that are associated with it. Therefore, a study with a larger number of patients is required to further analyze the factors, including, but not limited to, DNA methylation and DNA hydroxymethylation, that are associated with survival in terms of 1-year, 3-year, and 5-year rates.
REPLY:We thank Li et al. (1) for their interest in the study and for the opportunity to further discuss our findings to gain deeper insights.The first point raised by Li et al. addresses the issue of the influence of dietary, environmental, and psychological factors on DNA methylation and suggests the need to take these factors into account when determining survival rates, considering the likelihood that patients change their habits after liver surgery for primary liver cancer. It is certainly an appropriate observation, although one should recall that the relative short follow-up period for such cancer patients, considering also the high mortality rate, may not be sufficient to observe an effect of diet or other environmental factors on DNA methylation, as observed in other studies in which the objective was to determine the risk of cancer development. In fact, the effect of carcinogenesis on epigenetic phenomena is much stronger than the epigenetic effects of lifestyle factors. It is well known that global DNA methylation is strictly regulated in a narrow range under normal physiologic conditions. Yet, ongoing analyses on nutritional and other environmental factors will certainly add essential information in this regard.As for the other point of discussion, i.e., the evaluation of methylation on DNA from peripheral blood mononuclear cells, it is certainly considered more achievable than testing liver tissue as a possible epigenetic prognostic tool and it has been shown to give encouraging results in a number of studies thus far. (2,3) Regarding the remark that biological samples differ in neoplastic cell enrichment and that DNA methylation is tissue-specific, this is surely correct. As for testing DNA methylation from blood samples, it should be pointed out that, in the present study, we measured methylation from peripheral blood mononuclear cell DNA and not from DNA obtained from circulating blood.The question, appropriately brought up by Li et al.,(1) on how a blood-positive screening assay may be helpful to detect the precise tissue origin of cancer disease is indeed open and challenging. Global DNA methylation can represent the overall epigenetic effect. However, the tissue-specific DNA methylation profile is widely variable among individuals and cancer types. Thus, theoretically a tissue-specific DNA methylation profile is ideal but practically difficult to use for this purpose. In the present study, however, we did not propose the evaluation of peripheral blood mononuclear cell DNA methylation as a screening test, but we hypothesized it to be a possible prognostic tool for a malignancy of known origin. As for the potential suitability of measuring global DNA methylation and hydroxymethylation for diagnostic purposes, although C...