Viewed as a horse race, none of the four COVID-19 vaccine candidates whose developers released early clinical data this month fully broke away from the pack galloping toward regulatory approval.
However, a trio of distinguished experts interviewed by GEN, and two investment firms that track biopharma stocks, agreed that three of the four candidates showed sufficient promise in Phase I or Phase I/II studies to warrant interest from researchers, industry observers, and others in upcoming trials for:
- AZD1222 by AstraZeneca, the University of Oxford, and its spinout Vaccitech
- BNT162b1, the most advanced of four vaccine constructs by Pfizer and BioNTech
- mRNA-1273 by Moderna
- Ad5-nCoV by CanSino Biologics
All four are among the 18 “Front Runners” of the more than 270 COVID-19 therapeutics included in GEN’s updated “COVID-19 Drug & Vaccine Candidate Tracker.”
“You can’t bet on this race,” William A. Haseltine, PhD, the HIV research pioneer and chairman of ACCESS Health International, told GEN. “The vaccine that I’d bet on is the one that’s the safest, and I can’t tell you which that is at this time.”
Echoing that sentiment is Peter J. Hotez, MD, PhD, dean for the National School of Tropical Medicine at Baylor College of Medicine: “There’s no winner at this point and we’ll have many more vaccines coming through.”
Before that can happen, Hotez said, “we’re going to need to go through the hard work of a large Phase III clinical trial or multiple Phase III trials with 10,000 to 30,000 patients, to show that these vaccines actually work and they’re safe. And that’s what’s going to take time.”
All four vaccine candidates have such large-scale trials planned, with Moderna saying it anticipates launching its approximately 30,000-patient Phase III trial this month, and Pfizer and BioNTech planning to launch their own Phase IIb/III study very soon. AstraZeneca has already begun Phase II/III trials in the U.K., Brazil, and South Africa, with plans to start in the United States.
AZD1222: Seeing double
AZD1222 is a vaccine based on an adenovirus vector and the SARS-CoV-2 spike protein. After vaccination, the spike protein is produced, which primes the immune system to attack the coronavirus.
On Monday, researchers published preliminary Phase I/II data showing AZD1222 (formerly ChAdOx1 nCoV-19) to have an acceptable safety profile, and favorable immunogenicity against the virus: “The preliminary results of this first-in-human clinical trial supported clinical development progression into ongoing Phase II and III trials,” the researchers concluded in their study, published in The Lancet.
The Phase I/II trial enrolled 1,077 healthy adult participants aged 18–55 years, and assessed a single dose of AZD1222 against a comparator meningococcal conjugate vaccine, MenACWY.
But while 543 patients were randomized to AZD1222 and 534 to MenACWY, only ten patients were enrolled to receive two doses of AZD1222, a prime and a boost. All ten showed neutralizing antibody activity after receiving the booster dose.
AstraZeneca hinted that it will pursue two doses of AZD1222: “We saw the strongest immune response in participants who received two doses of the vaccine, indicating that this might be a good strategy for vaccination,” Andrew Pollard, MBBS, PhD, chief investigator and co-author of the Oxford Vaccine Trial, said in a statement released by AstraZeneca.
All the leading vaccine candidates showed better results after two doses compared with one—yet a single-dose vaccine would be optimal, Haseltine noted.
“This should be a vaccine that could be used around the world, for everybody, in rich and poor countries, as a constant. And for that, you need a vaccine in which a single dose will do you,” Haseltine said. “The reason for that, as any vaccinologist will tell you, is that in difficult to reach areas, it’s hard to get people once, it’s even harder to get them twice. Some of the vaccines are two weeks apart, some of them are 28 days apart, so it’s difficult. That’s not optimal.”
However, with the notable exception of the seasonal flu vaccine, many vaccines against infectious diseases work with initial “prime” and subsequent “boost” injections.
Among those receiving the single dose of AstraZeneca/Oxford’s vaccine, neutralizing antibody responses against SARS-CoV-2 were detected in 32 of 35 participants when measured via microneutralization assay (MNA80) and in all 35 participants when measured via the 50% plaque reduction neutralization assay (PRNT50).
“It makes you wonder, they’ve just done a two-dose study in ten people and now they’re talking about the capacity to make hundreds of millions of dollars. How about a little humility here?” said Paul A. Offit, MD, director of the Vaccine Education Center and an attending physician in the division of infectious diseases at Children’s Hospital of Philadelphia. “How about going through a Phase II trial, where you at least look at hundreds of people who got your two doses to make sure that it’s safe and to make sure that it’s consistently immunogenic?
“I don’t feel extremely confident that a paper basically that looks at ten people that were going to be getting the dose number that they’re planning on doing is in any sense a proof. So, go slowly and be a little more humble about all this,” urged Offit, who is also the Maurice R. Hilleman chair of vaccinology at the University of Pennsylvania.
