Reflections on pivoting our vaccine strategy


Reflections on pivoting our vaccine strategy

By A/Prof Michelle Ananda-Rajah MBBS (HONS) FRACP PhD, Consultant Physician Infectious Diseases & General Medicine, Melbourne, Monash University; 15 Jan 2021


Ideal characteristics of a pandemic vaccine:


  • Safe.
  • Highly efficacious in preventing infection.
  • If it does not prevent infection, it reduces severity of infection
  • Short interval between first and second dose.
  • Feasible storage conditions (cold chain).


There are three vaccines of relevance to Australia with the following with key characteristics:


  • Astra Zeneca (AZ) interim findings: 62.1% efficacy determined 14 days after the second dose, (2 standard dose schedule given at an intended schedule of 28 days apart, with the UK regulator approving dosing of up to 3 months apart; Storage at fridge temperature); n=8,895 people in the standard dosing group from the UK and Brazil; 18 years or more. Australia’s order is 53.8 million doses for the vast majority of our population. Made on-shore by CSL, who have indicated they do not have capacity to make any other vaccines in 2021. Cheap at €1.78 per dose. The AZ vaccine is an adenoviral vectored vaccine which delivers genetic code for the spike protein within a chimpanzee derived adenovirus shell that cannot replicate in the body.


  • Pfizer: 94.6% efficacy determined 7 days after the second dose, (2 doses, 21 days apart; stored at -70C +/- 10C then up to 5 days at fridge temperature); n=43,448 people 16 years or more, from the US, Argentina, Brazil, Sth Africa, Germany, Turkey. Australia’s order of 10 million doses for 5 million high risk people. Cost €12 per dose.


  • Moderna: 94.1% efficacy determined 14 days after the second dose, (2 doses, 28 days apart; storage -25 to -15C then up to 30 days at fridge temperature); n=30,420 from the US; 18 years or more. Australia’s order is nil (except through Covax, which Australia is part of). Cost USD18 per dose.


    • The mRNA vaccines deliver synthetic messenger RNA encoding the spike proten wrapped in lipid nanoparticles to the cell which then manufactures the spike protein and elicits an immune response.


Pre-purchasing agreements show a diversity in vaccine procurement noting that not all countries will exercise their right to purchase their reserved allocation (accessed Jan 14).

Image removed.

The following table is derived from the above (in millions of doses) with addition of the EU’s recent announcement of 300 million additional Pfizer doses:


Setting    Pfizer    Moderna    mRNA (Pfizer, Moderna)    AZ    Novavax
EU*         600       160            760                                     400    100
USA        200       200            400                                      500    110
UK             40        17               57                                     100      60
Canada*    20        40               60                                        20     76
Mexico*     77.4       0               77.4                                     34.4    0
Australia   10           0               10                                        53.8   51
Chile         10           0               10                                        14.4     0
Sth Korea  20         20              40                                         20       0
Taiwan       10           0              10                                         30       0
Costa Rica* 3           0                 3                                           1      0


*These settings have a higher ratio of the mRNA vaccines (Pfizer, Moderna) compared to the AZ vaccine.


All three trials had similar endpoints focussing on symptomatic infection. The AZ trial had a slightly more restrictive definition based on fewer symptoms compared to the Pfizer and Moderna trials which allowed a broader range of symptoms plus PCR positivity. These slight differences in endpoints are not likely to explain the differences in vaccine efficacy seen between the AZ and mRNA vaccines.

It is unlikely that the AZ vaccine, even with final results due at the end of Jan-March will achieve close to the same effectiveness of the mRNA vaccines with two standard doses. The 90% effectiveness reported with the low dose/standard AZ vaccine was confined to a small subgroup of people in the clinical trial (n= 2,741) who accidentally got two different doses. This finding generated some excitement but should be interpreted with caution. It was a dosing accident that led trial investigators to seek and obtain approval from the UK regulator before combining results with the standard dosing group. It may be statistical noise and needs confirmation in a larger trial. The pooling of data from two different dosing groups has attracted criticism because the groups were different with respect to age (the low dose group had younger people 18-55 years), they received the second dose a little later (median 84 vs 69 days) and the vaccine dosage was obviously different. Thus, a vaccine effectiveness of 62.1% represents two standard doses of the AZ vaccine as is being currently administered in the UK. The AZ vaccine also had relatively few older people in the trial, so there is less certainty about efficacy in older people.

Several commentators have suggested that giving one type of vaccine one year followed by another type the next year might be feasible, but there is no data on this. The effectiveness and safety of a mixed dose schedule is unknown and would be a precedent, emphasising the importance of having a robust vaccine strategy from the outset.

