Epidemiology, treatment patterns, and disease burden of cytomegalovirus in hematopoietic cell transplant recipients in selected countries outside of Europe and North America: A systematic review

Cytomegalovirus (CMV) disease impacts morbidity and mortality in hematopoietic cell transplant (HCT) recipients. This systematic review summarized data on the epidemiology, management, and burden of CMV post‐HCT outside of Europe and North America.

In patients with impaired cell-mediated immunity, such as hematopoietic cell transplant (HCT) recipients, CMV is an important cause of morbidity and mortality. 3 The uncontrolled replication of CMV is associated with direct effects such as CMV infection or reactivation, which can subsequently lead to CMV disease with manifestations such as pneumonitis, retinitis, colitis, and hepatitis. 1 In addition, indirect effects of CMV may include graft-versus-host disease (GvHD), opportunistic infections, or graft failure, which can negatively impact patient survival and contribute to non-relapse mortality. [4][5][6] Advances in the development of potent antivirals have decreased CMV-related mortality. Prophylaxis and preemptive therapy represent two distinct approaches to preventing CMV infection in HCT recipients. 7 Prophylaxis is the administration of antiviral agents in at-risk patients without any evidence of CMV infection or reactivation, while preemptive therapy consists of regular monitoring of patients to detect early CMV replication-for example, through quantitative polymerase chain reaction (PCR) testing once a week.
In the latter approach, treatment is initiated when the viral load exceeds a pre-defined threshold or when a positive antigenemia is detected.
Intravenous (IV) ganciclovir (GCV) and its valyl prodrug oral valganciclovir (VGCV) are the conventional first-line treatments for the management of CMV. 8 However, there are unmet needs for these drugs due to refractoriness to treatment and hematologic toxicities such as neutropenia or thrombocytopenia. 9 Furthermore, CMV refractoriness and resistance to standard therapy increase morbidity and mortality. Foscarnet or cidofovir, used as secondline treatments, are also associated with important adverse events (AEs), including nephrotoxicity or electrolyte imbalance. 10   health-related quality of life) and economic burden. Availability of guidelines on the clinical management of CMV was also explored. The review protocol was registered in PROSPERO (CRD42021271652)

Scope of the review
prior to the start of data extraction.

Study selection
Study selection and data extraction were performed independently by two assessors (with conflicts resolved by a third assessor). Eligibility criteria for the selection of relevant sources were based on the PICOTS and are listed in Table S6. At stage 1, literature search outputs were screened based on titles and abstracts, and the level of agreement between assessors was determined. A review of full-text articles was conducted at stage 2 to confirm the eligibility of sources retained after screening. Reasons for exclusion of sources at this stage were documented. For each retained study, information on study methods, populations, and results were recorded in a data extraction form that was piloted prior to the start of extraction. There was no attempt to contact study authors for Supporting Information.

Methodological quality assessment
The methodological quality of retained studies published as full-text articles was assessed using the Joanna Briggs Institute critical appraisal tools for observational studies. 14

Synthesis of findings
Synthesis of findings was qualitative and there was no attempt to pool results through a meta-analysis.

Search results
The flow of search results through the different stages of the selection process is presented in the PRISMA flow chart (Figure 1). The Therapy, and also covered Australia. 15 The characteristics, key findings, and methodological quality of retained studies are summarized in Table S9.

Patient characteristics
Across retained sources, the number of HCT patients included ranged from 18 in a single-center study 16    period (before or after 2012-corresponding to the year when the PCR assay was made available for the detection of CMV) did not uncover any apparent trends ( Figure S1). Although reported rates of infection varied extensively between studies conducted within the same country, results from the review suggested a trend toward lower estimates in Australia (25% 29 and 31% 30 ), while higher estimates were reported in South Korea (70.0%, 31 73.9%, 32 and 79.7% 33 ) ( Figure 3 and Table S8). A lower incidence was also found in young transplant recipients (9.6% in allogeneic HCT recipients with a median age of 25 years 34 and 18.6% in patients aged ≤20 years 35 ), although other studies conducted in the pediatric population reported higher estimates. As reported in two studies, the incidence of CMV infection varied according to the type of donor, with significantly higher rates found in adult patients who underwent haploidentical HCT (81.0% 19 and 87.5% 36 ) compared to those who underwent HCT from a matched sibling (23.5% 19 and 39.0% 36 ) or matched unrelated donor (55.6% 36 ).

Epidemiology of CMV infection and CMV disease in HCT recipients
In the population of autologous HCT recipients, four single-center studies, all from Turkey, reported incidence estimates of CMV infection within 1-year post-transplantation of 5.3%, 37 14.0%, 38 14.6%, 39 and 36.6%. 22 In all studies, almost all patients (from 95.6% to 98.8%) were CMV seropositive prior to transplantation. In addition, highest estimate was observed in a single-center study that actively monitored patients post auto-HCT and defined CMV infection as a CMV DNA viral load ≥70 IU/mL. 22 Reported incidence of CMV disease ranged between 2.9% and 15.7% (10 studies) in allogeneic HCT recipients, with results uniformly distributed within this range. 19 (four studies). 19,25,40,49 In other countries, higher proportions of gastroenteritis among patients with CMV end-organ disease were found (81.8% [9/11 patients] in Colombia, 18

