I know that there are lots of GWAS associations in regions with CNV.
Defensins, Fcy receptors, and complement proteins are all great examples of proteins whose loci contain copy number variation that is thought to contribute to disease pathogenesis.
However, I am not aware of any cases in which the CNV or SV detected was shown to account for the association reported in that locus. That seems to me to be a higher bar.
The context here is that there is a lot written about CNV and SV in the context of missing heritability. Most papers on the subject at least refer to SV as a potential source of missing heritability. However, there are other large studies that have concluded it probably is not a major source of MH.
What would be nice, however, is to show an example of a case in which a CNV or SV in linkage with a lead SNP has been shown to be the causal variant that is responsible for part or all of the association in the first place. Has that been done? I would prefer any autoimmune disease, but I am willing to take any complex disease if I cannot.