| Table 3. Results of multiple linear regression analysis of biochemical and hormonal markers on P53 gene mutation. | |||||
| Predictor Variable | Coefficient | Std. Error | t-value | p-value | 95% Confidence Interval |
|---|---|---|---|---|---|
| Superoxide Dismutase (U/mL) | -0.0233 | 0.0226 | -1.03 | 0.305 | -0.0680 to 0.0215 |
| Testosterone (ng/dL) | -0.0008 | 0.0005 | -1.62 | 0.108 | -0.00169 to 0.00017 |
| Follicle Stim. Hormone (FSH, mIU/mL) | 0.0087 | 0.0064 | 1.35 | 0.180 | -0.0040 to 0.0214 |
| Estradiol (pg/mL) | -0.0044 | 0.0046 | -0.95 | 0.342 | -0.0135 to 0.0047 |
| Prostate Specific Antigen (ng/mL) | 0.1172 | 0.0573 | 2.04 | 0.043 | 0.0038 to 0.2306 |
| Interleukin-6 (pg/mL) | 0.0346 | 0.0126 | 2.75 | 0.007 | 0.0097 to 0.0594 |
| Constant | 2.3510 | 0.6927 | 3.39 | 0.001 | 0.9806 to 3.7215 |
Multivariate linear regression analysis was performed to assess the relationship between biochemical and hormonal predictors and the dependent variable. Coefficients represent the change in the outcome variable for each unit increase in the predictor. Statistically significant associations were observed for prostate-specific antigen (PSA; p=0.043) and interleukin-6 (IL-6; p=0.007), with 95% confidence intervals not crossing zero. Other variables, including superoxide dismutase (SOD), testosterone, follicle-stimulating hormone (FSH), and estradiol, did not show significant predictive value (p>0.05), as shown in