It should, but it does not always require that, at least for some soft sciences. In my very niche field (German grammar and German language instruction) studies are almost never reproduced as part of the peer review process, though it is totally normal for people to publish new papers that are essentially reproductions of existing studies.