We investigated how the histological scoring system designed by our group (Quaglia et al J Hepatol 2001;34 Supplement 1) could be used to produce an algorithm for the histological assessment of HCC and its putative precursor nodular lesions. This scoring system was applied to 212 nodules (106 HCC median size 15, range 2-75 mm; 74 MRN median size 10, range 5-21 mm; 32 DN median size 12, range 3-24 mm). The data were analysed by multiple correspondence analysis (MCA). MCA redistributed the different types of nodules (i.e. HCC, DN, and MRN as defined by routine histological evaluation), in a "malignancy" scale with all HCC at one end and MRN at the other end. Individual variables were also distributed along a "malignancy scale". At the left side of the scale there were those groups of values present predominantly in HCC, and at the opposite end of the scale those values present predominantly in MRN and DN. These data were used to create an algorithm for the assessment of these nodular lesions. This algorithm produces a final assignment which does not rest on a single histological feature (or sum of features), but on the balance of all components, and categorize an individual nodular lesion as MRN, DN or HCC with a degree of probability. Conclusion. We have designed an algorithm for the histological assessment of HCC and its putative precursors in cirrhotic liver. This algorithm has been integrated and is available in a multidisciplinary, international liver tumour database (http://imt.itc.it/LTD_Website)
A novel algorithm for the histological assessment of hepatocellular carcinoma and its purative precursor nodular lesions in cirrhotic liver
Demichelis, Francesca;Fattore, Gianni;
2002-01-01
Abstract
We investigated how the histological scoring system designed by our group (Quaglia et al J Hepatol 2001;34 Supplement 1) could be used to produce an algorithm for the histological assessment of HCC and its putative precursor nodular lesions. This scoring system was applied to 212 nodules (106 HCC median size 15, range 2-75 mm; 74 MRN median size 10, range 5-21 mm; 32 DN median size 12, range 3-24 mm). The data were analysed by multiple correspondence analysis (MCA). MCA redistributed the different types of nodules (i.e. HCC, DN, and MRN as defined by routine histological evaluation), in a "malignancy" scale with all HCC at one end and MRN at the other end. Individual variables were also distributed along a "malignancy scale". At the left side of the scale there were those groups of values present predominantly in HCC, and at the opposite end of the scale those values present predominantly in MRN and DN. These data were used to create an algorithm for the assessment of these nodular lesions. This algorithm produces a final assignment which does not rest on a single histological feature (or sum of features), but on the balance of all components, and categorize an individual nodular lesion as MRN, DN or HCC with a degree of probability. Conclusion. We have designed an algorithm for the histological assessment of HCC and its putative precursors in cirrhotic liver. This algorithm has been integrated and is available in a multidisciplinary, international liver tumour database (http://imt.itc.it/LTD_Website)I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
