Are open-source approaches the most efficient way forward for COVID-19 drug discovery?

ABSTRACT

Introduction: The COVID-19 pandemic has catalyzed the production of potential antivirals and vaccines from research organizations across the globe. The initial step for all drug discovery models is the identification of suitable targets. One approach organizations may take to tackle this involves issuing raw data publicly for collaboration with other organizations in order to spark discussion, collectively experiment and stay up to date with advances in scientific knowledge.

Areas covered: Numerous organizations have released genomic data, amongst other tools, for the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and this has led to the development of growing datasets of knowledge for continued collaboration amongst different scientific communities. A different technique employs a more closed, market-driven method in order to stay ahead financially in the race for developing a suitable antiviral or vaccine. The latter allows sustained motivation for company ambitions and progress has been made toward clinical trials for potential drugs.

Expert opinion: A case can be made for both open and closed drug discovery models; however, due to the rapidly evolving nature of this deadly virus, organizations should collate their research and support one another to ensure satisfactory treatment can be approved in a timely manner.

KEYWORDS:

• COVID-19
• coronavirus
• antiviral
• treatment

1. Introduction

Open-source models are a relatively new approach for drug discovery. They rely on the collaboration and sharing of data and knowledge between different academia, organizations, and industries. Participation from any academic level for suggestions and open discussion is encouraged [1]. The Council for Scientific and Industrial Research’s Open Source Drug Discovery (OSDD) project and The Synaptic Leap’s Schistosomiasis (TSLS) project illustrate how this approach has worked effectively in the past to help develop drug innovation for tropical diseases [2,3]. The consortium first published research in 2009 detailing the genomic map of tuberculosis. Since then, 110 projects have been launched by 7,500 registered researchers from around the world. The main accomplishments of the organization to date include the development of eleven models that exhibit anti-tuberculosis activity and re-mapping of the M.tuberculosis genome [2]. These approaches are theorized to be efficacious due to the immediate availability of data, flexibility between those who participate, open communication and the utilization of community efforts where parties from different institutions help each other to accomplish related outcomes, rather than compete [4]. A nonprofit organization called Results for Development has outlined a set of features that projects must exhibit for them to be accessible for all users:

1. The data must be open access and uncharged.

2. A forum for collaboration should be opened.

3. The project must follow a set of rules that comply with ‘openness’ [2].

2. Current open-source approaches for COVID-19 vaccine and antiviral development

Weeks after its initial outbreak in December 2019, China published a draft of the genetic sequence of the novel coronavirus, allowing researchers across the globe to begin identifying antiviral and vaccine targets. Subsequently, multiple scientific organizations have publicly released data types ranging from dashboards and visualization tools to clinical studies, genomics, and chemical structure data of potential antivirals. Recursion, a clinical-stage biotechnology company, released the first human cellular morphological dataset of the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) containing 305,520 5-channel fluorescent microscopy images and their deep learning embeddings [5]. The data can be obtained openly allowing a vast number of scientific communities to analyze and experiment on. Similarly, the American Chemical Society’s data division issued a dataset consisting of 50,000 compounds with potential antiviral properties [6]. Whilst some groups have used computational approaches to identify antiviral compounds, other researchers from around the world advocate the use of experimental (in vitro) approaches, both viable options to accelerate drug discovery [7,8]. There is uncertainty as to when a suitable antiviral or vaccine will be approved but the collaboration orchestrated by these organizations may certainly expedite the recognition of substances which may be repurposed for treatments in the near future.

The difficulty in developing such treatment stems from the ongoing growth in the ability of the virus to mutate, contributing to viral resistance to therapeutic interventions. Coronavirus3D is a server designed to provide its users with details of how mutations in the SARS-CoV-2 proteins affect their overall 3-dimensional compositions and their functions [9]. The server is free for public access and allows individuals to visually analyze mutations and conduct their own experimentation using this data. A recent collaboration between the University of Cambridge and Seoul National University has led to the creation of the first open-source all-atom model of a complete spike (S) protein. This protein model has been made publicly available in a paper recently published by Woo et al. in The Journal of Physical Chemistry, formulated using the CHARMM-GUI membrane builder [10]. This can prove to be beneficial as the S protein is a component which plays a crucial role in the entry of a virus inside of its host cell. This allows for the potential of targeting the protein in future antivirals and vaccinations.

