HepatoPredict: stratifying liver cancer patients for transplantation

The HepatoPredict – a powerful predictive tool to accurately select hepatocellular carcinoma patients for successful liver transplantation.

Dr José Pereira Leal, the founder and executive director of Ophiomics – Precision Medicine, introduces HepatoPredict, a predictive test for accurate assessment of the five- and ten- year recurrence risk following liver transplantation in the context of hepatocellular carcinoma (HCC).

A simple gene expression signature of informative genes is integrated with clinico-pathological features into a clear prognostic score that can help physicians better decide on the likelihood of success of a liver transplant. HepatoPredict is developed by Ophiomics – Precision Medicine, a Lisbon-based biotechnology company developing genomics and bioinformatics based-solutions to support clinical decision in the context of precision oncology.

What is hepatocellular carcinoma?

HCC is the most common primary malignancy of the liver, with over half a million new cases diagnosed annually worldwide (782,000 in 2012).¹ It is the fifth most common cancer in men (554,000 cases, 7.5% of the total) and the ninth in women (228,000 cases, 3.4%). In Europe liver cancer is the 14th most common cancer, with around 63,500 new cases diagnosed in 2012 (2% of the total). Hepatocellular carcinoma is the second leading cause of cancer-related mortality in the world.¹ Chronic liver disease due to hepatitis B virus (HBV) or hepatitis C virus (HCV) accounts for the majority of HCC cases.²

Among all cancers, HCC is one of the fastest growing causes of death (second leading cause of cancer mortality worldwide) and poses a significant economic burden on healthcare. A SEER-Medicare database, between 1991 and 1999, showed that the annual economic burden of HCC was €524m in the USA. This included healthcare costs and costs due to loss of productivity, which were responsible for 89.2% and 10.8% of the total cost, respectively.³

Despite advances in prevention techniques, screening and new technologies in both diagnosis and treatment, incidence and mortality continue to rise. In North America and parts of Europe, rates are increasing due to chronic HCV infection. Epidemic metabolic syndrome and non-alcoholic fatty liver disease are expected to lead to further increases.⁴

Why liver transplantation?

There are currently three types of curative treatment for HCC in cirrhotic liver:

1. Liver transplantation (LT)
2. Partial liver resection (LR)
3. Arguably percutaneous destruction (using ethanol injection, radiofrequency or microwave ablation, irreversible electroporation, cryoablation and laser ablation).^5,6

Palliative treatments include trans-arterial chemoembolisation, trans-arterial radioembolisation, external beam radiotherapy (conventional or stereotactic), proton beam therapy and, finally, systemic therapy using sorafenib or other targeted agents.⁶ Liver transplantation for HCC is the best treatment option for cirrhotic patients with early-stage tumours and accounts for ∼20% of all liver transplantations performed at most centres worldwide.⁷

Although other treatment modalities exist, only orthotopic liver transplantation (OLT) or surgical resection are curative. While OLT remains the best curative surgical procedure, the shortage of available organs precludes this therapy for many patients with HCC. The best outcomes are achieved when patients are carefully selected for each treatment option. So, optimum candidate selection is crucial to limit surgical morbidity and mortality.

Why do we need better criteria?

Liver transplantation is decided for patients who meet or are down-staged into the Milan or University of San Francisco criteria, which are based on clinical parameters; both are available globally to clinicians. These criteria are based on a combined analysis of several physiological and clinical features of the disease, mainly size and number of tumour foci, as well as tumour aggressiveness. Additional factors are also considered depending on the guideline (vascularisation, blood levels of certain molecules, presence and abundance of immune cells).

