The primary objectives of this study were to describe: (a) the incidence of CMV infections (including but not limited to: reactivations, number of recurrences as well as incidence of resistant or refractory infection or disease); (b) incidence of CMV antiviral-related myelotoxicities, including leukopenia and neutropenia, and their impact on patient’s treatment adjustments (decrease/discontinuation of immunosuppressants and CMV antiviral agents); and (c) the association of CMV management with health care resource utilization (HCRU), including the length of hospital stay (LOS) and frequency of hospital admissions.

This retrospective, consecutive subject, multicenter cohort study was designed to examine the clinical and economic burden and unmet prophylactic needs in CMV seronegative adult recipients (R-) 18 years or older of a kidney transplant from a CMV seropositive donor (D+). Recipients who participated in an interventional trial in the prior 90 days, had received another organ or stem-cell transplant, or were exposure to letermovir were excluded. The study was conducted in seven major transplant centers in Canada ( Table 1 ) and included consecutive CMV seronegative recipients in each participating transplant program who received a kidney transplant from a CMV seropositive donor from January 1st, 2018 to December 31st, 2021, and for whom continuous follow-up data were available. A pragmatic sample of approximately 300 CMV D+/R- kidney transplant recipients was considered adequate to provide preliminary evidence of CMV treatment and burden.

The index date for each participant was the date of kidney transplant (Day 0) and data were recorded for the first year following transplantation. Primary data elements included, but were not limited to, patient demographics (age, age at time of first kidney transplant, race, proximity to transplant center); donor source (living/deceased), HLA matching; clinical therapies including induction regimen; CMV infection, syndrome, or disease; the incidence and management of CMV antiviral-related myelotoxicities; and healthcare resource use. Secondary data elements included time to CMV infection, viremia levels, viral load kinetics; event history (phlebitis, anemia, thrombocytopenia, pancytopenia, neutropenia, direct renal tubular toxicity, crystalline nephropathy, leukopenia, other); non-CMV opportunistic infections; graft outcomes; utilization patterns of CMV therapy; and incidence and management of other CMV antiviral-related toxicities.

The following characteristics were extracted from the patient charts to describe antiviral prophylaxis and treatment for CMV: whether anti-viral prophylaxis treatment was prescribed; the stage of use (primary [immediate post-transplant] or secondary [post-primary infection]); and the type of antiviral medication including the dose, unit; frequency and start and stop dates. The response of CMV infection to antiviral treatment was measured using the following data obtained from the patient chart: the date of the first episode of CMV infection; the duration of the episode (days); any change in treatment; (immunosuppression, antiviral prophylaxis, G-CSF, other); whether resistant or refractory CMV Infection or disease occurred (yes/no); and whether resistant infection or disease was confirmed (yes/no).

Myelotoxicity (determined as hematological values below the lower level of normal for each institution) and other adverse consequences of antiviral treatment were recorded using measures obtained from the patient chart. These included: whether CMV antiviral related myelotoxicities occurred (yes/no); the type of myelotoxicity (neutropenia, leukopenia, thrombocytopenia, pancytopenia); other antiviral toxicity recorded (nephrotoxicity, neurotoxicity, electrolyte disturbance, other); grading of the severity (using the NCI-CTCAE Grade 1 - 5) (29); whether toxicity was drug related (yes/no) and if so the drug name (ganciclovir, foscarnet sodium, immunoglobulin, valganciclovir, other); and any change in treatment (immunosuppression, antiviral prophylaxis, G-CSF, other).

All analyses were performed based on the Eligible Population (ELIG) which was a subset of all enrolled participants with complete data available for 365 days, or to the point of death or graft loss. Descriptive statistics were produced for all key study variables for the overall study population. Demographic data included baseline recipient characteristics. Therapeutic data included immunosuppressive therapy, anti-viral prophylaxis and treatment, and other principal therapies that could influence outcome or disease prevalence. Clinical and laboratory data included CMV Ig status at the time of transplant, quantitative viremia, CMV syndrome and disease, graft and participant survival and other outcome variables of interest.