Therapeutic management of CMV infection and disease
In the selected countries, preemptive therapy with CMV monitoring was the preferred approach to manage CMV infection and CMV disease in both adult and pediatric HCT recipients. 18,29,55,64,68,69 Across studies and centers, thresholds for initiating preemptive therapy ranged between >500 and >1000 CMV DNA copies per mL, 18,29,55 and the decision to end preemptive treatment was based on two consecutive negative viral load results, as evidenced by PCR 18 or a negative CMV antigenemia test. 68 Furthermore, in several countries  28 and a mean of 16 days (range: 5-32) in India. 21 For patients who were intolerant to GCV or VGCV and for those who did not respond to therapy after 2 weeks, foscarnet (180 mg/kg daily) or cidofovir were used as second-line agents. 28,55 However, in India, the use of lower cost alternatives, such as artesunate and leflunomide, was reported. 60 A major shortcoming of conventional first-line treatments is the frequent occurrence of serious AEs, as shown in Figure 5. GCV-associated neutropenia, myelosuppression, and nephrotoxicity led to premature treatment discontinuation in 13.6%, 10.0%, and 2.3% of patients, respectively. 64,68 times of the CMV DNA viral load). 28 No definition of refractory CMV was provided in the latter study. 73 A higher proportion of refractory CMV (49.4%) was observed in a prospective cohort study that included 107 consecutive patients diagnosed with hematological malignancy who underwent allogeneic HCT in China. 25 In this study, patients with refractory CMV had a lower absolute number of CMV-specific CD8+ T memory cells 30 days following HCT than those without CMV or with treatment-responsive CMV. After controlling for confounders, the lower absolute number of CMV-specific CD8+ T memory cells was found to be a significant predictor of refractoriness.   77 Third, methods for diagnosing CMV also differed across health care centers with regards to the type of specimen (e.g., whole blood or plasma) and diagnostic assay (antigenemia or PCR). Such differences in procedures are known to contribute to the variability in viral load values and thus the ability to detect CMV. 78  America. 83,84 The rate of resistant and/or refractory CMV was infrequently reported, with marked differences between available estimates. These results are consistent with those previously reported for Europe and North America, 77 where rates of CMV resistance were usually low (from 1.8% to 9.0% according to 10 studies), and the rates of refractory CMV tended to be high, ranging from 9.0% to 25.5% in Europe, 85,86 and from 19.9% to 47.4% in the US. 87,88 Such variability between estimates is largely attributable to the denominators used for the estimation (i.e., patients with suspected resistance, patients treated for CMV reactivation, or HCT recipients overall) as well as the triggers for testing (i.e., only patients with suspected resistance are usually tested). In the selected countries, preemptive therapy is the preferred approach to manage CMV infection and prevent CMV disease in both adult and pediatric HCT recipients. Conventional first-line treatments remain IV GCV or VGCV across all countries of interest, which is consistent with practices found in Europe and North America. 82,83,[95][96][97][98] There were, however, differences observed between countries in second-line treatments, consisting mainly of foscarnet or cidofovir but also high-dose GCV or artesunate. Heterogeneity in treatment patterns across Asia, South America, and the Middle East was documented, which may be due to differences in resources, clinical practices, drug approval, and reimbursement practices. Based on the current regional clinical practices, we note that in Brazil, GCV/VGCV are preferred as preemptive therapy, although foscarnet is sometimes the choice for patients with neutropenia or when treatment is initiated before engraftment. In South Korea, second/third-line agents, such as cidofovir and foscarnet, are typically not covered by payers.

PROs and economic burden of illness
A similar treatment strategy is being adopted/considered in Turkey whereby preemptive therapy with GCV is initiated when, based on the PCR assay, the absolute viral load is >1000 CMV DNA copies or when there is a progressive increase in the number of copies over two to three consecutive tests. In Australia, conventional treatments used for preemptive therapy include GCV or VGCV. Also, as letermovir is not subsidized in this country, a recent study reported on the use of high-dose valaciclovir for CMV prophylaxis in patients at high risk of CMV infection. 99 In this country, addition of foscarnet to the standard therapy is considered in patients with severe neutropenia at the onset of CMV DNAemia or at the initiation of preemptive therapy. In Turkey, foscarnet is not approved but can be reimbursed through offlabel use for patients who are refractory to standard of care or poor graft function.

CONCLUSION
This systematic review contributes to the synthesis of knowledge on the epidemiology, treatment, and burden of CMV infection and disease in HCT recipients outside of Europe and North America. Reported rates of CMV infection and disease were heterogeneous between studies, but overall, they remained high across all countries. First-line conventional treatment with GCV or VGCV is consistently used, in accordance with published guidelines. 102 However, similar to Europe and North America, CMV that is resistant or refractory to treatment was shown to be relatively frequent, and treatment toxicity is a major cause of treatment discontinuation that may negatively impact patient outcomes. These findings highlight critical unmet needs and call for the development of effective and safe treatments.
The synthesis of findings published over the past decade in the countries of interest was challenged by the absence of consistent definitions for CMV infection, disease, resistance, and refractory CMV, as well as the lack of data on PROs and economic burden. Further research is required in Asia-Pacific, Latin America, and the Middle East, given the reimbursement constraints for treatments in these regions.

AUTHOR CONTRIBUTIONS
Dirk Demuth, Anudeep Sandhu, and Inderjeet Singh contributed to the review conceptualization, design, and data analysis. All authors contributed to data analysis, writing of the manuscript, and commenting on previous versions of this document. All authors read and approved the final manuscript.

ACKNOWLEDGMENTS
The authors would like to thank YOLARX Consultants for providing the expertise in conducting the systematic review and medical writing for this manuscript. This support was funded by Takeda Pharmaceuticals International AG-Singapore Branch.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.