Roseoloviruses in transplant recipients: clinical consequences and prospects for treatment and prevention trials

Abstract

Roseoloviruses frequently reactivate in transplant recipients. We review the impact of Roseoloviruses in transplant recipients and highlight research priorities. Human herpesvirus 6A (HHV-6A) and HHV-6B were recently classified as distinct species with important differences. Both viruses can result in inherited chromosomally integrated HHV-6, which may cause complications after transplant. HHV-6B is the primary species associated with disease and appears to have pleiotropic effects on the central nervous system. Small preemptive and prophylactic studies have not shown a statistically significant impact on HHV-6 disease. Although Roseoloviruses are associated with diverse complications in transplant patients, studies providing strong evidence for a causal role are lacking. Trials focusing on prevention and treatment will be important to inform the significance of Roseolovirus reactivation.

Introduction

The Roseolovirus genus of the Betaherpesvirinae consists of three distinct but closely related species, human herpesvirus 6A (HHV-6A), HHV-6B, and HHV-7. Roseoloviruses cause ubiquitous human infection and have important differences in epidemiology and pathogenesis. They establish latency after primary infection, and viral reactivation is the usual cause of disease in immunocompromised patients, especially after allogeneic hematopoietic cell transplantation (HCT) and solid organ transplantation (SOT). This review highlights recent findings about the impact of Roseoloviruses in transplant recipients (Table 1), as well as research priorities to advance understanding of the clinical significance and management of these pathogens.

Table 1.

Clinical characteristics of Roseoloviruses in immunocompromised transplantation recipients

	HHV-6A	HHV-6B	HHV-7
Incidence of  reactivation	Infrequent (0–3%), consider ciHHV-6	Frequent (30–50%), more common after cord blood HCT	Frequent (40–60%)
Disease  associations	Rare case reports, consider ciHHV-6	More frequent after HCT than SOT CNS dysfunction (especially encephalitis, where evidence supports causal association) Fever and rash Myelosuppression Acute GVHD Allograft dysfunction, rejection CMV reactivation Pneumonitis Hepatitis All-cause mortality	Rare case reports of disease but not well studied
Treatment and  prevention	Not well studied	Antiviral treatment recommended for encephalitis (expert opinion) Prevention not shown to be beneficial in small studies	Not well studied
Research  priorities	Significance of ciHHV-6	Optimal treatment for encephalitis Diagnosis of end-organ disease Significance of ciHHV-6 Clinical trials of treatment and prevention strategies	Large, prospective study of HHV-7 epidemiology, disease associations, and relationship to HHV-6 and CMV

HHV, human herpesvirus; ciHHV-6, inherited chromosomally integrated HHV-6; HCT, hematopoietic cell transplantation; SOT, solid organ transplantation; CNS, central nervous system; GVHD, graft-versus-host disease; CMV, cytomegalovirus.

Inherited chromosomally integrated human herpesvirus 6A and B

An important consideration unique to HHV-6 is its ability to integrate into human chromosomal telomeres. When this occurs in germ-line cells, vertical transmission of chromosomally integrated HHV-6 (ciHHV-6) results in offspring with at least 1 copy of HHV-6 DNA in every nucleated cell of their body [1], independent of viral reactivation. Population-based studies have estimated this to occur in about 1% of people [2^••], and inherited ciHHV-6 has been described in both HCT and SOT recipients [3]. Although characteristics of routine HHV-6 testing can indicate inherited ciHHV-6, a lack of appreciation for this condition has caused considerable confusion and misreporting in the literature [4].

Whether patients with inherited ciHHV-6 can develop HHV-6 reactivation with associated pathogenicity, or other indirect complications, is a topic of considerable controversy. However, an accumulating body of in vitro and in vivo evidence suggests that inherited ciHHV-6 can be the source of pathogenic viral reactivation or other complications [5–7]. A recent report detailed in vivo molecular and virological evidence of functional HHV-6A reactivation from inherited ciHHV-6A in an immunocompromised boy with X-linked severe combined immunodeficiency who underwent HCT [8^••]. A study in liver transplantation recipients found that those with pre-transplant inherited ciHHV-6 had higher rates of bacterial infection [9]. However, a review of 21 reported cases of patients or recipients of donor cells with inherited ciHHV-6A and B found no clinical disease associations [3]. Novel approaches facilitating fast, cost-efficient, and clinically accessible testing for inherited ciHHV-6 [10^•] will be important to broaden our understanding of this condition and whether routine pre-transplant screening is warranted. Inherited ciHHV-6 is reviewed in detail in this section by Kaufer et al.

