NEC rises to the challenge of developing a COVID-19 vaccine
Three NEC Group companies collaborate to unlock the potential of AI drug discovery
Leveraging its cutting-edge AI technologies, NEC Corporation has thrown its hat into the ring to develop a coronavirus vaccine. This is NEC's second AI-powered drug development project. The first was for the development of neoantigen cancer vaccines, which are now being tested in clinical trials launched January 2020. In the global race for pharmaceutical companies and research groups to develop a COVID-19 vaccine, what is the NEC Group's strategy? We asked members of the project from NEC, NEC OncoImmunity AS (NOI) and NEC Laboratories Europe (NLE) about the possibilities of AI-powered drug development.
The COVID-19 coronavirus, first detected at the end of 2019, quickly swept around the globe and still shows no sign of abating. According to World Health Organization (WHO) data, the cumulative number of COVID-19 cases worldwide has reached 11,500,302 while the number of deaths is 535,759 (both figures are as of July 7, 2020). In light of the situation, NEC held an online press conference on April 23 to announce that it was developing COVID-19 vaccines using AI and was publishing its research findings in a research paper that same day.
The vaccines are not being developed independently by NEC but through close collaboration between NEC, NLE, located in Heidelberg, south-west Germany, and NOI, a biotechnology subsidiary located in Oslo, Norway , and the Group's development efforts have attracted attention as a new style of drug discovery research which transcends borders and time differences.
This latest announcement drew attention because it related to the development of COVID-19 vaccines. However, this is not the first time that NEC has tried its hand at drug discovery. It was in 2019 that NEC formally announced that it was entering the drug discovery business. NEC first set its sights on the development of neoantigen cancer vaccines. Neoantigens are not expressed by normal cells. They are newly formed antigens resulting from tumor mutations. Neoantigen vaccines are vaccines designed based on the prediction of such neoantigens using genetic information.
Clinical trials of neoantigen-based cancer vaccines are already underway and are off to a good start. And now, NEC is taking aim at vaccines for COVID-19, as recently announced.
Proposals submitted simultaneously by two European Group companies. Not an easy decision.
Q: Why did you decide to have a go at developing vaccines for COVID-19?
Akira Kitamura (Kitamura): The vision mission statement set out by NEC at the time of establishment of the HealthTech Business Development Office, the predecessor of the current AI Drug Development Division, also mentioned the AI-powered development of vaccines for infections and autoimmune diseases as well as cancer vaccines. However, in 2019, nobody had seen the COVID-19 outbreak coming and so it was not our intention to develop COVID-19 vaccines. At the time of the outbreak, NLE and NOI submitted proposals for the development of COVID-19 vaccines to NEC both at the same time.
Brandon Malone (Malone): NLE has experience of developing neoantigen cancer vaccines. Thinking we could apply the platform we had created to COVID-19, we suggested having a go at developing vaccines.
Trevor Clancy (Clancy): The development of vaccines, whether to prevent cancer, bacteria or viruses, requires precise immune profiling and the AI prediction platforms are the same. NOI felt it should definitely have a go at designing a blueprint for a vaccine since it had all the necessary technology at its disposal.
Q: So did you set about developing vaccines as soon as you received the proposals?
Kitamura: The proposals were made in March 2020. The COVID-19 outbreak had already been declared a pandemic and we were two months into the crisis. We wondered whether we would develop a vaccine in time if we started then... whether we would succeed in developing an effective vaccine... and whether we ought to spare valuable R&D resources. It definitely wasn't an easy decision.
Q: By March, pharmaceutical companies and startups all over the world had already thrown their hats into the ring to develop a vaccine, right? According to data published by PhRMA in April, there were already more than 70 COVID-19 research programs underway in the US alone, including six active clinical trials for potential coronavirus vaccines.
Kitamura: Yes, it was clear that the NEC Group was a late comer to the race even if only by two months. That said, nobody can tell whether the vaccines developed first will be successful or not. Figuring that even if we failed to develop a vaccine in time, the research would serve as validation of our cancer vaccine development technology and that this was a good opportunity to examine application of the technology to the infectious diseases domain, we decided to proceed with vaccine development, with NEC Corporation playing a coordinating role.
Aiming to develop personalized therapeutic cancer vaccines but broad-spectrum vaccines against infectious diseases
Q: Explain the roles of NOI and NLE in the vaccine development project.
Richard Stratford (Stratford): NOI plays a leading role in terms of the science and vaccine design. The NOI team, headed by myself and Clancy, was in charge of genome analysis of the virus and the subsequent epitope mapping analysis to identify epitopes that could be recognized by the human immune system (by a class of immune cells known as T-cells).. (Epitope: specific pieces of the virus which are recognized by T-cells or B-cells). The analysis performed by NOI identified hotspots in the viral proteome that contain multiple overlapping or closely located epitopes that represent optimal building blocks for developing a vaccine.
