CMV reactivation during pretransplantation evaluation: a novel risk factor for posttransplantation CMV reactivation

Abstract

Cytomegalovirus (CMV) disease occurs occasionally before allogeneic hematopoietic cell transplantation (HCT) and is associated with poor post-HCT outcomes; however, the impact of pre-HCT CMV reactivation is unknown. Pre-HCT CMV reactivation was assessed in HCT candidates from the preemptive antiviral therapy (2007-2017) and letermovir prophylaxis (2018-2021) eras. CMV DNA polymerase chain reaction (PCR) surveillance was routinely performed during the pre-HCT workup period, and antiviral therapy was recommended according to risk of progression to CMV disease. Risk factors for pre-HCT CMV reactivation were characterized, and the associations of pre-HCT CMV reactivation with post-HCT outcomes were examined using logistic regression and Cox proportional hazard models, respectively. A total of 1694 patients were identified, and 11% had pre-HCT CMV reactivation 14 days (median; interquartile range [IQR], 6-23) before HCT. Lymphopenia (≤0.3 × 103/μL) was the strongest risk factor for pre-HCT CMV reactivation at multiple PCR levels. In the preemptive therapy era, patients with pre-HCT CMV reactivation had a significantly increased risk of CMV reactivation by day 100 as well as CMV disease and death by 1 year after HCT. Clearance of pre-HCT CMV reactivation was associated with a lower risk of post-HCT CMV reactivation. Similar associations with post-HCT CMV end points were observed in a cohort of patients receiving letermovir prophylaxis. Pre-HCT CMV reactivation can be routinely detected in high-risk HCT candidates and is a significant risk factor for post-HCT CMV reactivation and disease. Pre-HCT CMV DNA PCR surveillance is recommended in high-risk HCT candidates, and antiviral therapy may be indicated to prevent post-HCT CMV reactivation.

Graphical abstract

Figure 1.

Proportion of patients with pre-HCT CMV reactivation and pre-HCT CMV DNA PCR levels by number of risk factors. Proportion of patients with detectable CMV DNAemia by PCR from 1 to 90 days before HCT (left) and corresponding levels of pre-HCT CMV DNAemia in these patients (right) stratified by the number of pre-HCT risk factors present, including HCT-CI score ≥3, absolute lymphocyte count ≤0.3× 10³/μL, and underlying lymphoid malignancy.

Figure 2.

Risk factors for pre-HCT CMV reactivation at multiple levels. Multivariable logistic regression of any (A), ≥150 IU/mL (B), or ≥500 IU/mL (C) viral load CMV in the 90 days before HCT transplant in patients from both eras. ⁺Denotes the lowest pretransplant lymphocyte counts within 7 to 90 days before transplant. ∗Denotes inclusion of unknown or unreported HCT-CI scores. Note that underlying disease categories were collapsed in the pre-HCT reactivation ≥150 and ≥500 IU/mL models due to a lower number of events in each. Specifically, categories include the following: (1) lymphoid malignancies (acute lymphocytic leukemia [ALL], chronic lymphocytic leukemia [CLL], Hodgkin lymphoma [HL], non-Hodgkin lymphoma [NHL], and plasma cell neoplasms); (2) myeloid malignancies (acute myelogenous leukemia [AML], myelodysplastic syndrome [MDS], and other myeloproliferative neoplasms [MPNs]); and (3) Nonmalignant diseases (a variety of diseases such as aplastic anemia [AA] or primary immune deficiencies). 95% CI, 95% confidence interval; PLL, prolymphocytic leukemia.

Figure 3.

Cumulative incidences of CMV reactivation in the first 100 days post-HCT stratified by the presence of pre-HCT CMV DNAemia. Cumulative incidence curve showing the time to any CMV infection (A,D), to ≥150 IU/mL CMV infection (B,E), or to ≥500 IU/mL CMV infection (C,F), among preemptive therapy (PET)–era patients (n = 1367; A-C) or among participants receiving letermovir (LET) prophylaxis (n = 332; D-F) within the first 100 days after transplant, separated by whether any pretransplant CMV was detected by PCR (yes in red [top line]; no in blue [bottom line]). Lighter colored shading indicates the 95% confidence interval. In each graph, the difference between those with any pretransplant CMV PCR detection was calculated using Gray’s test; associated P values are listed on the right of the graphs.

Figure 4.

Cumulative incidences of CMV reactivation in the first 100 days post-HCT stratified by the last pre-HCT CMV DNA PCR test result. Cumulative incidence curve showing the time to any CMV infection (A,D), time to CMV viral load ≥150 IU/mL (B,E), or time to CMV viral load ≥500 IU/mL (C,F), among PET era patients (n = 1367; A-C) or LET era patients (n = 332; D-F) in the first 100 days after HCT, dependent on the presence (green dashed line), absence (blue solid line), or clearance (red dotted line) of pre-HCT CMV DNAemia. Lighter colored shading indicates the 95% confidence interval. In each graph, the difference between those with any vs no pretransplant CMV PCR detection was calculated using Gray’s test; P values are listed on the right of the graphs.

Figure 5.

Associations of pre-HCT CMV reactivation (with or without clearance after preemptive antiviral therapy) and of last pre-HCT CMV DNA PCR test result with the risk of post-HCT CMV reactivation at multiple levels. Multivariable Cox regression of the risk of post-HCT CMV reactivation associated with the absence or presence of pre-HCT CMV with or without clearance after preemptive antiviral therapy (above demarcation line in each plot, as circles) or by the last pre-HCT CMV PCR test (below demarcation line in each plot, as diamonds). Evaluated post-HCT CMV end points include any level (A), ≥150 IU/mL (B), or ≥500 IU/mL (C). ˆReference (ie, no CMV detection within 7-90 days before HCT) includes a small number of patients with pre-HCT CMV reactivation ≥50 and <150 IU/mL who became negative before HCT without preemptive antiviral therapy (n = 17). ⁼Indicates inclusion of patients with pre-HCT CMV reactivation ≥50 IU/mL who received preemptive antiviral therapy and became negative before HCT. ^#Indicates inclusion of patients with pre-HCT CMV reactivation ≥50 IU/mL who received preemptive antiviral therapy and did not become negative before HCT. Models were also adjusted for recipient age, race, underlying disease, HCT-CI score, transplantation year, donor CMV serostatus, HLA matching, GVHD prophylaxis, and acute GVHD grade (time-dependent variable). 95% CI, 95% confidence interval.