Tissue biomarkers as predictors of outcome and selection of transplant candidates in hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is a common cause of cancer deaths worldwide and has a rising annual incidence. Liver transplantation (LT) is an accepted curative treatment for patients with tumors which satisfy Milan criteria (single tumor ≤ 5 cm or up to three tumors, each ≤ 3 cm, no macrovascular invasion). These criteria predict 5 year overall survival rates of 70% following liver transplantation¹. Since the introduction of the Milan criteria, subsequent studies have explored expanding transplant recipient selection to include individuals with tumors that exceed Milan criteria². A recent study demonstrated acceptable 5 year overall survival rates (71.2%) in patients transplanted with tumors beyond Milan criteria that satisfied the up-to-seven rule (seven being the sum of size of the largest tumor (in cm) and the number of tumors) in the absence of microvascular invasion³. This approach represents the best data available for understanding tumor behavior after LT, but still is based upon pathological data. Size/tumor number variables are unable of defining subclasses of patients with better biology and outcome, and thus it is expected that biomarkers will represent a major step forward in this setting during the next decade.

Numerous molecular pathways involved in the pathogenesis of HCC have been identified, including activation of pathways involved in angiogenesis (VEGF), cell proliferation and survival (EGF, IGF, HGF/Met), and cell differentiation and proliferation (Wnt/β catenin, Hedgehog signaling). Activation of VEGF ⁴, AKT ⁵ and MET ⁶ has been shown to correlate with aggressive phenotype and poor prognosis following liver resection. Similarly, several gene signatures have been involved in prediction of outcome in HCC⁷. Gene expression profiling from formalin-fixed, paraffin embedded tissue samples derived from HCC resection specimens has been described and validated in the prediction of survival outcomes in patients following resection for HCC ⁸. This profiling technique offers the ability to perform retrospective studies using stored histologic specimens. In addition, it offers a potentially practical clinical application through the ability to perform gene profiling using common, formalin-fixed biopsy specimens rather than frozen tissue.

This article summarizes 3 areas in which molecular tissue biomarkers should be considered in the management of HCC with liver transplantation. Serum markers, such as AFP, Ang2 or DCP are not analized herein.

1. Role of tissue biomarkers in the diagnosis of HCC.

2. Can gene signatures or tissue biomarkers predict prognosis, and thus aid in the extension of Milan criteria for HCC?.

3. How can biomarkers predict response to therapies.

1 Role of biomarkers in the diagnostic of HCC

Diagnosis of HCC is based upon pathology or non-invasive criteria⁹. Pathological differentiation of dysplastic nodules- particularly high grade- and very early HCC is sometimes difficult, especially on a cirrhotic background. Few studies have tested the accuracy of molecular diagnosis of early HCC in this setting. For instance, gene-signatures allowed molecular demarcation between low grade dysplastic nodule, high grade dysplastic nodule and early HCC both in Asian¹⁰ and in Western patients¹¹. More specifically, a 3-genes signature (including glypican 3, lyve-1 and survivin) has been reported to discriminate early HCC < 2cm from dysplastic nodules with an accuracy around 90%¹². Nonetheless this signature has not yet been externally validated. More recently, a immunohistochemistry study testing the expression of glypican 3, HSP70 and glutamine synthetase appeared to be a useful tool to detect well-differentiated HCC in biopsy¹³, and is currently being considered in guidelines of management of HCC⁹.

2 Role of biomarkers in predicting prognosis

Patients with HCC developed on cirrhosis treated by resection have a high rate of recurrence (around 70% at 5 years) ^2,14. In HCC, the molecular assessment of prognosis could determine the type of patients that will benefit of adjuvant therapy after resection of radiofrequency, two curative treatments with high risk of relapse. Moreover, it could refine the group of patients that should be transplanted for HCC beyond the Milan criteria. It is still an area of discussion whether the risk of tumor seeding counterbalances the advantages of obtaining tissue-based molecular profiling. In a recent meta-analysis, the risk of tumour seeding after liver biopsy is 2.7% with a median time interval between biopsy and seeding of 17 months ¹⁵. These data includes also large tumors, and thus it is expected that the risk of complications in early small tumors should be significantly lower, and thus acceptable.

Biomarkers predicting prognosis or response to therapy are crucial in modern oncology. Novel prognostic biomarkers enabling tumor classification and/or monitoring of disease state could advance efforts toward realizing the potential of personalized medicine in cancer.¹⁶ Aside from reports on alpha-fetoprotein (AFP) level in relation to outcome,^17–19 recent studies have correlated various types of markers—such as gene expression, microRNAs (miRNAs), and methylation changes—with survival in HCC patients, a topic reviewed elsewhere (see Table 1)²⁰. Among them, five markers or signatures—epithelial cell adhesion molecule (EpCAM), a hepatic stem cell marker in tumor tissue ^21,22, G3-proliferation subclass²³, expression status of the miR-26 miRNA precursor²⁴; and two gene prognostic signature in non-tumor hepatic tissue^8,25—have emerged as the more consistent ones. Finally, both VEGF and Ang 2 have shown to have independent prognostic value in a large cohort of patients with advanced HCC²⁶. Although these results support the possibility of utilizing these genetic and molecular markers as prognostic biomarkers in patients with HCC, they require external validation before they can be included in staging systems and/or incorporated into guidelines of clinical management. Fraction of allelic imbalance (FAL) measuring chromosomic instability has been associated with outcome in HCC and with recurrence after liver transplantation, an observation that requires attention in future studies. ^27,28. Similarly, data from CD90+ circulating cells may provide a tractable supply of tissue for molecular characterization, but it is still under investigation ²⁹.

