Clinical and healthcare burden of disease associated with cytomegalovirus in allogeneic hematopoietic stem cell transplantation - A retrospective single-center study.

List of Abbreviations

cytomegalovirus

clinically significant

donor

graft‐versus‐host disease

hazard ratio

hematopoietic stem cell transplantation

incidence rate ratio

odds ratio

polymerase chain reaction

pre‐emptive therapy

lower and upper quartiles

recipient

INTRODUCTION

CMV infection is a common complication in allogeneic hematopoietic stem cell transplantation (HSCT). 1 , 2 CMV infection occurs in 30%–70% of the patients, usually within 0–100 days after the transplantation, 3 , 4 , 5 , 6 , 7 but late CMV infection (after day 100 post allogeneic HSCT) may occur in 10%–30% of the patients. 3 , 8 When untreated, CMV infection leads to CMV disease, most commonly pneumonia or gastroenteritis, in 10%–40% of the patients and CMV disease is associated with higher mortality. 2 Both early and late CMV infection are associated with increased morbidity and non‐relapse mortality. 4 , 5 , 9

Allogeneic HSCT patients are commonly monitored for CMV replication by polymerase chain reaction (PCR) in blood or urine samples. If clinically significant CMV (CS‐CMV) infection is observed, pre‐emptive therapy (PET) with antivirals is often initiated to prevent CMV disease. Established anti‐CMV drugs include ganciclovir 10 or valganciclovir, 11 , 12 foscarnet, 13 and cidofovir as second‐line treatment. 14 The introduction of PET has reduced the incidence of CMV disease down to 1%–10%. 4 , 5 , 15 , 16

Donor (D) and recipient (R) CMV serostatus are significant risk factors for CMV infection in allogeneic HSCT, 2 , 4 , 5 and D/R status is associated with CMV‐related outcomes, such as delayed CMV‐specific immune reconstitution, repeated CMV infections, additional anti‐CMV treatment courses, late CMV infection and development of CMV disease. 3 , 17 , 18 , 19 , 20 , 21 Other risk factors include the use of high‐dose corticosteroids, T‐cell depletion, acute and chronic graft‐versus‐host‐disease (GVHD), and the use of mismatched or unrelated donors. 8 , 15 , 21 , 22 , 23 , 24 , 25

CMV infection, CMV disease, and PET have been reported to associate with higher clinical burden and increased healthcare resource utilization. Retrospective single‐center studies have reported that a greater economic burden is associated with CMV treatment. 26 , 27 , 28 Three recent studies from the United States reported the association of CMV infection with higher readmission rates and greater healthcare costs, driven by medication costs, additional hospital days, and outpatient visits. 29 , 30 , 31

There are no published data on CMV seroprevalence or CS‐CMV infection among allogeneic HSCT patients in Finland. A study from the Finnish Maternity Cohort serum bank reported a decline in CMV seroprevalence from 84.5% in 1992 to 71.5% in 2012. 32 Pregnant women are typically younger than allogeneic HSCT patients, and they may not reflect the seroprevalence in allogeneic HSCT patients. Also, we don't have published data on the healthcare resource utilization associated with CMV infection in allogeneic HSCT in Finland. In this retrospective study from secondary patient data, we investigated the association of CS‐CMV infection with clinical outcomes and healthcare resource utilization in the treatment of CMV in allogeneic HSCT patients in Finland.

MATERIALS AND METHODS

The study population and data collection

This single‐center, retrospective, observational cohort study with Finnish allogeneic HSCT patients investigated the association of CS‐CMV infection with clinical outcomes and healthcare resource utilization. The study included adult patients who received their first allogeneic HSCT [10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD‐10) code: Z94.8, Nordic Classification of Surgical Procedures 2015 codes: WW202/204/206/210, ≥18 years of age] between January 1, 2013, and December 31, 2018, at the Hospital District of Southwest Finland. Patients who had participated or were considered to participate in the letermovir phase 3 CMV reactivation prophylaxis study 33 were excluded. Patients were followed for 3 years after the transplantation. Data were collected from the Turku University Hospital data lake.

The data collected at baseline and/or during follow‐up included: age, gender, date of allogeneic HSCT, R and D CMV serostatus, combined D/R CMV serostatus, donor type (matched unrelated, HLA identical sibling, and haploidentical), the primary reason for allogeneic HSCT (acute leukemias, myeloproliferative malignancies, lymphoproliferative malignancies, aplastic anemia), preconditioning regimen (myeloablative, 34 , 35 reduced intensity, 36 and sequential 37 ), stem cell source (peripheral blood, bone marrow), anti‐CMV medication (foscarnet, ganciclovir, and valganciclovir), CS‐CMV infection episode, number of outpatient visits, number and length of hospital admissions and intensive care, date and type of GVHD, and dates of CS‐CMV infection, relapse and death. Anti‐CMV medication, number of outpatient visits, number and length of hospital admissions, and intensive care were collected only from the first year of follow‐up. For further details of definitions and categorizations, see Supporting Information.

