Treatment for First Cytomegalovirus Infection Post-Hematopoietic Cell Transplant in the AURORA Trial: A Multicenter, Double-Blind, Randomized, Phase 3 Trial Comparing Maribavir With Valganciclovir

Abstract

Background: Neutropenia may limit the use of valganciclovir treatment for cytomegalovirus (CMV) infection following hematopoietic cell transplant (HCT). A phase 2 study indicated efficacy of maribavir with fewer treatment-limiting toxicities than valganciclovir.

Methods: In this multicenter, double-blind, phase 3 study, patients with first asymptomatic CMV infection post-HCT were stratified and randomized 1:1 to maribavir 400 mg twice daily or valganciclovir (dose-adjusted for renal clearance) for 8 weeks with 12 weeks of follow-up. The primary endpoint was confirmed CMV viremia clearance at week 8 (primary hypothesis of noninferiority margin of 7.0%). The key secondary endpoint was a composite of the primary endpoint with no findings of CMV tissue-invasive disease at week 8 through week 16. Treatment-emergent adverse events (TEAEs) were assessed.

Results: Among patients treated (273 maribavir; 274 valganciclovir), the primary endpoint of noninferiority of maribavir was not met (maribavir, 69.6%; valganciclovir, 77.4%; adjusted difference: -7.7%; 95% confidence interval [CI]: -14.98, -.36; lower limit of 95% CI of treatment difference exceeded -7.0%). At week 16, 52.7% and 48.5% of patients treated (maribavir and valganciclovir, respectively) maintained CMV viremia clearance without tissue-invasive disease (adjusted difference: 4.4%; 95% CI: -3.91, 12.76). With maribavir (vs valganciclovir), fewer patients experienced neutropenia (16.1% and 52.9%) or discontinued due to TEAEs (27.8% and 41.2%). Discontinuations were mostly due to neutropenia (maribavir, 4.0%; valganciclovir, 17.5%).

Conclusions: Although noninferiority of maribavir to valganciclovir for the primary endpoint was not achieved based on the prespecified noninferiority margin, maribavir demonstrated comparable CMV viremia clearance during post-treatment follow-up, with fewer discontinuations due to neutropenia. Clinical Trials Registration. NCT02927067 [AURORA].

Keywords: cytomegalovirus; hematopoietic cell transplant; maribavir; valganciclovir.

Graphical Abstract

Figure 1.

Study design. Abbreviations: BID, twice daily; BL, baseline; Rand, randomization. ^aUnless dose adjustment was required for renal impairment. ^bVisit 2A was required only for patients receiving tacrolimus, cyclosporine, everolimus, or sirolimus at baseline.

Figure 2.

Patient disposition at enrollment, randomization, and follow-up. Percentages are based on the number of patients randomized per arm as the denominator (maribavir, n = 276; valganciclovir, n = 277). In the maribavir arm, the reasons for the 3 patients who were randomized but not dosed were withdrawal from the study (n = 2) and adverse event due to neutropenia (n = 1). In the valganciclovir arm, the reasons for the 3 patients who were randomized but not dosed were withdrawal from the study (n = 2) and other (1 patient was randomized by interactive response technology by error). Abbreviations: CMV, cytomegalovirus; COVID-19, coronavirus disease 2019; N, number of patients enrolled; n, number of patients through trial; PCR, polymerase chain reaction. ^aOther reasons for treatment discontinuation in the maribavir arm included patient decision (n = 1), poor condition of the patient (n = 1), negative CMV virology (n = 1), difficulties in swallowing (n = 1), investigator decision (n = 1), and starting another treatment for BK virus with a prohibited medication (n = 1). ^bOther reasons for study discontinuation in the maribavir arm included patient request (n = 2), COVID-19 pandemic (n = 1), misinterpretation of the protocol by site staff (n = 1), poor condition of the patient (n = 1), and COVID-19 infection (n = 1). ^cOther reasons for treatment discontinuation in the valganciclovir arm included investigator decision due to negative CMV PCR or favorable CMV levels (n = 4), unspecified investigator decision (n = 1), investigator decision due to patient hospitalization (n = 1), issue with study drug dispensing/assignment (n = 1), persistent cytopenias (n = 1), patient decision (n = 1), and investigator decision due to treatment being no longer clinically necessary (n = 1). ^dOther reasons for study discontinuation in the valganciclovir arm included reason unknown (n = 1), screen failure (n = 1), patient decision (n = 1), and physician decision due to a third reactivation of CMV (n = 1).

Figure 3.

Confirmed CMV viremia clearance at week 8 (primary endpoint) and prespecified subgroup analyses of confirmed CMV viremia clearance at week 8 by treatment arm (Modified Randomized Population). The Cochran–Mantel–Haenszel weighted average approach was used for the adjusted difference in proportion (maribavir−valganciclovir) and the corresponding 95% CI, adjusting for acute GVHD and baseline plasma CMV DNA concentration. Abbreviations: CI, confidence interval; CMV, cytomegalovirus; GVHD, graft-versus-host disease; N, number of patients in a treatment arm; n, number of responders; NA, not applicable.

Figure 4.

Maintenance of confirmed CMV viremia clearance with no clinical findings of tissue-invasive disease achieved at week 8 through weeks 12, 16 (key secondary endpoint), and 20 (Modified Randomized Population). Abbreviations: CI, confidence interval; CMV, cytomegalovirus; N, number of patients in a treatment arm; n, number of responders.

Figure 5.

Cumulative probability of first CMV viremia clearance at any time on study by treatment arm (Modified Randomized Population). Abbreviation: CMV, cytomegalovirus.

Figure 6.

Percentage of patients with confirmed CMV viremia clearance by study week and treatment arm (Modified Randomized Population; post hoc analysis). A post hoc analysis of patients with confirmed CMV viremia clearance by study week for the 2 treatment arms was performed by calculating the proportion of patients with CMV DNA below the lower limit of quantification at that visit and consecutive prior visits spanning 5 days (ie, confirmed viremia clearance by central laboratory) for each treatment arm. Abbreviation: CMV, cytomegalovirus.