Human herpesvirus 6A

The epidemiology of HHV-6A is not well described and appears to be geographically distinct from HHV-6B. Primary infection occurs later in life, usually without clinical symptoms, in the USA, Europe, and Japan [11], but it appears to be more common during infancy in Sub-Saharan Africa [12]. HHV-6A is found in about a third of individuals with inherited ciHHV-6 [2^••].

No disease has been causally linked with HHV-6A, which is infrequently described in immunocompromised patients and only accounts for 0–3% of reactivation after transplantation [13^••,14^•]. There are a few case reports of HHV-6A encephalitis after HCT [15–17]. However, with a contemporary understanding of HHV-6, these cases were most likely due to unrecognized inherited ciHHV-6, and whether HHV-6A was contributory to the patients’ illness is hard to know with the information provided. Ultimately, HHV-6A detection in the setting of transplantation is suggestive of inherited ciHHV-6. Due to the low incidence of HHV-6A reactivation in HCT and SOT patients, it remains unclear if it shares an association with the diversity of complications seen in the setting of HHV-6B reactivation.

Human herpesvirus 6B

HHV-6B infects most children within the first 3 years of life and is found in approximately 2/3 of patients with inherited ciHHV-6 [2^••]. HHV-6B accounts for the majority of HHV-6 reactivation in transplant recipients and can be detected in plasma and/or serum samples from 30 to 50% of patients [13^••,14^•,18^•,19^••,20^•]. HHV-6B reactivation typically occurs during the first 2–4 weeks after transplantation and has been associated with a variety of complications (Table 1), which occur with greater frequency after HCT than SOT.

HHV-6B as a cause of encephalitis was first noted after HCT in 1994 [21]. A large body of literature has since convincingly described a distinct syndrome of limbic encephalitis after transplant as described in Table 2. These patients have detectable HHV-6B DNA in cerebrospinal fluid (CSF) [19^••,20^•,22–24], and studies have demonstrated viral protein expression in the mesial temporal lobes of affected patients [25,26]. HHV-6B-associated limbic encephalitis has significant morbidity and mortality despite antiviral treatment [23,24,27,28]. This syndrome occurs in approximately 1% of all HCT patients and as many as 10% of cord blood HCT recipients, who have an increased incidence of HHV-6 reactivation and disease [20^•,23]. A recent meta-analysis of 19 papers found that HHV-6B encephalitis was significantly higher in CBT recipients (8.3%) than other allogeneic HCT recipients (0.5%) [19^••]. In a prospective cohort of CBT recipients transplanted without antithymocyte globulin, reactivation was documented in 94% of patients [29]. The authors reported a low level (1.6%) of encephalitis and no association with clinical outcomes, but almost 1/3 of patients received preemptive antiviral therapy. This syndrome has also been described after SOT but is much less common [14^•,30].

Table 2.

Clinical features of HHV-6B encephalitis

	Typical findings
Incidence	~1% in all allogeneic HCT recipients, rare case reports in SOT recipients ~8% in cord blood HCT recipients
Disease onset	~2–6 weeks after HCT
Symptoms/signs	Confusion, encephalopathy, anterograde amnesia, SIADH, seizures, insomnia
Brain MRIa	Circumscribed, non-enhancing, hyperintense lesions in the medial temporal lobes (especially the hippocampus and amygdala)
Cerebrospinal fluid	HHV-6B DNA, +/− mild protein elevation, +/− mild lymphocytic pleocytosis
Diagnosis	HHV-6B DNA detection in cerebrospinal fluid, consistent clinical findings, exclusion of inherited ciHHV-6 and other causes
Prognosis	Memory deficits and neuropsychological sequelae in up to 50%
	Death due to progressive encephalitis in up to 25% of all HCT recipients and 50% of cord blood HCT recipients

HHV-6B, human herpesvirus 6B; HCT, hematopoietic cell transplantation; SOT, solid organ transplantation; SIADH, syndrome of inappropriate antidiuretic hormone secretion; MRI, magnetic resonance imaging.

^aFeatures of T2, fluid attenuation inversion recovery (FLAIR) and diffusion weighted-imaging (DWI) sequences.

HHV-6B may have protean effects on the central nervous system (CNS) beyond encephalitis. A large prospective study of 315 allogeneic HCT recipients that systematically assessed for CNS dysfunction found an independent and temporal association of HHV-6B reactivation with delirium and neurocognitive decline after HCT [31]. Two other studies show that HHV-6B DNA detection in CSF in the absence of other explanatory etiologies may be associated with a spectrum of disease, ranging from headache alone to fulminant encephalitis [32,33^•]. Our understanding of the significance of HHV-6B DNA detection in the CSF continues to evolve but is limited by the lack of routine CSF sampling in asymptomatic controls. Although previous reports suggest that HHV-6B detection in the CSF in the absence of CNS symptoms is very unusual (<1%) [24,34], a recent study found that this may occur with greater frequency [33^•].