CMV‐serostatus analysis, CMV‐monitoring, CS‐CMV infection, and CS‐CMV infection episode

CMV‐serostatus was analyzed in donors and recipients by enzyme immunoassay according to the local laboratory procedure. After the transplantation, peripheral blood plasma CMV levels were monitored weekly for the first 60 days, and then every 2–4 weeks until the end of the first post‐transplant year by quantitative PCR using GeneProof CMV PCR Kit, 1 copy/ml corresponding to 1 IU/ml. The lower detection limit of the assay is 250 copies/ml, and a sample containing <50 copies/ml is reported as negative. CS‐CMV infection was defined as ≥1000 copies/ml, or the initiation of PET at any level of CMV DNAemia. PET was initiated with CMV DNA ≥1000 copies/ml by local protocol, or with CMV DNA levels <1000 copies/ml by clinical choice. CS‐CMV infection episode was defined as the occurrence of CS‐CMV infection, the start of any anti‐CMV medication, and the change of CMV DNAemia below the detection limit during anti‐CMV medication, followed by a period of two weeks with undetectable CMV nucleic acid levels in blood samples after cessation of anti‐CMV medication.

Outcome measures

The clinical outcomes included CMV seroprevalence, incidence and the number of CS‐CMV infections, time to the first CS‐CMV infection and CMV infection‐free survival, overall survival, relapse of the underlying malignancy and relapse‐free survival, acute and chronic GVHD and GVHD‐free survival.

Healthcare resource utilization outcomes included hospital readmission and intensive care unit admission rates, length of stay in the hospital and intensive care units, the number of outpatient hospital visits, and the use of anti‐CMV medication.

Statistical analyses

The continuous variables in the data were described using means and standard deviations if normally distributed and as medians along with lower and upper quartiles (Q 1 –Q 3 ) if non‐normally distributed. The normality of the empirical distributions was tested using the Kolmogorov‐Smirnov test. Discrete variables were presented using observed frequencies and proportions.

Differences in patient characteristics (age, gender, recipient and donor serostatus, primary reason for allogeneic HSCT, preconditioning regimen, donor type, and HLA matching, and stem cell source) between patients with and without CS‐CMV infection were analyzed using Mann‐Whitney U‐test for numerical variables and Pearson's chi‐squared of Fisher exact test categorical attributes. The association of different patient characteristics and the occurrence of CS‐CMV infection (in general or multiple episodes) were analyzed using simple logistic regression.

Simple and multiple Cox proportional hazards models were used to analyze the overall survival of the patients. CMV infection‐free survival, relapse‐free survival, and GVHD‐free survival were analyzed using a simple and multiple proportional hazards model for the subdistribution of a competing risk because death or non‐relapse‐related death were considered competing risks. The Schoenfield residuals were checked to confirm the proportional hazards assumption, and if the assumption was not met, the Kaplan‐Meier estimator and log‐rank test were used for overall survival and the cumulative incidence function, and Gray's test for events with competing risks. Healthcare resource utilization due to CS‐CMV infection was further analyzed with a simple negative binomial model. The use of anti‐CMV medication between patients with one versus multiple infections as well as acute GVHD grade 0–2 versus 3–4 was analyzed using Pearson's chi‐squared or Fisher's exact test and the differences in total dosage were analyzed with Mann‐Whitney U ‐test.

In all analyses where CS‐CMV infection or GVHD were used as independent variables, sensitivity analyses were performed and the model selection was based on the smallest value of the Akaike information criterion. For further details of the sensitivity analyses, see Supporting Information.

The results of the Cox proportional hazard models were presented using the average hazard ratios (HRs), and the results of logistic and negative binomial regression were presented as odds ratios (ORs) and incidence rate ratios (IRRs), respectively. The results of the log‐rank and Gray's test were reported using the chi‐squared test statistic together with the degrees of freedom. All results were presented with p ‐values and corresponding 95% confidence intervals when applicable. p ‐values less than 0.05 were considered statistically significant. Statistical analysis, tables, figures, and listings were all produced with R version 3.6.3 (R Core Team, 2018) using RStudio Server in Auria Clinical Informatics’ secure operating environment.

Ethical considerations

This study was approved by the Administration of the Hospital District of Southwest Finland (T94/2020) in the role of the data controller and was conducted in compliance with applicable national legislation. According to the Act of Secondary Use of Health and Social Data, informed consent is not required for studies based on patient records. MSD had oversight of the study. MSD had an initiative in the conception of the study, and MSD participated, together with other investigators, in the design of the study, and interpretation of the results, and MSD was responsible for the publication development. MSD did not participate in the data collection and analysis.

RESULTS

Baseline characteristics

Between January 1, 2013, and December 31, 2018, a total of 276 patients were identified. Twenty‐five patients were excluded from this study: 22 patients participated or had been considered for participation in the letermovir CMV reactivation prophylaxis study, 33 one patient received syngeneic HSCT and two patients had not received HSCT. Altogether, 251 allogeneic HSCT patients were included in the study.

Baseline characteristics are presented in Table 1 . The median age was 55 years, and the proportions of male and female patients were 56.2% and 43.8%, respectively. CMV seroprevalence was 69.7%. D/R CMV serostatus is detailed in Table 1 . Thirty‐three (13.1%), 38 (15.1%), and 180 (71.7%) patients received a transplant from a haploidentical donor, HLA identical sibling donor, or matched unrelated donor, respectively. The reasons for allogeneic HSCT were acute leukemias (51.4%), lymphoproliferative diseases (29.5%), myeloproliferative diseases (15.9%), and (3.2%) aplastic anemia. Ninety‐seven, 90, and 64 patients received sequential, reduced intensity, or myeloablative preconditioning, respectively.

CMV infection

One‐hundred forty‐six patients (58.1%) and 148 (59.0%) had CS‐CMV infection by week 24 and one year, respectively, and the median time to the first CS‐CMV infection was 34.5 (27.0, 45.0) days (Figure 1 ). CS‐CMV infection occurred in 77.1% and 17.1% of the seropositive and seronegative patients, respectively. One hundred twenty‐seven (85.8%) patients had one CS‐CMV infection episode, and 21 (14.2%) had ≥2 CS‐CMV infection episodes within one year (Table 2 ). One patient had the first CS‐CMV infection after one year.

