Skip to main content
Vai all'homepage della Commissione europea (si apre in una nuova finestra)
italiano italiano
CORDIS - Risultati della ricerca dell’UE
CORDIS

Machine learning software to design personalized neoantigen vaccines tailored to specific vaccine delivery systems

Periodic Reporting for period 1 - MEDIVAC (Machine learning software to design personalized neoantigen vaccines tailored to specific vaccine delivery systems)

Periodo di rendicontazione: 2019-05-01 al 2020-07-31

Cancer remains a leading cause of disease and death worldwide, causing an enormous societal and economic burden. An estimated 14 million new cancer cases were diagnosed in 2012 and 8.26 million people died from the disease. The annual cost of treating cancer in the EU is estimated to be over €150 billion. While the global incidence of cancer has more than doubled over the past 30 years, the future outlook is even more pessimistic, primarily due to increasing living standards and an ageing population. Thus, there is an urgent need for more effective methods of treatment to reduce this massive global burden.

Immunotherapies are currently regarded as the most promising avenue to delivering the holy grail of medicine i.e. providing a cure for cancer. Despite their Nobel-winning status, personalization of immunotherapies remains a key challenge to which no cost-effective solution currently exists. Current methods for identifying the immunogenic neoantigens required to design patient-specific cancer vaccines typically utilize next generation sequencing (NGS) analysis of DNA and RNA coupled with mass spectrometry and/or immunological methods, such as tetramer staining. However, these methods are time consuming to perform (up to three months required to develop a single vaccine), not readily scalable (involving cumbersome manual operations in wet lab environments) and expensive –which currently prohibits the mass roll-out of personalized cancer vaccines.

NEC OncoImmunity (NOI) already developed a state-of-the-art software solution (the Immune Profiler prediction engine) that addresses the key bottleneck of cancer vaccine development, i.e. cost-effective accurate identification of immunogenic tumor neoantigens. The Immune Profiler has already been validated using retrospective clinical data and demonstrates an unprecedented level of accuracy. The objective of the MEDIVAC project was to undertake the required activities to transform the current Immune Profiler prototype into a fully-fledged, client-tailorable solution (MEDIVAC) and demonstrate its performance in a large clinical test to thoroughly document its performance. This would have enabled NOI to validate the technology and comply with all regulations, and to gather the requisite medical evidence to support a strong communication strategy directed at gaining wide industrial endorsement for our product. Moreover, the MEDIVAC project would have generated a significantly improved graphical user interface (GUI), which would have enabled vaccine companies (our clients) to rapidly design personalized vaccines that contained an optimal constellation of immunogenic neoantigen customised for their specific vaccine system (VDS), whilst simultaneously removing off-target epitopes or neoantigens specifically expressed in key critical organs that could generate unwanted auto-immune responses. The MEDIVAC software had the potential to help revolutionize the personalized cancer vaccine field, enabling these life-saving therapies to be generated in a scalable, cost-effective and clinically actionable time frame, bringing hope to the 1.3 million patients diagnosed with late stage cancer every year in the EU and US.
The project implementation followed the work plan without significant delays during the first eight months, from May to December 2019 and was progressing according to the plan until the project was terminated by the EASME following the acquisition of OncoImmunity by the Japanese company NEC Corporation. Significant progress has been made on the vaccine design module, the graphical user interface (GUI), regulatory aspects supporting the technology and its deployment in GDPR compliant manner. In particular, NOI has further developed it algorithmic framework which when complete will enable the MEDIVAC software to select optimal 27-mers containing the right constellation of class I and class II neoepitopes to induce a potent clinically relevant anti-tumor response and to combine these together in an optimal manner to minimise the risk of off-target or autoimmune epitopes. This framework will generate “presentation” heat maps that can be visualized using the improved GUI enabling our clients to rapidly design safe and efficacious vaccine constructs. NOI has also got their core technology that underpins the MEDIVAC software the “Immune Profiler” CE-marked as an in vitro diagnostic, and has also generated a deployment strategy that leverages the Google cloud platform that will allow the MEDIVAC software to be deployed anywhere in the world and analyse the clients data in a GDPR compliant manner.
During the first 8 months of the project, prior to its termination by the EASME, NOI implemented the project according to plan and made significant strides towards transforming the current Immune Profiler prototype into a fully-fledged, client-customizable solution. In particular, NOI made significant advances in progressing the client customization of the platform and developing the machine learning framework and GUI for optimized and customized vaccine design i.e. selecting the optimal constellation of clinically beneficial immunogenic neoantigen whilst removing unwanted off-target and autoimmune generating epitopes. In addition, the company made significant progress in making the software compliant with the necessary regulatory certifications, including getting the Immune Profiler CE-marked as an in vitro diagnostic medical device. Finally, NOI also progressed the commercial aspects of the project entering into agreements with a number of prominent players in the personalized vaccine field to provide the MEDIVAC solution to support their personalized cancer vaccine clinical trials and is in late stage discussions with several others both in Europe and the US.
noi-image.png
Il mio fascicolo 0 0