Madhu S. Kumar, PhD, senior research analyst with Baird, noted in an investor report that the AZD1222 trial revealed a challenge common to all leading COVID-19 vaccine candidates, namely containing adverse events. Of particular interest is that adverse events have proven less of a problem for adenovirus-based candidates than for mRNA vaccines.
AstraZeneca and the University of Oxford addressed adverse events at two of their trial’s five sites (Oxford and Southampton) with a protocol amendment allowing paracetamol (the active ingredient in Tylenol<sup>®</sup>), to be administered before vaccination.
Researchers reported 56 severe adverse events in participants who received both the AZD1222 and paracetamol, compared with 487 among participants dosed with the vaccine alone—of which 328 (67%) were mild to moderate pain after vaccination.
mRNA-1273: “Warning sign” seen
Moderna’s mRNA-1273 is a lipid nanoparticle (LNP)-encapsulated mRNA vaccine encoding for a prefusion stabilized form of the spike protein. The vaccine candidate showed positive interim data in a Phase I study (NCT04283461) whose results were published in The New England Journal of Medicine (NEJM) on July 14, about two months after they were announced by the company—an example of how much more aggressively than ever developers have used press releases and publicity to shape public perception of their COVID-19 vaccines before publication of data, according to Hotez.
The study evaluated a two-dose vaccination schedule of mRNA-1273 given 28 days apart across three dose levels (25, 100, 250 µg) in 45 healthy adult participants ages 18–55 years, with results reported through day 57.
mRNA-1273 induced binding antibodies to the spike protein in all participants after the first vaccination, with all participants seroconverting by Day 15. Dose dependent increases in binding titers were seen across the three dose levels, and between prime and boost vaccinations within the dose cohorts.
As a result, Moderna said it will give the 100 µg dose to participants randomized to treatment in its upcoming Phase III trial. At the 100 µg dose, they reported, the geometric mean titers seen among patients were 2.1-fold higher than those seen in convalescent sera obtained from 38 individuals with confirmed COVID-19 diagnoses (and 4.1-fold above those seen in reference convalescent sera).
At the highest (250 µg) dose level, three of 14 participants (21%) reported one or more severe adverse events. A fourth severe event was a fever of 39.6°C (103.3°F) reported in a patent in the 100 µg group. Lesser solicited systemic adverse events were more common after the second vaccination.
“The Moderna vaccine knocked a couple of young healthy people on their ass, and it was only 2.5 times the actual dose they plan to use. That’s not good. That is a warning sign,” Haseltine said. “If I were the CEO of that company, I wouldn’t have proceeded until I knew a lot more about the safety profile in target populations: Older people, and people with some comorbidities, say mildly obese and relatively healthy diabetic.”
Researchers from Moderna, the National Institute of Allergy and Infectious Diseases (NIAID)—which led the trial—and their clinical research partners reported in the NEJM that the adverse events were not trial-limiting (the systemic adverse events after the first vaccination were all graded mild or moderate). They emphasized that mRNA-1273 induced rapid and strong immune responses against SARS-CoV-2: “These safety and immunogenicity findings support advancement of the mRNA-1273 vaccine to later-stage clinical trials.”
BNT162: Dose-response links vary
BioNTech and Pfizer are evaluating four vaccine constructs of BNT162 in an mRNA-based clinical program the companies have dubbed “Project Lightspeed.” The most advanced of the four is BNT162b1, a lipid nanoparticle-formulated, nucleoside-modified mRNA (modRNA) vaccine candidate that encodes a trimerized SARS-CoV-2 spike glycoprotein receptor-binding domain (RBD) antigen.
BNT162b1 and a second modRNA construct, BNT162b2, have both received the FDA’s Fast Track designation.
“What makes RNA technology attractive is that it’s easy to make. The minute that you knew the sequence of this virus in January, then you knew the genetic sequence of the spike protein, so it was plug and play,” Offit said. “All you had to do is take that generic sequence and then plug it into an mRNA strategy or plug it into a DNA strategy, or plug it into a replication defective adenovirus strategy or plug it into a vector vaccine strategy, the way we make the dengue or Ebola vaccine.”
Ease of making and scale-up are the reasons Offit predicts “those are going to be the leading candidates, and they know that will be the first vaccines that are used. It doesn’t make them the best vaccines, but I think we’ll find that out over time. It certainly makes the fastest to make vaccine.”
Pfizer and BioNTech just published initial data from an ongoing German Phase I/II, open-label, non-randomized, non-placebo-controlled, dose-escalation trial of BNT162b1 (EU Clinical Trials Registry EudraCT number 2020-001038-36). The preliminary data—from a non-peer-reviewed preprint posted on medRxiv—showed BNT162b1 to have elicited high, dose level-dependent SARS-CoV-2-neutralizing titers and RBD-binding immunoglobulin G (IgG) concentrations after the second dose.