The most prominent criteria are the Milan Criteria for HCC, which has been adopted by the Organ Procurement and Transplantation Network (OPTN) in the evaluation of patients for potential transplantation. The Milan criteria have a predictive value of 67% for five-year disease-free survival. Currently, given the limitations of the original study, including the small number of patients and limited inclusion criteria, there is ongoing discussion and controversy regarding the convenience of using this criteria for transplant.⁸

In recent times, efforts have been made to improve this rate and go beyond the Milan criteria – for example with the proposal of the San Francisco criteria, an expanded set of criteria proposed by the University of California San Francisco that represents only a modest expansion of the Milan criteria, is also based on clinical parameters and experiences, and is still focused on physiological and clinical factors.⁹

Under these clinical criteria only 20% of HCC patients are candidates for OLT. There is thus an urgent need to better define the tumour biology based on molecular signature with the aim of successfully incorporating them into the eligibility criteria for the transplant listing, thereby increasing the survival rate. In addition to this problem, up to 33% of liver transplantations fail to cure the disease because there are no tools for predicting a successful curative liver transplantation. Apart from the important health consequences of such failed interventions, there is an economic problem associated with the absence of successfully curative predictions.

The cost of each transplantation varies across countries, the average cost in OECD countries being €120,000. More than 6,500 OLTs are currently performed in Europe per year; 20% of them are related to HCC patients. As about 33% of these transplantations fail to cure, we estimate that about €52m is wasted annually in Europe alone. Thus, even though the current criteria are free, the average cost per successful transplant is higher, as they have poorer survival rate.

In summary, there is a pressing need for better selection criteria that can reduce the cost per transplantation, increase the likelihood of success of each transplant, and identify all the patients who will benefit from this procedure.

The HepatoPredict advantage

We have developed a powerful predictive tool to accurately select HCC patients for a successful liver transplant. HepatoPredict was developed based on an extensive database of patients transplanted over many years at the Liver Transplantation Unit of the Hospital Curry Cabral in Lisbon, Portugal, which includes many patients transplanted beyond the Milan criteria as described in reference.¹¹

In short, following an extensive meta-analysis of the literature, we selected a number of candidates for pilot testing in a small cohort of archived samples of HCC patients that underwent transplant and with more than five years’ follow-up. Gene expression was assessed using real-time PCR in FFPE tumour biopsies. Markers that confirmed its predictive value in this test set were then further tested in a cohort of hundreds of HCC transplanted patients, again with more than five years’ follow-up, for predictive power under multiple clinically relevant scenarios.

Expression levels of these markers in the different groups were then integrated with clinical variables using machine learning approaches to produce a predictive tool – HepatoPredict – a simple molecular signature and clinical variables, with a predictive value for successful transplantation currently ranging between 82% and 95%, and actively optimised for improved predictive power.

While there are academic studies describing complex gene expression signatures as predictive tools for hepatocellular carcinoma,^12,13 in our development of HepatoPredict we focused, in addition to positive and negative predictive values, on implementation simplicity, reproducibility and cost-effectiveness, as without these it is unlikely that any solution can be adopted by any health systems which are increasingly under financial pressures for better and cheaper approaches to counterbalance the rising costs of novel pharmaceuticals.

HepatoPredict includes a proprietary algorithm that integrates clinical and molecular variables and provides a single, objective value for the probability of success of transplantation. It is simple to implement in the laboratory – it is based on a biopsy of the tumour followed by a digital PCR-based quantification of gene expression for a small number of informative genes. Because of this simplicity, it can provide a result in few days.

HepatoPredict is being developed to be available in the following format:

1. Sample collection kit is sent to customer (pathology lab associated to transplantation centres)
2. The lab inserts the sample (biopsy) into the kit and sends it by courier service to our central lab (run by Ophiomics)
3. The central lab performs the test and feeds its result into the algorithm for integration with clinical information
4. The ordering clinician downloads a web-based report summarising the results and a clear, easy-to-interpret result.

The simplicity of the procedure makes it possible for the report to be available in about a week after sample reception. Furthermore, HepatoPredict will be able to avoid long waiting times in doubtful cases as clinicians will have a powerful tool to make the decision instead of waiting to see how the tumour behaves and evolves.

