Don’t hype the hope for Oxford-AstraZeneca Covid-19 vaccine


AFTER months of hype, the world finally has human trial data from a front-running vaccine collaboration between the University of Oxford and AstraZeneca plc. Spoiler alert, it’s good news.

The data, published in The Lancet on Monday, showed that the vaccine produced an encouraging immune response. Just as crucially, perhaps, no significant safety issues emerged. Investors took these developments as a cue to bid up AstraZeneca shares as they did with Moderna Inc’s stock last week on its positive vaccine news. And there are indeed elements of the Oxford-Astra shot’s profile that may make it especially promising.

But it’s important to remember that this early stage in the process, every piece of vaccine data is still just part of a thesis that needs confirmation.

The piece of data garnering the most attention is the vaccine’s “dual immune responses” — its ability to produce both an antibody and T-cell response in volunteers.

There is some evidence that antibodies may decline over time in recovered Covid-19 patients and that T-cells — a type of immune cell that can remember and hunt viruses — may be key to durable protection. This theory is one possible explanation for the lack, so far at least, of many reports of reinfection, even as antibodies have declined.

While evidence of a T-cell response is undoubtedly better than the alternative — new data from Pfizer Inc and BioNTech SE’s vaccine collaboration and CanSino Biologics Inc on Monday also produced such data — there are still tremendous gaps in knowledge. It’s unclear, for example, what exactly declining antibody levels might mean. Individuals may still be primed to generate new antibodies even after measurable levels fall.

On the T-cell side, scientists know little about the longevity and protective abilities of natural responses and even less about what vaccine developers should be measuring. Immune responses measured in the lab don’t always correlate to real-world protection, a risk that’s especially acute for rapidly developed vaccines against a novel virus.

There isn’t enough data to declare one vaccine effort firmly ahead of others. We know too little, and cross-trial comparisons are fraught; different research groups measure different things in different ways. The only answer to these questions will come from big real-world confirmatory trials, something I’ve said many times and will keep repeating until the data arrive.

The same uncertainty holds for safety data. This vaccine and others have been tested in mostly young, healthy and undiverse groups so far. Safety and efficacy in a broader population, including older adults and people with health issues, will be crucial in determining how useful they are.

With that cautionary note, there is some good news. The shot is already in large-scale trials in the UK, South Africa and Brazil that could generate more evidence in the next few months even as some rivals are still designing studies.

It’s not clear exactly when this data will arrive; the trial in the UK is the largest and started in May, late in the country’s lockdown as infections were on the decline. Data may be slow to accrue. The trials in Brazil and South Africa don’t have that problem, though both smaller efforts started later.

It’s unclear whether those three groups add up to a data package that will meet approval standards in the US or Europe, which is likely why AstraZeneca plans a further large study in the US starting in August. However, the need is such that the vaccine could see limited early use if the results are very compelling. At the very least, it would provide more secure grounds for optimism and scaling up manufacturing capacity, while waiting for more information.

It’s tempting to leap on every piece of vaccine news as a firm step forward or the clincher for a preferred candidate. At this stage, excesses of both optimism and odds-making can get both investors and policymakers in trouble. — Bloomberg

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.