When analyzing the CMV‐free survival with the simple Cox proportional hazards model, D/R CMV serostatus was associated with the risk of CS‐CMV infection. Donor CMV serostatus did not influence on the risk of CS‐CMV infection in CMV seropositive recipients (D‐/R+ vs. D+/R+: HR 0.96, 95% confidence interval [CI] 0.67–1.39, p = .841) (Figure 1B and Table S1 ), but recipient CMV seronegativity was associated with lower risk for CS‐CMV infection (D+/R‐ vs. D+/R+: HR 0.35, 95% CI 0.18–0.66, p = .001) (Figure 1B and Table S1 ). The haploidentical graft was associated with higher CS‐CMV infection risk as compared to matched unrelated donor (HR 1.54, 95% CI 1.02–2.34, p < .001) (Figure S1A and Table S1 ). Other covariates, such as type of preconditioning, and the primary reason for allogeneic HSCT or GVHD, were not associated with differences in the CS‐CMV infection‐free survival.

In the multiple Cox proportional hazards model, donor CMV serostatus did not have effect on the risk of CS‐CMV infection in CMV seropositive recipients (D‐/R+ vs. D+/R+: HR 0.87, 95% CI 0.59–1.27, p = .463) (Table 3 ). Recipient CMV seronegativity was associated with a lower risk for CS‐CMV infection when the graft was received from a seropositive donor (D+/R‐ vs. D+/R+: HR 0.25, 95% CI 0.13–0.51, p < .001) (Table 3 ). Lymphoproliferative diseases, with borderline statistical significance, increased the CS‐CMV infection risk (HR 1.46, 95% CI 1.00–2.13, p = .050) (Table 3 ), even though the reason for allogeneic HSCT was not associated with CS‐CMV infection in the simple Cox proportional hazards model (Figure S1B and Table S1 ). A graft from an HLA identical sibling donor was associated with lower CS‐CMV infection risk (HR 0.55, 95% CI 0.34–0.88, p < .013) (Table 3 ). D‐/R‐ patients had a low risk of CS‐CMV infection in the simple (Figure 1B and Table S1 ) and multiple (Table 3 ) Cox proportional hazards models.

Association of baseline and time‐dependent factors with ≥2 CS‐CMV infections during the first year after the transplantation was analyzed with a simple logistic regression. Factors associated with a higher risk for ≥2 CS‐CMV infections included myeloproliferative diseases and aplastic anemia as the primary reason for the allogeneic HSCT (OR 3.38, 95% CI 1.09–10.70, p = .034), presence of acute GVHD (OR 4.90, 95% CI 1.70–17.75, p = .006), and grade 2–3 chronic GVHD (OR 12.62, 95% CI 4.31–46.38, p < .001) (Table S2 ). Age was associated with ≥2 CS‐CMV infections but the effect size was small (Table S2 ).

CS‐CMV infection and mortality

All‐cause mortality, relapse mortality, and non‐relapse mortality were investigated at one year and three years, and the mortality rates are detailed in Table 2 . CS‐CMV infection did not increase the risk of all‐cause mortality at one year (HR 0.86, 95% CI 0.52–1.43, p = .569) or three years (HR 1.16, 95% CI 0.77–1.76, p = .485) (Figure 2A ). Also, CMV infection did not increase the risk of relapse‐related deaths at one year ( X ( 1 ) 2 1.17, p = .279) or three years (HR = 0.98, 95% CI 0.53–1.81, p = .938) of follow‐up (Figure 2B ).

CS‐CMV infection, GVHD, and relapses

CS‐CMV infection did not increase the risk of GVHD at 24 weeks (HR 1.01, 95% CI 0.75–1.37, p = .936) or at one year (HR 1.00, 95% CI 0.73–1.35, p = .982). However, the risk of GVHD was higher in patients with ≥2 CS‐CMV infections compared to those without GVHD by one year (HR 1.61, 95% CI 1.03–2.51, p = .035) (Figure S2 ). In the multiple Cox proportional hazards model, GVHD risk was not higher in patients with one CS‐CMV infection (HR 1.06, 95% CI 0.74–1.53, p = .742), but the risk of GVHD was higher in patients with ≥2 CS‐CMV infections, (HR 1.65, 95% CI 1.02–2.66, p = .041) compared to those without CS‐CMV infection. CS‐CMV infection was not associated with relapses at one year (HR 1.21, 95% CI 0.68–2.16, p = .521) or three years (HR 1.23, 95% CI 0.77–1.98, p = .385).

CS‐CMV infection and healthcare resource utilization

In the study cohort, 208 (82.9%) patients were readmitted to the hospital. One hundred thirty‐three (89.8%) and 75 (72.8%) patients with and without CS‐CMV infection, respectively, were readmitted during the first year after the transplantation (Table 4 ). The number of hospital readmissions was higher in patients with CS‐CMV infection compared to those without CS‐CMV infection [median (Q 1 –Q 3 ) 3.0 (2.0–5.0) vs. 2.0 (1.0–3.5), IRR 1.38, 95% CI 1.10–1.73, p = .005]. Correspondingly, the number of hospital readmissions was higher in patients with ≥2 CMV infections compared to those without CMV infection [median (Q 1 –Q 3 ) 6.0 (3.0–7.0) vs. 2.0 (1.00–3.5); IRR 2.15, 95% CI 1.53–3.03, p < .001] (Table 4 ).

