Population-Based Factors Associated With Death After Liver Resection For Hepatocellular Carcinoma In Queensland, Australia

Paul CLARK, QIMR Berghofer Medical Research Institute, Australia
O'ROURKE T. 4 , FAWCETT J. 4 , STUART K. 2 , MACDONALD G. 2 , CRAWFORD D. 3 , LEGGETT B. 3,5 , WHITEMAN D. 1 , BAADE P. 6 , VALERY P. 1

1 Population Health, QIMR Berghofer Medical Research Institute, Brisbane, Australia
2 Department of Gastroenterology, Princess Alexandra Hospital, Brisbane, Australia
3 School of Medicine, University of Queensland, Brisbane, Australia
4 Department of Surgery, Princess Alexandra Hospital, Brisbane, Australia
5 Department of Gastroenterology, Royal Brisbane and Women's Hospital, Brisbane, Australia
6 Cancer Council Queensland, Brisbane, Australia

Purpose: Accurately identifying patients with poorest prognosis from hepatocellular carcinoma (HCC) may prevent futile, quality-of-life impairing surgery. Little is known of the social determinants of poor survival after liver resection for HCC. Using a population-based-linked cohort of HCC patients we examined factors associated with early death after resection.
Methods: Cancer registry (1584 HCC cases), hospital admission and viral hepatitis notification data were used.
Results: Bivariate analysis showed that the proportion of patients dying within 12 months of diagnosis reduced over time (66% (1996-2000), 61% (2001-2005), 53% (2006-2011), P<0.001). Poorer HCC survival was seen with more social disadvantage (P<0.001) and living more remotely (P<0.001). Patients with hepatitis B and C infection had better survival than non-infected patients (P<0.001). While there was no significant evidence that Indigenous people had poorer HCC survival than non-Indigenous Australians (p=0.225), they had poorer overall survival (all deaths) (p=0.012).  The 236 patients who underwent liver resection had better survival than those who did not (median HCC survival 1537 vs 173 days; P<0.001). 50/236 (21%) patients who had resection died within 12 months of diagnosis (42 HCC deaths/8 non-cancer deaths). On multivariate analysis, poorer HCC survival after resection was associated with social disadvantage (HR=2.5 95%CI 1.15-5.53 most disadvantage vs most affluent), earlier diagnosis period (2001-2005 vs 2006-2011, HR=1.6 95%CI 1.02-2.61), and male sex (HR=2.0 95%CI 1.12-3.44). Remoteness, hepatitis B diagnosis, Indigenous status and age were not associated with HCC-specific or overall survival.
Conclusions: Social determinants are important in early HCC death. Early HCC death (<12 months) after HCC diagnosis and survival after resection have improved over time. Reduced early HCC death-rates may reflect improved screening. With better recognition and assessment of underlying liver disease in patients with HCC, better patient selection, operative techniques and post-operative care may underlie improved post-resection survival. 
Funding: NHMRC Fellowships (1083090/1052622/105534/1058522).