HHV-6B reactivation has been associated with many other conditions in transplant patients (Table 1), although its role is less clearly defined. There is an association of HHV-6B reactivation with myelosuppression and delayed engraftment, particularly involving platelets [35]. A recent large prospective trial in 235 consecutive allogeneic HCT recipients showed delayed platelet engraftment among the 48% of patients who developed HHV-6B viremia within 100 days [18^•]. This study also showed an association between HHV-6B reactivation and acute graft-versus-host disease (aGVHD) in univariate and multivariable models, a finding corroborated in more detail in another large prospective trial [13^••] as well as other recent studies [23,36]. Interestingly, a retrospective study found that either initial HHV-6 reactivation or aGVHD predisposed patients to subsequent development of the other [37^•]. They also show that HHV-6B reactivation was associated with rash that imitated skin aGVHD in 17 patients; histopathology revealed a lymphoid infiltrate but no evidence of aGVHD. These patients were treated with steroids for presumptive aGVHD, which may have been unnecessary. Given that fever and rash are the most frequent manifestation of HHV-6B reactivation after HCT [38], study of the causal pathways connecting these manifestations with HHV-6 and aGHVD is warranted.

Increased risk for cytomegalovirus (CMV) reactivation and disease has been variably demonstrated in HCT [39] and SOT [40,41] recipients. In a large prospective study of 315 allogeneic HCT patients, HHV-6 reactivation was independently and quantitatively associated with increased risk of subsequent CMV reactivation [13^••]. A small study of 9 CMV seropositive and negative patients undergoing kidney transplantation showed that both groups had increased risk for HHV-6 viremia when the donor was CMV seropositive [42]. Additionally, HHV-6 viremia was a sign of impending primary CMV infection in CMV sero-discordant transplantations. A recent study of 41 liver transplant patients found that recipients of liver allografts with detectable HHV-6 DNA in pre-transplantation samples had increased risk for post-transplantation CMV disease and allograft loss [43^•].

Allograft dysfunction and rejection is gaining more traction as a serious complication of HHV-6B reactivation [44] or primary infection [45], particularly in liver transplantation. Retrospective testing for HHV-6 DNA in 26 liver biopsies was positive in 39% of patients with graft hepatitis of unclear etiology, and confluent periportal necrosis was associated with high viral loads in 4 patients who responded to antiviral therapy [46]. After lung transplantation, HHV-6 testing of transbronchial biopsies found a significant association between HHV-6 detection and interstitial pneumonia, although there was no association with graft rejection [47].

Cases of what appear to be HHV-6B directly affecting organs other than the brain have been reported in the HCT population as well. A case of HHV-6B-associated hepatitis in an allogeneic HCT recipient established the link between HHV-6B and clinical hepatitis. This report demonstrated temporally associated HHV-6B reactivation in blood and liver samples, along with immunohistochemical evidence of active viral replication in the liver [48^•]. A number of studies have attempted to address a potential association between HHV-6 and pneumonitis. Findings have been conflicting and the issue remains unresolved. While additional studies are clearly needed, HHV-6 may be an underappreciated cause of hepatic and pulmonary disease in HCT and SOT patients.

The importance of attaining a better understanding of the significance of HHV-6 detection in end organ diseases is underscored by an association between HHV-6 reactivation and increased mortality after HCT [13^••,18^•,49]. However, whether HHV-6 directly contributes to this and other clinical outcomes or is simply an indirect marker of the severity of patient illness and immunologic deficiency remains unclear.

Human herpesvirus 7

Infection with HHV-7 typically occurs in childhood, and most adults are seropositive. Primary infection rarely comes to clinical attention and is less well understood than with HHV-6B. HHV-7 DNA detection in blood samples after transplantation is primarily due to reactivation and is found in approximately 40–60% of patients [50–52]. This typically occurs within the first 2–4 weeks after transplant. In contrast to HHV-6B, there is no obvious peak time of reactivation, and the vast majority of reactivation events are transient, low-level, and not associated with any clinical manifestations. However, HHV-7 is not as well studied in immunocompromised patients, and most reports are small, often with conflicting results.

HHV-7 reactivation has been associated with a variety of diseases in a few case reports. These include CNS disease [53], myelosuppression [54], aGVHD [55], and allograft dysfunction and rejection [56]. Data from a number of studies suggest a potential indirect effect of HHV-7 to increase rates of CMV reactivation and disease in HCT [57] and SOT [58,59] recipients, but this is not a consistent finding [52]. Future studies of HHV-6B should incorporate testing for HHV-7 to advance our understanding of its possible significance in transplantation.

