Treatment Options for HHV-6A & HHV-6B
An increased awareness of diseases associated with HHV-6 acute infection (in both immunocompetent and immunocompromised patients has resulted in a growing interest in the evaluation of the best treatment for HHV-6 disease. A number of drugs used for cytomegalovirus infection have shown in vitro efficacy against HHV-6. Dozens of case reports and several small studies have been conducted to demonstrate efficacy of the compounds against HHV-6 in vivo. However, no compound has yet been approved for the specific treatment of HHV-6 infection.
A review of the published data suggests that foscarnet is the most selective in vitro inhibitor of HHV-6 among these compounds with an average selectivity index (SI) of 52.4 in 12 studies. The SI for cidofovir averaged 18.1 in 11 studies and the SI for ganciclovir was surprisingly poor at 4.5 in 7 seven studies. Efficacy did not differ substantially between HHV-6A and HHV-6B for any of the three compounds. Newer compounds such as brincidofovir and the antimalarial drug artesunate show excellent antiviral activity with limited in vitro toxicity. Immunotherapy with adoptive T cells and natural compounds that stimulate cellular immunity also show promise for controlling HHV-6 infection.
Introduction
Reactivation of latent Human herpesvirus-6B (HHV-6B) in the post-transplant setting can result in the development of fatal HHV-6 encephalitis and other neurological sequelae including exanthematous rash, fever, seizures [Yamashita 2005], encephalopathy [Ichiyama 2009], limbic encephalitis [Hill 2012] and amnesia [Gorniak 2006], cognitive dysfunction [Zerr 2011], lymphadenopathy [Maric 2004], colitis [Revest 2011, Lamoth 2008, Amo 2003], and hepatitis [Chevret 2008, Yoshikawa 2006]. Less established associations have been reported for renal failure [Chapenko 2009,], hemophagocytic syndrome [Marabelle 2010, Dharancy 2008], myocarditis [Leveque 2011], pneumonitis [Yamaguchi 2003], hypogammaglobulinemia [Kano 2004], and arteriopathies [Takatsuka 2003].
HHV-6A reactivation has been found in subsets of patients with multiple sclerosis [Garcia-Montojo 2011], HIV infection [Lusso 2007], encephalitis [Crawford 2007], and syncytial giant-cell hepatitis [Potenza 2008].
Although the pharmacological treatment of HHV-6 acute infection (AI) is currently administered only in a limited number of clinical settings, an increased awareness of HHV-6 diseases associated with acute infection in both immunocompetent and immunocompromised patients has resulted in a growing interest in the evaluation of the best treatment for HHV-6 disease [Agut 2011].
Antiviral Drugs for HHV-6A and HHV-6B
No antiviral agents have been officially developed for the treatment of HHV-6. In the absence of specific anti-HHV-6 agents available for treatment, the purpose of this report is to present a review of the existing literature on the in vitro efficacy of the three available anti-cytomegalovirus (CMV) drugs physicians most use for the treatment of HH-6 disease. The two most commonly used drugs are the nucleoside analog ganciclovir (or oral version valganciclovir) and the pyrophosphate analog foscarnet (Table 1), either alone or in combination. The nucleotide analog cidofovir (CDV) has been used less commonly often in combination with either foscarnet or ganciclovir. An oral version brincidofovir (CMX-001) is currently in a Phase III trial for the prevention of CMV disease in stem cell transplant patients. The antimalarial drug artesunate, has potent antiviral properties and been used successfully for HHV-6 viral myocarditis in one case report. Unlike the polymerase inhibitors, artesunate works at an early stage, inhibiting immediate early activity (Milbradt 2009).
Several dozen case reports and several small studies suggesting that both pre-emptive and prophylactive treatment with foscarnet and/or ganciclovir can improve outcome for HHV-6 encephalitis (Ljungman 2006 Ishiyama 2010, Ishiyama 2011, Hill 2012).
We reviewed the English literature database PubMed for all publications with information regarding the in vitro anti-HHV-6 efficacy of ganciclovir, cidofovir, and foscarnet. Data from each of these sources was gathered and sorted according to the assay, measurement, and viral strain/cell line used in each study to determine the average anti-HHV-6 effectiveness of each compound and determined that the selectivity index was significantly higher for foscarnet than for ganciclovir or cidofovir. See Supplemental Table 1.
Table 1. Current & Phase III drugs with activity against HHV-6 | |||||
Drug Name | Brand Name(s) | Main Use | In Vitro Activity against HHV-6 | Cross BBB? | Clinical Risks |
(Val)Ganciclovir | Cytovene, Valcyte | HCMV | Moderate | Yes | Bone Marrow Suppression |
Cidofovir | Vistide | HCMV | Good | No (a) | Renal Toxicity |
Foscarnet | Foscavir | HCMV | Excellent | Yes | Renal Toxicity |
Artesunate | Malartin,Artesor | malaria | Excellent | Yes | Minimal |
Brincidofovir (CMX-001) | ** | HCMV | Excellent | Yes (b) | GI Bleeding |
(a) The product insert for Vistide says it does not cross the BBB, but data is limited to one unpublished case report.
