In this multi-part series, we step back from the flood of day-to-day pandemic news, and put our attention back on the big questions. This series is divided into three major parts: disease biology, therapies, and the progress of the pandemic. We’ll close by discussing the future. This is Part II, on therapies and vaccines. Read Part I: disease biology.
Since it became clear in late January that COVID-19 posed a threat to global health, scientists and doctors around the world have been working to develop and prove therapies, on a number of fronts. There are now literally hundreds of therapy projects underway.
While some studies may not “read out” (i.e. show results) for years, some of the most potent are expected to begin arriving this year. And a limited number of repurposing options have already demonstrated some efficacy. Vaccine programs give every sign of being able to prevent COVID-19 worldwide on approximately the timescale originally surmised.
Here’s what you need to know about therapies and vaccines:
- Some drugs have already been repurposed to treat COVID-19 successfully. Remdesivir and dexamethasone are notable successes, we have our eye on famotidine, and others will probably emerge.
- We have hope for antibody treatments in the year to come, though we’ll have to keep COVID-19 at bay so companies can actually manufacture these treatments.
- Vaccines do seem likely, and scientists are at work to make vaccination a speedy, global project.
Next, we’ll summarize the situation up to now, dividing all of the projects in the works into the categories of repurposed drugs, novel small molecules, antibodies and plasma, and vaccines.
Therapies: drugs to treat COVID-19
Pharmaceutical repurposing offers the first opportunity to make a difference in the pandemic, because existing drugs don’t need to be invented or proven in all the same ways as new drugs. Starting with mechanistic research, real world evidence, and in-vitro screens, researchers around the world have been identifying existing drugs that might help, and then testing them in early stage clinical trials.
Some drugs that appeared to have potential based on early case studies have fizzled out completely, and are now believed to be ineffective in fighting the coronavirus. These include the HIV protease inhibitors lopinavir and ritonavir, an IL-6 antibody from Sanofi. This category also includes hydroxychloroquine, which has failed a lengthy series of randomized trials and RWE analyses around the globe, and as a result has seen its use authorizations withdrawn. The failure of hydroxychloroquine was a particular disappointment, because it is a cheap and widely available drug. If it worked, it could have made a big impact. It was already being widely used in the anticipation of positive results.
However, at least three drugs are known to help with COVID-19. There’s remdesivir, an RNA polymerase inhibitor originally developed for Ebola, which reduces the duration of COVID-19 infection and may help with fatality rates (or may not). It’s passed a couple of randomized trials, and while it’s expected to go into shortage this summer, Gilead and its licensing partners are scaling production, and expect to be able to provide enough to treat millions of patients in a matter of months. Gilead has also announced it is working on inhaled and other forms of remdesivir, which may allow it to be used outside the inpatient setting to which it is currently restricted.
In addition, a randomized trial in China demonstrated that a cocktail of ribavirin plus interferon beta-1b China showed that compared to the control group, these two drugs shortened the time to negative PCR testing. While both of these effects are relatively modest compared to what could be hoped for, these drugs will surely find wider use.
Finally, in June a randomized trial in England read out, demonstrating that dexamethasone, a corticosteroid widely used for a variety of conditions, is effective at reducing death among seriously ill, hospitalized COVID-19 patients. Dexamethasone does not act against SARS-CoV-2 directly, but suppresses the human immune system, blunting the worst effects of cytokine storm in seriously ill patients. In the English trial, it reduced the risk of death by 35% among ventilated patients and 20% among oxygenated patients, but didn’t help patients of moderate severity at all. Dexamethasone also didn’t help patients discharge faster. Due to dexamethasone’s wide availability around the world, low cost, and well understood properties, it’s likely this innovation can be rapidly deployed worldwide. We covered this possibility in February.
And these modest successes are not the end of the story. Many additional drugs are being tried, and some may well prove more effective than those we already know about. Two that we at The Prepared are keeping a close eye on are famotidine and TMPRSS2 inhibitors.
