Introduction

Cytomegalovirus (CMV) is a DNA virus which is a member of the Herpesviridae family. The prevalence of this infection is high worldwide, ranging between 60 and 100% [1]. The prevalence of CMV demonstrated by IgG positivity in our population is as high as 95–97% [2]. Primary CMV infection is mostly asymptomatic and self-limiting in nature in immunocompetent hosts and is characterized by persistence of the virus in a latent form in the body. However, CMV infection in the immunocompromised host, such as in patients undergoing allogeneic HSCT can be a cause of significant morbidity and mortality.

CMV has both direct and indirect effects on the body. Direct effects are caused by viral proliferation in various host tissues leading to manifestations such as gastroenteritis, pneumonitis, hepatitis, retinitis, encephalitis, etc. On the other hand, increased risk of graft rejection and dysfunction, GvHD, opportunistic infections and malignancies are a sequela of the indirect effects of the virus due to persistent low-level replication in the body [3]. Activation of the immune system by viral proteins predisposes to graft rejection and acute or chronic GvHD, whereas immunosuppressive effects of CMV in turn lead to an increased risk of other opportunistic fungal and bacterial infections. Boeckh and Nichols demonstrated a higher incidence of acute GvHD in CMV-seropositive recipients after allogeneic HSCT than seronegative ones [4]. CMV is associated with GvHD and rejection; on the other hand, both rejection and GvHD promote CMV replication.

Despite several advances in the field of HSCT, CMV reactivation post-transplant remains a significant factor influencing transplant outcomes, especially in a developing country like India, where the majority of patients and donors are CMV seropositive. The factors predisposing to CMV reactivation include pre-transplant donor and recipient serological status, the degree of human leucocyte antigen match, the presence of GvHD, the degree and duration of immunosuppression, the conditioning regimen, and the use of T cell-depleted grafts [5, 6]. With the progressive increase in the number of alternate donor transplants being performed, the burden of CMV infection post HSCT has increased manifold. Reported CMV reactivation rates after HSCT are variable, ranging between 30 and 70% and is associated with a higher non-relapse mortality rate (relative risk (RR), 1.61–1.95) [7].

This study aimed to analyze the burden of CMV reactivation in patients undergoing allogeneic HSCT at our center and to study its correlation with all-cause mortality. In addition, we analyzed the impact of CMV reactivation on all-cause mortality and overall survival (OS).

Patients and Methods

This study was a single-center retrospective study conducted at the Blood and Marrow Transplantation Unit at Apollo Specialty Cancer Hospital, Chennai. We included all children up to 18 years of age who underwent allogeneic HSCT between February 2002 to December 2021. We obtained institutional ethics committee approval before the study. We collected the data through a retrospective review of patient charts and medical records. The data analyzed included the incidence of CMV reactivation, its correlation with donor and graft type, source of stem cells, underlying diagnosis, and acute and chronic GvHD.

CMV Monitoring Protocol

CMV viral load was monitored in blood using real time quantitative DNA PCR analysis (Argene SA assay, Biomerieux) every week starting from the time of engraftment until day + 100 for all children undergoing MUD and haploidentical HSCT. Based on a previous study conducted at our center [8], patients undergoing MFD HSCTs were monitored for CMV only if they met any of the following criteria—underlying diagnosis of inborn errors of immunity, presence of GvHD, or when they received granulocyte transfusions. Beyond day + 100, CMV monitoring was individualized and was performed in those patients with GvHD requiring prolonged immunosuppression. CMV viremia greater than 1000 copies/ml was considered to be positive, and pre-emptive treatment was started with ganciclovir or valganciclovir.

Definitions

Acute GvHD was diagnosed and graded according to Glucksberg criteria [8] and chronic GvHD was diagnosed according to the National Institutes of Health consensus criteria [9]. All-cause mortality was defined as death due to any cause. Overall Survival (OS) was calculated as the time from transplantation to death due to any reason.

Statistical Analysis

The continuous variables were represented as means ± SDs and as medians if they were usually or nonnormally distributed, respectively. The comparison of typically, nonnormally, categorical variables was performed by independent sample t-test, Mann–Whitney test, chi-square test, or Fisher exact test, respectively. Kaplan–Meier curve was drawn to assess the survival pattern. Data validation and analysis were carried out by SPSS version 25.0 (SPSS, Inc., Chicago, IL). P values < 0.05 were considered statistically significant.

Results

We included 1035 patients in the study, with a median age of 6 years (0.1–18 years). The study group comprised 647 males and 388 females (M: F ratio 1.66:1). Table 1 highlights the details of the patient demographics, disease, and HSCT data.

