A retrospective analysis of clinical outcomes among patients with infiltrative hepatocellular carcinoma (iHCC): A single institution study.

Abstract

e15667 Background: iHCC is an aggressive disease with poorly characterized treatment outcomes. We sought to describe clinical and radiographic features of iHCC, describe overall survival (OS), and compare outcomes of sorafenib (SOR) use with a cohort of non-infiltrative hepatocellular carcinoma (niHCC) patients. Methods: Radiology reports from the University of Washington with an Li-RADS (LR) assignment, 2013-2017, were reviewed (5037 reports, 1954 patients (pts)) for terms indicative of iHCC. Results: : 102 cases were identified, with 43% arising from known niHCC. iHCC size is difficult to measure but when reported, median was 8cm (interquartile range 5-10cm). Other radiographic features: vascular invasion 72%, R lobe only 62%. Clinical data were obtained via medical record review. The cohort was primarily male (83%), White (65%), age > 55 (88%), HCV (74.5%), heavy alcohol use (67%). At diagnosis: BCLC stage C (76%), ALBI grade 2 (62%) and 3 (21%). Median OS after diagnosis was 5.7 months (m). 38 (37%) pts were too ill to receive therapy. OS for de novo iHCC and for iHCC arising from niHCC did not differ (7.5m vs 5.1m, p = NS). All niHCC pts who received SOR in any line of therapy were identified among pts with a LR 5 lesion in at least one of the aforementioned radiology reports, 2013-2017. Pts who received SOR on trial and/or combined with another agent were excluded. Clinical data for the 83 niHCC pts were abstracted and compared to the 25 iHCC pts who received SOR, any line of therapy. Based on χ2 analyses, the 2 groups did not differ by age, sex, race, BCLC stage at SOR initiation, ECOG status at SOR initiation, HCC etiology, ALBI grade at SOR initiation, or previous locoregional therapy. OS after initiation of SOR was 14.1m for niHCC, 10.5m for iHCC, p = NS. PFS after SOR was 3.7m for niHCC, 7m for iHCC, p = NS. AFP at the time of SOR initiation was higher in iHCC than niHCC (median 1331 vs 48, respectively, p = 0.004), with median %AFP change after 30 days -13% in iHCC and +6% in niHCC, p = NS. Decrease in AFP after 30 days of therapy was associated with improved OS among niHCC (21m versus 7m, p = 0.001) but not among iHCC pts (p = NS). Conclusions: iHCC is a poorly understood subset of HCC with underlying biology not well defined. Compared to niHCC, clinical outcomes with sorafenib do not appear to differ in this cohort. Future characterization of outcomes with Y-90 and immune checkpoint inhibitors is warranted.