Surveillance over cytomegalovirus (CMV) reactivation following hematopoietic stem cell transplantation: a single-center experience

Abstract

Cytomegalovirus (CMV) infection is widespread and usually asymptomatic in immunocompetent individuals. However, during immune suppression, especially after hematopoietic stem cell transplantation (HSCT), viral reactivation may result in symptomatic infection and end-organ disease. Because there is high incidence of CMV seropositivity in Bulgaria, CMV infection/reactivation is an expected common complication in HSCT recipients which requires frequent monitoring. CMV reactivation monitoring via qRT PCR has a significant role in terms of the moment of initiating pre-emptive therapy in these high-risk patients in order to avoid the development of CMV disease and better clinical management. The aim of this study was to determine and analyze the incidence of CMV reactivation in HSCT recipients in a new transplant center in Bulgaria. The incidence of early CMV reactivation after HSCT in our center is very high (76.7%) most likely related to the high-risk donor/recipient (D/R) profile of our recipients. The mean period from HSCT to the first positive PCR result was quite short, 21.5 days on average. The highest-risk group donor-/recipient+ (D-/R+) experienced the longest and the highest levels of viremia, as well as the highest number of reactivations. The nature of the D/R status most significantly affected the degree of viral load. Our data demonstrate the essential role of monitoring of CMV with reliable and sensitive laboratory methods, such as qRT-PCR to avoid development of CMV disease in high-risk patients.

Keywords:

• HSCT
• CMV
• monitoring
• qRT PCR
• CMV reactivation

Introduction

Cytomegalovirus (CMV) causes one of the most common chronic viral infections worldwide. CMV acquisition in a population is associated with an age-dependent rise in seropositivity, showing correlation with socioeconomic status and differences between countries [1]. CMV-exposed healthy individuals devote a median of 9.1% and 10.2% of their circulating CD4+ and CD8+ memory T cell repertoires, respectively, to this virus [2]. The condition is referred to in the literature with the term ‘memory inflation’ describing the progressive expansion of virus-specific T cells, particularly in the context of chronic latent CMV infection [3]. CMV can take over much of the immune defenses even in subjects who are otherwise healthy; thus, it can be devastating for people with compromised immune systems as hematopoietic stem cell transplantation (HSCT) recipients [4]. High seropositivity in а country implies a considerable risk of CMV reactivation under immunosuppression in HSCT recipients. However, CMV seropositivity is not considered as a definite predictive factor for non-relapse mortality in HSCT patients [4,5]. Many allogeneic and autologous HSCT are performed annually in transplant centers across the globe for the complex treatment of otherwise incurable chemo- and immune-sensitive malignant and non-malignant diseases [6]. Transplantation is preceded by intensive myeloablative chemoradiotherapy, followed by stem cell ‘rescue’. Survival after allogeneic transplantation depends on many factors: the success of engraftment, the graft-versus-host disease severity and the presence of relapse [7]. Various studies show that HSCT recipients with early CMV reactivation have poor survival and higher mortality [5,8]. The risk of late CMV disease after allogeneic HSCT has long been debated. It was estimated as 18%, with a mortality rate of 46% [9]. Clinically significant CMV reactivation, however, is still the main cause of viral complications after allogeneic HSCT [10]. An optimal CMV prevention strategy has not been established yet. Polymerase chain reaction (PCR)–guided preemptive therapy requires a lot of effort for regular PCR monitoring and initiating preemptive therapy if needed. However, this strategy combined with an antiviral prophylaxis regimen could significantly decrease the incidence of late CMV disease and mortality rate [11]. At present, a practical approach is PCR–guided preemptive therapy or risk-based modern prophylaxis with new antivirals [12].

The aim of this study was to determine and analyze the incidence of CMV reactivation in HSCT recipients in a new transplant center in Bulgaria. The following parameters were used as endpoints: Suffered DNA-emia or not; Median time to first reactivation after the HSCT; Number of reactivations for the observed period; Duration of DNA-emia expressed in number of consecutive visits with a positive PCR; and CMV viral load alongside with the predetermined CMV Donor/Recipient status. Data from the study will help to better predict and manage this common problem in HSCT recipients.

