Were thrombocyte counts or bleeding events monitored at all during ChAdOx1 (AZD1222) clinical trials?

Question

Maybe I'm just missing something, but on a quick look, I find it a bit strange that since issues with thrombocyte (platelet) counts in adenovirus-vectored therapeutics have (1) been known for about two decades and (2) the ChAd3 (Ebola vaccine) trials have have looked at these over and over and the papers had something to say on the topic, albeit reassuring things like "No clinically meaningful thrombocytopenia was reported" or "None of the prolongations [of partial-thromboplastin time] were associated with symptoms or clinical features of coagulopathy"... the ChAdOx1 (Covid-19 vaccine) trials don't seem to mention such issues at all, at least in the two Lancet papers I found (Aug 15; Dec 8).

After all the brouhaha with temporary suspension of the AZ vaccine (turned permanent in some countries like Denmark or Norway), there's now a draft paper based on a survey in Norway that suggests there's even more besides the severe cases namely that

The hypothesis is that milder symptoms indicating altered thrombocyte function may affect a larger proportion of vaccinated individuals than the recently reported severe cases with thrombosis and thrombocytopenia. [...]

More than 80 000 cohort participants responding to electronic questionnaires about covid-19 vaccination and potential side effects during weeks 11-13, 2021. [...]

Four of the 3416 subjects (0.2%) who were vaccinated with a single dose of mRNA vaccine reported skin bleeding as a side effect, as opposed to 163 of 5132 subjects (3.2%) vaccinated with a single dose of the adenovirus-vectored vaccine, OR (odds ratio)=16.0 (95% confidence interval (CI) 7.5-34.1). Corresponding ORs for nose and gingival bleeding were 8.0 (4.0-15.8) and 9.3 (4.3-20.0), respectively.

This was a survey conducted on March 17 after the suspension of the AZ vaccine in Norway (on March 11), so it's possible that the respondents were influenced by the news etc. But 3.2% is a high enough percentage of side effects that some of it should/would have been spotted in clinical trials, had these particular kinds of events been measured back then.

So, my question is: were bleeding events or thrombocyte counts measured at all in any ChAdOx1 clinical trials?

Answer 1

Apparently some testing was conducted; it's mentioned on p. 49 in the MHRA assessment report, albeit not in a lot of detail:

Laboratory testing was only conducted in a subgroup of subjects up to 28 days after each dose. The incidence of decreases in white blood cells, neutrophils, and platelets was slightly higher in the AZD1222 group compared with control. However, there were very few unsolicited haematology or biochemistry adverse events reported and these were balanced between the 2 study groups.

There's also a table with a line "Thrombotic, thromboembolic, and neurovascular events"; there 4 in the treatment group and 8 in the control one.

In a (very long) EMA report [no. 19683] it's noted by one reviewer somewhere towards the end that one weird event was apparently noted in the animal trials:

In a non-clinical study in monkeys one animal (26Z) presented with very low (zero) thrombocytes (and other haematological parameters) following vaccination with ChAdOx nCOV-2 vaccine and before viral challenge. The MAH [marketing authorisation holder -- i.e. AZ] should comment on this finding in relation to the current issue of rare thrombosis and thrombocytopenia in humans following vaccination with Vaxevria.

It's not clear if that was a measurement or reporting error though... (Can a monkey even live at all with absolutely zero thrombocytes? Actually, the answer to that seems to yes.) Alas there was no MAH response included in time for the report on that specific issue.

If there's more detail somewhere else, I'd accept that as a better answer.