Famotidine is an H2 receptor antagonist, which reduces stomach acid by interfering with the H2 receptor’s activation of the proton pumps which generate acidity in the stomach. It was first identified as a potential COVID-19 therapy in a comorbidity analysis in China, which showed that patients with some gastric conditions had lower risk of dying of COVID-19. More investigation showed that this effect was confined to those patients taking famotidine. On the strength of this result, a number of clinical trials were initiated, and some are expected to read out soon. An RWE analysis in the United States shows the same correlation as the Chinese study. Famotidine, an over the counter drug which normally sells for at little as $6 per bottle, has already sold out in some channels due to the possibility that it is effective against COVID-19.
TMPRSS2 is a serine protease involved in cleaving SARS-CoV-2 proteins, which is necessary for the virus to infect cells after its contents have been introduced. Along with ACE2, it is the other major member of the collection of proteins most directly involved in the SARS-CoV-2 infection cycle. While ACE2 inhibitors aren’t effective against SARS-CoV-2 in vitro, TMPRSS2 inhibitors are effective in vitro, and are already being tried in humans to treat COVID-19.
Perhaps also worth mentioning: low dose x-ray irradiation of the lungs in severe cases to damp inflammation, a treatment used historically on bacterial pneumonia in desperate cases. This has only been published once that we know of, with just five patients, but is being looked into in several registered trials. The hour is early and it could easily prove not to be effective, but if it is effective it could be used widely without much delay.
Watch for trials of these and other repurposed drugs to read out in the coming months, and for more news about how the ones that work influence the course of the pandemic.
Novel small molecule drugs
Another potential category of COVID-19 therapies would be new small molecule drugs designed to either directly interfere with viral processes (antivirals) or block the virus’s activity in and with the host cell (host-target drugs).
While a range of drug companies have announced that they are developing both categories of drugs for COVID-19, small molecule drug development is a lengthy, time-consuming process. Small molecule development hasn’t benefited from some of the biotechnologies which have sped the development and manufacturing of biological products like vaccines and antibodies. No significant announcements of milestones on this front have taken place, e.g. no such drugs have yet entered trials. There is no realistic prospect for new small molecules to arrive, e.g., this year.
Antibodies and convalescent plasma
Another class of COVID-19 therapies are those based on antibodies. These can be either natural anti-SARS-CoV-2 antibodies extracted from the blood of recovered patients (convalescent plasma) or synthetic antibodies manufactured for use as drugs. Infused or injected, they stay in the blood for weeks or months, functioning either as a treatment for COVID-19 or as a temporary vaccine.
In theory these are faster to develop, because the natural antibodies which serve as the jumping-off points for sequence optimization are generated in animal challenge experiments, and because as biological products they can use the manufacturing capacity already built for other antibodies. As a result of this, a number of antibody projects have already reached significant milestones.
Regeneron pharmaceuticals, based in New York, announced in March it had finalized the sequence of a SARS-CoV-2 antibody and was committing to manufacture 200,000 doses per month beginning in August even before achieving any clinical trial success. They’ve since announced that their therapeutic, dubbed REGN-COV2, will be composed of two neutralizing antibodies targeting different regions of the SARS-CoV-2 spike protein, a strategy intended to prevent viral mutations from allowing the virus to resist the antibody treatment. The success of this strategy is shown by an upcoming publication in Science, Regeneron has announced. They’ve begun dosing patients in an adaptive phase 1/2/3 trial that will test severe and mild COVID-19 patients as well as pre-exposure and post-exposure prophylaxis, and is intended to carry the antibodies all the way to approval. If things go as they expect, approval may come as early as the fall.