The additional length of hospital stay was longer in patients with CS‐CMV infection compared to those without CMV infection [median days (Q 1 –Q 3 ) 27.0 (14.8–53.0) vs. 14.0 (4.0–37.0); IRR 1.48, 95% CI 1.12–1.94, p = .005]. Also, patients with ≥2 CS‐CMV had longer additional length of hospital stay [median days (Q 1 –Q 3 ) 75.0 (56.0–92.0) vs. 14.0 (4.0–37.0); IRR 2.71, 95% CI 1.76–4.35, p < .001].

CS‐CMV infection was not associated with increased intensive care admission (Table 4 ) or its length. We did not see differences in the number of outpatient visits, either (Table 4 ).

One hundred forty‐eight patients had CS‐CMV infection, and 145 (98.0%) patients received PET. One hundred twenty‐four patients received PET for one CMV infection, and 21 patients received PET for ≥2 CMV infections (Table S3 ).

Frequencies and the medians of total doses for all anti‐CMV drugs are detailed in Table S3 . In patients with CS‐CMV infection, 73.0% and 62.2% received valganciclovir and foscarnet, respectively. Ganciclovir (36.5%) was less often used.

Foscarnet was used more often in patients with ≥2 CS‐CMV infections compared to patients with one CS‐CMV infection (90.5% and 57.5%, respectively, p = .004) (Table S3 ). There was no difference in the proportions of the patients with one CS‐CMV infection versus ≥2 CS‐CMV infection episodes who received ganciclovir or valganciclovir. The total doses did not differ between anti‐CMV drugs (Table S3 ).

Patients with grade 3–4 acute GVHD received more often foscarnet (51.0% vs. 33.2%, p = .020), and the total dose of foscarnet was higher compared to the patients with grade ≤2 acute GVHD (Table S3 ). There was no difference in the frequencies or total doses of ganciclovir or valganciclovir between patients with grade ≤2 or grade 3–4 acute GVHD (Table S3 ).

DISCUSSION

This retrospective single‐center study investigated clinical and healthcare burdens due to CS‐CMV infection in allogeneic HSCT patients by one year after allogeneic HSCT, and mortality rate up to three years. This is the first study reporting CMV infection rate and CMV seroprevalence in Finnish allogeneic HSCT patients. CMV seroprevalence was 69.7%, and CS‐CMV infection occurred in 59.0% of all patients, and 77.1% of the CMV seropositive patients. Also, 14.2% of the patients with CS‐CMV infection experienced ≥2‐CMV infections during the first year after the transplantation. CS‐CMV infection was associated with greater healthcare resource utilization, indicated by higher readmission rates and longer length of hospital stay.

D/R CMV serostatus has a significant impact on the risk of CMV infection and clinical outcomes after allogeneic HSCT. 2 , 4 , 5 Recipient seropositivity is associated with CMV‐related outcomes, such as delayed CMV‐specific immune reconstitution, repeated CMV infections, additional anti‐CMV treatment courses, late CMV infection, and the development of CMV disease. 3 , 17 , 18 , 19 , 20 , 21 In this study, recipient CMV seropositivity was associated with CS‐CMV infection, and D+/R‐ and D‐/R‐ patients had lower CS‐CMV infection rates in the 1‐year follow‐up, and these findings are in line with the published literature. Other risk factors include high‐dose corticosteroids, T‐cell depletion, acute and chronic GVHD, and the use of mismatched or unrelated donors. 8 , 15 , 21 , 22 , 23 , 24 , 25 In our study, donor type and HLA matching, and the primary reason for allogeneic HSCT had independent effects on the CMV infection risk. We also investigated the rate of subsequent CMV infections, and ≥2 CMV infections occurred in 14% of the study patients. In a recent study, 23% of all seropositive allogeneic HSCT recipients and 31% of those who received PET for their first CMV reactivation, had ≥2 CMV reactivations during the first year after transplantation. 6

CMV seroprevalence was 69.7% in the study cohort. Observed CMV seroprevalence is in line with the CMV seroprevalence reported from the Finnish biobank cohort of pregnant women, with CMV seroprevalence of 84.5% in 1992 and 71.5% in 2012. 32 Of note, the population in our study covers approximately 40% of the allogeneic HSCT patients during the study period and represents Finnish population in the limited geographical regions. Also, the population in our study was older than pregnant women typically. These factors should be considered when extrapolating the observed CMV seroprevalence to all allogeneic HSCT patients in Finland.

CMV seropositivity and CMV infection have been reported to associate with negative clinical outcomes in allogeneic HSCT patients. 2 , 4 , 5 In the previous studies, positive donor or recipient CMV serostatus were associated with lower survival among allogeneic HSCT recipients. 38 , 39 In a large retrospective study, positive donor or recipient CMV serostatus was associated with lower overall survival acute myeloid leukemia and acute lymphoblastic leukemia patients, 4 and CMV infection has been associated with higher non‐relapse mortality contributing to lower overall survival. 4 , 7 , 40 PET has reduced the incidence of CMV disease and CMV disease‐related mortality. 4 , 5 , 15 , 16 However, CMV replication at any level has been reported to associate with higher mortality even after adjusting the use of PET. 5 A post‐hoc analysis of the phase 3 trial investigating a 100‐day CMV prophylaxis with letermovir in CMV seropositive recipients suggested that all‐cause mortality by week 48 was lower in the letermovir group (12%) compared to that in the placebo group (17%) and the difference was associated with a decrease in CMV reactivation‐related mortality in letermovir group, despite of PET. 41 In our study, mortality rates were 14.7%, 24.7%, and 38.2% at week 24, one year, and three years, respectively, and we did not see the difference in mortality rates among patients with or without CS‐CMV infection. Also, relapse mortality was not associated with CS‐CMV infection. However, looking at the mortality curves, one may see higher early all‐cause and non‐relapse mortality among patients without CMV infection. The mortality follow‐up was started from the transplantation. We retrospectively grouped the patients by CS‐CMV infection, which is a time‐related event, and the mortality analyses may contain an inherent bias due to the clustering of early transplantation‐related deaths in the group of patients without CS‐CMV infection. Green et al. 5 analyzed the association of CMV infection at day 0–60 after transplantation with the mortality by one year among day 100 survivors. This landmark analysis suggested that CMV reactivation with high viral loads was associated with all‐cause and non‐relapse mortality even if the early transplantation‐related mortality had been excluded. 5

