Liver transplantation as a treatment for patients with hepatocellular carcinoma : a study using electronic health care data

Background: Annually over 4,000 new cases of hepatocellular carcinoma (HCC) are being diagnosed in the UK. Overall survival of HCC patients is poor with less than 30% alive at one year. Prognosis is best for patients who receive a liver transplant. The emergence of liver transplantation as a curative option for HCC has resulted it now being the most common indication for liver transplantation. This development has increased the disparity between the number of patients waiting for transplantation and the availability of suitable livers. In response, livers from donation after circulatory death (DCD) are increasingly being used to expand the donor pool, although outcomes have been previously proven to be worse than with livers from donation after brain death (DBD). The overall aim of my research was to use linked large-scale electronic health care data to improve the current evidence base on the best strategy for using liver transplantation in patients with and without HCC. Methods: In this thesis, national datasets were linked at patient level and analysed to investigate outcomes of patients following liver transplantation. The Standard National Liver Transplant Registry (SNLTR) was used to identify the donor and recipient characteristics of patients with and without HCC and the Hospital Episode Statistics (HES) dataset was used to capture information on clinico-sociodemographic characteristics that included deprivation status, co-morbidity, treatments and length of hospital stay. Throughout the thesis, posttransplant mortality and graft failure for HCC and non-HCC patients were analysed in distinct posttransplantation time-periods ('epochs'). Stepwise case mix adjustment for recipient and then donor factors was used to ascertain determinants of post-transplant outcomes. Multiple imputation methods were used to handle missing data items. The analyses in this thesis were split into three related clinical areas; DCD livers, pre-transplant performance status (PS), and the administration of transarterial chemoembolization (TACE) on the transplant waiting list. Results: In the last two decades, mortality after liver transplantation in the UK has more than halved for HCC patients despite the increased and preferential use of DCD livers in these patients. However, one in four HCC recipients still die within five years compared to only one in six non-HCC patients. Improvements in survival for HCC and non-HCC recipients has been driven by significant improvements both in short-term mortality and in improvements in outcomes of patients who receive a DCD liver. In fact, mortality in recipients of a DCD liver is now comparable to those receiving a DBD liver. In non-HCC recipients, impaired PS at the time of transplantation was found to be associated with posttransplant mortality whilst in both HCC and non-HCC recipients poorer PS was also associated with an increased incidence of major post-transplant complications and length of hospital stay. However, the impact of PS on each of the measured post-transplant outcomes was limited only to the early post-transplantation timeperiod. In the final analysis, 40% of HCC recipients in the UK received at least one treatment of TACE prior to their transplant. However, there was no evidence that TACE increased the risk of post-operative complications or affected the risk of post-transplant mortality or graft failure. These results did not depend on the number of TACE treatments, the type of donor organ (DBD or DCD), or the time-period after transplantation. Conclusions: Improvements in peri-operative care and in the use of DCD livers has led to significant improvements in posttransplant mortality for HCC and non-HCC patients. Countries who experience high waiting list mortalities should now consider increasing their utilisation of DCD livers. Measurements of PS scores can improve the ability to predict post-transplant survival, post-transplant complications and length of hospital stay, which may contribute to the assessment of the suitability of HCC and non-HCC patients for transplantation. The use of TACE in HCC patients prior to transplantation does not increase the risk of post-operative complications nor does it improve post-transplant mortality. The benefit of TACE is therefore restricted to its impact on modulating tumour growth before transplantation.