New promising, but provisional, data for two of the leading COVID-19 vaccine candidates were released on Monday, prompting inevitable comparison by the market.
Caution on this front is paramount, warned AstraZeneca CEO Pascal Soriot, noting that use of different assays to measure the effectiveness of these vaccines not only makes such comparisons largely meaningless, but undermines an ideal scenario where both (and other candidates) prove effective at preventing infection.
Preliminary Phase I/II data were published on Monday for AZD1222 showing that the COVID-19 vaccine being co-developed by AstraZeneca and the University of Oxford increased levels of both neutralising antibodies and immune T-cells  that target the SARS-CoV-2 virus. The data showcase a relatively clean safety profile with no severe adverse events reported.
Elsewhere, more data for BioNTech and Pfizer's experimental mRNA COVID-19 vaccine BNT162 were released , showing it elicited strong CD4+ and CD8+ T-cell responses. Previously released results show BNT162 also increases levels of neutralising antibodies .
What it means
T-cell response (alongside antibody response) is increasingly seen as a potentially important clinical characteristic for potential COVID-19 vaccines that could confer durable immunity. Speaking to reporters on Monday afternoon, AstraZeneca's executive vice president of BioPharmaceuticals R&D Mene Pangalos caveated "we are not sure if one is more important than the other, but we are pleased to be seeing both types of response (with AZD1222)."
In terms of neutralising antibody responses, data show that of the 35 patients evaluable (from 1077 enrolled in the Phase I/II study), all showed a detectable response after both a prime and booster injection, versus 91% of patients after one injection. Pangalos confirmed that these data mean AstraZeneca and the team at Oxford are leaning towards a two-dose strategy, at least initially in a bid to secure approval of AZD1222. The levels of neutralising antibodies seen in participants receiving either one or two doses were in a similar range to those seen in convalescent COVID-19 patients.
Data from 43 evaluable patients show that a T-cell response was induced by AZD1222 in all participants, peaking by day 14 and maintained two months after injection, noted AstraZeneca. Details at this stage are relatively scant with no analysis provided of specific CD4+ or CD8+ responses, for example. Further T-cell response was not seen after a second boost injection, it was added.
The bigger picture
Announcing its own T-cell data, BioNTech and Pfizer were able to steal some thunder with detailed results, highlighting in particular the ability of BNT162 to elicit a functional CD8+ T-cell response, which was not dose dependent, suggesting effectiveness even at low doses.
Pangalos was quick to suggest, however, that cross study comparisons are essentially meaningless at this stage given the relatively small datasets available and, in the case of comparing neutralising antibody titers, the different assays that are being used.
In short, it is positive that more than one vaccine has now shown promise in terms of neutralising antibody and T-cell response, because the world will likely require multiple vaccines, added Soriot.
AZD1222 could nevertheless hold some advantages specific to it being an adenovirus vector vaccine (which uses a modified chimpanzee adenovirus to circumvent pre-existing immunity among humans).
If effective, it will be easier to produce at scale – important if the majority of developers pursue a two-dose immunisation strategy as Pangalos predicts – and easier to distribute and store versus mRNA vaccines, which require freezing instead of refrigeration. Tolerability also improves on receiving the second dose of AZD1222, a clinical characteristic that Pangalos hinted may not hold true for mRNA vaccines.
Based on data in MERS patients using the same modified adenovirus-based approach, scientists at the University of Oxford saw a sustained immune response of 18-22 months. This suggests that AZD1222 could be administered yearly or at a slightly longer interval, said Pangalos, but more data are required to show the vaccine actually works.
Initial Phase II/III results that could support approval are anticipated sometime between September and November, depending on the infection event rate in those countries (Brazil, South Africa and the UK) where patients are being enrolled, said Soriot. A Phase III trial in the US is due to start shortly.