Since the complete sequencing of the genome for SARS-CoV-2 in January 2020, there has been an unprecedented scientific effort to translate that knowledge into a usable vaccine for large-scale immunization. Vaccination represents our best hope to bring the pandemic to an end and return to pre-pandemic activity and life. In this column, we will present updates for four of the more promising vaccine candidates: BNT162b1/BNT162b2 from Pfizer and BioNTech, mRNA-1273 from Moderna, Ad26.COV2.S (also called JNJ-78436735) from Janssen, and AZD1222 from AstraZeneca and Oxford University.

In a recent news article in Nature, Gaebler and Nussenzweig highlighted the similarity between the vaccine development and the steps required for producing other novel drugs and biologics (Nature 2020;586:501-2). Initial preclinical studies evaluate candidate vaccines using in vitro methods such as molecular or cellular assays. Subsequent preclinical studies evaluate candidates in vivo, typically starting with rodent models and often culminating in studies involving non-human primates. Early phase 1 trials slowly escalate the dose, looking for complications. Although safety in primate models is a prerequisite for moving novel therapeutics into humans, the initial studies must be done very slowly, at very low doses that are slowly increased. This “first-in-man” stage is associated with considerable risk. For example, in 2006, a novel monoclonal antibody, TGN1412, shown to be safe in primates was given to six healthy men in England (N Engl J Med 2006;355:1018-28). All six became critically ill within hours and nearly died over the subsequent weeks (all survived). In phase 1 trials of vaccines, investigators look for expected complications (e.g., injection site pain, local swelling or tenderness), common systemic effects (e.g., myalgia, headache, or fever), or unexpected toxicity (e.g., unexplained rises in liver enzymes or liver function tests, evidence of bone marrow depression, or evidence of organ toxicity).

With completion of the phase 1 studies, the vaccine moves to phase 2: what dose works? With continued vigilance for any safety signal, investigators identify the immunogenicity (i.e., the ability of the vaccine candidate to induce an immune system-based response) with increasing doses. In these studies, immunogenicity is often assessed via measurable responses of the adaptive immune system, including the production of T helper (CD4) cells, cytotoxic (CD8) T cells, and neutralizing antibodies. The production of neutralizing antibodies is a particularly promising development for a vaccine candidate, as they tend to confer long-lasting immunity against the targeted infectious agent. With the completion of phase 2 studies, the drug innovator has determined in a small number of subjects a dose that appears safe and effective. Drug development proceeds to phase 3 clinical trials, which are required to meet the regulatory agency definition of “adequate and well-controlled trial” for drug approval. These trials may involve tens of thousands of subjects and cost hundreds of millions of dollars for a single trial. This process typically takes five to 15 years. The fact that we appear near emergency use authorization of several vaccine candidates in roughly 10 months is a testament to human achievement when enough money and intellectual firepower is directed toward a scientific goal.


BNT162b1 and BNT162b2 are lipid-nanoparticle encapsulated, nucleoside-modified mRNA vaccine candidates that are being jointly developed by Pfizer and BioNTech. BNT162b1 encodes a secreted trimerized SARS-CoV-2 receptor-binding domain (the so-called RBD) of the S protein, while BNT162b2 encodes a membrane-bound SARS-CoV-2 full-length S protein, which is stabilized in the prefusion conformation.

In a report in Nature, researchers presented initial safety, tolerability, and immunogenicity data for BNT162b1 from the phase 1/2 placebo-controlled dose escalation clinical trial (NCT04368728) (Nature 2020;586:589-93). BNT162b1 is an mRNA vaccine, a novel and previously unproven technology for rapidly developing new vaccines. The report included data from 45 participants between 18 and 55 years of age who were randomized to receive two doses of 10 micrograms, 30 micrograms, or 100 micrograms 21 days apart. Local and systemic adverse events were transient, dose-dependent, and mild to moderate. The study highlights how safety is evaluated in phase 1 and 2 trials, as the second inoculation with the 100 micrograms dose was not given because of heightened reactogenicity. SARS-CoV-2 neutralizing antibody titers increased dose-dependently. A second report in Nature documented robust CD4+ and CD8+ T cell responses (Nature September 2020).

A recent report in the New England Journal of Medicine compared BNT162b1 and BNT162b2 (N Engl J Med October 2020). The results indicated that BNT162b2 was associated with less frequent and lower-severity systemic reactogenicity, especially in older adults. Based on the data presented in this analysis and safety and immunogenicity data obtained for studies of younger adults in Germany, the companies selected BNT162b2 as the candidate vaccine.

In a November 9, 2020, press release, Pfizer presented preliminary results from the phase 3 evaluation of BNT162b2. The release observed that there were 94 confirmed cases among the 43,538 trial participants. Of these, 90% were in the patients who received placebo, suggesting 90% efficacy. The study is still enrolling and will only be completed when the number of cases reaches the prespecified endpoint of 164 cases. Pfizer has stated that they will not seek Emergency Use Authorization until they have had at least two months of safety data, so that is not expected until December 2020 or January 2021.

Pfizer hopes to have 1.5 billion doses prepared by the end of 2021. However, the vaccine requires two shots separated by three weeks, and must be stored at -70°C. Both may complicate efforts for wide distribution.