The German study recruited 60 participants in four dose groups (1, 10, 30, and 50 μg). For each dose group, a dozen participants received a first dose on Day 1 and a boost dose on Day 22. In addition, 12 participants in the 60 μg group only received a single injection.
At Day 43, SARS-CoV-2 neutralizing geometric mean titers ranged from 0.7-fold (1 µg) to 3.2-fold (50 µg) compared to sera from patients recovered from SARS-CoV-2 infection. The sera of participants vaccinated with BNT162b1 showed broadly neutralizing activity in assays across a panel of 16 SARS-CoV-2 RBD variants represented in publicly available sequences and against the D614G strain. The German trial results also showed for the first time high levels of CD4+ and CD8+ T cell responses, indicating a strong potential for cell mediated anti-viral activity, according to BioNTech and Pfizer.
However, a team of SVB Leerink analysts led by Geoffrey C. Porges, MBBS, director of therapeutics research, observed that the German study did not show clear dose dependency of the T-cell response strength using dose levels ranging from 1 μg to 50 μg. (Even 1 μg of the mRNA can induce T-cell responses that are similar to 50 μg.)
“The authors suggested this could indicate stimulation and robust expansion of T cells might be accomplished at the lowest mRNA-encoded immunogen level. We find this phenomenon quite intriguing and not exactly convinced [sic] by the explanations provided by the authors,” the SVB Leerink analysts wrote.
Pfizer and BioNTech also disclosed encouraging immunogenicity and a favorable safety profile in preliminary Phase I/II data posted July 1 in a preprint study reporting on their U.S. Phase I/II trial (NCT04368728). The U.S. trial enrolled 45 healthy adults ages 18–55—of which 24 participants received two injections, 21 days apart, of 10 µg or 30 µg; 12 participants received a single injection of 100 µg on day 1; and nine participants received two placebo control doses.
Ad5-CoV: Disappointing data
The most disappointing data of the four lead vaccine candidates comes from CanSino Biologics for Ad5-nCoV, a recombinant vaccine incorporating a non-replicating adenovirus type 5 vector for prevention of COVID-19.
Researchers from CanSino and partners published results Monday in The Lancet from a Phase II study (NCT04341389), conducted with China’s Beijing Institute of Biotechnology, Academy of Military Medical Sciences. The results showed Ad5-nCoV to be safe, with its effectiveness varying depending on the measure.
Both doses of the vaccine resulted in significant neutralizing antibody responses to live SARS-CoV-2, the investigators reported. RBD-specific ELISA antibody responses induced by the vaccine were detected from day 14 onward, with geometric mean titers (GMTs) of 94.5 and 85.1 in the 1×1011 and 5×1010 viral particles per mL dose groups, respectively. Seroconversion of binding antibodies were seen in 96% and 97% of participants.
However, the vaccine induced seroconversion of neutralizing antibodies in just 59% and 47% of participants in the 1 × 1011 and 5 × 1010 dose groups, respectively, and positive specific T-cell responses measured by IFNγ-ELISpot were found in 90% and 88% of participants.
“The antibody and T-cell responses were both very low,” Brad Loncar, CEO of Loncar Investments and an investor in CanSino, told Bloomberg News. “I thought the data was unimpressive.”
CanSino’s adverse event rates were 9% (24/253) and 1% (1/129) at the 1E11 and 5E10 doses, respectively. Those rates are higher than those published by Moderna, though Moderna’s cohorts had lower numbers of patients. Baird’s Kumar noted: “All in all, we believe the existing clinical data from these Ad COVID vaccine trials provide a much higher safety bar for new COVID-19 vaccine entrants than that required by the mRNA-1273 data set so far.”
The need to address adverse events, and assess the vaccines in much larger populations explains why Hotez doesn’t expect one or more vaccines to emerge until at least the third quarter of 2021—which would still be a record for vaccine development, which usually takes a decade or more to achieve
“We’ll have the first vaccines come out, but they’ll eventually get replaced by potentially better vaccines over time,” Hotez said. “And that’s not unusual. We saw this happen with rotavirus, we saw this happen with homofluous influenza type B, with HPV. It’s actually the rule rather than the exception that the first vaccines are often not the best vaccines and they get replaced.
“That’s why I look upon this business of a race for a vaccine with some amusement,” Hotez said. “Be careful what you wish for, because history says that if you’re the winner, it has a built-in obsolescence and will be replaced.”
Haseltine said the scramble by biopharmas to develop a COVID-19 vaccine brought to mind the words of the Biblical book of Ecclesiastes:
“I returned, and saw under the sun that the race is not to the swift, not the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favor to men of skill; but time and chance happeneth to them all.”