Sorrento, a startup in California, announced in May that it had identified an antibody which fully neutralized SARS-CoV-2 in in-vitro culture, and that it intended to work with Mt. Sinai hospital to incorporate it into an antibody triplet which it intends to call Covi-Shield. The triplet is intended to have very long kinetics, so that it could persist in the body for up to a year. Sorrento has the ability to manufacture 200,000 doses per month at its facility in California starting this summer, and its manufacturing partners in China have capacity for up to 10 million doses per month. Sorrento intends for the triplet to enter the clinic in August and see an efficacy endpoint before the end of the year.
These are just two of the many antibody therapies which are in development for COVID-19. Another comes from AstraZeneca and is expected to enter the clinic in August. A startup called Distributed Bio announced last week in a preprint publication that one of its early stage antibody candidates reduced COVID-19 symptoms in immunocompromised hamsters. One of the more exotic entrants is Sab Biotherapeutics, which intends to manufacture a polyclonal antibody mixture against SARS-CoV-2 using cows genetically modified to have a humanized immune system. While other antibodies usually originate new sequences with animal challenge studies and modify them for optimal properties before manufacturing them in a cell culture, Sab intends to let a polyclonal antibody mix from a challenge study function directly as a therapeutic, manufacturing it in a herd of free-range cattle donating antibodies by plasmapheresis in the same manner as human plasma donors. The industry is truly trying everything.
While Regeneron and Sorrento are currently the leaders in the synthetic antibody area, the use of natural human antibodies may make a large impact in the clinic before either one. The method of administering plasma from recovered patients originated in the late 19th century, before molecular genetics, genetic engineering, and monoclonal antibodies. It has been used with some success in the 1918 flu pandemic, the 2003 SARS epidemic, and other challenging disease environments. Notably, if it proves successful, manufacturing could theoretically scale very quickly, because of the high preexisting capacity for plasma donation and the refining of intravenous immunoglobulins, which are already administered to millions of patients for immunosuppression and other conditions. This would depend on recovered COVID-19 patients’ willingness to donate, of course.
Convalescent plasma has already undergone one randomized controlled trial, in China. This trial, published in the Journal of the American Medical Association, was not able to accrue its intended number of patients because it accrued in the midst of a rapid reduction in the number of COVID-19 cases in China. The patients on convalescent plasma had better outcomes than the control patients, but the results were not statistically significant and could have been due to random chance. Similar trials have already begin in the United States and Europe, and are intended to read out soon. If they read out well, convalescent plasma could be used against COVID-19 at some scale shortly thereafter.
Vaccines: The most important COVID-19 drugs of all
Of all categories of COVID-19 therapies, by far the most transformative in potential are vaccines, which have the potential to end the pandemic, possibly permanently, by themselves. Of our three scenarios for the long term, one of them is entirely defined by the success of vaccines. Unless we manage to eradicate SARS-CoV-2 solely with public health measures, which seems like a remote possibility at this point, humanity will be destined to attain herd immunity either by vaccination or by a big burn. That is, the relative timing of vaccine development and the pandemic will decide by itself whether tens of millions live or die.
This humanitarian imperative, and the floodgate of funding opening up from governments, philanthropists and investors, have combined to result in an astonishing profundity of vaccine development projects coming from seemingly every large pharmaceutical company, plus a large number of smaller firms, startups, research institutes, universities, and government laboratories. They are based on many different underlying technologies, from tried-and-true options like dead viruses, expression of viral proteins on harmless viruses, and recombinant proteins, to exotic options like engineered cells and raw nucleic acids. At least a hundred and thirty vaccine candidates have entered development, several already have efficacy readouts in animals, and at least twelve have already entered human clinical trials.
These projects are accelerating many of the phases of vaccine development to put a vaccine in the field potentially six times faster than any prior campaign in global history. Among many other time-saving measures, many have already committed to cumulatively billions of dollars worth of manufacturing “at risk,” i.e. speculatively on behalf of vaccines which have not yet been proven. It’s difficult to know which projects will succeed on safety and efficacy enough to reach approval, and which will reach these milestones and scaled manufacturing fastest.
