Page 200 - EJMO-9-2

Eurasian Journal of
Medicine and Oncology Vitamin D and HNC: Causal association

the within-family GWAS consortium, which includes IVW approach was applied in the MVMR analysis for
99,996 and 83,626 participants, respectively. The GWAS causal effect estimation.
24
datasets analyzed in this study exclusively encompassed In this study, statistical analyses were performed using
participants of European ancestry. the “TwoSampleMR” and “MRPRESSO” packages in R
2.3. IV selection 4.3.2, with p < 0.05 indicating statistical significance.
To ensure the accuracy and reliability of the MR analysis, 3. Results
this study employed a stringent screening process to select
suitable IVs. First, SNPs exhibiting strong associations 3.1. Genetic instruments
with exposure factors (p < 5 × 10 ) were extracted from This study involved screening of eight SNPs as IVs to
-8
the pooled GWAS data. To avoid potential bias that investigate the potential causality of Vitamin D for
may result from linkage disequilibrium, r < 0.01 and HNC (Table 1). Among them, two SNPs (rs11940196
2
kb = 10000 were set. Second, the screened SNPs were and rs10745742) were missing in the genetic association
further examined in the Phenoscanner database to rule database for HNC. In addition, one SNP (rs8018720) was
out known confounders linked with the IVs. Finally, the excluded due to the presence of palindromic sequences.
efficacy of the IVs was evaluated using the F-statistic, In addition, the F-statistic of all SNPs included in the MR
and only SNPs with F > 10 were retained to address the analysis was >10, showing a low risk of weak IV bias.
problem of weak IVs. Simultaneously, palindromic
25
SNPs with moderate allele frequencies were eliminated 3.2. Two-sample MR analysis
to enhance robustness. The IVW method, used as the main analytical approach,
2.4. MR analysis suggested a possible causal relationship between Vitamin D
levels and HNC risk, indicating that higher Vitamin D levels
The primary methodology employed was the inverse may decrease the HNC risk (odds ratio [OR] = 0.9958, 95%
variance weighted (IVW) approach to assess the confidence interval [CI] = 0.9934 – 0.9983, p = 0.0007). The
possible causality of Vitamin D for HNC risk. The IVW results from the weighted median analysis further support
method is based on the premise that all IVs are valid and the conclusion that Vitamin D lowers the HNC risk.
estimates the causal effect by calculating the Wald ratio Although the MR-Egger, weighted mode, and simple model
for each SNP and combining these values. This method analyses did not yield significant results, the direction of
provides the advantage of a more stable estimation their results was consistent with that of the IVW analysis.
of causality. In addition, to evaluate the robustness The results are displayed in Figures 2-4.
26
of our findings, MR-Egger, weighted median, simple
mode, and weighted mode were used as supplementary Cochran’s Q test was utilized to evaluate heterogeneity,
analytical methods. revealing the lack of notable heterogeneity in both
MR-Egger (Cochran Q = 4.5369, p = 0.6044) and IVW
To ensure the reliability of the MR analysis, several (Cochran Q = 6.0413, p = 0.5349) analyses. The MR-Egger
sensitivity analyses were performed. Cochran’s Q test intercept indicated the lack of significant evidence of
was performed to evaluate the heterogeneity of the IVs. horizontal pleiotropy (Egger intercept = −0.00012,
In addition, the MR-Egger intercept method assesses p = 0.2659), and the MR-PRESSO analysis did not detect
horizontal pleiotropy, ascertaining its presence or absence any substantial outliers (p = 0.6050). In addition, the leave-
through the calculated intercept term. The MR pleiotropy one-out analysis, which involved sequentially excluding
27
residual sum and outlier (MR-PRESSO) analysis was each SNP to evaluate the effect of the remaining SNPs,
employed to identify and rectify outliers in the MR revealed that no single SNP significantly influence the
analysis caused by pleiotropy. Moreover, the leave-one- overall results (Figure 5).
28
out analysis was employed to assess whether individual SN
Ps biased the total causal effect. 29 3.3. Multivariate MR analysis
To account for the possible effect of ever-smoking and After adjustment for ever-smoking alone, a significant
alcohol consumption on results and improve the precision negative association between Vitamin D levels and HNC
of the causal effect, this study incorporated these two risk was observed (OR = 0.9959, 95% CI = 0.9937 – 0.9982,
variables into the MVMR analysis, as they are recognized p = 0.0004). Similarly, this causality was confirmed when
risk factors for HNC. This strategy aimed to minimize the only alcohol consumption was considered (OR = 0. 9952,
potential effect of lifestyle factors on causality estimation, 95% CI = 0.9926 – 0.9979, p = 0.0004). When both ever-
thereby enabling a more accurate evaluation of the direct smoking and alcohol consumption were considered
causal relationship between Vitamin D and HNC. The (OR = 0.9956, 95% CI = 0.9933 – 0.9979, p = 0.0002),

Volume 9 Issue 2 (2025) 192 doi: 10.36922/ejmo.7099