From what I understand, ketamine's rapid antidepressant effect was once thought to have been mediated by NMDA antagonism. However, countless NMDA antagonists have been studied for depression, and their efficacy was nowhere near the S-enantiomer of ketamine, suggesting that NDMA antagonism alone isn't responsible for the full effect.
DXM acts as a dirty serotonin reputake inhibitor before it's metabolized into DXO, which is actually an NDMA antagonist like ketamine.
By combining DXM with bupropion, the bupropion prevents the DXM from rapidly metabolizing into DXO through enzyme inhibition, leaving more SSRI-like DXM in the blood to affect the brain. Bupropion itself is a potent inhibitor of enzymes that metabolize many pharmaceuticals.
So the idea behind combining DXM+bupropion to treat depression doesn't really come from the same mechanism of action that esketamine has. It's more about the SSRI-like properties DXM has in the body as long as it isn't metabolized into DXO.
Also, combining two medications that are off-patent for a new treatment means that you can patent the new combination as if it's a new drug. I suspect that's why it's being investigated. Both DXM and bupropion have pretty good safety profiles, they're cheap to manufacture, and the combination can be patented.