Outcomes of Cytomegalovirus Monitoring in Autologous Transplantation: A Single Institution Experience

Introduction: Identification of patients seropositive for cytomegalovirus (CMV) prior to stem cell transplant (SCT) is a well-accepted practice across institutions designed to reduce a known cause of morbidity and mortality in this population, but the role of monitoring and preemptive approaches to CMV identification and treatment are controversial and not standardized in autologous transplantation. The preemptive approach necessitates the use of significant resources and requires persistent patient involvement. Patients undergoing autologous SCT are at a relatively low risk for CMV reactivation, especially those seronegative for CMV at the time of transplant. Here, we show that the necessity of routine monitoring of autologous transplant patients is of minimal clinical value.

Methods: To determine the efficacy of the CMV monitoring protocol currently in place at our institution in detecting patients who would later develop CMV reactivation and disease following autologous SCT, we retrospectively analyzed the charts of 218 adult patients between 11/1/14 and 8/1/19 who underwent transplant at St. Louis University Hospital. No patients underwent CD34 selected stem cell infusions. The protocol stipulated the following: CMV IgG/IgM and CMV DNA PCR prior to preparative regimen followed by weekly CMV DNA PCR to day +30 . We correlated the predictive ability of positive results on any of these screening tests to identify whether patients would later develop quantifiable CMV DNA PCR positivity, clinical manifestations of CMV disease, and/or require pharmacologic treatment for CMV.

Results: Quantifiable pre-BMT DNA PCR was positive in only 0.46% of patients, and 97.79% of patients were DNA PCR negative prior to transplant. CMV IgG was positive in 56.4% patients, and only 22.1% of patients in this group went on to develop a quantifiable post-transplant PCR. Of the remaining 43.6% of patients initially testing negative for CMV IgG, no patients went on to develop a quantifiably positive post-transplant PCR. Regardless of seropositivity, only 0.08% of the 1,191 PCRs performed during the study period were found to be quantifiable. Further, no patients in our cohort developed CMV disease or required CMV treatment during the monitoring period. This trend persisted despite stratification by age, diagnosis, transplant number, and preparative regimen.

Conclusion: When clinically-significant CMV is defined by cases requiring treatment or the development of end-organ disease, no screening tests performed elicited clinical action. Laboratory-based CMV surveillance, based on our data, has minimal diagnostic implications and represents an overly-stringent practice in a set of patients already utilizing a substantial share of healthcare resources. We believe that pre-transplant screening for CMV IgM serology and CMV DNA PCR can be safely eliminated in the autologous SCT population at our institution while CMV IgG still plays a role in determining candidacy for CMV-negative blood products. We also propose the elimination of serial post-transplant monitoring with DNA PCR in patients without clinical signs, symptoms, or pathologic findings suggestive of CMV disease. We have changed the protocol to test for CMV PCR only if there are clinical scenarios that indicate a utility, such as prolonged fever post-transplant, unexplained cytopenias, or unexplained pneumonitis, colitis, or hepatitis. By extension, other centers should consider determining the necessity of CMV screening in their autologous transplant population given the potential resource conservation and reduction in healthcare expenditures.