Reply: Regarding the Predictive Role of CXCL16 in Liver Inflammation in Chronic Hepatitis B Patients

Abstract

We have received your letter and thank you very much for your hard work. We are very pleased that the reader has carefully read the article, and we also greatly appreciate the valuable feedback provided by Zhang. I provide the following response to Zhang's question.

Firstly, we conducted an univariate logistic regression analysis to preliminarily identify the associated factors with severe liver inflammation. Considering the close relationship between these parameters and liver inflammation in univariate analysis, factors with p < 0.05 were adjusted for further multivariate analysis. However, we totally agree with Zhang's comments that a larger sample size is more appropriate for this analysis, but we encountered great difficulties in collecting serum samples. Due to the fact that liver puncture is an invasive examination and our inclusion criteria are strict, therefore it is difficult to obtain samples that meet the standards. I hope to gain the reader's understanding regarding this matter.

Secondly, how I adjust which factors to enter into multivariate logistic regression analysis and choose which factors to build the model is a key question. I would like to discuss my viewpoint and hope to receive Zhang's support. (1) This factor must have statistical significance in univariate logistic regression analysis. (2) This factor is of great significance in practical clinical work, such as alanine aminotransferase (ALT), platelet (PLT) and albumin (ALB), which can evaluate the progression of liver diseases from different aspects. (3) We have considered the correlation between factors. If there is a strong correlation between factors, it may affect the stability of the model. (4) We have done a lot of statistical work, arranging and combining these factors separately for statistical analysis. We compared the final results and came to a conclusion. We found that CXCL16, ALT, PLT and ALB were statistically significant in the multivariate logistic regression analysis, and the predictive model was the most representative. Due to the large and complex statistical information, we did not present all statistical results in the article. (5) Our mouse experiments confirmed a strong correlation between CXCL16 and liver inflammation. I apologize for the confusion caused to Zhang and other readers.

Thirdly, Zhang raised the issue of the impact of antiviral drugs on patients with chronic hepatitis B, which is indeed worth considering. However, we have excluded patients who underwent antiviral therapy at enrollment, as clearly indicated in exclusion criteria.

In future research, we will gradually expand the sample size and consider more factors that affect the outcomes of patients with chronic hepatitis B. We look forward to collaborating with Zhang and other readers. We sincerely thank you again for your valuable letter.

Sincerely yours,

Chao Wu, MD, PhD.

All authors reviewed this manuscript and agreed to submit this manuscript.

The authors have nothing to report.

The authors declare no conflicts of interest.