Concepedia

Abstract

Human herpesvirus 6 (HHV-6) is a ubiquitous herpesvirus that infects most children during early years of life, and, like most other herpesviruses, establishes latency after primary infection (De Bolle et al, 2005; Zerr, 2006). In the immunocompromised host, the virus can reactivate and cause disease, such as encephalitis (Zerr, 2006; Ljungman et al, 2008). HHV-6 can integrate its genome in a persistent latent state into the host’s chromosomes (Ward et al, 2006). Due to the high prevalence of infection, the persistence of the virus and potential chromosomal integration (CI), there are considerable diagnostic problems for HHV-6 disease, which, in turn, influences the decision to institute and to stop antiviral therapy. For the assessment of HHV-6 DNA concentration, an internal controlled HHV-6 real-time polymerase chain reaction (PCR) using the Quantitect probe PCR kit (Qiagen, Hilden, Germany) and the ABI7900 sequence detection system (Applied Biosystems, Darmstadt, Germany) was performed (Preiser et al, 2003). As internal control, a defined number of cell culture supernatant-derived murine CMV (mCMV; strain Smith; ATCC VR-1399) virions were added (Preiser et al, 2003). Quantification was achieved through an external standard curve. For the analysis of hair follicles, DNA was extracted from hair follicles lysed in NaOH 2% for 10 min at 70°C. The human genomic sequence was quantified using the GAPDH gene. A 17-year-old patient was diagnosed with acute lymphoblastic leukaemia (ALL) in May 2006. The central nervous system (CNS) was not affected [acellular cerebrospinal fluid (CSF), normal magnetic resonance imaging (MRI)]. The patient was treated according to the trial ALL-BFM 2000; induction therapy consisted of prednisone, vincristine, daunorubicin and asparaginase and intrathecal methotrexate. On day 20, the patient presented with depressed consciousness, disorientation and focal seizures. MRI revealed bilateral subcortical white matter oedema with cortical involvement in the occipital lobes (Fig 1). Haemorrhage and thrombosis were excluded. Investigation of the CSF revealed normal protein level and a slight pleocytosis of 6 cells/μl, which increased to 96 cells/μl in the control performed several days later. HHV-6 DNA was detected in CSF (21 000 and 1 000 000 copies/ml, respectively), whereas all other PCRs and cultures remained negative. HHV-6 DNA was also detected in plasma (20 000 copies/ml), and positivity of anti-HHV-6 IgG in the absence of IgM excluded primary infection. Electroencephalography, which was normal at the time of diagnosis of ALL, revealed diffuse abnormalities. The clinical condition rapidly improved under antiviral treatment with ganciclovir and anticonvulsive therapy. Magnetic resonance imaging in a 17-year-old boy undergoing therapy for acute lymphoblastic leukaemia and suffering from encephalitis due to human herpesvirus 6. FLAIR axial section shows bilateral subcortical white matter oedema with cortical involvement in the occipital lobes (top). Correlated T2 axial image (bottom). Over the next months, HHV-6 DNA levels in CSF and plasma were monitored to guide antiviral therapy (Fig 2). As no further CNS symptoms were noted, antiviral therapy was stopped during maintenance treatment for ALL. Top: Course of human herpesvirus 6 (HHV-6) DNA levels in cerebrospinal fluid (CSF) (left Y-axis), CSF leucocyte count (right Y-axis) and antiviral therapy. HHV-6 DNA in CSF; leucocytes in CSF. Bottom: Course of HHV-6 DNA levels in plasma (left Y-axis), peripheral leucocyte count (right Y-axis) and antiviral therapy. HHV-6 DNA in plasma; leucocytes in peripheral blood * after development of neurological symptoms. In order to assess a possible HHV-6 CI, HHV-6 DNA concentration was analysed in the patient’s blood, serum and hair follicles. In addition to a significantly higher viral load in the patient’s whole blood compared to plasma (5 622 000 vs. 3900 copies/ml), HHV-6 DNA was found in the hair follicles (10 300 copies/five hair follicles), indicating HHV-6 CI (Leong et al, 2007). To exclude that detection of HHV-6 DNA in the CSF may be a common phenomenon in immunocompromised