Thanks. As an experiment, could you trying the same with this /etc/hosts entry?
104.92.230.170 acme-v01.api.letsencrypt.org
The thinking is that there is a history of some clients having problems with some Akamai edges when POSTing a large request body, which can be avoided by using an edge in a different region.
The following may also help diagnostically:
mtr -c 20 --no-dns acme-v01.api.letsencrypt.org --report
curl -v --data "$(dd if=/dev/urandom bs=32767 count=1 | base64)" -m 10 -H "Pragma: akamai-x-get-cache-key, akamai-x-get-true-cache-key, akamai-x-get-request-id, akamai-x--meta-trace, akamai-x-get-nonces, akamai-x-get-ssl-client-session-id" https://acme-v01.api.letsencrypt.org/acme/new-authz