Previous studies report that D+/R+ serology is associated with a lower incidence of acute GVHD in chronic myeloid leukemia, whereas D/R serology or CMV infection is not associated with acute GVHD in other diseases. 4 We did not see the difference between the CS‐CMV infection and the incidence of GVHD. Also, the incidence of CS‐CMV infection did not differ in patients with grade 3–4 acute GVHD compared with those with grade ≤2 acute GVHD, but the association between ≥2 CMV infections and GVHD remained significant after multiple adjustments. From this data, it is difficult to say the effect of ≥2 CMV infections on the incidence of GVHD, because CMV infection and GVHD often coincide, and we did not carry out a competing event analysis here. Previously, it has been reported that patients with acute GVHD and no CMV infection had a long time to fatal outcome compared with the patients hospitalized with or because of CMV infection. 31

In our study, the malignancy relapse rate was 19.5% and 29.9% by one year and three years, respectively, and CS‐CMV infection was not associated with relapse. CMV infection has been postulated to reduce relapses in certain hematological malignancies, possibly by activating the graft and tumor immunity, 42 , 43 and a recent meta‐analysis suggested a lower risk of acute myeloid leukemia relapse in patients with CMV infection. 44 A large retrospective study, however, did not find an association between CMV infection and relapses. 4

Healthcare costs due to CMV infection have been compared with costs in patients with acute GVHD. 31 CMV infection was associated with higher hospital readmission rates, a longer length of hospital stay, and increased total costs. 31 Patients hospitalized with CMV infection received more often foscarnet and cidofovir, contributing to the higher pharmacy costs. 31 Another study showed that CMV infection and PET were associated with a longer length of stay for HSCT procedures, a higher number of hospital readmissions, a longer total length of hospital stay, and higher inpatient costs. More detailed analysis suggested that higher inpatient costs originated from hospital facility and procedure costs, laboratory services, and pharmacy costs. 45 Two retrospective registry‐based studies in the United States showed that CMV infection caused increased healthcare resource use already within 100 days 29 and by one year post allogeneic HSCT. 30 In our study, CS‐CMV infection was associated with a higher readmission rate and longer length of hospital stay. When CS‐CMV infection is observed, patients are admitted to the hospital, and PET is initiated with intravenously administered ganciclovir or foscarnet, which in our study may contribute to the observed higher hospitalization rate of patients with CS‐CMV infection. When clinical improvement is observed, PET is continued after hospital discharge either with intravenously administered ganciclovir or foscarnet in the outpatient clinic, or with per os valganciclovir, and this is seen as a relatively high proportion of patients receiving valganciclovir. Patients with ≥2 CMV infections received more often foscarnet, but there was no difference in the proportions of the patients receiving other anti‐CMV drugs with respect to the number of CMV infections.

In contrast, a single‐center retrospective study investigating costs in 310 allogeneic HSCT patients did not report an association between CMV infection and changes in healthcare costs. 28 Studies investigating costs, as our study also, often are based on retrospective data, which may contain bias difficult to control. Differences in healthcare and reimbursement policies inevitably impact cost formation, and therefore studies reporting costs must be interpreted with caution. Yet, analysis of the healthcare resource utilization in the prospective 100‐day CMV prophylaxis trial with letermovir suggested that a lower CMV reactivation rate with letermovir prophylaxis was associated with fewer readmissions and shorter total readmission length of stay at weeks 14, 28, and 48. 46

Allogeneic HSCT patients report a negative impact on the quality of life in many dimensions of quality‐of‐life questionnaires, including physical, psychological, social, and role functions, especially during the first year after the transplantation. 47 The quality of life is lower during the transplantation inpatient stay and gradually improves during the first post‐procedure year after hospital discharge. 48 Impacted dimensions of the quality of life during the inpatient stay included physical functioning, pain, appetite, and sleep among others. 48 Although not reported in the cited study and not investigated in our study, one may presume that hospital readmission during the first year after the transplantation has a negative impact on the quality of life.

In conclusion, CMV seroprevalence is relatively high among allogeneic HSCT patients in Finland. Clinically significant CMV infection was common, occurred within three months after the transplantation, and repeated CMV infection was associated with GVHD. CMV infection was associated with a higher hospital readmission rate and longer length of hospital stay.

CONFLICT OF INTEREST

JR received the Principal Investigator fee, and Turku University Hospital District received institutional research funding for this study. RK is an employee of MSD Finland Oy and owns stocks of Merck & Co., Inc., Rahway, NJ, USA.

Supporting information

Supporting information

Click here for additional data file.

Supporting information

Click here for additional data file.

Graphical Abstract

Click here for additional data file.

