This is one of the challenges with using a carrier for vaccines, often a strong immune response can be generated towards the carrier rather than the target portion of the vaccine. This would inevitably mean a strong immune response towards the natural infection by the carrier. Also, this may mean that use of that particular carrier in any future vaccines would be hampered as the immune response may clear those particles too quickly for the immune response to generate a response against the new target molecule

Here's an example of a study using a Salmonella carrier for a HIV vaccine.Article Safety and immunogenicity of attenuated Salmonella enterica ...

Here's a review of mine looking at different vaccine technologies

Article Novel approaches for the design, delivery and administration...

Thank you much Mr. Wallis for the reply. Indeed, carrier use not only changes the immunity against those, otherwise not very dangerous, bacteria/viruses, but as you mentioned endangers using the same technologies for future vaccines. However, answering that question also will be interesting looking from already adenovirus vaccinated people reaction against those as a carrier. In US, military recruiters are vaccinated against adenovirus. Would be interesting to check their response during the time agains a carrier.

The major challenge in the use of viral vectors has been the presence of pre-existing immunity against the vector, particularly against Adenoviruses.

Most people have been exposed to Adenoviruses, which cause common cold. Additionally, due to development of immunity to the vector after first dose, prime-boost strategy for multiple doses from different vectors/platform

may be required, for example, the Russian COVID-19 vaccine uses two different Adenovirus strains for primary and secondary doses. To avoid pre-existing antibodies to human Adenoviruses, chimpanzee-derived adenovirus vector (ChAd) has been used for one of the COVID-19 vaccines. Viral vectors can cause inflammatory responses or innate immunity which can be a problem with COVID-19 patients, which is a disease of hyper innate immunity.

Whether to use replicating or non-replicating vectors? In principle, non-replicating vector pose lesser safety risks than replicating vector, but very high doses for replicon defective vectors are required, for example a dose of the Russian Adenovirus based COVID-19 vaccine is 10^11 viral particles (100 billion virus particles in comparison to a few million virus particles for the live polio vaccine). This requires growing large quantities of vectors in cell cultures.

As you mentioned, US army has been immunized with adenovirus vaccines, which would interfere in getting protection for vaccines made in Adeno vectors.