Continuous variables were summarized using the number of non-missing observations, mean, standard deviation (SD), median, minimum, and maximum values and interquartile ranges. Categorical variables were summarized using the number and percentage of participants belonging to each category. Event rates (summarized as the number of participants experiencing that event) and incidence rates (summarized as the total number of events in a group divided by the time at risk), were summarized within the study population and sub- populations of interest for all the dependent variables (e.g. CMV viremia, CMV resistance, developing refractory CMV). Time to event endpoints were calculated using the Kaplan-Meier estimator, and differences between groups were evaluated using the log-rank test. Analysis was performed using SAS version 9.4.

This study including all relevant documentation was approved by the Research Ethics Boards at each of the participating sites.

A total of 311 consecutive CMV D+/R- kidney recipients within the index period were enrolled across seven Canadian centers ( Table 1 ). There were no screen failures or exclusions. Of these, 103 received a transplant from a living donor (LD) and 208 from a deceased donor (DD); the median age (IQR) at baseline was 58 (46-67) years and at the first kidney transplant was 55 (42, 64) years; and 69.5% were male and 52.7% were Caucasian. Most participants (68%) lived within 100 km of their transplant center. Primary reasons for kidney failure were diabetes (19.0%), polycystic kidney disease (13.8%), IgA nephropathy (12.5%), other glomerulonephritis (12.2%), or for reasons not specifically defined (26.0%). Most grafts were HLA mismatched (97.7%) with typically two mismatches at HLA-A (44.5%), HLA-B (60.1%), HLA-C (45.4%), HLA-DR (49.6%), HLA-DQ (53.5%) and HLA-DP (47.2%) for those reporting these values. No clinically meaningful differences between those receiving an LD or DD graft were observed with regards to participant baseline characteristics.

Most patients received induction with basiliximab (61.4%), ATG (24.4%) or another agent (14.1%), principally alemtuzumab, and 93.9% received mycophenolate mofetil (MMF) or mycophenolic acid (MPA) as maintenance immunosuppression, 87.1% received tacrolimus and 77.2% prednisone ( Table 2 ). Most participants (62.4%) received all 3 maintenance agents, while 33.4% received two and 4.2% only one. All participants received antiviral CMV prophylaxis, principally with valganciclovir (293, 90.2%), started a median of 1 day post-transplant and lasting for a median of 180 days (range: 2–465 days). Most patients received only one course of antiviral therapy (76.2%), though almost one quarter received two (19.0%) or more courses (4.8%). The duration of each course is shown in Table 2 . There were no clinically meaningful differences between LD and DD recipients with regards to antiviral prophylaxis. Less than 2% of participants had documented toxicities related to antiviral therapies, of which 28.6% were related to valganciclovir, 14.3% to ganciclovir and 57.1% to other agents (tacrolimus) with antivirals. Outside of myelotoxicity (discussed below), the most common toxicities were neurotoxicity, nephrotoxicity, electrolyte disturbances and other toxicities that were mild (grade 1 – 2). No participants experienced symptoms related to G-CSF administration.

A total of 106 (34.1%) participants developed post-transplant CMV infection, of whom 15 (14.2%) (had two and 12 (11.3%) had three or more episodes of infection ( Table 3 ). Of these 106 patients, 47 (44.3%) had asymptomatic viremia during follow-up while 59 (55.6%) had CMV disease presenting principally with fever (21.9%), malaise (33.3%), leukopenia (28%) and transaminitis (12%). CMV infection occurred approximately 90 days post-transplant and increased in frequency to a plateau at the end of the first year ( Figure 1 ). Breakthrough CMV infection occurred during prophylaxis in approximately 23% of these patients for whom precise dates were available (n=95). CMV disease was diagnosed a median of 223 days (IQR 165, 257) post-transplant, lasted for a median of 34 days (IQR 23, 41)and resolved by 261 days (IQR 192, 306) days post-transplant. Only 7 of these patients (11.9%) had end-organ disease, manifested principally by gastrointestinal or neurological symptoms. CMV infection was slightly more common in DD than LD recipients (viremia: 36.5% vs 29.1%, symptomatic: 21.2% vs 14.6%) but there was no clinically meaningful or statistically significant difference between LD and DD recipients. Peak viral load during the first infection occurred at a median of 218 days (IQR: 154, 260) days post-transplant with a median value of 14,224 IU/ml (IQR: 3,900, 127,050 IU/ml). Peak viral loads measured in subsequent viremic episodes were later (up to 352 days) and lower (maximum 32,007 IU/ml). Eighty-one patients received a second course of antiviral treatment, including 69 (85.2%) with valganciclovir and 9 (11.1%) with ganciclovir.