Table 1.

Main mRNA, miRNA-based, epigenetic and structural alterations, which prognostic impact in HCC patients needs to be tested or confirmed (Modified for Villanueva et al CCR 2010¹⁸).

Molecular alteration	Clinical significance	REMARK recommendations*	Status †
mRNA based (gene signatures)¶
Poor-survival signature	Poor survival	OK	EV
EpCAM signature	Poor survival	OK	EV
Venous metastasis signature	Hepatic metastasis	OK	EV
Class A / Hepatoblast signature	Poor survival	OK	IV, EV
G3 subclass	Poor survival	-	IV, EV
AFP, Ang2,	Poor survival	Ok (unclear cut-off)	EV
miRNA-based
Down-regulation miR-26a	Poor survival	OK	EV
20-miRNA signature	Venous metastasis, overall survival	OK	EV
Down-regulation miR-122	Poor survival	-	IV, EV
Down-regulation Let-7 members	Early recurrence	-	IV, EV
Up-regulation miR-125a	Better survival	-	IV, EV
19-miRNA signature	Poor survival	-	IV, EV
Epigenetics
Genome wide hypomethylation	Tumor Progression, Survival	-	IV, EV
Hypermethilation of E-cadherin or GSTP1	Poor survival	-	IV, EV
Chormosomal Instability (CI)
Allelic imbalance (FAL)/CI	Recurrence/survival	OK	EV

^†Current status in terms of clinical implementation (T: need further preliminary prognostic evaluation, IV: lacks internal validation, EV: lacks external validation)

^¶Molecular classifications (mRNA-based) with prognostic impact are thoroughly discussed elsewhere^{5, 6, 20}

In an era of limited organ availability, better predictors of HCC recurrence are needed in selecting appropriate liver transplant candidates whose tumors exceed Milan criteria. Identification of a subgroup of patients who are beyond Milan criteria but have favorably low risk for recurrence following transplant offers potential cure to those who would otherwise be excluded based on current organ allocation policies. Whether any of the biomarkers or gene-signatures previously described are able to identify those patients with better biological profile needs to be elucidated in molecular studies addressing this point. Only a small study has addressed this question in a specific manner, and found that chromosomic instability (measured by fraction of allelic imbalance) was independently predicting patients with low-risk of recurrence beyond Milan criteria²⁷. Similarly, preliminary reports describing surrogates of microvascular invasion- the main predictor of HCC recurrence after LT- required independent validation in the setting of transplantation³⁰

3. Role of biomarkers in predicting response to molecular targeted therapies

Biomarkers of treatment response still represent a rarity in oncology; only a few have made their way into routine clinical use. Those that are well-defined are believed to characterize an oncogenic addiction loop—the proposed mechanism by which a tumor cell becomes largely reliant on a single activated oncogene³¹—and define particular tumor subtypes that respond to specific molecular targeted therapies. Examples of oncogenic addiction include amplification of Her2 in breast cancer responding to trastuzumab³², mutations in EGFR defining responders to erlotinib in non–small cell lung cancer,³³ and c-KIT–positive gastrointestinal stromal tumors responding to the multikinase inhibitor imatinib.³⁴ In addition, wild type KRAS has recently emerged as a marker of response to cetuximab and panitumumab in colorectal cancer, albeit via an entirely different mechanism involving the downstream regulation of EGFR signaling³⁵. Moreover, a new step in personalized medicine has been achieved recently with the development of a specific inhibitor of mutated V600E BRAF, which has shown impressive clinical efficiency with few adverse events in a recent phase 2 study in melanoma³⁶. So, in the future, mapping the genetic alterations of the tumor, before the treatment or after treatment failure, will improve the clinical care of cancer ³⁷.

The case of biomarkers in HCC is somewhat more complex because HCC is a very heterogeneous disease for which oncogenic addiction loops have yet to be characterized. Initial approaches to define a molecular classification have not yet been linked to specific treatment responses^38,39. So far, only one small molecule- sorafenib- has shown to improve survival of HCC patients⁴⁰. Sorafenib is a multikinase inhibitor that targets a number of kinases, including VEGFR-2 and -3, PDGFR-β, c-KIT, RET, FLT-3, and RAF⁴¹. Isolated reports have reported the use of sorafenib in the adjuvant setting after LT. In a companion biomarker study of the pivotal SHARP trial, ten serum markers and one tissue marker were tested, but none of them succeeded in identifying subclasses of responders²⁶. Nonetheless, it is expected that the fast development of new biotherapy and the growing numbers of clinical trials in HCC will lead us to use the molecular features of the tumors to define the type of treatment. In this setting, we have to reevaluate the utility and the frequency of tumor biopsy to have easy access to tissue.

FUTURE PROSPECTS

Novel molecular data can change the approach to diagnosis, staging and prognosis of HCC during the current decade. Regarding prognosis assessment, recently reported prognostic gene signatures and miRNA can enter and complement clinical variables in staging systems, once they have been externally validated by independent studies. These advancements in the understanding of HCC have to be ultimately transferred into the clinical practice as daily tools for management and treatment selection. On the other hand, predictors of treatment response will emerge along with novel drugs in the treatment of HCC. Sorafenib positive results⁴⁰ have opened a new era in HCC research. Future trends in drug development will pivot on accurate assessment of genetic traits in human disease on an individual basis (i.e., personalized medicine). In HCC, the identification of these singularities will allow maximizing therapeutic response by selecting the best drug for the ideal candidate.