ACKNOWLEDGMENTS

We would like to acknowledge the patients, the doctors, and the nurses whose lives and efforts made this evaluation possible. We also thank Eeva Kronqvist at Auria Clinical Informatics for the administrative support for this study. This study was sponsored by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.

REFERENCES

1 Ljungman P , Hakki M , Boeckh M . Cytomegalovirus in hematopoietic stem cell transplant recipients . Hematol Oncol Clin North Am . 2011 ; 25 ( 1 ): 151 ‐ 169 . 21236396 10.1016/j.hoc.2010.11.011 PMC3340426

2 Ljungman P , de la Camara R , Robin C , et al. Guidelines for the management of cytomegalovirus infection in patients with haematological malignancies and after stem cell transplantation from the 2017 European Conference on Infections in Leukaemia (ECIL 7) . Lancet Infect Dis . 2019 ; 19 ( 8 ): e260 ‐ e272 . 31153807 10.1016/S1473-3099(19)30107-0

3 Ozdemir E , Saliba RM , Champlin RE , et al. Risk factors associated with late cytomegalovirus reactivation after allogeneic stem cell transplantation for hematological malignancies . Bone Marrow Transplant . 2007 ; 40 ( 2 ): 125 ‐ 136 . 17530009 10.1038/sj.bmt.1705699

4 Teira P , Battiwalla M , Ramanathan M , et al. Early cytomegalovirus reactivation remains associated with increased transplant‐related mortality in the current era: a CIBMTR analysis . Blood . 2016 ; 127 ( 20 ): 2427 ‐ 2438 . 26884374 10.1182/blood-2015-11-679639 PMC4874224

5 Green ML , Leisenring W , Xie H , et al. Cytomegalovirus viral load and mortality after haemopoietic stem cell transplantation in the era of pre‐emptive therapy: a retrospective cohort study . Lancet Haematol . 2016 ; 3 ( 3 ): e119 ‐ e127 . 26947200 10.1016/S2352-3026(15)00289-6 PMC4914379

6 Zavras P , Su Y , Fang J , et al. Impact of preemptive therapy for cytomegalovirus on toxicities after allogeneic hematopoietic cell transplantation in clinical practice: a retrospective single‐center cohort study . Biol Blood Marrow Transplant . 2020 ; 26 ( 8 ): 1482 ‐ 1491 . 32315708 10.1016/j.bbmt.2020.03.019 PMC8220837

7 Akahoshi Y , Kimura SI , Inamoto Y , et al. effect of cytomegalovirus reactivation with or without acute graft‐versus‐host disease on the risk of nonrelapse mortality . Clin Infect Dis . 2021 ; 73 ( 3 ): e620 ‐ e628 . 33341890 10.1093/cid/ciaa1871

8 Einsele H , Hebart H , Kauffmann‐Schneider C , et al. Risk factors for treatment failures in patients receiving PCR‐based preemptive therapy for CMV infection . Bone Marrow Transplant . 2000 ; 25 ( 7 ): 757 ‐ 763 . 10745262 10.1038/sj.bmt.1702226

9 Boeckh M , Leisenring W , Riddell SR , et al. Late cytomegalovirus disease and mortality in recipients of allogeneic hematopoietic stem cell transplants: importance of viral load and T‐cell immunity . Blood . 2003 ; 101 ( 2 ): 407 ‐ 414 . 12393659 10.1182/blood-2002-03-0993

10 Boeckh M , Gooley TA , Myerson D , Cunningham T , Schoch G , Bowden RA . Cytomegalovirus pp65 antigenemia‐guided early treatment with ganciclovir versus ganciclovir at engraftment after allogeneic marrow transplantation: a randomized double‐blind study . Blood . 1996 ; 88 ( 10 ): 4063 ‐ 4071 . 8916975

11 Busca A , de Fabritiis P , Ghisetti V , et al. Oral valganciclovir as preemptive therapy for cytomegalovirus infection post allogeneic stem cell transplantation . Transplant Infect Dis . 2007 ; 9 ( 2 ): 102 ‐ 107 . 10.1111/j.1399-3062.2006.00183.x 17461994

12 Ayala E , Greene J , Sandin R , et al. Valganciclovir is safe and effective as pre‐emptive therapy for CMV infection in allogeneic hematopoietic stem cell transplantation . Bone Marrow Transplant . 2006 ; 37 ( 9 ): 851 ‐ 856 . 16532016 10.1038/sj.bmt.1705341

13 Reusser P , Einsele H , Lee J , et al. Randomized multicenter trial of foscarnet versus ganciclovir for preemptive therapy of cytomegalovirus infection after allogeneic stem cell transplantation . Blood . 2002 ; 99 ( 4 ): 1159 ‐ 1164 . 11830461 10.1182/blood.v99.4.1159

14 Ljungman P , Deliliers GL , Platzbecker U , et al. Cidofovir for cytomegalovirus infection and disease in allogeneic stem cell transplant recipients. The infectious diseases working party of the European group for blood and marrow transplantation . Blood . 2001 ; 97 ( 2 ): 388 ‐ 392 . 11154213 10.1182/blood.v97.2.388

15 Marty FM , Bryar J , Browne SK , et al. Sirolimus‐based graft‐versus‐host disease prophylaxis protects against cytomegalovirus reactivation after allogeneic hematopoietic stem cell transplantation: a cohort analysis . Blood . 2007 ; 110 ( 2 ): 490 ‐ 500 . 17392502 10.1182/blood-2007-01-069294 PMC1924486

