that both the AST and ALT levels are signi cantly higher in the HCC patients when compared to the non-HCC cirrhotic and CHC patients. The raised AST is signi cantly associated with HCC (OR: 1.03; 95% CI; 1.01-1.05). At a cut off value of > 50 U/L, AST has sensitivity 76% and speci city 65.71%. This nding is supported by large cohort study that has been done on 1108 patients with HCC by Carr and Guerra 21 , who have found an association between increasing levels of liver enzymes (AST, ALP and GGT) and HCC aggressiveness.
Our data present signi cantly lower levels of vitamin D3 in the HCC group compared to the cirrhotic and the CHC groups. Furthermore, vitamin D3 de ciency is signi cantly more frequent in patients with HCC (96%) compared to the cirrhotic and chronic hepatitis patients, (p < 0.001). These results are similar to those of a study done by Finkelmeier et al. 22 , who have measured vitamin D3 in 200 patients with HCC and cirrhosis and have compared its level to the stages of both HCC and stages of cirrhosis based on MELD, CTP, BCLC, and The Cancer of the Liver Italian Program (CLIP) scores. In their study, vitamin D3 levels are negatively correlated with the stages of cirrhosis as well as stages of HCC. Furthermore, their patients with severe vitamin D3 de ciency had the highest mortality risk. Our study indicates that the mean vitamin D3 levels are not signi cantly different between the cirrhotic and the CHC patients. However, vitamin D de ciency is more frequent in the cirrhotic than in the CHC groups (22.5% vs zero%, respectively; p = 0.004). This later nding is in agreement with Duarte et al. 23 , who have studied vitamin D3 level in 100 patients with chronic viral hepatitis, (49 noncirrhotic and 51 with cirrhosis), where they have found that vitamin D3 was low in only 3 cirrhotic patients without signi cant difference between the cirrhotic and the non-cirrhotic groups. However, Miroliaee et al. 24 have reported that vitamin D de ciency is signi cantly more frequent in cirrhotic patients compared to non-cirrhotic patients (76.5 vs.17.9%; p = 0.001).
Vitamin D3 and its derivatives have immune, neuroendocrine activities, anti-carcinogenic properties. 25,26 Recently, Diaz et al. 27 have reported the ability of vitamin D3 to enhance the anti-tumor activity of chemotherapeutic drugs by activating apoptosis. However, a causal relationship has remained mostly unclear because most of the studies were small or concentrated on the assessment of vitamin D3 serum levels at the date of HCC occurrence which may result in false statistical associations due to the in uence of impaired liver function on circulating vitamin D3. Another study done by Caputo et al. 28 has reported an inhibitory effect of vitamin D3 on the growth of the human liver cancer cell lines which express functional receptors able to speci cally bind vitamin D3.
Although our results indicate signi cant very low levels of vitamin D3 in HCC patients with more signi cant frequency of its de ciency, yet on multiple stepwise logistic regression analysis, vitamin D3 is not a signi cant independent predictor for HCC. It is not clear whether this re ects a type two statistical error due to the small sample size or this de ciency may re ect a functional synthetic error for vitamin D3 by the HCC diseased liver rather than a risk factor for HCC development.
We also investigated serum Zinc levels in our study groups. Our results show that Zinc level is signi cantly lower in patients with hepatocellular carcinoma compared to levels in liver cirrhosis and CHC patients This nding is in agreement with results reported by other studies. 29,30 HCC malignancy is ZIP14de cient tumor, so with Zinc de ciency, the abolished cytotoxic effects of Zinc on malignant cells cannot be ruled out and low Zinc levels may predispose patients to HCC development. At the same time, the malignant cells derive adaptation mechanisms through which they lower down the concentration of zinc to avoid its cytotoxic effects on them at normal zinc levels 30 . A recent study has reported data about the protective role of long-term Zinc supplementation against development of HCC. 31 However, again, in spite of the signi cant low levels of zinc in our HCC patients, yet on multiple stepwise regression analysis, zinc is not a signi cant predictor for HCC.
Our results indicate a signi cant higher level of serum PTH in patients with HCC compared to its levels in the cirrhotic and the CHC patients. In addition, the serum PTH signi cantly increases in patients with liver cirrhosis in comparison to CHC patients. When multiple stepwise logistic regression analysis was done in relation to CLD rather than cirrhosis, PTH, age, AST and corrected calcium levels were signi cantly associated with HCC, Table (6). PTH is considered a predictor for HCC (OR: 1.11; 95% CI; 1.004-1.02; P < 0.014) at cut off value of > 100 ng/l with sensitivity 84% and speci city 74.29%.
On the other hand, our study shows statistically signi cant higher level of corrected calcium in patients with hepatocellular carcinoma compared to levels in liver cirrhosis and CHC patients. Our data show also that the corrected calcium is one of HCC independent predictive factors (OR: 4.0; 95%CI: 2.3-6.9; p < 0.001). Corrected calcium at a cut-off value of (> 9.63) had sensitivity 82% and speci city 75.71% in the prediction of HCC in relation to chronic liver disease. Hypercalcemia associated with malignant disease is not uncommon. This is ascribed to both bony metastatic lesions and the production of parathyroid hormone-related protein (PTHrP) from the malignant cells. 32 Two case reports have described cases with hypercalcemia that is caused by HCC secreting intact parathyroid hormone (iPTH). 33,34 When effective, Trans Arterial Chemoembolization (TACE) against the HCC results in stepping down the serum iPTH level and calcium to within the normal range, suggesting a correlation between the carcinoma and the iPTH. 35

Conclusion
This study indicates the prevalent de cient levels of vitamin D3 and Zinc in HCV-induced HCC; however, a causal relationship is not established. This is associated with signi cantly higher levels of corrected calcium and PTH serum levels. It is not clear whether these changes re ect a mere signi cant association or otherwise these metabolic parameters may have a risk for development of HCC.

Limitations Of The Study
The small sample size of HCC group of patients and the cross sectional nature of the study are limitations that do not enable us to conclude if vitamin D3 and Zinc are hazard risk factors for development of HCC in HCV infected patients or their de cient levels are just an association. Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Con ict of Interest: All authors declare that they have no competing interest.
Funding: This research has not received any speci c grant from funding agencies in the public, commercial, or not-for-pro t sectors.   -Kruskal Wallis test for non-parametric data (expressed by median) between the three groups followed by Mann Whitney test between each two groups.
-Chi square test and Fisher exact test for qualitative data between groups.
-*: Signi cant difference at P value < 0.05. -Kruskal Wallis test for non-parametric data (expressed by median) between the three groups followed by Mann Whitney test between each two groups -( ¶) One-way ANOVA test for parametric quantitative data between the three groups followed by post hoc Tukey analysis between each two groups -*: Signi cant difference at P value < 0.05 -Kruskal Wallis test for non-parametric data (expressed by median) between the three groups followed by Mann Whitney test between each two groups -( ¶) One-way ANOVA test for parametric quantitative data between the three groups followed by post hoc Tukey analysis between each two groups -*: Signi cant difference at P value < 0.05  Figure 1 ROC curve presenting speci city and sensitivity for corrected calcium, PTH, AST and age