Overall, HepatoPredict will enable clinicians to better decide the right eligibility of patient candidates for a successful liver transplantation at the right time. It thus follows that HepatoPredict will have significant societal value, as 20-25% more lives can be saved, representing 180-220 additional cured HCC patients annually in Europe alone.

The team at Ophiomics

Ophiomics – Precision Medicine was established in 2015 to develop molecular diagnostic solutions to support precision medicine in the context of chronic diseases, with a particular emphasis on cancer. We have developed a precision oncology portfolio currently deployed as clinical routine in Portugal and in the international network of the Germano De Sousa Laboratories (www.germanodesousa.com).

Our precision oncology portfolio is focused on solid tumours and includes genetic risk tests, liquid biopsy-based somatic tests for patient stratification, and a standard molecular pathology portfolio based on the latest technologies, such as next-generation sequencing and digital PCR. Bioinformatics is a major focus of the company, with our in house-developed genomics laboratory management and clinical reporting system OphiDx (www.ophiomics.com/en/pipeline-en/ophidx-en/) a cornerstone of our clinical activity.

Ophiomics – Precision Medicine was founded by the Germano de Sousa family and scientists from the Instituto Gulbenkian de Ciência in Oeiras, Portugal. José Germano de Sousa, MD, former president of the Portuguese Medical Association, is our clinical director and co-founder. He established and scaled up the Centro de Medicina Laboratorial Germano de Sousa (CMLGS), the largest Portuguese group of clinical pathology with over 900 employees and multiple laboratories in Portugal, as well as an international presence in 14 other countries. Dr Joana Cardoso Vaz, PhD is scientific and technical director and co-founder.

During her PhD at the Erasmus University Medical Center, Rotterdam, the Netherlands, she specialised in cancer genomics. She has married a research programme in markers and mechanisms of cancer progression with extensive clinical genomics experience. In the development of HepatoPredict, Dr José Pereira-Leal, PhD, CEO and co-founder, is also the scientific director of Healthcare City, a new startup focused on the health sector. Previously, he co-ordinated the Computational Genomics Laboratory and the Bioinformatics Unit of Instituto Gulbenkian de Ciência, where he helped establish the European biological data infrastructure ELIXIR and its national counterpart, the Portuguese biological data infrastructure BioData.pt.

Ophiomics’ services and portfolio are supported by intensive R&D activity. The development of HepatoPredict derives from a long-standing collaboration with the Liver Transplantation Unit of the Hospital Curry Cabral, the reference liver transplantation centre in Portugal, as well as a major hepatobiliary surgery centre. Hugo Pinto Marques, MD, PhD and Eduardo Barroso, MD, PhD are the key clinical collaborators and advisors.

Barroso is a worldwide known hepato-biliary-pancreatic (HPB) surgeon, as well as past-president of the Portuguese and European surgical associations. He is the founder of the HBP and Transplantation Centre of Lisbon, currently the most active liver resection centre in Europe and one of the top five centres for liver transplantation in Europe. He is a full professor at NOVA University Lisbon and a key opinion leader in his field.

Marques is a surgery specialist focusing on liver diseases and transplantation, and works, since 2004, at the Hepato-Bilio-Pancreatic and Transplantation Centre (Curry Cabral Hospital, Portugal). He is also a professor at NOVA Medical School and holds several positions in surgical societies: he is currently a member of the board of the European Surgical Association, secretary general of the Compagnons Hépato-Biliaires, and adjunct secretary general of the Portuguese Surgical Association. Both Barroso and Marques are principal investigators for HPB oncology at the Centre for Chronic Diseases of NOVA
Medical School.

Award-winning in the EU

Ophiomics – Precision Medicine was recently awarded an exploratory Phase I grant from Horizon 2020’s SME Instrument to develop HepatoPredict. This award is a prestigious achievement as its success rates are very low. This grant endorses the innovation and business potential of HepatoPredict and is being used to prepare an application for a Phase II SME Instrument grant to fund the final development of HepatoPredict to market-ready stage.

