Treatment with Brivudine in Immunocompromised Pediatric Patients with Herpes Zoster

Hematopoietic stem cell transplantation (HSCT) is an important curative treatment option for hematological malignancies, solid tumors, and genetic diseases. However, the success of HSCT is still compromised by complications. Herpes zoster (HZ) is a common and serious complication after HSCT. Especially, latent varicella-zoster virus (VZV) can reactivate and results in HZ because of the immunosuppressive conditions. Berman et al. [1] showed that among 90 VZV seropositive children and adolescents, who underwent allogeneic HSCT, the incidence of infection was 54% by 12 months and 69% by 60 months after HSCT.

During primary infection, VZV manifests as varicella (chickenpox). Subsequently, the virus enters a state of latency and persists in the dorsal root, cranial, and enteric ganglia [2]. Endogenous reactivation of VZV leads to HZ infection, a painful skin rash with a dermatomal distribution [3]. The mechanism controlling the latency of VZV is not clearly understood. However, it is known that the reactivation is accompanied by a weakened VZV-specific T-cell response [4]. The natural decline in T-cell function with age (immunosenescence) is the most important factor influencing VZV reactivation in otherwise healthy individuals. It has already been demonstrated that a severely weakened immune system caused by disease-induced immunosuppression or by immunosuppressive therapy predisposes to a higher rate of VZV reactivation and a severe course of the disease [5]. Specifically, regarding children, HZ develops frequently in patients with weakened immune systems. A Danish study provides further evidence that nearly one-third of VZV seropositive children experienced HZ infection during primary chemotherapy. The risk of getting sick was related to the intensity of the treatment [6].

Currently, the drug of choice for antiviral treatment of VZV infections in children is acyclovir. Many studies showed that therapy with intravenous acyclovir in patients with normal renal functions (500 mg/m² every 8 h) reduces the duration of viral replication and stops disease progression [7‒9]. However, the terminal elimination half-life of approximately 3 h at intravenous application necessitates high doses, frequent administration, and hospitalization [10]. Moreover, acyclovir is poorly absorbed after oral administration with a bioavailability of only 20% [11]. Therefore, in children, an orally given drug would be preferred to provide outpatient treatment.

It was previously demonstrated that in cell culture studies brivudine proved to be 200–1,000 times more effective in inhibiting VZV replication than acyclovir or penciclovir [12]. A double-blind, randomized study comparing oral brivudine and oral acyclovir showed similar results. Both drugs were administered for 7 days. Brivudine presented a similar safety profile and a significant improvement in efficacy in immunocompetent patients with HZ [13].

In this retrospective study, we aimed to investigate the clinical outcome of VZV infections in children with hematological malignancies, solid tumors, and genetic diseases who were treated with brivudine. Pediatric patients would clearly benefit from an oral antiviral drug with high potency that could provide outpatient treatment.

In this retrospective study, we analyzed 64 immunocompromised children suffering from HZ in the period from January 2004 to December 2019 at the Department of Pediatrics, Jena University Hospital, Jena, Germany. All patients received immunosuppressive treatment due to hematological malignancies, solid tumors, and genetic diseases (Table 1). The vesicular exanthema was localized on thoracic and abdominal dermatomes in 32 patients (50.0%). In 23 patients (35.9%), HZ lesions were localized on the extremities. Nine patients (14.1%) showed a typical rash on the head. In all patients, the rash was limited to one, two, or three dermatomes. There was no evidence of cutaneous or visceral dissemination. We excluded patients who received treatment with 5-fluorouracil and its prodrugs within the last 4 weeks. A detailed characterization of the study population is presented in Table 1.

The diagnosis of HZ was performed by clinical examination. Typical skin manifestations included unilateral vesicular lesions within one to three adjacent dermatomes. Burning pain usually preceded the rash. To confirm the clinical diagnosis, vesicular fluid was investigated in 34 patients and blood samples were collected from 16 patients. These samples were analyzed for the presence of VZV DNA using the polymerase chain reaction (PCR) [14].