3. Recent advances in COVID-19 drug discovery

In addition to collaborative efforts, research organizations and pharmaceutical companies are also understandably opting for ‘closed-door’ approaches in order to be the first to produce an effective treatment. Having identified drug targets, progression has been made toward the funding of clinical trials to determine safety and efficacy. Gilead Sciences are supporting clinical trials for the use of remdesivir, a nucleotide analogue prodrug which inhibits viral RNA polymerases. One trial demonstrated clinical improvement in 68% of patients and appeared to have a favorable clinical safety profile [11]. In order to accurately measure efficacy, ongoing trials will need to be conducted. Roivant Sciences is developing gimsilumab, a monoclonal antibody targeting granulocyte-macrophage colony-stimulating factor (GM-CSF), in phase II of clinical trials of patients with Acute Respiratory Distress Syndrome (ARDS) [12]. An appreciative safety and tolerability profile has been exhibited so far on the data that has been collated.

Breakthroughs have been made with the use of messenger RNA (mRNA) based vaccines. Moderna is working on the formation of a messenger RNA-based vaccine labeled mRNA-1273. Following its preliminary phase I trials with 45 healthy adults, the vaccine has shown to provide a neutralizing activity in all participants, due to the ability of the mRNA to induce an anti-SARS-CoV-2 immune response [13]. The vaccination is currently in phase III of its clinical trials and corroborated by the National Institute of Health and the Biomedical Advanced Research and Development Authority. AstraZeneca has been working on using adenovirus vectors in order to help develop an immune response to the S proteins produced, helping reduce its number and preventing the entry of the SARS-CoV-2 virus into cells. Vaccinations made from the ChAdOx1 virus were given in a phase I/II trial and it was found that T cell responses had peaked by 14 days and IgG antibody responses peaking at 28 days, both being S protein specific [14]. Inovio Pharmaceuticals have also been targeting the spike glycoprotein of SARS-CoV-2 using a deoxyribonucleic acid (DNA) plasmid pGX9501 in order to help generate an immune response to the protein [15]. The INO-4800 vaccine has recently commenced its phase I trials and is currently awaiting its results.

4. Conclusion

There is a dilemma faced by drug developers when determining whether to approach the process of finding a vaccine or antiviral in a public or privatized demeanor. By making information about potential drug targets, obtained from clinical trials through private companies publicly available, there can be an increase in the information attainable from topics which may currently be limited in understanding. This benefits other professional bodies who may be delving in the wrong avenues and can help to inform them in prioritizing where to focus their efforts on. While the pecuniary outgrowth may be limited through collaborative work, the desperation for a working vaccine by international governments still holds economical worth, and hence the financial incentive is by no means lost.

5. Expert opinion

In the face of a global pandemic, time is of the essence. In comparison to other infectious diseases, which may potentially prefer a privatized approach for drug development from organizations, the prevalence rates of COVID-19 are exceptionally high and are only increasing. The utilization of open-source models for COVID-19 drug discovery has led to the development of numerous everblooming datasets of knowledge, conveying a great deal of potential in the timely approval of a suitable antiviral or vaccine. This can drastically increase the speed at which a vaccine can be developed, trialed, and brought out into mass production. There is also a sense of camaraderie in the working together of medical professionals, helping combine areas of expertise for the greater good of ensuring the health of the general public. With the ultimate goal being eradication of this disease caused by SARS-CoV-2 and the prevention of future outbreaks, it is fair to argue that an open-source stance toward compound discovery is in the best interests of public health. Independent organizations must convene as a global community and share knowledge in order to be able to overcome the challenge of keeping up with the rapidly evolving nature of this virus. Competition and market-driven approaches may increase attrition rates amongst organizations reaching terminal standstills, creating unnecessary economic burdens amongst the setting of an already deadly and devastating infliction.

Despite this, a focus has to be made on the potential financial loss that comes with collaboration. When undertaking the development of a drug, by operating independently one is entitled to the complete share of profits behind any sales of the product. With governments all over the world willing to pay billions for these drugs, there is a lucrative incentive for companies to privatize their information for personal gain. Unfortunately, this potential increase of finance and fame comes with an increased workload in lieu of the reduced time and demands associated with collaborative, open-sourced approaches. As a result, an informed decision must be made on a case-by-case basis, depending on the needs and goals of the company, alongside which aspects they prioritize the most. It is difficult to predict what may happen in the coming months with regards to drug development and further outbreaks. A combination of both open-source and closed approaches may be a practical solution in order to benefit both commercial interests of companies and public health. However, this is a strategy which is yet to be successfully implemented.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript has not been funded.