Table 1 Demographic, disease, patient and HSCT data in the cohort n = 1035
Full size table

CMV Reactivation

CMV reactivation was documented in 258/1035 (24.9%) patients. Among the 258 children with CMV reactivation, the incidence was highest among those with haploidentical donor − 142 (51%), followed by 77 (36.1%) patients in the MUD group and 39 (7.2%) in the MFD group (p-value < 0.0001). Incidence of CMV reactivation was 216/745 (28.9%) among those receiving peripheral blood, 18/201 (9%) in those receiving bone marrow, and 22/45 (48.8%) in those receiving cord as stem cell source (p-value < 0.0001). There was no significant difference between T replete −106 (49.7%), and T deplete—36 (54.5%) haploidentical HSCT (p-value 0.49). Table 2 highlights the association of CMV reactivation with demographic and HSCT parameters.

Table 2 Correlation of CMV reactivation with demographic and HSCT parameters in the cohort n = 1035
Full size table

On subset analysis in 205 children who underwent alternate donor HSCT including MUD and haploidentical HSCT, CMV reactivation was documented in 84/205 (40%) children. Among the 84 children, donor recipient serostatus for CMV prior to HSCT was noted to be D + R + in 65/84 (77%) children followed by D + R− in 13/84 (15%).

Relation of CMV Reactivation with All-Cause Mortality

There was a total of 255 deaths (24.6%). The overall mortality rate was significantly higher in the CMV positive group compared to the CMV negative group (39.9 versus 19.6%, p-value = 0.0001). Among the 103 mortalities in the CMV positive group, only 13 (12.6%) deaths were attributed to CMV disease, of which 9 (69.2%) were in patients diagnosed with inborn errors of immunity. Figure 1 highlights our finding that GvHD as a cause of death was significantly higher among those with CMV infection (31 versus 10%, p-value = 0.0001). The mean survival time was 13.7 years (range 13.1–14.4) in the CMV negative group versus 9.2 years (range 7.5–11) in the CMV positive group (p-value = 0.0001) (Fig. 2).

Fig. 1
figure 1

Comparison of causes of death in relation to CMV reactivation

Full size image
Fig. 2
figure 2

Kaplan Meier survival curve analysis depicting mean survival time of 13.7 years (range 13.1–14.4) in the CMV negative group, versus 9.2 years (range 7.5–11) in the CMV positive group (p value = 0.0001)

Full size image

Discussion

In our study, the overall CMV infection rate post HSCT in the entire cohort was 25%. The risk factors for CMV reactivation were the source of stem cells (peripheral blood and cord), the type of donor (MUD and haploidentical), and the presence of acute and chronic GvHD. In addition, CMV was associated with increased overall mortality but low attributable mortality.

In a previous study done at our center, we documented CMV reactivation in 3% of children undergoing MFD transplants, 33.3% undergoing MUD HSCT using peripheral blood as stem cell source, 17.4% undergoing unrelated donor cord blood HSCT and 36.5% undergoing mismatched or haploidentical grafts (P =  < 0.0001) [10]. Other studies from India have reported variable CMV reactivation rates of 43.8% [11] and 9.56% [12] post HSCT. Hugo Sousa [13] observed CMV infection in 60.3% of their patients undergoing HSCT in Portugal, with an increased risk in those undergoing mismatched related or unrelated donor transplants. However, they found no significant association between stem cell sources and the presence of GVHD.

Broers et al. [14] noted CMV seropositivity to be a significant adverse risk factor for transplant-related morbidity and mortality (TRM) despite the very efficient prevention of CMV disease by preemptive therapy. Patients with CMV seropositivity experienced an increased incidence of acute GvHD, which was associated with increased TRM and decreased OS. GvHD itself and the drugs used to treat it both cause immunosuppression. On the other hand, CMV can lead to the production of inflammatory cytokines, which play a role in the initiation of GVHD [15]. Jain et al. [16] and Meet et al. [11] both found a significant relationship between CMV reactivation and GvHD.

Studies have shown CMV reactivation to be associated with decreased OS [13]. In addition, any level of CMV viremia in seropositive patients had a significant impact on all-cause mortality and non-relapse mortality despite the use of pre-emptive therapy [17]. However, Meet et al. [11] found no differences in the OS of patients with or without CMV reactivation in their study.

The fact that CMV replication is associated with mortality is the principal reason to advocate the use of prophylaxis against CMV infection after allogeneic HSCT. Pre-emptive therapy has no effect on the indirect effects of CMV caused by persistent low-level viremia in the body [18]. Letermovir is an important advance, breaking the paradigm of preemptive therapy with the shift to prophylaxis. Letermovir led to lower all-cause mortality in CMV seropositive patients at high risk of reactivation. However, pediatric data is still scarce about this drug [19], and we need to wait for more information regarding its efficacy and safety in children before it can be incorporated into clinical practice.

Conclusion

Patients undergoing haploidentical and MUD transplants are at high risk of CMV reactivation, highlighting the importance of regular surveillance and early pre-emptive therapy in these patients. Patients with underlying inborn errors of immunity are a vulnerable group at high risk of CMV disease. CMV reactivation is associated with an increased risk of GvHD and transplant-related morbidity and mortality. Pre-emptive therapy may not be sufficient to mitigate the indirect effects of CMV reactivation. Hence, prophylaxis may be the way forward to decrease transplant-related mortality and improve survival.