16 Green ML , Leisenring W , Stachel D , et al. Efficacy of a viral load‐based, risk‐adapted, preemptive treatment strategy for prevention of cytomegalovirus disease after hematopoietic cell transplantation . Biol Blood Marrow Transplant . 2012 ; 18 ( 11 ): 1687 ‐ 1699 . 22683614 10.1016/j.bbmt.2012.05.015 PMC3467354

17 Avetisyan G , Aschan J , Hagglund H , Ringden O , Ljungman P . Evaluation of intervention strategy based on CMV‐specific immune responses after allogeneic SCT . Bone Marrow Transplant . 2007 ; 40 ( 9 ): 865 ‐ 869 . 17724444 10.1038/sj.bmt.1705825

18 Ganepola S , Gentilini C , Hilbers U , et al. Patients at high risk for CMV infection and disease show delayed CD8+ T‐cell immune recovery after allogeneic stem cell transplantation . Bone Marrow Transplant . 2007 ; 39 ( 5 ): 293 ‐ 299 . 17262060 10.1038/sj.bmt.1705585

19 Moins‐Teisserenc H , Busson M , Scieux C , et al. Patterns of cytomegalovirus reactivation are associated with distinct evolutive profiles of immune reconstitution after allogeneic hematopoietic stem cell transplantation . J Infect Dis . 2008 ; 198 ( 6 ): 818 ‐ 826 . 18666855 10.1086/591185

20 Zhou W , Longmate J , Lacey SF , et al. Impact of donor CMV status on viral infection and reconstitution of multifunction CMV‐specific T cells in CMV‐positive transplant recipients . Blood . 2009 ; 113 ( 25 ): 6465 ‐ 6476 . 19369230 10.1182/blood-2009-02-203307 PMC2710937

21 Ljungman P , Perez‐Bercoff L , Jonsson J , et al. Risk factors for the development of cytomegalovirus disease after allogeneic stem cell transplantation . Haematologica . 2006 ; 91 ( 1 ): 78 ‐ 83 . 16434374

22 Ljungman P , Aschan J , Lewensohn‐Fuchs I , et al. Results of different strategies for reducing cytomegalovirus‐associated mortality in allogeneic stem cell transplant recipients . Transplantation . 1998 ; 66 ( 10 ): 1330 ‐ 1334 . 9846518 10.1097/00007890-199811270-00012

23 Martino R , Rovira M , Carreras E , et al. Severe infections after allogeneic peripheral blood stem cell transplantation: a matched‐pair comparison of unmanipulated and CD34+ cell‐selected transplantation . Haematologica . 2001 ; 86 ( 10 ): 1075 ‐ 1086 . 11602414

24 Miller W , Flynn P , McCullough J , et al. Cytomegalovirus infection after bone marrow transplantation: an association with acute graft‐vs‐host disease . Blood . 1986 ; 67 ( 4 ): 1162 ‐ 1167 . 3006831

25 Walker CM , van Burik JA , De For TE , Weisdorf DJ . Cytomegalovirus infection after allogeneic transplantation: comparison of cord blood with peripheral blood and marrow graft sources . Biol Blood Marrow Transplant . 2007 ; 13 ( 9 ): 1106 ‐ 1115 . 17697973 10.1016/j.bbmt.2007.06.006

26 Jain NA , Lu K , Ito S , et al. The clinical and financial burden of pre‐emptive management of cytomegalovirus disease after allogeneic stem cell transplantation‐implications for preventative treatment approaches . Cytotherapy . 2014 ; 16 ( 7 ): 927 ‐ 933 . 24831837 10.1016/j.jcyt.2014.02.010 PMC4051841

27 Svahn BM , Remberger M , Alvin O , Karlsson H , Ringden O . Increased costs after allogeneic haematopoietic SCT are associated with major complications and re‐transplantation . Bone Marrow Transplant . 2012 ; 47 ( 5 ): 706 ‐ 715 . 21874063 10.1038/bmt.2011.162

28 Webb BJ , Harrington R , Schwartz J , et al. The clinical and economic impact of cytomegalovirus infection in recipients of hematopoietic stem cell transplantation . Transplant Infect Dis . 2018 ; 20 ( 5 ): e12961 . 10.1111/tid.12961 29975816

29 Schelfhout J , Brown H , House JA , Raval AD . Cytomegalovirus infection and associated hospitalization and costs among individuals undergoing allogeneic hematopoietic stem cell transplant . Curr Med Res Opin . 2020 ; 36 ( 1 ): 43 ‐ 50 . 31490097 10.1080/03007995.2019.1664827

30 Schelfhout J , Bonafede M , Cappell K , Cole AL , Manjelievskaia J , Raval AD . Impact of cytomegalovirus complications on resource utilization and costs following hematopoietic stem cell transplant . Curr Med Res Opin . 2019 : 1 ‐ 9 . 10.1080/03007995.2019.1664826 31490093

31 El Haddad L , Ghantoji SS , Park AK , et al. Clinical and economic burden of pre‐emptive therapy of cytomegalovirus infection in hospitalized allogeneic hematopoietic cell transplant recipients . J Med Virol . 2020 ; 92 ( 1 ): 86 ‐ 95 . 31448830 10.1002/jmv.25574 PMC6842396

32 Puhakka L , Sarvikivi E , Lappalainen M , Surcel HM , Saxen H . Decrease in seroprevalence for herpesviruses among pregnant women in Finland: cross‐sectional study of three time points 1992, 2002 and 2012 . Infect Dis (Lond) . 2016 ; 48 ( 5 ): 406 ‐ 410 . 26654892 10.3109/23744235.2015.1123290