Following the diagnosis, treatment with brivudine was immediately initiated in all patients. Brivudine was administered orally for a minimum of 7 days. Duration of therapy was determined individually and extended up to 21 days depending on how well the vesicular rash regressed. All patients received brivudine once daily at a dose of 2 mg/kg. To monitor the patient’s response to the drug, we examined all patients twice a week. During this period, we documented the time to full crusting of lesions and loss of crusts. Complications or adverse effects were also recorded.

We investigated the course of HZ in 64 patients treated with brivudine, of whom 36 patients (56.2%) were males and 28 patients (43.8%) were females during a period of 16 years. All patients reported a painful, burning, and itchy feeling in the involved dermatome before the typical vesicular lesions became visible. Some patients also experienced hyper- or paresthesia. In the following days, patients described the pain to be more intense. Unilateral and dermatomal distributed lesions were visible and developed into taut blisters filled with serous fluid. All analyses of the fluid of lesions (34 out of 34 samples) showed positive results. Furthermore, VZV DNA was detected in 93.8% of the patients in whom the blood samples were tested (15 out of 16 samples). The PCR results are presented in Table 2.

The first dose of brivudine was started immediately after the diagnosis of HZ. Thirty patients (46.9%) were treated for 1 week and 26 patients (40.6%) were treated for 2 weeks. Eight patients (12.5%) received the drug for 3 weeks. In addition to therapy with brivudine, we prescribed supportive treatment consisting of drying preparations (zinc oxide paste), antipruritics (dimetindene), and analgesics (ibuprofen, metamizole, or paracetamol). Sixteen patients (25.0%) did not need any additive treatment. All other patients took advantage of additional drying preparations. Eight patients (12.5%) received medication to reduce itching and only 4 patients (6.3%) took pain-alleviating medication.

To evaluate patients’ response to the treatment, we focused on the time to crusting of the lesions. This allowed us to confirm that the process of healing had begun. Full crusting occurred within a median time of 6 days (Fig. 1). Fifty patients (78.1%) showed full crusting within the first week of treatment. In all remaining patients (21.9%), full crusting occurred during the second week. Our second hallmark of patients’ response to brivudine was the time to the complete healing of lesions. This was observed between the 7th and the 21st day of treatment. Eleven pediatric patients (17.2%) showed complete healing within the first week, 39 patients (60.9%) during the second, and 14 patients (21.9%) during the third week. The results of the course of the disease are demonstrated in Table 3 and Figure 1. Figure 2 shows an HZ infection on the left scrotum. Figure 3 shows its complete healing after 7 days of brivudine therapy.

In summary, brivudine was very well tolerated. Interestingly, 1 patient showed a special course of the disease, briefly summarized as follows. At the age of 14 years, he received the VZV vaccination as he had not suffered from chickenpox yet. Three weeks later, he was diagnosed with acute myeloid leukemia (AML). It can be assumed that he was already ill at the time of the vaccination. Since his immune system was weakened and he was further immunosuppressed by chemotherapy, he suffered from chickenpox 4 weeks after the diagnosis of AML. Since then, the VZV persisted in his dorsal root and cranial nerve ganglia. Two months after bone marrow transplantation, his VZV-specific cellular immunity declined, which led to HZ only 59 days after his first transplantation and 56 days after his second HSCT. The VZV DNA taken from the fluid of the lesions and blood was tested and sequencing showed that only the vOka DNA of the live attenuated Oka vaccine was found. PCRs with primers for single-nucleotide polymorphisms (SNPs) located in open reading frame 6 (SNP 5745) and open reading frame 62 (SNPs 105544, 105705, 107136, 107252) of the VZV genome were used to discriminate vaccine from wild-type strains [15, 16]. PCR products were sequenced and evaluated at the indicated variable positions. The analyzed variants carried most stable vaccine markers and matched with the SNP profile of Oka/Merck varicella vaccine strain as published by Tillieux et al. [17]. This means that the patient suffered two times from vaccine-associated HZ because he was vaccinated shortly before his diagnosis of AML and start of chemotherapy.

