P115 Radiological-histological discordance may predict outcomes in patients transplanted for HCC

<h3>Background and Aims</h3> Hepatocellular carcinoma (HCC) remains a leading cause of liver-related morbidity and mortality. Although liver transplantation (LT) is a radical therapy that improves survival, there remains a significant risk of HCC recurrence post-LT. Contrast-enhanced (CE) imaging plays an important role in diagnosis and staging of patients with HCC and radiological-histological discordance may adversely the affect outcomes of these patients. Our study aimed to determine radiological-histological discordance post-LT and its association with cancer recurrence. <h3>Method</h3> All patients who underwent a LT for HCC at King’s College Hospital between 2016 and 2023 were included in this retrospective study. Demographic, clinical, laboratory and radiological data were collected. Pearson correlation was applied for the number of nodules and the size of the largest nodule pre and post LT. Univariable and multivariable logistic regression models were performed for cancer recurrence. <h3>Results</h3> One hundred and ninety patients were included, 157 (82.3%) were male, with a median age of 60 (IQR 55.8 - 65) years. The most common aetiology was ARLD (39.5%), followed by HCV and MASLD (32.3%). The median time from HCC diagnosis to LT was 353.5 (IQR 10–3258) days while median waitlist time was 106.5 (IQR 46.8–207.5) days. CE CT was the most common imaging modality (80.2%) and the median time from last scan to LT was 46.4 (IQR 21–75) days. On pre-transplant imaging, the mean number of lesions identified was 1.54 (SD 0.788) compared with 3.29 (SD 3.349) nodules on explant (correlation coefficient 0.404; p &lt; 0.01). The mean size of the largest lesion pre-LT was 28.60 (SD 13.08) mm compared with 26.34 (SD 10.60) mm on explant (correlation coefficient 0.549; p &lt; 0.01). Based on explant pathology, 68.1% of patients met standard UK criteria for LT. HCC recurred in 15 LT recipients (7.9%). The mean overall survival was 74 (95% CI 68.8–79.9) months: 75.9 (95% CI 70.7 – 81) months for non-HCC recurrence vs 48.6 (95% CI 31.8 to 65.3) months for HCC recurrence; Log rank &lt;0.001). In the multivariate analysis, HCC recurrence was higher, although not statistically significant, in those who had radiological-histological discordance (OR 2.65 95% CI 0.915–7.688, p= 0.072). <h3>Conclusions</h3> Radiological-histopathological concordance remains poor in our cohort, with the HCC burden being underestimated in one third of patients. HCC recurrence is likely to occur in patients who had discordance which highlights the need a more effective method of defining the best modality for cancer diagnosis and staging.