Our next challenge in the development of HepatoPredict is the multi-centre retrospective and prospective validation of the predictive algorithm. To this end, collaborations have been established with multiple reference liver transplantation centres in Europe, North America and South America. We expect this work to start at a limited scale early in 2019, and to scale up towards the end of the same year.

Ophiomics pipeline for liver cancer

Ophiomics has two other products in development to support clinical decision making in
liver cancer. Both products are aimed at the hepatology/gastroenterology clinic. The first one is an extension of HepatoPredict to identify those patients that will benefit from a liver resection. Depending on the centre, even more patients may be eligible for this approach that is not limited by organ availability.

The second one, at an earlier stage of development, is a liquid biopsy test to aid in diagnosing liver cancer when image data is unclear and/or when lesions are still very small. In the more distant future, we aim to complement this portfolio of precision hepato-oncology tests with an early detection test suitable to use for monitoring high-risk cases.

References

1. Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. Wiley-Blackwell; 2015 Mar 1;136(5):E359-86
2. Mittal S, El-Serag HB. Epidemiology of HCC: Consider the Population. J Clin Gastro. 2013 Jul 1;47(0):S2-S6
3. Lang K, Danchenko N, Gondek K, Shah S, Thompson D. The burden of illness associated with hepatocellular carcinoma in the United States. J Hepatol. 2009 Jan;50(1):89-99
4. Stewart BW, CP Wild – World Health Organization, 2014. World Cancer Report 2014. International Agency for Research on Cancer
5. European Association for the Study of the Liver, European Organisation for Research and Treatment of Cancer. EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. Vol. 56, J Hepatol. 2012. 908-43
6. Dhir M, Melin AA, Douaiher J, Lin C, Zhen WK, Hussain SM, et al. A Review and Update of Treatment Options and Controversies in the Management of Hepatocellular Carcinoma. Ann Surg. 2016 Jun;263(6):1112-25
7. Bellavance EC, Lumpkins KM, Mentha G, Marques HP, Capussotti L, Pulitano C, et al. Surgical management of early-stage hepatocellular carcinoma: resection or transplantation? J Gastrointest Surg. 2008 Oct;12(10):1699-708
8. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver Transplantation for the Treatment of Small Hepatocellular Carcinomas in Patients with Cirrhosis. N Engl J Med. 1996 Mar 14;334(11):693-700.
9. Yao FY. Expanded criteria for liver transplantation in patients with hepatocellular carcinoma – PubMed – NCBI. Hepatology Research. 2007 Sep;37(s2 Fourth JSH S):S267-74
10. van der Hilst CS, IJtsma AJC, Slooff MJH, TenVergert EM. Cost of Liver Transplantation: A Systematic Review and Meta-Analysis Comparing the United States With Other OECD Countries. Medical Care Research and Review. SAGE PublicationsSage CA: Los Angeles, CA; 2008 Sep 16;66(1):3-22
11. Marques HP, Ribeiro V, Almeida T, Aniceto J, Silva S, Sobral M, et al. Long-term Results of Domino Liver Transplantation for Hepatocellular Carcinoma Using the “Double Piggy-back” Technique: A 13-Year Experience. Ann Surg. 2015 Nov;262(5):749-56-discussion-756
12. Wang Z, Teng D, Li Y, Hu Z, Liu L, Zheng H. A six-gene-based prognostic signature for hepatocellular carcinoma overall survival prediction. Life Sci. 2018 Jun 15;203:83-91
13. Miltiadous O, Sia D, Hoshida Y, Fiel MI, Harrington AN, Thung SN, et al. Progenitor cell markers predict outcome of patients with hepatocellular carcinoma beyond Milan criteria undergoing liver transplantation. J Hepatol. Elsevier; 2015 Dec 1;63(6):1368-77

Dr José Pereira Leal
Founder, Executive Director
Ophiomics
+351 210 416 538
jleal@ophiomics.com
http://www.ophiomics.com

This article will appear in issue 7 of Health Europa Quarterly, which will be published in November 2018.