A total of 121 (38.9%) patients experienced antiviral related myelotoxicities classified principally as leukopenia or neutropenia, with a frequency that was similar in LD (39.8%) and DD (38.5%) recipients. Cumulative incidence analysis showed that myelotoxicity was observed within 30 days post-transplant, rising rapidly during the first 3 months to 34.5% at day 100 and 46.8% by the end of the first year of follow-up ( Figure 2 ). The median onset was 90 (IQR 65, 133) days post-transplant, median duration was 30 (IQR 14, 74) days and only 6 patients (4.9%) had an NCI-CTCAE severity grade above 3. Immunosuppressive therapy was reduced in almost 80% of the patients with myelotoxicity, being decreased on 65(53.7%) or discontinued or interrupted on 43 (35.1%) of the 134 changes recorded, and antiviral therapy was similarly decreased or discontinued in 34 of the 50 (68%) patients on treatment at the time of myelotoxicity. Other agents such as trimethoprim-sulfamethoxazole were discontinued in 9.1% of patients and G-CSF was implemented in 22 (18.2%) patients during the episode of myelotoxicity. A second episode of myelotoxicity occurred in 14 (11.5%) of these patients at a median of 136 (IQR 89, 167) days, with a duration of 39 (IQR 24, 46) days, of which none had an NCI-CTCAE severity over grade 3. Immunosuppressive therapy was decreased, interrupted or discontinued in 94.5%, antiviral therapy discontinued or interrupted in 28.6% and G-CSF was started in 14.3% of these patients. A third or greater episode of myelotoxicity occurred in 5 (4.1%) of patients a median of 174 (IQR 161, 247) days post-transplant, lasting for 70 (IQR 60,80) days, in which no events were above Grade 2 severity. Immunosuppression was decreased or discontinued/interrupted in 2 (40%) patients, and antiviral prophylaxis discontinued or interrupted in 1 (20%) patient. No patients received G-CSF.

A total of 119 (38.3%) patients experienced opportunistic infections post-transplant, which were significantly more common in DD (44.2%) than LD (26.2%) recipients (p=0.002) ( Table 4 ). Cumulative incidence analysis ( Figure 3 ) showed that infections were first observed within 30 days post-transplant, rising to an inflection at 60–90 days and increasing slowly thereafter to approximately 38% by the end of the first year, with a median onset of 53 (IQR: 14, 142) days post-transplant and a median duration of 28 (IQR: 15, 62). Recurrent opportunistic infections were recorded in 45 (14.5%) of these patients, of whom 15 (4.8%) had three or more episodes. Median onset for the second and third or more episodes was 137 (IQR 77, 213) days and 181 (IQR 109, 236) days respectively, with a duration of 22 (IQR 14, 49) days and 29 (IQR 18, 50) days. Of the 179 total episodes of opportunistic infection reported, 98 (54.7%) were bacterial, 56 (31.3%) were non-CMV viral infections and 30 (16.7%) were fungal in origin. Almost all (97.8%) were Grade 1–3 in severity.

