The short answer is that it does. Technically, that is a form of graft-versus-host since the leukaemic cells are kind of "self". However, cancer cells exhibit tumour (specific - I don't like this word because most of the antigens are not specific to tumours) associated antigens which may be picked up by healthy immune cells thereby why adoptive cell therapy (ACT) works. This the the short explanation. There are reasons why ACT is still limited in its usage. GVHD and off-target killing is definitely one of the main problems.

Please refer to our study on NK cell infusion.

Dear Darelleynn Oo, thank you for your answer, I understand that the NK cells are going to attack above all the leukemic cells.

But I do not get:

-How it is possible that a donor NK cell, expressing self KIR receptors, is not going to kill some healthy recipient cells that are expressing not-donor HLA? Eventhough these healthy cells do not express activating-NK ligands, they will express different HLA and as far as I know then, the KIR receptors will not inhibit the cell killing.

I will answer it based on your current questions.

There is no guarantee and as I have answered before the risk of GVHD does exist. Other factors will affect the risk and that will result in whether patients present with varying levels of GVHD (if any). I do not know what prompted you to think otherwise. Since you mentioned KIR, there are several combinations depending on whether it is activating or inhibitory and it is a matter of balance regardless of whether you are using HLA match or mismatch. Mismatched HLA will obviously carry a higher risk of GVHD. Perhaps this will best illustrate it.

Link: https://www.fredhutch.org/en/news/spotlight/imports/nk-cell-activation-through-kir-hla-interactions-decreases-leukem.html

Thanks, the link was very explanatory!