But governments around the world are already placing their bets, jockeying to sign deals with vaccine concerns for tens and hundreds of millions of doses of the vaccines they think will land first. The USA has already agreed to pay $1.2 billion to Astrazeneca for 400 million doses of its vaccine (notionally 100 million in 2020 and 300 million in 2021), while the EU has agreed to pay $843 million for 300 million doses. Israel has agreed to purchase in bulk from Moderna. And this competition has already led to tension between the governments of major powers, as the USA and France have sparred verbally over the rights to early lots of vaccine from Sanofi, a French pharmaceutical giant with a large research arm in the USA.
With such a large number of projects underway, it would be impossible for us to summarize all of them, but we’ll do our best to go through some salient points about the six about which most information is available, and which have progressed furthest through the development process. They are the projects of Moderna, CanSino Biologics, AstraZeneca/Oxford, Pfizer/BioNTech, Sinovax, and Novavax. While many more projects are underway, these six are emblematic of the general categories of vaccines into which most of the rest will fall, and they will yield information first about how availing each line of inquiry is likely to be.
Oxford/AstraZeneca: ahead in the horse race
The Oxford/AstraZeneca candidate has grabbed a great deal of attention because it’s arguably in the lead for first deployment. The team at Oxford which developed it had already been developing a MERS vaccine, and made the decision to switch to SARS-CoV-2 vaccine development very early in the pandemic. They inked their production deal with AstraZeneca swiftly, including an agreement to manufacture at risk at a massive scale. As a result, they were able to begin dosing humans among the earliest of all vaccine efforts, and estimate they will have an efficacy readout and mass production at risk as early as September 2020, raising the possibility that they will be deploying a significant (but short of global scale) number of doses of an approved vaccine in 2020. AstraZeneca and its manufacturing and packaging partners, including Catalent, CEPI, GAVI, and the Serum Institute of India have said they ultimately intend to build capacity to produce two billion doses per year. There has been discussion of producing about 800 million doses before the end of 2020, and many countries have already made agreements to buy almost a billion doses. This is a campaign to vaccinate the world.
The vaccine candidate itself is an adenovirus vector, a harmless engineered virus which is geared to attack dendritic cells. Dendritic cells are a type of antigen-presenting cell (APC), whose function in the human immune system is to take up genome fragments of pathogens and express them on its cell surface, thereby amplifying antigen proteins and entraining the rest of the immune system against them. Infecting these cells with a virus which delivers a genome fragment directly to them theoretically allows these vaccines to get the immune system’s attention using its own mechanisms.
Animal trials of the Oxford candidate have shown that it elicits a robust neutralizing antibody response in both mice and rhesus monkeys. A challenge trial in monkeys showed that while vaccinated monkeys were capable of getting infected with SARS-CoV-2, they were uniformly asymptomatic and exhibited low viral loads and short infection timelines. This has brewed a controversy, with some analysts suggesting the Oxford candidate may be only partially effective.
The AstraZeneca team have already finished dosing a thousand patients in their Phase 1 trial, and although the results are not yet public, the safety data satisfied regulators enough for them to allow a phase 2/3 trial in over 6000 patients,which is already enrolling and whose efficacy results AstraZeneca hopes will lead to approval this fall.
If you get vaccinated this year, you’ll most likely get this vaccine.
Cansino: Another adenovirus candidate
The CanSino candidate is also based on an adenovirus vector similar to the Oxford one. While CanSino has not announced manufacturing plans, its clinical program initially moved along at a comparable speed to AstraZeneca. Its first Phase 1 trial has already been published in the Lancet, the first COVID-19 vaccine trial results to pass peer review. It showed that in 108 patients, mild and moderate side effects like injection site pain, mild fever, fatigue, and headache were common, but no serious side effects were observed. They observed both neutralizing antibodies and T-cell responses, which is a good sign.
