Cytomegalovirus Reactivation and Relapse Following Allogeneic Stem Cell Transplantation Using a Reduced Intensity Conditioning Regimen.

Background: Recent studies have reported CMV reactivation rates of 42% to 65% in patients treated with allogeneic stem cell transplantation using reduced intensity conditioning regimens (RIT). However, published data on RIT patients who experience CMV reactivation, are treated successfully with antiviral therapy to eliminate detection, and who subsequently develop CMV relapse, is sparse.

Methods: We performed a retrospective cohort analysis of 106 patients who underwent RIT at Tufts-New England Medical Center using a preparative regimen of pentostatin, extracorporeal photophoresis, and reduced total body irradiation, from 1997–2003. All patients received identical graft-versus-host disease (GVHD) prophylaxis, which consisted of IV cyclosporine and PO methotrexate. CMV serostatus was determined on all patients prior to transplant. All patients were screened weekly by CMV antigen capture assay after day +14. Patients did not receive CMV prophylaxis. CMV reactivation was defined as 2 consecutive positive (>2.1 pg/mL) CMV DNA measurements. CMV reactivation was treated with either Ganciclovir 5mg/kg IV daily or Valganciclovir 450mg PO BID until whole blood CMV DNA levels were no longer detectable. Patients were treated with antiviral therapy until a documented negative CMV DNA assay. Those found to have detectable CMV DNA after adequate therapy were then defined as having CMV relapse. Patients were also assessed for incidence of GVHD and mortality. Attributable mortality was defined as mortality in patients who had CMV relapse compared to those who had CMV reactivation without relapse. Fisher’s exact test was used to compare proportions, Kruskal-Wallis was used to compare means, and survival and time to reactivation and relapse were analyzed by Kaplan-Meier

Results: Of 106 patients, 49 (46.2%) were CMV seropositive prior to transplant. Twenty -five (51%) of forty-nine CMV positive patients developed CMV reactivation at a median of 43 days (range 26 – 312 days) after receiving stem cells. Among patients with CMV reactivation, 36 were MRD and 13 were MUD. Nine (36%) of 25 patients with CMV reactivation developed CMV relapse. CMV relapse occurred at a median of 16 days (range 4 – 77 days) after CMV reactivation. CMV reactivation occurred earlier among those who relapsed (median 34 days, range 26 – 70 days) compared to those who did not relapse (median 55.5 days, 27 – 312 days, p=.03). Peak viral load was significantly higher in CMV relapsers (median 55.3 pg/mL, range 14.5 to 486.8) compared to non–relapse patients (median 4.4 pg/mL, range 2.1 – 58.2, p=. 0007). There was no difference in acute GVHD in the groups (100% vs. 75%, p=.26). However, those who did relapse had a higher incidence of chronic GVHD than those who did not (89% vs. 38%, p=.03). There was no difference in median survival between non-relapse and relapse patients (13 months vs. 16 months, p=. 99). The attributable mortality rate due to CMV relapse was 23%.

Conclusions: Our results suggest there is a subgroup of patients who are at high risk for CMV relapse in the post RIT setting. Risks for CMV relapse include early reactivation and higher peak CMV viral loads. In addition, there was a higher risk of chronic GVHD in CMV relapse patients. We have identified a high- risk subset of patients who reactivate CMV for whom additional therapeutic strategies may be warranted.