To me there is a natural split between 何・も just as there is between 変わり・者. も adds meaning (compared to 何 by itself). It has a distinct meaning/role in this and similar words which would also naturally split this way (いつ・も、どこ・も、だれ・も).
The speaker's accent does this の→ん change, and in both cases to show this 者・何 get switched to kana, and in both cases the whole of that "unit" gets katakana. The other parts of the words in question, (変わり、も) don't get touched, and remain in their standard formats.