A Large Cohort Analysis of CMV Viral Load and Ganciclovir-Related Neutropenia in Nonmyeloablative (NM-HCT) and Myeloablative (M-HCT) Allogeneic Hematopoietic Cell Transplant Recipients.

Background. The “true” nonmyeloablative allogeneic stem cell transplantation (NM-HCT) allows initial establishment of mixed T-cell chimerism for up to 6 months. The prolonged presence of host memory immune response after NM-HCT may play a role in protection against CMV infection especially early after NM-HCT. Furthermore, we previously reported that the incidence and poor outcome of ganciclovir-related neutropenia (GCV-N) is strongly associated with myelotoxic conditioning. Therefore, non-myeloablative HCT may also contribute to a lower risk of GCV-N. In this to date largest cohort we examined the risk of CMV infection and GCV-N after NM conditioning.

Methods. We retrospectively analyzed 537 NM-HCT recipients (median age 54.2 yrs) and 2489 M-HCT recipients (median age 39.8 yrs) transplanted between 1/1995 and 12/2005. The conditionings for NM-HCT mostly consisted of 2 Gy TBI with/without fludarabine. Post-grafting immunosuppression consisted of most commonly mycophenolate mofetil and cyclosporine (CSP) for NM-HCT recipients and methotrexate/CSP for M-HCT recipients. CMV surveillance was performed weekly by antigenemia (AG) or plasma PCR testing. GCV/VGCV was given for CMV AG/PCR positivity. We evaluated

1. any AG/DNA detection by day 100,

2. AG >10/200,000 PBL or PCR> 1000 copies/ml (high-grade CMV AG/DNA) by day 100 and

3. GCV-N defined as non-relapse-related neutropenia (ANC<500 μL) after AG/PCR positivity with ANC>1000 μL at time of viremia using multivariable Cox regression.

Results. There was no significant difference in the incidence of CMV AG/DNA at any level between NM-HCT and M-HCT (39% vs.37%, HR 1.1, P=0.42). However, there was less high-grade CMV infection in NM-HCT compared to M-HCT patients (9% vs. 14%, adj. HR 0.6, P < 0.01). Risk factors for high-grade CMV infection other than M-HCT were advanced recipient age, CMV serostatus, transplantation year and acute GVHD. The incidence of GCV-N was similar between NM-HCT and M-HCT recipients (28% vs. 20%, adj. HR 1.3, P=0.27). Risk factors for GCV-N were advanced recipient age and acute GVHD. On the other hand, the incidence of post-engraftment neutropenia was higher in NM-HCT (29% vs. 13%, adj. HR 2.1, P<0.0001). Other risk factors for post-engraftment neutropenia included advanced recipient or donor age, cord blood, HLA mismatched/unrelated donor, female donor to male recipient, acute GVHD, recipient CMV seropositivity and CMV infection.

Conclusion. Our results suggest that the risk of CMV reactivation is not affected by NM conditioning but that progression to higher levels of viral load is reduced. This may be due to residual host memory immunity after NM-HCT. Despite less toxic conditioning, the incidence of GCV-N in NM-HCT was similar to that in M-HCT. Notably, the incidence of post-engraftment neutropenia was more frequent in NM-HCT compared to M-HCT recipients. This may be due to higher recipient and/or donor age, lower capacity to metabolize drugs, or the use of myelotoxic co-medication such as MMF or TMP-SMX.