The large difference in effectiveness between the AZ and mRNA vaccines is striking but poorly understood. mRNA technology is new and this finding will prompt further research. We do know that humans are naturally exposed to adenoviruses, and pre-existing immunity to adenoviruses may render the adenovirus vectored vaccines less effective. The AZ vaccine uses a chimpanzee adenovirus to try and overcome this problem, but cross-reactivity may still be an issue, as well as the second dose (because the first dose may generate antibodies to the adenovirus). This means that the immune system, retaining a knowledge of adenoviruses, may attack the adenovirus carrier and reduce the load of vaccine particles making the vaccine less effective. The mRNA vaccines, tested in much larger group of people (n=43,448 for Pfizer vs n=30,420 for Moderna) than the AZ standard dosing cohort (n=8,895), with more data on older people, had consistently high vaccine responses of 95% (Pfizer) and 94% (Moderna) in preventing symptomatic infection. The size of the mRNA trials compared to the AZ trial lends confidence to high vaccine efficacy as does the fact that these findings were seen across two large independently conducted trials. We do not have data on long term immunity or safety from these trials because follow up was short at 2 months. However, given the speed and scale of vaccine roll out globally, this data is rapidly accruing.

Reassuringly, all three vaccines appear to prevent severe infection with consistently high effectiveness also seen in non-white subgroups among the mRNA trials. This has implications for Australia where non-white populations (like elsewhere) are at higher risk for worse outcomes and often employed in service industries, including aged care, transport, retail and hotel quarantine.

How well clinical trial results apply to the broader population outside of the trial is very important. The mRNA vaccines were tested in a more gender balanced diverse population compared to the AZ standard dosing subgroup which was skewed towards healthcare workers (74% overall), women (59%) and younger age. The higher proportion of elderly people in the mRNA trials (42% for Pfizer vs 25% for Moderna vs 16% for AZ) is highly relevant to Australia which has an ageing population where 16% are 65 years or more. The elderly along with other vulnerable groups (i.e. patients with cancer or impaired immune systems, First Nations people) must be ring fenced with the most effective vaccine because they do not have the physiologic, social or economic reserve to withstand COVID-19. Important chronic diseases like diabetes was also higher in the mRNA trials at 8.3% in Pfizer vs 8.4% in Moderna (compared to 2.7% in the AZ trial), more closely representative of the 5% prevalence in Australia. Both mRNA trials were enriched with people at high risk of severe infection and found to be effective even in these subgroups. Non-white populations were similar for all trials at 20% for AZ and 17% each for Moderna and Pfizer but the Pfizer trial recruited from more geographically diverse countries (US, Argentina, Brazil, South Africa, Germany, Turkey). Ethnic diversity is relevant to building vaccine confidence in culturally diverse communities, relevant to Australia where 26% of the population was born overseas (2016 census). Good vaccine uptake among ethnically diverse or marginalised groups is important because this pandemic has shown that when our minority or vulnerable groups are not protected, everyone’s lives and livelihoods are jeopardised.

Preventing infection is important because SARS-CoV-2 is transmitted by people who are unknowingly infected, accounting for 40-50% of transmission. Unfortunately, infection without symptoms was not well examined in the clinical trials but we do have some clues. The AZ vaccine at standard dosing (the schedule approved in the UK) did not prevent asymptomatic infection; in the smaller low dose/standard dose group, there was some reduction seen (0.6% with vaccine vs 1.5% with placebo) but it may be a chance finding. Of interest, there was reduction in asymptomatic infection in the Moderna trial. This was measured by a surveillance swab taken just before the second dose, with 39 positive for SARS-CoV-2 in the placebo group vs. 15 in the vaccine group, noting that this was seen after only one dose which is encouraging. There are some interesting data emerging from Israel who agreed to share patient-level data with Pfizer in exchange for weekly vaccine consignments. Early reports from Israel on 800,000 people suggest that the Pfizer vaccine reduces infection by 33%-60%, 14 days after the first dose. Pfizer are currently considering providing vaccines to Iceland for a herd immunity experiment (Jan 12). Arguably, Australia with its more ethnically diverse population would be well positioned for similar negotiations should it desire.