Subjects and methods

Ethics statement

The study is a retrospective analysis of laboratory results. The tests performed were part of the standard diagnostic procedures for all HSCT recipients monitored in ‘St. Marina’ University Hospital, Varna. Written informed consent was obtained from patients at each admission. No additional patients’ intervention (laboratory visits or sampling) was performed. Ethical approval for publication was obtained from the Ethics Committee of the Medical University Varna (Protocol Approval Number 140/01.02.2024).

Subjects

We repeatedly examined 86 patients with predetermined CMV serostatus for CMV reactivation. Eight of them underwent autologous HSCT and 78 allogeneic HSCT at ‘St. Marina’ University Hospital, Varna, from November 2017 to April 2023. No exclusion/inclusion criteria were applied for sampling. Following the transplant procedure, we monitored the recipients at regular one-week intervals by quantitative CMV RT-PCR for reactivation. During such episodes, the recipients were tested at least twice a week. For the observed period, we collected 1359 plasma samples or a mean of 15.8 PCR tests per recipient. The blood samples were collected in EDTA vacutainers and the resulting plasma was stored at −20 °C before DNA extraction and PCR analysis.

The demographic and clinical characteristics of the enrolled participants are summarized in Table 1.

Table 1. Characteristics of 86 patients undergoing HSCT.

Total number of patients	86
Gender (Male/Female)	46/40
Median age (range) years	42.9 (5-67)
Age of the recipient (years)	Number
<18	5
19-30	12
31-40	14
41-50	29
51-60	19
>60	7
Underlying disease	Number (%)
Acute Lymphoblastic Leukemia	24 (27.9%)
Hodgkin’s Disease	6 (7%)
Aplastic Anemia	5 (5.8%)
Acute Myeloid Leukemia	40 (46.5%)
Acute Monocyte Leukemia	6 (7%)
Acute Myelomonocyte Leukemia	5 (5.8%)
Stem cell source	Number (%)
Bone marrow (BM)	0
Cord blood (CB)	0
Peripheral blood stem cell (PBSC)	86 (100%)
Cell dose	No CD34 cell/kg (range)
PBSC	3.73x10^6 (0.6-11.60)
Donor/Recipient (D/R) CMV status	Number (%)
D+/R+	67 (77.9%)
D−/R+	15 (17.4%)
D+/R−	3 (3.5%)
D−/R−	1 (1.2%)

Enzyme-linked immunosorbent assay (ELISA)

Commercial indirect ELISA kits for measuring anti CMV IgG (Euroimmun, Germany: REF: EI 2570-9601 G) were used, according to the standard instructions of the manufacturer.

Polymerase chain reaction (PCR)

DNA was extracted from 150 µL plasma using SaMag Viral Nucleic Acid Extraction kit (Sacace Biotechnologies S.r.l., Italy REF: SM003). All PCR reactions were performed with Taq-man Quantitative RT PCR for CMV-DNA detection – CMV Real-TM Quant (Sacace Biotechnologies S.r.l., Italy REF: V7-100/2FRT) in the presence of Internal Control to identify possible inhibition of the reaction and endogenous internal control (β-globin gene) to evaluate the adequacy of the material and its storage. The amplification was performed with a PCR instrument Sa-cycler (Sacace Biotechnologies S.r.l., Italy) in a final volume of 25 µL reactions.

Data analysis

The results obtained were processed with non-parametric tests because of the abnormal distribution of data (almost all groups were right-skewed): Mann–Whitney test when comparing two groups and Kruskal Wallis test when comparing three or more groups. The statistical program used was the R project for statistical computing (version 4.0.4/2021-02-15).

Results

The serological CMV status of 86 donors and recipients was specified via ELISA one month prior to HSCT. The recipients were predominantly anti-CMV IgG positive, 82/86 (95.3%).

We found a total of 494/1359 (36.4%) of the collected plasma samples to be CMV positive in qRT PCR. The viral load ranged from 20 copies/mL to 88 000 copies/mL, with an average viral load of ∼3 000 copies/mL.

