Cytomegalovirus After Autologous Hematopoietic Cell Transplantation: Impact on Outcome and A Personalized...

Autologous hematopoietic cell transplantation (auto-HCT) is standard consolidation therapy, but the incidence and impact of cytomegalovirus (CMV) infection post-auto-HCT remain poorly defined. Current guidelines do not recommend CMV surveillance for unmodified grafts due to insufficient evidence.

To evaluate the frequency of CMV reactivation and disease in a routinely monitored CMV-seropositive population, identify associated risk factors, assess the impact on mortality and develop a rational algorithm for CMV surveillance in non-CD34⁺ selected auto-HCT recipients.

We retrospectively analyzed 1,000 CMV-seropositive auto-HCT recipients (2011-2021) prospectively monitored for CMV reactivation; 846 CMV-seronegative recipients served as comparators for survival. Preemptive antiviral therapy was initiated at standardized viral load thresholds using a risk-adapted approach. Univariable and multivariable time to event analyses were used to assess predictors of CMV endpoints and transplant outcomes, and to develop a risk-prediction model.

By day +100, CMV viremia, clinically significant CMV (requiring antiviral therapy), high-risk CMV infection (≥500 IU/mL/ ≥7 Ag or CMV disease), and disease developed in 29%, 13.7%, 4.4%, and 1%, respectively. CMV serostatus did not affect non-relapse mortality or overall survival. High-risk CMV infection or disease was associated with pre-HCT CMV viremia, prior HIV exposure, and receipt of anti-thymocyte globulin in patients with non-malignant disorders. Among patients with lymphoma and myeloma, pre-HCT creatinine (>1.5 mg/dL), hypoalbuminemia (<3.4 g/dL), and lymphoma diagnosis were significant risk factors for high-risk infection, yielding a 9.4% high-risk CMV incidence among patients with >2 risk factors versus 1.4% patients with zero risk factors. Post-transplant use of corticosteroids independently increased the risk of CMV reactivation. High-risk CMV infection was associated with non-relapse mortality in multivariable models.

A subgroup of CMV-seropositive autograft recipients continue to be at risk for CMV complications. The proposed risk-adapted surveillance provides a rational strategy to target PCR surveillance while minimizing the risk of severe CMV complications.