When it comes to the development of COVID-19 vaccines we initiated using the cancer vaccine platform, the direction of development is slightly different. In the case of cancer vaccines, it is necessary to develop personalized vaccines tailored for each patient. However, in the case of the COVID-19 vaccines, we needed to generate the optimal immune response with the broadest coverage of the human population.
Malone: After NOI completed the selection of epitopes and corresponding epitope hotspots, NLE was responsible for calculating coverage of human population and searching for optimal combinations of the epitope hotspots selected by NOI. While the development direction is different, the process of selecting epitopes for vaccine purposes is the same for both cancer and infectious diseases. The aim is to identify promising epitopes and the process of using bioinformatics and machine learning to achieve this purpose is the same.
Harnessing the advantages of being a latecomer to the vaccine development race
Q: In which aspects of vaccine development can NEC utilize its strengths?
Stratford: To develop an effective vaccine, it is necessary to identify promising epitopes which are recognized by and activate T cells, which mediate immunity. More specifically, even when pathogens such as a virus enters the body, T-cells are unable to recognize the antigens directly. The T-cells are only able to recognize antigens presented as small fragments bound to HLA (human leukocyte antigens). These HLA alleles are extremely diverse and in order to develop a broadly effective vaccine, it is necessary to identify epitopes which respond to the most frequent HLA alleles.
In other words, development of a truly effective vaccine requires information not only about COVID-19 but also about HLA alleles and the NEC Group's technologies are being used to study the optimal epitope/HLA combinations.
Malone: In order for COVID-19 to enter host cells, spike proteins (S-proteins) protruding from the virus surface must bind with ACE2, a cell surface protein which serves as the receptor for the virus. Researchers all over the world are still conducting research into COVID-19 and the majority of such efforts are focused on the S-proteins and very little research is being done into the other viral proteins. We believe we have a winning chance there despite our late start.
Q: We heard that NEC's data processing technology is also useful in the analysis part of the process.
Clancy: NOI analyzed all the proteins in the COVID-19 virus not just the S proteins and identified potential epitopes that could bind with 100 HLA alleles to create an epitope map. In addition to performing computer simulations, NOI selected virus epitopes predicted to have the best HLA-binding affinity and presentation potential. Then, to prevent the immune response stimulated by the vaccine from attacking normal human cells, we excluded human epitopes and epitopes with high similarity to human tissue at the amino acid sequence level.
Saverio Niccolini (Niccolini): And the NEC Group's data processing technology proved useful in this process. We studied 3,400 coronavirus strains and HLA genes in 22,000 individuals and we deployed "digital twin" technology which uses cloud computing to replicate and assess in advance physical objects in our analysis.
Stratford: Binding epitopes to HLA one by one in a laboratory takes a great deal of time and effort. But, by using data processing technology, we were able to identify an optimal combination of 8 potential epitope hotspots using the "digital twin" analysis after just a month of research, which could cover as much as 90% of the world's population.
Q: Are the next steps creation of the vaccines and confirmation of their preventive effect in clinical trials?
Kitamura: Yes, that is correct. NEC is currently holding discussions with pharmaceutical companies both in Japan and overseas.
Proudly committed to a common goal
Q: How have you managed to continue your research amid the COVID-19 pandemic?
Malone: Many of our researchers were forced to work from home and so we have held most of our discussions and meetings online. Despite working remotely, we have been able to conduct meetings face-to-face and share data and have somehow managed to continue with our research.
Niccolini: We have managed to push ahead with the research and produce results under such difficult conditions because the team shared the common goal of wanting to put an end to the COVID-19 outbreak and this made us pool our technologies and skills. By solving the problem of developing a vaccine we can solve a global problem. We put our heart and soul into solving the problem and this culminated in these latest findings.
Stratford: Our determination to fight back against COVID-19 held the key to the success or failure of the project. While a global pandemic was not the ideal research environment, the research itself was extremely exciting and we were strongly motivated to do it.
Clancy: We have demonstrated through our latest findings that informatics is extremely important for vaccine design. A pandemic is in some respects a race against time and so I firmly believe that this technology has great potential.
Q: And finally, what is your outlook for the future?
Kitamura: There are still a number of steps that need to be completed before a vaccine is developed but I believe that the strong sense of mission of our members who want to make some kind of contribution to society facing an unprecedented crisis is evident in these latest research findings. NEC, NOI and NLE will continue pushing ahead with their collaborative research efforts in the firm belief that we will succeed in developing a vaccine.
(July 22, 2020)