Healthcare resource utilization of maribavir versus investigator‐assigned therapy in transplant recipients with cytomegalovirus infection refractory (with or without genotypic resistance) to prior treatment: Exploratory analysis of the Phase 3 SOLSTICE trial

Cytomegalovirus (CMV), a common post‐transplant infection, is associated with increased healthcare resource utilization. In the Phase 3 SOLSTICE trial, maribavir was superior to investigator‐assigned therapy (IAT; valganciclovir/ganciclovir, foscarnet, and cidofovir) for CMV viremia clearance at Week 8 in transplant recipients with confirmed refractory CMV infection with/without resistance. This exploratory analysis evaluated hospital admissions of patients during the SOLSTICE trial.


INTRODUCTION
Cytomegalovirus (CMV) infection is a common complication in hematopoietic stem cell and solid organ transplant recipients. 1 CMV infection in transplant recipients has been associated with poor outcomes such as increased morbidity and mortality. 2,3 Conventional treatment of CMV infection includes intravenous ganciclovir, oral valganciclovir, intravenous foscarnet, and intravenous cidofovir. The use of these treatment strategies may be complicated by treatment-limiting toxicities such as nephrotoxicity with foscarnet or cidofovir, and myelosuppression with valganciclovir or ganciclovir. [4][5][6][7] In addition, CMV infection in transplant recipients has been associated with higher hospitalization-related costs and increased inpatient days compared with those without infection, and multiple antiviral courses have been shown to further increase costs. 8 Maribavir is a UL97 protein kinase inhibitor with multimodal anti-CMV activity. 9 In the Phase 3 SOLSTICE trial (NCT02931539; EudraCT: 2015-004725-13) in transplant recipients with CMV infection refractory to prior treatment with or without resistance, maribavir demonstrated superiority to investigator-assigned therapies (IATs; valganciclovir/ganciclovir, foscarnet, and cidofovir) for the primary endpoint of CMV clearance at Week 8, and the secondary endpoint of CMV clearance with symptom control at Week 8 was maintained through Week 16. 10 Maribavir was also associated with fewer treatmentemergent adverse events of acute kidney injury versus foscarnet and of neutropenia versus valganciclovir/ganciclovir. This exploratory healthcare resource utilization (HRU) analysis of patients in the SOLSTICE trial aimed to evaluate hospital admission data including hospitalizations and length of hospital stay (LOS) by treatment arm.

METHODS
An exploratory HRU analysis of the

Data analysis
The number and timing of hospital admissions, the number of days spent in the hospital setting (general ward and/or intensive care unit), and the reason for hospital admission were analyzed. Reasons for hospital admissions were recorded by site investigators, categorized using the Medical Dictionary for Regulatory Activities 11 classification codes, and summarized descriptively by treatment groups.
Statistical analyses were conducted using SAS (version 9.4; SAS Institute, Cary, NC). Hospitalization rates (admissions/person/year) and LOS (days/person/year) were estimated using negative binomial models adjusting for the time in the relevant study period. Adjusted yearly incidence rates (IR), 95% confidence intervals (CI), IR ratios (IRR), and percent reduction in IRRs were calculated. All data were summarized descriptively.
Subgroup analyses were conducted for the maribavir rescue arm and individual IATs. For the main analyses, patients in the IAT group who received maribavir rescue treatment had their HRU data included up to the point of the treatment switch. Healthcare encounters for patients in the maribavir rescue arm were compared before maribavir rescue medication (pre-rescue) and on or after receiving maribavir rescue treatment (on/after rescue).

RESULTS
Overall, 235 patients were randomized to the maribavir arm and 117 F I G U R E 1 Hospitalization rates and LOS by treatment during the treatment and follow-up phases. Treatment-phase hospitalization rates and LOS were adjusted for the duration of time on treatment, which was longer for the maribavir arm (52.0 days for maribavir, 35.7 days for IAT). Follow-up phase hospitalization rates and LOS were adjusted for the duration of follow-up, which was longer for the maribavir arm (80.8 days for maribavir, 58.1 days for IAT). As the number of patients who received cidofovir was so small, the adjusted yearly rate and LOS are not interpretable and therefore not shown. CI, confidence interval; IAT, investigator-assigned therapy; IR, incidence rate; IRR, incidence rate ratio; LOS, length of hospital stay; SD, standard deviation.

76.3
On/after rescue

F I G U R E 2
Adjusted hospitalization rates and LOS in the maribavir rescue arm (n = 22). Hospitalization rates and LOS were adjusted for the duration of time on each treatment (30.9 days pre-rescue, 132.8 days on/after rescue). IRRs were calculated as the pre-rescue incidence rate divided by the on/after rescue incidence rate; therefore, an IRR >1 represents an increased rate of hospital admissions for pre-rescue compared with on/after rescue. CI, confidence interval; IRR, incidence rate ratio; LOS, length of hospital stay. 12.8%, respectively), a lower proportion of patients in the maribavir arm were hospitalized due to CMV treatment-related reasons (6.8% vs. 14.5%, respectively). Patients treated with maribavir had a lower rate of hospital admissions due to neutropenia compared with valganciclovir/ganciclovir (1.7% vs. 10.7%, respectively) and renal disorders compared with foscarnet (4.7% vs. 12.8%, respectively).

DISCUSSION
In this exploratory analysis of the Phase 3 SOLSTICE trial, patients who received maribavir had a reduced rate of hospitalization compared with patients treated with IAT during the treatment phase.
For both treatment arms, hospitalization rates during the treatment phase were higher than during the follow-up phase. The total time spent in the hospital per person/year was also significantly reduced (54%) for patients treated with maribavir compared with those receiving IAT. Further, patients in the rescue arm were less likely to be The higher hospitalization rate observed in patients treated with foscarnet during the treatment phase could be related to the intravenous administration of this therapy requiring hospital admission, which was one of the more common reasons for hospitalization reported in the IAT arm. In addition, a higher proportion of patients who received foscarnet were hospitalized due to renal and urinary disorders, which is reflective of the previously reported association of foscarnet treatment with nephrotoxicity. 4,6 However, due to the small individual IAT sample sizes in this study and wide CIs, these results need to be interpreted cautiously.
General limitations of the analysis included reliance on sites to report/record HRU events and the small number of patients in some subgroups; therefore, the interpretability of some analyses is limited.
Further, this study was based on data from a Phase 3 trial, and therefore, data may not reflect real-world evidence. Additional studies that explore the real-world evidence of different CMV treatments will be valuable.
These findings collectively show that patients treated with maribavir may experience a lower hospitalization rate and shorter LOS compared with those treated with IAT. Similar findings were observed in the rescue arm. Overall, the results of these descriptive analyses contribute to the evidence quantifying the HRU experience of patients with post-transplant CMV infection and the potential opportunities to reduce the burden of hospital admissions in the post-transplant setting.

AUTHOR CONTRIBUTIONS
All authors participated in the conception or design, acquisition, analysis, and interpretation of the study, critically revised the manuscript for important intellectual content, and approved the final manuscript for publication. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

ACKNOWLEDGMENTS
The authors would like to thank Leighann Litcher-Kelly and Benjamin

DATA AVAILABILITY STATEMENT
The datasets, including the redacted study protocol, redacted statistical analysis plan, and individual participant's data supporting the results reported in this article, will be made available within 3 months from the initial request to researchers who provide a methodologically sound proposal. The data will be provided after their de-identification, in compliance with applicable privacy laws, data protection, and requirements for consent and anonymization.