children with acute leukaemia, we tested a total of 80 CSF samples from 75 patients without neurological symptoms. All patients [45 boys, 30 girls; median age (range), 7·7 (0·4–22) years] had acute leukaemia (ALL = 67, acute myeloid leukaemia = 8). CSF samples were obtained before patients received intrathecal prophylaxis for CNS leukaemia as part of their chemotherapy. Anti-HHV-6 IgG titres were seen in 49/55 (89%) patients tested for HHV-6 at the time of diagnosis of acute leukaemia. However, HHV-6 DNA was not detected in any of the CSF samples analysed. Whereas HHV-6 encephalitis is not uncommon in recipients of haematopoietic stem cell transplants (HSCT), this disease is rarely seen in immunocompromised patients not undergoing HSCT (De Bolle et al, 2005; Zerr, 2006). In most of these patients, the detection of HHV-6 DNA in the CSF led to the diagnosis of HHV-6 encephalitis. Whereas CSF findings were reported to be normal in up to half of the patients (Zerr, 2006; Ljungman et al, 2008), our patient had increasing pleocytosis. Furthermore, not all patients exhibit abnormalities upon MRI, but, as seen here, most patients present with diffuse abnormal electroencephalograms (Zerr, 2006; Ljungman et al, 2008). When other infectious agents and underlying causes are excluded, antiviral treatment with ganciclovir or foscarnet should be started (Ljungman et al, 2008). HHV-6 differs from most other human herpesviruses because of the ability of their genomes to integrate in a persistent latent state into the host’s chromosomes. It therefore can be transmitted from parent to child in the germ line (Ward et al, 2005; Leong et al, 2007). CI of HHV-6 DNA was also demonstrated in our patient with characteristic DNA concentrations in full blood, serum and hair follicles (Ward et al, 2006). The estimated frequency of HHV-6 CI is below 2% (Leong et al, 2007). Given that, every leucocyte and non-haematopoietic cell contains viral sequences in individuals with HHV-6 CI, persistently high levels of HHV-6 DNA in blood and CSF are characteristically found, which might cause diagnostic problems (Ward et al, 2006). Whereas CSF HHV-6 DNA levels due to primary infection differ in immunocompetent patients from those due to CI, the situation in immunocompromised patients with HHV-6 reactivation is not clear (Ward et al, 2007). None of the 80 samples from 75 asymptomatic patients with acute leukaemia tested contained HHV-6 DNA, which is in contrast to a report on transplant recipients, where HHV-6 DNA was detected in the CSF in up to 25% of symptomatic haematopoietic stem cell recipients (Wang et al, 1999). Unfortunately, this study did not analyse HHV-6 CI. Although it is recommended to exclude HHV-6 CI for the diagnosis of HHV-6 encephalitis (Ljungman et al, 2008), we strongly believe that the virus was the causative agent for the disease in our patient, because extensive testing did not reveal any other underlying condition for the clinical symptoms. In addition, whereas it is well known that asymptomatic elevation of HHV-6 DNA load correlates to leucocyte engraftment in transplant patients whose donor has HHV-6 CI, (Ljungman et al, 2008), we did not observe a striking correlation of HHV-6 DNA concentration and the leucocyte count in blood and CSF respectively (Fig 2). Whereas most experts agree to institute antiviral therapy in a symptomatic patient with presumed HHV-6 encephalitis, monitoring of HHV-6 DNA levels was misleading in our patient and led to prolonged antiviral therapy. This pitfall was also recently reported in a patient with aplastic anaemia (Hubacek et al, 2007). Therefore, HHV-6 CI has to be excluded before long-term administration of potentially toxic antiviral drugs. We conclude from our analysis that (i) HHV-6 encephalitis has to be considered in symptomatic patients undergoing treatment for leukaemia, even when HHV-6 CI has been shown (ii) HHV-6 DNA is usually not present in CSF of immunocompromised children with leukaemia (iii) HHV-6 CI has to be excluded before long-term administration of potentially toxic antiviral drugs. None of the authors has to disclose a conflict of interest.

References

YearCitations

Page 1