Despite the unknowns, herd immunity is not likely attainable with the AZ vaccine even with 100% uptake. This uncomfortable scenario is dawning in the UK. The Financial Review (Jan 14) reported that vaccination (referencing the AZ and Pfizer vaccines) may not stop disease transmission or the lifting of restrictions noting that this puts pressure on the British campaign which like Australia, is heavily reliant on the AZ vaccine (100 million doses). A highly efficacious vaccine like Pfizer and Moderna products, may help us attain the holy grail of herd immunity and this has become more pressing with the emergence of highly contagious strains globally. We will need to wait for the final AZ clinical trial results due in Feb-March but it would seem unlikely that efficacy will jump from interim findings of 62% to 94-95% or above.

The AZ vaccine may be an option for low to middle income countries that cannot afford better vaccines, but a country like Australia should consider pivoting to the more efficacious mRNA vaccines. We should also be seeking contracts with several other vaccine manufacturers, as many more good vaccines will become available through the year. There are over 60 in late stage development and over 170 in early stage development.  Currently Australia’s only other agreement is for Novovax, a protein subunit vaccine for which the data will be available later in 2021.

At a minimum Australians expect that vaccination will restore some semblance of normality to their lives, but this is unlikely with the AZ vaccine. It speaks volumes that the mRNA vaccines have been preferenced by several countries including Singapore, Israel, Canada, the European Union and indeed Costa Rica and Mexico who are managing ultra-cold chain logistics under trying circumstances in many cases. It remains to be seen whether the US increases its Pfizer or Moderna procurement with the Biden administration. The existing 53.8 million doses of AZ vaccine Australia has procured could be deployed to our regional neighbours as a gesture of good will.

Pfizer expects to produce 2B billion doses (announced 12 Jan) and Moderna, 600 million to 1 billion doses in 2021 suggesting that supply may not be the barrier to procurement we once believed. On Jan 13, Pfizer CEO Albert Bourla in a CNBC interview said “that we have much more [vaccine] than they [countries] can use right now”. It is not a given that Australia is taking vaccine from more deserving countries, when production is clearly ramping up and our consignment would be a modest 40-50 million doses. Furthermore, Australia is in the enviable position of being able to wait for the more effective mRNA vaccines after first vaccinating our essential workers and vulnerable groups. The term “essential workers” warrants re-definition, as it is far wider than health care, border control or hotel quarantine. Education, emergency services, transport, retail and large industry are all essential to a strong Australia.

Pivoting vaccine strategy would not be first for Australia, with the European Union only recently (Jan 8) announcing that it had purchased an additional 300 million Pfizer doses taking its consignment to 600 million, in response to community pressure. With good uptake, as is feasible in Australia, preventing sustained community infections with the highly efficacious mRNA vaccines may be within reach, putting Australia on the road to a speedy economic and social recovery. The additional expense needs to be seen in the context of all vaccines being sold relatively cheaply compared to routinely used vaccines (the meningococcal vaccine is AUD250-500), with overall costs likely eclipsed by the over AUD170 billion we spent on the pandemic in 2020.

Aside from preventing severe disease which all vaccines should do, the mRNA vaccines will prevent more breakthrough infections and this should translate to higher productivity with fewer days off work, fewer long-term complications such as chronic fatigue or cardiac complications with societal and economic benefits. The alternative with more infections, is pulsed restrictions and ongoing outbreaks with social and economic disruption. Instead of limping out of economic recovery, the best performing vaccines will help us accelerate out of this pandemic. There are always trade-offs however, and Australia needs to weigh up rapid vaccination with the less efficacious AZ vaccine vs waiting for something better.  Stakeholders including industry partners and community groups should be included in these discussions.


Arguments around prevention of death and hospitalisations, though very important, fail to account for Australia’s outlier status. Compared to several peer countries, Australia does not have rampant infections, rationed care and a collapsing healthcare system thanks to a strong public health response, high trust in public institutions and a white hot streak of civic mindedness in our community. For our efforts, we have been rewarded with thinking time to re-evaluate our strategy. After some labour pains, the government announced (Jan 14) that they will “continue to ensure that we have a whole of population protection, at the same time as pursuing herd immunity” with no guarantees of long-term durability or a time frame. Our North Star should be herd immunity and a reduction of sustained community transmission. This would align with community expectations for a vaccination program tailored to our unique context. Procuring the most effective vaccines and diversifying our portfolio will help us get within striking distance of our North Star. We may not make it but we have a duty to try.


Add new comment

Restricted HTML

  • Allowed HTML tags: <a href hreflang> <em> <strong> <cite> <blockquote cite> <code> <ul type> <ol start type> <li> <dl> <dt> <dd> <h2 id> <h3 id> <h4 id> <h5 id> <h6 id>
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.