As shown in Table 5 , there were no significant differences in the incidence, number of episodes, causal agents or severity of opportunistic infections between patients who did, or did not, experience post-transplant CMV infection (p=NS), Opportunistic infection occurred significantly later in patients with CMV infection (median: 86, IQR: 20, 180 days, p=0.037) than in those without CMV infection (median: 43, IQR: 14, 127 days) though there was no difference in duration of opportunistic infections between the two groups (median: 28 days).

A total of 141 (45.3%) participants were re-hospitalized post-transplantation, 103 (36.9%) in the living donor and 38 (49.5%) in the deceased donor group. Of these admissions, 25 (17.7%) were related to CMV infection or leukopenia/neutropenia. Cumulative incidence analysis ( Figure 4 ) showed that re-hospitalization commenced early, rising to 28% by day 30 and continuing to 45% by the end of the first year of observation. The median time to first re-hospitalization was 58 (IQR: 18, 158) days and the median duration was 5 (IQR: 2, 11) days. Of these patients, 9 (6.3%) required admission to ICU, for a median of 14 (IQR: 5, 14) days. Fifty (35,4%) of these 141 patients were re-hospitalized a second time, of which 6 (12%) were related to CMV infection or leukopenia/neutropenia. The median time to re-admission was 114 (IQR: 49, 205) days, and median duration was 6 (IQR: 4, 10) days. Two patients (4%) were admitted to ICU for a median of 2 days. And 31 (21.9%) patients were re-hospitalized 3 or more times of which 3 (9.6%) were for CMV infection or neutropenia. The third admission was a median of 145 (IQR: 95, 249) days post-transplant, lasting for a median of 7 (IQR: 3, 12) days and 2 were admitted to ICU for a median of 3 days.

As shown in Table 6 , the proportion of patients who were re-hospitalized did not differ significantly between patients with or without CMV infection post-transplant (p=NS), But re-hospitalization occurred substantially later in patients with CMV infection (median: 122, IQR: 21, 236 days, p<0.0001) than in those without CMV infection (median: 37, IQR: 16, 130 days) but there was no difference in duration of hospital stay between the two groups (median: 5 days, p=NS).

In the year following transplantation, all participants (100%) saw a nephrologist (total 5,285 visits, mean 17.0), 82.6% a urologist (total 425 visits, mean 1.4), 57.7% another healthcare provider (total 866 visits, mean 2.8), 39.9% an ambulatory nurse (total 1,293, mean 4.1) and 20.1% a general practitioner (total 162, mean 0.5). There were no clinically meaningful differences between LD or DD recipients, with regards to health care visits.

Biopsy-proven acute graft rejection (BPAR) occurred in 20 patients (6.4%) overall (LD 4.9% and DD 7.2%, p=NS) after a median of 107 (IQR: 36, 202) days. Three of these patients received biologic therapy with rituximab or ATG. Ten patients (3.2%) lost their graft (LD 1.0%, DD 4.3%), all for reasons other than rejection. By 1 year post-transplant, 10 patients (3.2%) had died (LD 1.0%, DD 4.3%). Death occurred at a median of 152 days (IQR: 140, 201) days post-transplant due principally to cardiovascular events (n=5) and other causes (n=3). Only 1 case was reported related to CMV infection. Overall, 293 of the 311 patients (94.2%) were alive with a functioning graft by 1 year (LD 98.1%, DD 92.3%) ( Figure 5 ).

Graft and patient survival were similar in patients with or without post-transplant CMV infection ( Figure 6 ). BPAR was reported in 8 (7.5%) participants with post-transplant CMV infections compared to 12 (5.9%) without, although rejection occurred later in the former at a median of 150 days (IQR: 98, 248) days compared with a median of 64 days (IQR: 14, 199) days in those without infection. Graft loss was higher in patients without CMV infection (4.4% vs 0.9%) and 4 participants (3.8%) with CMV and 6 participants (2.9%) without CMV had died by the end of the follow-up period. Overall, 101 (95.3%) and 192 (93.7%) in each group were alive with a functioning graft by 1 year post-transplant. [Post-text Table 14.10.2].