The majority of the recipients (66/86, 76.7%) experienced at least one DNA-emia during the observed period (Table 2). The period from HSCT to the first laboratory detectable reactivation had a mean of 21.5 days, with a range of 2 to 80 days. The most frequently repeated period (mode) for the majority of the recipients was 19 days.

Table 2. Distribution of the analyzed parameters according to the predefined D/R status.

N (%)	D/R status	Median period to first reactivation after the HSCT (days)	Number of reactivations for the observed period (days means)	Duration of DNA-emia/ number of consecutive visits with a positive PCR (days means)	CMV viral load (c/mL mean)
67 (77.9%)	+/+	22.7	2.2	4.7	2 300
15 (17.4%)	−/+	18	2.9	5.2	6 000
3 (3.5%)	+/−	16.5	2.5	4.5	400
1 (1.2%)	−/−	0	0	0	0

There was no statistically significant difference between the period to first reactivation in D+/R + and D-/R + groups (means 22.7 and 18 or medians 19 and 15.5, respectively). The p-value of Mann Whitney U non-parametric test was 0.2.

We found no statistically significant difference between the number of reactivations in D+/R + and D-/R + groups (means 2.2 and 2.9 or medians 2 and 2.5, respectively). The p-value of the Mann–Whitney U non-parametric test was 0.23. Nevertheless, the highest-risk group D-/R + experienced the highest number of reactivations for the observed period, compared to the other groups (Table 2).

There was no statistically significant difference between the durations of the DNA-mia in D+/R + and D-/R+ (means 4.7 and 5.2 or medians 4 and 5, respectively). The p-value of Mann–Whitney U non-parametric test was 0.17.

However, there was significant difference when comparing the viral loads. The Kruskal–Wallis Test gave a p value of 0.004 when comparing the three groups: D+/R+, D-/R+, D+/R-. The medians were as follows: 482 (D+/R+), 234 (D+/R-) and 623 (D-/R+) and the means were 2296, 5986 and 471 c/mL. When only D+/R + and D-/R + were considered, the Mann–Whitney test calculated a p-value of 0.01.

Discussion

It is generally recognized that Cytomegalovirus disease is an important cause of morbidity and mortality, especially in patients after HSCT. A low incidence of CMV DNA-emia in HSCT recipients in a transplant center could be due to the lower risk profile of the recipients, but it also attests to the efficacy of CMV surveillance. After accumulating several years of experience with HSCTs in our center, here we summarize and analyze the data of the most frequently monitored parameters of CMV reactivation.

Recipients included in this study were predominantly anti CMV IgG positive (95.3%), as expected for the Bulgarian population. The overall seroprevalence in Northeastern Bulgaria is quite high and has been found to be 78.4% in a recent large-scale study [13].

According to the literature, about 30% of patients after allogeneic HSCT reactivate latent CMV infection, usually in the post-transplant period [14]. In our cohort, the proportion was much higher − 76.7% of the recipients experienced at least one viremia. This could be related to a different risk profile of our recipients. It is widely accepted that the most significant factor for the success of HSCT (regarding the possible viral reactivation) is the serologic status of the donor and recipient [15]. The D-/R- transplantation is with the lowest risk while D-/R + is accepted as the most uncertain [16–18]. Previous studies showed the CMV-specific immune reconstitution, CMV reactivations, late CMV recurrence and development of CMV disease to be more frequent after D-/R + transplantation [15,19–22].

The issue about the tool to monitor CMV reactivation in HSCT recipients remains questionable. The pp65 antigenemia test has been accepted for a long time as a reliable assay for CMV infection. A positive result is considered an indication for the initiation of pre-emptive therapy in such clinical cases. However, this test has some disadvantages because of the need of immediate sample processing, the time- and labor-consuming procedure, the inability to be automated and the lack of objectivity in the quantification. A major drawback is its inapplicability during episodes of severe neutropenia [23–26].