CanSino announced a Phase 2 trial in April, but it’s not clear if it’s begun dosing yet. It was set to occur in Wuhan, which now has no cases. Unlike the Oxford Phase 2/3 trial, this trial is smaller and not intended to yield an efficacy readout. CanSino has not announced a firm plan for the timing of an efficacy readout. However, they have announced that they are also pursuing an RNA-based vaccine with Canadian partners, leading some to speculate that their adenovirus platform may not be bearing fruit as expected.
Moderna and Pfizer/BioNTech: RNA vaccines
Two other pharmaceutical companies are pursuing RNA vaccines. Invented recently, these also work by directly targeting dendritic cells, but unlike adenoviral vectors, they present RNA in the blood to be taken up by dendritic cells in the same manner as they would take up genome fragments from a pathogen. There are veterinary RNA vaccines, but no RNA vaccine has ever been approved in humans.
The Moderna candidate is an mRNA corresponding to the SARS-CoV-2 spike protein, which is a common target for SARS-CoV-2 vaccine and antibody projects due to its uniqueness and envelope surface presentation. Moderna was the first to enter a human clinical trial for COVID-19, aided by the rapidity of small scale manufacturing of RNA, which is one of the fastest and most flexible manufacturing operations of modern biotechnology, due to the legacy of the nucleic acid supply chain. However, it may be harder for Moderna to scale to global scale because there is no legacy of RNA vaccine production. Initially, Moderna planned to scale to 100 million doses per year, but later signed a deal with Lonza to partner on capacity of up to 1 billion doses per year. Subsequently, however, Moderna acknowledged it probably would fall significantly short of this capacity, due to a combination of difficulty scaling manufacturing and the need for a higher dose.
Moderna’s candidate has passed a series of early trial results. It demonstrated safety and the production of neutralizing antibodies in a Phase 1 human trial. At 100ug, the highest dose level used, Moderna said, the antibody titer from its trial patients was even higher than that observed in recovered COVID-19 patients. The dosage is a significant concern, because the higher the dosage, the fewer doses Moderna will be able to make. In addition, the Moderna candidate offered full protection from infection in a mouse challenge model. Challenged mice did not become infected, a more robust result than the Oxford monkey results, although not in a directly comparable system. Phase 2 trials are now enrolling.
Moderna has announced its vaccine will enter a Phase 3 trial intended to facilitate approval, dosing first patients in July. The Phase 3 trial will use the highest dose from the Phase 1 trial, 100ug, and read out some time in the fall. While this timeline is not as aggressive as AstraZeneca’s plan to have an efficacy readout in September, it is compatible with approval in the fourth quarter of 2020.
BioNTech, a German startup, has a partnership with Pfizer to develop a COVID-19 mRNA vaccine. Unlike Moderna, which committed early to a strategy of using a single spike protein RNA, the Pfizer partnership is advancing four candidates to a simultaneous joint Phase 1 trial. The four use different RNA formats and express different targets, and Pfizer intends to narrow them down to one before approval. The first patients have already been dosed in a Phase 1 trial, and a Phase 2 has been registered, but no results are yet available. Pfizer hasn’t announced an estimated time for an efficacy readout.
Pfizer intends to manufacture the vaccine internally, but has outsourced manufacturing of some legacy products to free up resources to scale up mRNA vaccine manufacturing. It has announced that it intends to produce “millions” of doses in 2020, and “hundreds of millions” in 2021. BioNTech’s own manufacturing sites in Germany will allow it to produce additional doses separately from Pfizer.
Sinovac and Novavax: Protein and inactivated virus vaccines
While adenoviral vaccines and RNA vaccines—flashy new technologies—have grabbed headlines with achievements like being first to hit the clinic, likely first to get an efficacy readout, being able to test many vaccines in parallel, and setting themselves up to manufacture billions of doses, they’re also novel and untested. Much more proven in human clinical practice are more traditional vaccine modalities like inactivated viruses and recombinant proteins.
