ABSTRACT
BACKGROUND: Nuclear changes are typical in the carcinogenesis of hepatocellular carcinoma (HCC). Morphometry and chromatin texture analysis are quantitative methods for their quantification. In this study, we analyzed nuclear morphometry and chromatin texture parameters in samples of hepatocellular carcinoma from liver transplant patients and their associations with clinicopathologic variables. METHODS: Samples of HCC and adjacent tissue from 34 individuals were collected in tissue microarray blocks. Stained slides were microphotographed using an optical microscope and nuclear parameters analyzed in ImageJ (FracLac plug-in). ROC curve analysis was used to find accurate cut-offs for differentiation of neoplastic and non-neoplastic cells. The inter-rater agreement was also evaluated. RESULTS: Nuclear morphometric and textural differences were observed between the samples of HCC and adjacent tissue of liver transplant patients. Lower mean gray value (p = 0.034) and Feret diameter (p = 0.024) were associated with higher Model for End-Stage Liver Disease (MELD) scores. Nuclei with larger area (p = 0.014) and larger Feret diameter (p = 0.035) were associated with lower survival. Lower aspect ratio was associated with HCC recurrence after the transplant (p = 0.048). The cut-off of 1.13 µm (p = < 0.001) for aspect ratio and cut-off of 21.15 µm (p = 0.038) for perimeter were established for the differentiation of neoplastic and non-neoplastic cells. The morphometric analysis was reproducible to area, circularity, Feret diameter, mean gray value and aspect ratio between observers (p = < 0.001). CONCLUSIONS: Nuclear morphometric differences between the HCC and the adjacent tissue samples were associated with prognostic variables (MELD scores, recurrence and survival) and may predict liver transplant patients' outcomes.