Finally, we did not observe any myelo-, hepato-, or nephrotoxic side effects during the treatment with brivudine. We explored the following laboratory parameters of aspartate transaminase, alanine transaminase, cholinesterase, glutamyl transpeptidase, lactate dehydrogenase, alkaline phosphatase, ferritin, albumin, total bilirubin, C-reactive protein, and international normalized ratio. None of the patients treated with brivudine suffered from postherpetic neuralgia.

Our results demonstrate the highly beneficial effects of treating HZ with brivudine in immunocompromised children. The course of HZ is self-limiting in otherwise healthy patients [18]. Therefore, antiviral therapy is usually not indicated in immunocompetent children. Antiviral therapy is indicated if risk factors are already known, and complications could occur during the course of the disease. This requires careful consideration of whether the potential benefits outweigh the potential risks.

Previously published data have shown that brivudine is significantly more effective in inhibiting VZV replication than comparable nucleoside analogs such as acyclovir, famciclovir, or penciclovir (Andrei et al. [12], 1995). Wassilew and Wutzler [13] confirmed these results and compared the efficacy and safety of oral brivudine and acyclovir in a double-blind randomized study with immunocompetent patients.

Another small Belgian study with four immunocompromised patients showed excellent treatment results. All patients showed regression of the lesions within the first days and full recovery without any adverse effects [19]. Additionally, patients suffered only from little or no side and long-term effects and reported a significantly lower level of pain when receiving treatment with brivudine [20].

Apart from enhanced effectiveness, the convenience for patients is much higher with brivudine and patient compliance hence induced. Compared to acyclovir, brivudine can be taken per os as a single dose once daily. This is due to greater potency and long plasma elimination half-life [13]. In contrast, acyclovir is an intravenous antiviral drug, which must be given three times daily for at least 5 days. The oral alternative of acyclovir, which needs to be administered five times daily, is poorly absorbed by the body. This requires hospitalization of immunocompromised children.

Treatment with a nucleoside analog should be started as early as possible to reduce complications and adverse effects. Schmader [21] showed that early antiviral treatment significantly decreases pain intensity and increases lesion regression in the elderly. Based on this knowledge, we administered the drug as soon as possible. In our cohort, treatment was started immediately after the onset of the rash in all 64 children. We could not observe liver, kidney, or myelotoxicity during and after treatment.

A secondary aim of early treatment was to reduce the incidence of postherpetic neuralgia, defined as the persistence of dermatomal pain of more than 3 months. None of our patients reported pain beyond the 3-month mark. Standard treatment recommends brivudine therapy for 7 days. Antiviral medication should be continued if crusting and healing have not been completed after the 7th day of therapy.

Due to the absence of large cohort studies, brivudine is not currently approved for antiviral treatment of HZ in children. However, if risk factors are known or complications occur, it should be considered if the benefit of therapy (shortened duration and spread of symptoms, reduced intensity, and duration of HZ-associated pain) is greater than the risks of therapy. Brivudine is contraindicated in immunocompromised patients who receive treatment with 5-fluorouracil or 5-fluoropyrimidine-containing drugs within the past 4 weeks or are planned to receive treatment within the next 4 weeks. Lethal drug interactions can occur if both drugs are taken.

In conclusion, our findings confirm previous small studies on treating children suffering from HZ with brivudine [22‒24]. Compared to those studies, we investigated a significantly larger cohort of patients and could demonstrate that immunocompromised children can be safely treated with brivudine. This group of patients benefits significantly from an outpatient treatment with a once-daily dosing regimen. Further studies in larger cohorts are necessary to confirm our results.

All procedures were in accordance with the ethical standards of the institutional research committee. The study was reviewed and approved by the Jena University Hospital Ethics Committee (2077). Written informed consent was obtained from the participants’ parent/legal guardian/next of kin to participate in the study.

The authors declare to have no potential conflicts of interest.

No funding was received for this work.

Conception and design of the work: Bernd Gruhn and Peter Wutzler. Interpretation of data of the work: Bernd Gruhn and Clara Vogel. Drafting the work and/or revising it critically for important intellectual content and final approval of the version to be published: Clara Vogel, Laura Wetzel, Peter Wutzler, and Bernd Gruhn.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.