Diagnostic considerations

Studies have used a variety of non-standardized techniques to detect Roseoloviruses in a wide range of specimens, making direct comparisons problematic. This, along with the challenge implicit in attributing disease to commensal viruses, has led to many questions about the actual role Roseoloviruses play in many associated clinical conditions following transplantation. The historical lack of a clear distinction between HHV-6A and HHV-6B in the literature further limits a comprehensive understanding of their epidemiologic differences and etiologic associations. Notably, HHV-6A and HHV-6B were only recently classified as distinct species in 2012 [60], so previous reports lacked a scientific imperative to distinguish between them. Differentiating HHV-6A from HHV-6B has also been hampered by the lack of specificity of serologic assays [11]; distinguishing between the two viruses is best accomplished at this time with polymerase chain reaction assays and sequencing [61^•]. Unfortunately, these tests continue to be expensive and are not widely available. Diagnosis of Roseolovirus infections is reviewed in this section by Hill et al.

Treatment

Early diagnosis, treatment, and prevention are important concepts when considering Betaherpesviruses in immunocompromised transplant patients. Although some recommendations exist for diagnosis and management of specific HHV-6-associated diseases [62,63], there are no widely adopted standard practice guidelines. Furthermore, there are no FDA approved medications for treating Roseoloviruses, and optimal treatment approaches (i.e. single or combination antivirals, dose, duration) have not been rigorously studied. Although a detailed review of treatment is beyond the scope of this review, studies attempting to screen for and prevent HHV-6-associated diseases merit brief discussion.

The use of prophylactic ganciclovir in HCT and SOT patients can reduce HHV-6B reactivation [40,64] and may reduce associated morbidity in high-risk patients [65,66], but large scale adoption is limited by ganciclovir-induced myelosuppression. The use of low dose foscarnet for 10 days post-engraftment to prevent HHV-6B reactivation was recently explored in a retrospective cohort study of 118 allogeneic HCT recipients (unrelated or cord blood donors) [67^••]. High-level HHV-6B reactivation (>10,000 copies/ml) occurred in 19.4% of patients receiving foscarnet versus 33.8% of those not, and HHV-6B encephalitis occurred in 4.5% of patients receiving foscarnet versus 9.9% of those not; neither finding was statistically significant. Two small prospective studies of preemptive foscarnet in HCT recipients [68,69] did not significantly reduce the complications of HHV-6B reactivation and highlight the challenge of initiating preemptive therapy after a viral threshold has been reached but prior to onset of HHV-6-associated disease. The development of HHV-6 specific cytotoxic T cells for adoptive immunotherapy has been encouraging [70^•] but is in early stages of development. These topics are reviewed in detail by Prichard et al. and Becerra et al. in this section.

Research priorities

This review highlights the recent findings and continued deficiencies in our clinical understanding of the significance of Roseoloviruses in transplant recipients. To advance our ability to quickly and accurately assess patients who will be or are adversely affected by these pathogens, future studies should focus on strategies to diagnose HHV-6B-associated end organ diseases, determine an optimal treatment approach for HHV-6B encephalitis, identify possible HHV-6A or B disease in the setting of inherited ciHHV-6, and determine whether prevention of HHV-6B reactivation improves outcomes (Table 1). Testing should distinguish between species in all cases. Clinical trials of new antivirals (i.e. brincidofovir, previously CMX001) [71] and immunotherapeutic treatments with less toxicity will be imperative to establish support for a causal role of Roseolovirus pathogenicity in immunocompromised patients and to potentially improve outcomes. The clinical understanding of HHV-7 is much less developed, and additional large, well-designed prospective studies are warranted. Ultimately, a more complete understanding of the clinical impact of Roseoloviruses requires large, multicenter prospective studies using well-defined disease criteria and standardized viral assays capable of discriminating between species.

Conclusion

HHV-6B reactivation from latency is a frequent occurrence after transplantation, and a preponderance of evidence suggests a causal association between HHV-6B and encephalitis. Although a growing body of literature supports the involvement of Roseoloviruses, especially HHV-6B, in additional diseases, potential pathways of pathogenesis are not well defined. Thus, the precise role that Roseoloviruses play in causing these clinical outcomes in immunocompromised transplant recipients remains unclear. The widespread prevalence of Roseoloviruses in diverse human cell types complicates an understanding of its role in this myriad of associated diseases, and there is significant controversy regarding whether viral infection and replication cause, or merely correlate with, associated pathologies. Complex interactions between these viruses and the host immune system further confound attempts to implicate them in pathologic processes. Further study is needed to establish causality and explore prevention and treatment options.