Ash OtterDelighted to say our #Monkeypox/#MPOX paper is now out as a preprint:
https://www.medrxiv.org/content/10.1101/2022.12.22.22283648v1
It's a bit of a beast, but a thread of the results we've found...🧵
We decided to look at individual recombinant antigens to see if antibodies induced were different between smallpox vax & MPOX infection
So we used 24 MPXV & 3 VACV antigens, with all 3 VACV being homologues of some of the MPXV antigens...the importance of that is later!
So we got a bunch serum: negative samples (paediatric or pre Smallpox vax samples), IMVANEX vaccinated (post dose 1, 2 and 3), ACAM2000 vaccinated (thanks @CDCgov!) and of course MPOX convalescent (Clades IIa and IIb) and then did 27 different ELISAs with all of these antigens:
This way we could see which antigens were useful and different at multiple timepoints after vaccination, but more importantly which ones were different between vaccinated and MPOX infected.
A lot of the antigens we used had minimal antibody binding 👇
Running all this data through a Pearson correlation matrix we see quite nice groupings, but the most interesting and significant: almost all vaccinated (except dose 1) correlate with one another and really strongly with MPOX 👇
But dose 1 Smallpox vaccinated people had quite a distinctive yet variably significant grouping (see pointer below), some correlation to negatives, but also correlation to dose 2 vaccinated or MPOX infected - which we thought was quite surprising...
Whilst we do see low antibody responses in dose 1 vaccinated, this doesn't mean there isn't protection, as two other papers have demonstrated a robust cellular response:
https://www.cell.com/cell-host-microbe/fulltext/S1931-3128(22)00559-5
https://www.thelancet.com/journals/laninf/article/PIIS1473-3099(22)00662-4/fulltext
Whilst Pearson correlation showed that there was correlation between Smallpox vaccinated and MPOX infected, we used principal component analysis to see if we could see more of a distinctive grouping between these two groups, which we did - but there was overlap
Interestingly, the ACAM2000 vaccinated people are grouped in with the MPOX convalescent as well...but more on this later...
But what we see is that the further time point post-dose 2 of a Smallpox vaccine, the PCA shows grouping towards the dose 1/negative samples... Which makes differentiation between MPOX infected and Smallpox vaccinated quite difficult!
Thankfully PCA tells you factors that contribute to a particular grouping, whereby the variables & the impact of each of those variables on the grouping can be plotted...
Here we see that a number of antigens are skewed towards the MPOX group like A27L & M1R towards vaccinated:
Using these antigens individually, we find that the MPXV antigen A27 can discriminate between IMVANEX vaccinated and MPOX infected
But ACAM2000 vaccine throws a spanner in the works, as they also show antibody binding to this antigen as A27 is missing from the MVA-BN strain (used in the IMVANEX vaccine) but present in the MVA ACAM2000 strain and also MPOX MPXV M1R antigen may also provide us with a way to determine if people have recent IMVANEX vaccination, but we're not sure why...more soon...
In addition to specific antigens being reactive in vaccinated or MPOX infected, we see that there is definitely some preferential binding Using homologous VACV/MPXV proteins, we see MPOX infected marginally bind MPXV antigens better than VACV Ag, whilst Vaccinated are opposite!
This reinforced that there are analogous antibody responses between vaccinated/MPOX infected, but definitely subtle differences in immunology responses/Ab binding
This was similarly confirmed for neuts by Luca Zaeck & colleagues in their great paper:
Low levels of monkeypox virus-neutralizing antibodies after MVA-BN vaccination in healthy individuals - Nature Medicine
Historic smallpox vaccination and monkeypox virus (MPXV) infection elicit MPXV-neutralizing antibodies, but MPXV-neutralizing antibodies are less frequent and of lower magnitude after vaccination with MVA-BN—the vaccine approved and in use for protection against MPXV and smallpox.
Using this data together we also developed a pooled Ag ELISA using commercially available antigens to study poxvirus/vax antibody responses, rather than needing whole MVA
When we looked at antigen responses over time, only some show reactivity, dropping off after dose 1 or 2.
So we made an ELISA using 5 recombinant antigens all pooled together: MPXV antigens B6, B2, E8L and A35, and one VACV antigen B5 (homologue of B6) We see it mirrors the strong antigens when tested individually:
But also then see a potentiation in responses when using the pool antigens (e.g. more antigens present for Abs to bind) compared to individual antigens:
..and this pool gives some excellent sensitivity (97.14%) and specificity (98.23%), better than individual antigens (some being ~50% sens)
Using this assay we can also perform endpoint titrations to allow quantification of responses to Smallpox vaccine or MPOX infection 👇
So overall:
• Similar Ag recognition by antibodies induced by either Smallpox vaccin. or #MPOX infection
• Some distinction between Smallpox vaccinated and MPOX infected e.g. A27 or preferential binding
• Pooled Ag ELISA is accurate and useful in measuring pox Ab responses
• Similar Ag recognition by antibodies induced by either Smallpox vaccin. or #MPOX infection
• Some distinction between Smallpox vaccinated and MPOX infected e.g. A27 or preferential binding
• Pooled Ag ELISA is accurate and useful in measuring pox Ab responses
But the main takeaway is that now we know a subset of antigens that are immunodominant in MPOX infection and Smallpox-vaccinated, so this is a bunch of data on which Ags could be used as the target for new vaccines (MPOX or broad pox) or even therapeutics like mAbs
We're now looking in further detail into Ab/Ag binding in MPOX, with comparisons to Smallpox-vaccinated, as a way to understand these subtle differences but also what these mean for pan-pox immunity.. Hopefully more soon!
This was a massive collaborative piece of work, with thanks to colleagues across @UKHSA, @EPS_Porton, @RIPL_Porton, @CDCgov, @ImperialNHS & @ChelwestFT!
Mastodon definitely needs a thread option for sure....sorry to everyone's clogged timelines!
Gabriele Pollara@asherichia well, in a way each message merits to stand on its own feet, so not that spammy. I enjoyed the thread.
I guess a solution is to pack more into each message (default limit is 500 characters, but apparently some instances go even higher!)
Dr Pieter Peach@asherichia if you’re worried about clogging people’s timelines with the follow up posts simply have the first Toot as public and the follow up toots as “unlisted”. This way people can click on and follow your thread from the beginning but don’t see the others on their timeline. #feditips #feditip #NewbieTip