Establishment and analysis of the prediction model for cervical squamous cell carcinoma.

OBJECTIVE

This study aimed to construct a prediction model for cervical squamous cell carcinoma and evaluate its accuracy in diagnosing cervical squamous cell carcinoma.

PATIENTS AND

METHODS

Fifty patients with initially histopathologically confirmed cervical squamous cell carcinoma and 150 patients with initially histopathologically confirmed cervical intraepithelial neoplasia (CIN) were enrolled. The high-risk human papillomavirus (HR-HPV) infection, human telomerase mRNA component (hTERC) gene and cell-myc (c-myc) gene amplification, and minichromosome maintenance protein 5 (MCM5) protein expression were detected. The indicators related to cervical cancer were screened. The regression model was established to predict cervical squamous cell carcinoma with backward logistic stepwise regression method, and the accuracy of the model was evaluated.

RESULTS

Histograms for HR-HPV infection and viral load, hTERC and c-myc gene amplification, and MCM5 protein expression were constructed. There was a linear relationship between hTERC (X1), HR-HPV viral load (X2), MCM5 (X5) and the regression equation. Also, hTERC (X1), HR-HPV viral load (X2) and MCM5 (X5) were correlated with cervical squamous cell carcinoma. The regression model Logit (p) = -66.283 + 0.042 X1 + 0.061 X2 + 0.052 X5 was established. The model-fitting effect and prediction accuracy were evaluated, HL test p = 1 (p > 0.05). The model fitting effect was good, Cox-Sn ell R2 was 0.643 and Nagelkerke R2 was 0.958. The high accuracy of the model was 98.5%.

CONCLUSIONS

The fitting-effect of the regression model established by hTERC gene expression, HR-HPV viral load and MCM5 protein was good. The prediction accuracy of the model for cervical squamous cell carcinoma was high. The combined test of hTERC gene amplification, HR-HPV viral load and MCM5 protein could be used to predict and evaluate cervical squamous cell carcinoma.