33 Marty FM , Ljungman P , Chemaly RF , et al. Letermovir Prophylaxis for Cytomegalovirus in Hematopoietic‐Cell Transplantation . N Engl J Med . 2017 ; 377 ( 25 ): 2433 ‐ 2444 . 29211658 10.1056/NEJMoa1706640

34 Gooptu M , Kim HT , Ho VT , et al. A comparison of the myeloablative conditioning regimen fludarabine/busulfan with cyclophosphamide/total body irradiation, for allogeneic stem cell transplantation in the modern era: a cohort analysis . Biol Blood Marrow Transplant . 2018 ; 24 ( 8 ): 1733 ‐ 1740 . 29555313 10.1016/j.bbmt.2018.03.011

35 Dulery R , Bastos J , Paviglianiti A , et al. Thiotepa, busulfan, and fludarabine conditioning regimen in T cell‐replete HLA‐haploidentical hematopoietic stem cell transplantation . Biol Blood Marrow Transplant . 2019 ; 25 ( 7 ): 1407 ‐ 1415 . 30871978 10.1016/j.bbmt.2019.02.025

36 Reshef R , Porter DL . Reduced‐intensity conditioned allogeneic SCT in adults with AML . Bone Marrow Transplant . 2015 ; 50 ( 6 ): 759 ‐ 769 . 25730186 10.1038/bmt.2015.7

37 Rodríguez‐Arbolí E , Labopin M , Tischer J , et al. FLAMSA‐based reduced‐intensity conditioning versus myeloablative conditioning in younger patients with relapsed/refractory acute myeloid leukemia with active disease at the time of allogeneic stem cell transplantation: an analysis from the acute leukemia working party of the European society for blood and marrow transplantation . Biol Blood Marrow Transplant . 2020 ; 26 ( 11 ): 2165 ‐ 2173 . 32717436 10.1016/j.bbmt.2020.07.020

38 Nichols WG , Corey L , Gooley T , Davis C , Boeckh M . High risk of death due to bacterial and fungal infection among cytomegalovirus (CMV)‐seronegative recipients of stem cell transplants from seropositive donors: evidence for indirect effects of primary CMV infection . J Infect Dis . 2002 ; 185 ( 3 ): 273 ‐ 282 . 11807708 10.1086/338624

39 Schmidt‐Hieber M , Labopin M , Beelen D , et al. CMV serostatus still has an important prognostic impact in de novo acute leukemia patients after allogeneic stem cell transplantation: a report from the Acute Leukemia Working Party of EBMT . Blood . 2013 ; 122 ( 19 ): 3359 ‐ 3364 . 24037724 10.1182/blood-2013-05-499830

40 Kaito S , Nakajima Y , Hara K , et al. Heterogeneous impact of cytomegalovirus reactivation on nonrelapse mortality in hematopoietic stem cell transplantation . Blood Adv . 2020 ; 4 ( 6 ): 1051 ‐ 1061 . 32191806 10.1182/bloodadvances.2019000814 PMC7094017

41 Ljungman P , Schmitt M , Marty FM , et al. A mortality analysis of letermovir prophylaxis for cytomegalovirus (CMV) in CMV‐seropositive recipients of allogeneic hematopoietic cell transplantation . Clin Infect Dis . 2020 ; 70 ( 8 ): 1525 – 1533 . 31179485 10.1093/cid/ciz490 PMC7146004

42 Litjens NHR , van der Wagen L , Kuball J , Kwekkeboom J . Potential beneficial effects of cytomegalovirus infection after transplantation . Front Immunol . 2018 ; 9 : 389 . 29545802 10.3389/fimmu.2018.00389 PMC5838002

43 Hilal T , Slone S , Peterson S , Bodine C , Gul Z . Cytomegalovirus reactivation is associated with a lower rate of early relapse in myeloid malignancies independent of in‐vivo T cell depletion strategy . Leukemia Res . 2017 ; 57 : 37 ‐ 44 . 28279876 10.1016/j.leukres.2017.02.009

44 Zhang YL , Zhu Y , Xiao Q , Wang L , Liu L , Luo XH . Cytomegalovirus infection is associated with AML relapse after allo‐HSCT: a meta‐analysis of observational studies . Ann Hematol . 2019 ; 98 ( 4 ): 1009 ‐ 1020 . 30666434 10.1007/s00277-018-3585-1

45 Fang J , Su Y , Zavras PD , et al. Impact of preemptive therapy for cytomegalovirus on hospitalizations and cost after hematopoietic stem cell transplantation . Biol Blood Marrow Transplant . 2020 ; 26 ( 10 ): 1937 ‐ 1947 . 32640313 10.1016/j.bbmt.2020.06.025 PMC8248281

46 Golan Y , Tang Y , Mt‐Isa S , et al. Impact of letermovir use for cytomegalovirus prophylaxis on re‐hospitalization following allogeneic hematopoietic stem cell transplantation: an analysis of a phase III randomized clinical trial . PharmacoEconomics ‐ Open . 2021 ; 5 : 469 ‐ 473 . 33871830 10.1007/s41669-021-00264-9 PMC8333192

47 Bevans M . Health‐related quality of life following allogeneic hematopoietic stem cell transplantation . Hematology Am Soc Hematol Educ Program . 2010 ; 2010 : 248 ‐ 254 . 21239801 10.1182/asheducation-2010.1.248 PMC3460526

48 Grulke N , Albani C , Bailer H . Quality of life in patients before and after haematopoietic stem cell transplantation measured with the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Core Questionnaire QLQ‐C30 . Bone Marrow Transplant . 2012 ; 47 ( 4 ): 473 ‐ 482 . 21602898 10.1038/bmt.2011.107