Abstract
BackgroundLiver cancer, of which hepatocellular carcinoma (HCC) is by far the most common type, is the second most deadly cancer (746,000 deaths in 2012). Currently, the only curative treatment for HCC is surgery to remove the malignancy (resection) or to remove the entire diseased liver followed by transplantation of healthy liver tissue. Given the shortage of healthy livers, it is crucial to provide transplants to patients that have the best chance of long-term survival. Currently, transplantation is determined via the Milan criteria—patients within Milan (single tumor < 5 cm or 2–3 tumors < 3 cm with no extrahepatic spread nor intrahepatic vascular invasion) are typically eligible for transplantation. However, combining microRNA expression profiling with the Milan criteria can improve prediction of recurrence.HCC often presents with multiple distinct tumor foci arising from local spread of a primary tumor or from the oncogenic predisposition of the diseased liver. Substantial genomic heterogeneity between tumor foci within a single patient has been reported; therefore, biomarker development must account for the possibility of highly heterogeneous genomic profiles from the same individual.MethodsMicroRNA profiling was performed on 180 HCC tumor samples from 89 patients who underwent liver transplantation at the University of Rochester Medical Center. The primary outcome was recurrence-free survival time, and patients were observed for 3 years post-transplantation.ResultsMicroRNA expression profiles were used to develop a biomarker that distinguishes HCC patients at greater risk of recurrence post-transplantation. Unsupervised clustering uncovered two distinct subgroups with vast differences in standard transplantation selection criteria and recurrence-free survival times. These subgroups were subsequently used to identify microRNAs strongly associated with HCC recurrence. Our results show that reduced expression of five specific microRNAs is significantly associated with HCC recurrence post-transplantation.ConclusionsMicroRNA profiling of distinct tumor foci, coupled with methods that address within-subject tumor heterogeneity, has the potential to significantly improve prediction of HCC recurrence post-transplantation. The development of a clinically applicable HCC biomarker would inform treatment options for patients and contribute to liver transplant selection criteria for practitioners.Electronic supplementary materialThe online version of this article (doi:10.1186/s12920-016-0179-4) contains supplementary material, which is available to authorized users.
Highlights
Liver cancer, of which hepatocellular carcinoma (HCC) is by far the most common type, is the second most deadly cancer (746,000 deaths in 2012)
Visualization of microRNA Expression Reveals Two Subgroups The data consist of 847 microRNAs measured across 176 samples. t-Distributed Stochastic Neighbor Embedding (t-SNE) facilitates visualization of high dimensional data in a low-dimensional space [20]
A Cox proportional hazards model shows a statistically significant difference between the Within Milan group and the Outside Milan & Cluster 1 group as well as the Outside Milan & Mixed group; there was no discernable difference between the Within Milan group and the Outside Milan & Cluster 2 group (Additional file 1: Table S6). This latter group, Outside Milan & Cluster 2, represents a group of patients who would be deemed unfit for liver transplantation based on the Milan criteria but would appear to have a good chance of recurrence free survival if they received a liver transplant. It appears that there is a subset of patients who would benefit from microRNA profiling when used with the Milan criteria to determine transplantation
Summary
Of which hepatocellular carcinoma (HCC) is by far the most common type, is the second most deadly cancer (746,000 deaths in 2012). The only curative treatment for HCC is surgery to remove the malignancy (resection) or to remove the entire diseased liver followed by transplantation of healthy liver tissue. Combining microRNA expression profiling with the Milan criteria can improve prediction of recurrence. The only curative treatment is surgery: either tumor resection or liver transplantation Patients undergoing these treatments still have a high risk of recurrence. Both resection and transplantation result in 80 % 5-year patient survival rates in appropriately selected patients. Even though the 5-year HCC recurrence rate after transplantation seems acceptably low at 15–20 %, improved selection criteria would further optimize outcomes and more efficiently use the precious resource of donor liver tissue. While the Milan criteria alone perform reasonably well, combining the Milan criteria with a microRNA biomarker has been shown to improve prediction of recurrence [6]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