The vaccine candidate mRNA-1273 is a lipid nanoparticle encapsulated messenger RNA vaccine, which encodes the stabilized prefusion SARS-CoV-2 spike protein. A recent article in the New England Journal of Medicine presented results from preclinical studies of mRNA-1273 in non-human primates (N Engl J Med 2020;383:1544-55). Inoculation with mRNA-1273 induced antibody levels exceeding those present in the plasma of patients who recovered from COVID-19 (termed convalescent plasma). While low or undetectable CD8 T cell responses were noted with inoculation, CD4 T cell responses were observed over time. One important observation was that no detectable viral replication was noted in bronchoalveolar lavage samples two days post-challenge in seven of eight primates in both vaccinated groups. The investigators concluded, “Vaccination of non-human primates with mRNA-1273 induced robust SARS-CoV-2 neutralizing activity, rapid protection in the upper and lower airways, and no pathologic changes in the lung.”

“Since SARS-CoV-2 will likely evolve quickly in response to vaccines, there will probably not be one ‘winner’ in the vaccine race. Instead, multiple vaccines should continue their expedited trajectory to provide an armamentarium of epitopes necessary to bring the pandemic to a close.”

A second study in the New England Journal of Medicine documented an open label phase 1 dose escalation study in 40 older adults (N Engl J Med 2020;383). Older adults are a particularly important age group, as they are especially susceptible to the effects of COVID-19. Participants received two injections of either 25 micrograms or 100 micrograms mRNA-1273 28 days apart. Adverse events were generally mild to moderate, including fatigue, muscle ache, chills, headache, and injection site pain. Rapid increases in the binding-antibody responses were present after the initial injection, reaching antibody levels greater than the median level for a group of controls who had donated convalescent plasma. Strong CD4 type 1 helper T cell (Th1) responses were elicited in these participants by mRNA-1273. These data supported the use of the 100 micrograms dose for use in the subsequent phase 3 study (NCT04470427). Like the Pfizer vaccine, the Moderna vaccine likely will require two doses and refrigeration at -70°C.

In a November 16, 2020 press release, Moderna announced that a 90 of the first 95 COVID-19 cases in vaccinated patients occurred in the placebo group, suggesting 95% efficacy at p < 0.0001 ( There were no safety concerns.


Another SARS-CoV-2 vaccine currently being evaluated in clinical studies is Ad26.COV2.S (also referred to as JNJ-78436735). This vaccine candidate is a recombinant non-replicating adenovirus type 26 vector (Ad26) that encodes the spike (S) protein of SARS-CoV-2.

Preclinical studies from both in vitro and in vivo assays with spike protein variants identified an optimal candidate based on a membrane-bound stabilized form of the wild-type S protein (NPJ Vaccines 2020;5:91). These were followed by non-human primate studies that evaluated 26 possible combinations of adenovirus vectors and SARS-CoV-2 S protein variants (Nature 2020;586:583-8). The chosen candidate, Ad26.COV2.S, elicited robust neutralizing antibody production in the rhesus macaques and provided either complete or almost complete protection against SARS-CoV-2 post-challenge, as displayed by subsequent nasal swabs or bronchoalveolar lavage samples. The data provided the necessary preclinical evidence to advance development to a “first-in-man” study as described above.

Initial positive results for immunogenicity and safety from the phase 1/2 study have been published on medRxiv ( Adverse events included pain at the injection site, fatigue, headache, and myalgia, although some subjects developed brief fevers. Antibodies and T cell immunity were observed in nearly all subjects.

In September 2020, Janssen initiated the phase 3 ENSEMBLE trial to assess the safety and efficacy of a single vaccine in 60,000 patients. The trial was paused in October based on a single adverse event but resumed on October 27.

AZD1222-AstsaZeneca/Oxford University

AZD-1222 (ChAdOx1 nCoV-19) is similar to the Janssen vaccine in that it is an adenovirus-vectored delivery system of the SARS-CoV-2 S protein. Following demonstration of immunogenicity in mice and rhesus monkeys (, 1,077 volunteers participated in a phase 1/2 single blind randomized controlled trial in the U.K. (Lancet 2020;396:467-78). Pain, “feeling feverish,” myalgia, headache, and malaise were sufficiently common that the protocol was amended to include a pre-vaccination dose of paracetamol. There were no serious adverse events. Neutralizing antibodies were detected in 91% of subjects after a single dose, and 100% of the subjects following a booster dose. T cell responses were also seen in 100% of the subjects following the booster dose.

AstraZeneca launched a phase 3 study in August 2020, with a planned enrollment of 30,000 U.S.-based participants aged 18 years. A month later, AstraZeneca halted its clinical study temporarily as a result of one participant developing neurological symptoms. As of late October 2020, the FDA had authorized resumption of the phase 3 study.

Not necessarily one winner

One of the common SARS-CoV-2 variants, N439K, retains full function but sufficiently alters the spike protein to evade existing humoral defenses ( The N439K mutation has arisen spontaneously multiple times since the start of the pandemic. Since SARS-CoV-2 will likely evolve quickly in response to vaccines, there will probably not be one “winner” in the vaccine race. Instead, multiple vaccines should continue their expedited trajectory to provide an armamentarium of epitopes necessary to bring the pandemic to a close.