Detection of viral load in plasma by qRT PCR is considered to have many advantages because it allows early detection of CMV reactivation, monitoring of clinical response to antiviral therapy, risk prediction, emergence of resistant strains and eventual development of CMV disease [27,28]. qRT-PCR can be used reliably to determine when to start and stop pre-emptive therapy. In a randomized clinical trial [29], a comparison was made between DNA cut-off of 10 000 c/mL in whole blood and the first positive pp65 antigenemia test to determine the initiation of preventive therapy for CMV infection in adult patients after HSCT. Its results showed that the use of such a cut-off significantly reduces the number of patients requiring antiviral treatment, thus saving unnecessary medication and avoiding the possible development of resistant strains [29,30]. However, there is no a widely-accepted viremia level threshold for starting anti CMV preemptive therapy – several thresholds (from 1 000 to 10 000 c/mL of whole blood) were used [25,31,32]. To date, no DNA level cutoffs for plasma specimens have been clinically validated.

Different patients probably have a different DNA-emia level threshold based on their different risk profile. On the basis of the experience we have accumulated in our center, we believe that a threshold of 2 000 c/mL is the right time to start preemptive therapy and found it to be clinically safe.

When considering the D/R status, we observed that most often, the longest duration and the highest viral load were detected in the group D−/R+, as expected by literature data. A study shows the survival and non-relapse mortality is worst for CMV (D−/R+), followed by CMV (D+/R+) [33]. Unfortunately, the proportion of these two riskiest combinations was greatest (95.3%) in our recipients (Table 2). This obliged us to be strict despite the difficulties in carrying out many clinical and laboratory visits. We monitored the recipients for CMV reactivation weekly, and in case such reactivation was established, twice each week. The highest risk D-/R + profile is likely related to the nature of the CMV latency sites. A recent study confirms that CMV virions enter CD34+ cells by macropinocytosis but fail to fully uncoat or disassemble their tegument layers, leading to the establishment of latency [34]. In the case of HSCT with the riskiest D-/R + profile, naïve stem cells are infused into the recipient’s body, which are about to be subjected to primary infection in the background of severe immunosuppression.

Generally, we found a fairly high proportion of samples positive in PCR − 36.4% of the total number of samples. The viral load ranged from 20 c/mL to 88 000 c/mL, with an average of 3000 c/mL.

The mean time to the onset of the first positive PCR result after HSCT was 21.5 days, with a range of 2 to 80 days. The most frequently repeated period to first reactivation for the majority of the recipients was 19 days. There was no statistically significant difference between the period to first reactivation in D+/R + and D-/R + groups. The reactivation occurred within quite a short period as compared to a previous study which reported 47 days [35]. In that study, the authors found 94% of HSCT recipients to develop CMV viremia in the first 3 months after transplantation, which is a higher reactivation rate compared to our results of 66/86 (76.7%).

We found no statistically significant difference between the number of reactivations and the durations of the DNA-mia in D+/R + and D-/R + groups. However, there was an important difference when comparing the viral loads. The nature of the D/R status most significantly affected the degree of viral load.

Conclusions

The incidence of early CMV reactivation after HSCT in our center was very high (76.7%), most likely related to the high-risk D/R profile of our patients. The mean period from HSCT to the first positive PCR result was quite short, 21.5 days on average, compared to other studies. The highest-risk group D-/R + underwent the highest number of reactivations for the observed period. They experienced the highest levels of viremia, as well. We conclude that the nature of the D/R status most significantly affects the degree of viral load. Our data demonstrate the essential role of monitoring of CMV with reliable and sensitive laboratory methods, such as qRT PCR in terms of initiating pre-emptive therapy in these high-risk patients in order to avoid the development of CMV disease.

Authors’ contributions

ZS, conceptualization, study design, execution, analysis, data interpretation, writing – original draft preparation; ZS, TK and TT, experimental analysis; TT, statistical analyses; KK, data entry and processing, computations; TK, study design, writing – critical revision for important content and approval of the final version. All authors have accepted responsibility for the entire content of this manuscript and approved its submission.

Disclosure statement

The authors declare absence of conflict of interests.

Data availability statement

The anonymized data that support the findings of this study are available upon reasonable request from the corresponding author [ZS]. The data are not publicly available due to restrictions, e.g. their containing information that could compromise the privacy of research participants.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.