Instead of harnessing the immune system in a targeted way by putting a precise sequence directly into dendritic cells, these more traditional methods present an entire protein or even a complete virus, and let the human immune system run its course from end to end. Sinovac and Novavax are two of the first vaccines in these categories to begin yielding data we can examine, but will be are far from the last. A number of others, including some backed by pharmaceutical giants and cutting edge research institutes, will enter trials soon.
Sinovac, a Chinese biotech firm with commercialized vaccines against hepatitis and the flu, is developing a candidate vaccine based on an inactivated SARS-CoV-2 virus. This method is in some sense the most conservative. It has been the basis of proven vaccines for well over a century, and it’s the closest analogue to a live virus, offering the immune system both RNA and protein to develop against, and all the viral proteins, not just a chosen portion of one protein.
Sinovac has some notable successes. Its vaccine passed a challenge study in monkeys, preventing replication. Some have compared this result favorably to the results of the Oxford vaccine in monkeys, but there were some important differences that make it hard to compare directly. Oxford dosed the monkeys twice, and Sinovac, only once, while Oxford used a higher challenge dose and tested its monkeys more thoroughly for SARS-CoV-2. In addition, Sinovac has announced that in its Phase 1 trial in China, the new vaccine was safe and induced a neutralizing antibody response in over 90% of patients.
On the strength of this result, Sinovac is proceeding with Phase 3 trials in both China and Brazil, and building a new factory with a planned capacity of 100 million doses of its coronavirus vaccine per year, in addition to potential use of capacity in existing facilities. In theory, Sinovac could also announce manufacturing partnerships. No estimated dates have been announced for efficacy readout, approval, or the beginning of mass manufacturing.
The Novavax candidate is a protein vaccine based on a recombinant partial spike protein stabilized by attachment to a nanoparticle, coupled with their own adjuvant designed to increase immune response. Protein vaccines often use adjuvants. Sponsored mainly by nearly $400 million from CEPI, Novavax has already initiated a Phase 1 trial of its new candidate, and intends to give a safety readout and proceed to Phase 2 in July. It has also bought a vaccine factory in the Czech Republic and begun adapting it to manufacture its vaccine product, with the intention of being able to manufacture up to a billion doses per year.
The overall picture of COVID-19 therapies: There is a future
So, putting this all together, we can say a couple things about the likely course of future therapies for COVID-19.
Firstly, repurposed drugs are already making a difference in the treatment of COVID-19 and the prevention of COVID-19 deaths, and it’s likely that as time advances this improvement will grow. Just the wider deployment of remdesivir and dexamethasone is likely to help, and they may work cooperatively. Famotidine, TMPRSS2 inhibitors, and others as yet unknown may help further.
Secondly, antibody treatments may make a significant difference in 2020, particularly the Sorrento cocktail and convalescent plasma, both of which may be available in the fall. These are more likely to make a difference if the number of patients involved in the pandemic doesn’t escalate too much more, due to manufacturing constraints.
Thirdly, there is reason for optimism about the future of vaccines for COVID-19. There is a realistic prospect of them arriving at scale in 2020, and even if not, the profundity of diverse projects means scaled deployments in 2021 are likelier still. If even a couple of the many large projects bear out, 2021 and 2022 could see the planet awash in a wave of vaccines sufficient to immunize most humans and bring the pandemic to a definitive end.
That is, there is a future. The “sideways until therapies” scenario we anticipated in March, wherein humans keep the pandemic under control with most humans uninfected using non-pharmaceutical interventions until therapies arrive to save the day remains plausible. Effective vaccines are likely coming, if we can hold on until they arrive, and repurposed drugs and antibodies will help us in the interim.
Part 3 of this series will discuss the events that have elapsed since the pandemic went global at scale in March, and what they mean for the likelihood that we will succeed in holding off the pandemic until the arrival of new therapies.