Abstract

CMV remains the most common viral infection after AHSCT. NK and T cells provide protection against CMV reactivation. The interaction of inhibitory killer immunoglobulin-like receptors (KIRs) with target cell HLA class I molecules regulates NK cells and some T cell populations. Donor activating KIR genotype has been suggested to influence CMV reactivation after myeloablative AHSCT (

Cook et al,
Blood
2006
;
107
:
1230
–1232
). However, the effect of donor activating or inhibitory KIR genotype on CMV reactivation after T-cell replete RIC AHSCT has not been reported. We analyzed 64 consecutive patients (pts) who underwent T-cell replete matched sibling donor RIC AHSCT at our institution from 1/16/00-4/24/07 with fludarabine and low-dose total body irradiation (200 cGy: 36 pts; 400 cGy: 28 pts). All pts received cyclosporine and mycophenolate mofetil for GVHD prophylaxis. CMV monitoring was performed by polymerase chain reaction testing. 49 (77%) pts were CMV seropositive or had a CMV seropositive donor; 25/49 (51%) had CMV reactivation post-AHSCT. 15 (23%) pts were CMV seronegative along with their donors and 1 (7%) of them had CMV reactivation. Donor activating KIR genotypes (KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, KIR3DS1) and inhibitory genotypes (KIR2DL1, KIR2DL2, KIR2DL3, KIR3DL1, KIR3DL2) were determined by PCR-SSOP and/or PCR-SSP. Patient HLA KIR ligands including HLA-Cw groups C1 or C2, HLA-Bw4 and HLA-A3/11 (reviewed in
Farag et al
Blood
2002
;
100
:
1935
–1947
) and donor inhibitory KIR genotypes were analyzed. Missing patient KIR ligand was observed in 15 (23%) of the C1 group, 27 (42%) of the C2 group, 25 (39%) of the HLA-Bw4+ pts and 43 (67%) of the HLA-A3/11+ pts. There were no differences observed for CMV reactivation between any of these groups when comparing those with or without missing ligands. However, when the number of donor activating KIRs were considered a difference in CMV reactivation patterns was appreciated. Pts whose donor KIR genotype contained 5 or 6 activating KIR genes (N=16) had less CMV reactivation compared to those with only 1 to 4 activating KIR genes (N=48) (19% vs. 48%).

This finding remained significant on multivariable analysis (p=0.038). No specific activating KIR was found to be associated with CMV reactivation. There were no differences between the groups (5–6 vs. 1–4 activating KIR genes) with regards to patient/donor CMV seropositivity, age, diagnoses, gender, race, number of prior chemotherapy regimens, prior radiation, time from diagnosis to RIC AHSCT, TBI dose, donor to patient gender, CD34+ and CD3+ cell doses. We conclude that donor activating KIR genotype influences CMV reactivation after matched sibling donor T-cell replete RIC AHSCT. These results may guide the selection of donors as well as identify patients who may benefit from closer CMV monitoring and additional strategies to prevent CMV reactivation post transplant.

Author notes

Disclosure: No relevant conflicts of interest to declare.