(b) This drug passes the BBB in animal studies, and one case report showed clearance of JC virus in PML. **Phase III trial underway for HCMV |
In vitro efficacy studies
The effective concentration (EC) values from in vitro studies on selected antiviral studies for HHV-6 are summarized in Table 2. Although valganciclovir is the only oral drug with activity against HHV-6, the in vitro efficacy of ganciclovir is surprising poor (Reyman 1995). This limited activity is likely due to the low level of phosphorylation by U69 kinase and reduced susceptibility of the DNA polymerase (DeBolle 2002, DeBolle 2004). Ganciclovir resistant strains have been isolated that map to the U38 DNA polymerase and U69 kinase (Manichanh 2001, Safronetz 2003). In vitro drug resistance has also been reported for cidofovir mapping to U38 (Bonnafous 2008), and for foscarnet that maps to the DNA polymerase (Bonnafous 2007). In vitro drug resistance to all three drugs has likewise been reported for HCMV (Chou 2011, Chou 2012).
Differing methods, cell lines, and viral strains have been used in the HHV-6 studies, making comparisons difficult. The most commonly used HHV-6A strain for testing is GS (Salahhuddin 1986, Ablashi 1991) and it is grown up in either HSB-2 or Sup-T1 cells. For HHV-6B, the commons strains are HST (Aubin 1991) and Z29 (Lopez 1988); both of these grow up in a MOLT-3 cell line. HSB-2, Sup-T1, and MOLT -3 are lymphoblastic T-cell lines. Cord blood lymphocytes (CBL) have also been used to grow and test these viral strains (Kern 2005) and Ganciclovir shows more antiviral activity when the virus is grown up in CBL than in one of the immortalized cell lines (Yoshida 1998, Naesens 2006B). A variety of methods have been used to test the efficacy of the drugs. Among these methods are fluorescence activated cell sorting (FACS) [Williams 2003, Reymen 1995, Long 2003, Manichanh 2000, De Bolle 2004, Kushner 2003, Kern 2005], immunofluorescence assay (IFA) [Long 2003, Akesson-Johansson 1990], quantitative or real time polymerase chain reaction (qPCR) [Naesens 2006, Bonnafous 2008], and DNA hybridization [Prichard 2011, Williams-Aziz 2005, De Clercq 2001]. These methods are described in detail in Supplemental Table 2.
Artesunate is a drug with a long history of use against malaria and also shown to be effective against HCMV (Effort 2002, Kaptein, 2006); it has been shown to be effective against drug resistant strains of HCMV (Shapira 2008, Wolf 2011). Artesunate was also reported to be successful in treating a child with HHV-6 viral myocarditis (Hakacova 2013). It has an inhibitory concentration-50% (IC50) of approximately 3.8 µM against HHV-6 in vitro and reduces early and late viral protein synthesis without affecting the host cells adversely. HHV-6A was particularly sensitive to artesunate (Milbrandt 2009). Artesunate has been in long use clinically against malaria showing that is safe for use in humans (Gomes 2008)
Hexadecyloxypropyl-CDV (Brincidofovir/CMX001) is a prodrug of cidofovir it has proven highly effective against many DNA viruses and is much more effective against HHV-6A in vitro with an EC50 of 3 nM (Williams-Aziz 2005, Keith 2004). It has also shown excellent activity in animal studies (Bidanset 2004, Quenelle 2004).
Several analogs of benzimadazole, including maribavir (MBV) were tested against both HHV-6A and B in vitro (Prichard 2011). Two of the L analogs of BDCRB (1H-β-D-ribofuranosyl-2-bromo-5,6-dichlorobenzimidazole), compounds labeled 3 and 4 in the paper, showed very good activity against HHV-6A and although the activity against HHV-6B was not as good they did inhibit HHV-6B replication. Maribavir, tested because of its effectiveness against HCMV, showed disappointing results in this study when compared with the successful analogs; however, maribavir did show increased activity against HHV-6A under reduced serum conditions. Phase II and III studies for activity against HCMV (Marty 2011A, 2011B) have shown that the drug is well tolerated and although the phase III study failed the authors think it could be successful with different parameters. Maribavir, therefore, might be a good candidate for a clinical trial against HHV-6A.
CMV 423, a non-nucleoside inhibitor, was found to be very effective against both HHV-6A and B. It shows high activity with an EC50 of between 49-53 nM and exhibits high antiviral activity combined with low cytotoxicity. It acts at the early stage of viral replication and seems to effect protein tyrosine kinase (Naesens 2006, DeBolle 2004).
Cyclopropavir has proven potent against many herpesviruses but is particularly potent against HHV-6A and B with EC50 of 7.8 and 0.7 µM respectively. Its mechanism of action is similar to that of ganciclovir, and it is more effective than ganciclovir in vitro. (Kern 2005). Prichard (2013) studied several analogs of cyclopopavir and found many of them to be extremely effective against both A and B with EC50 values below 10 µ; especially effective were ethers with 3 or more carbon atoms. Thioester analogs with up to 5 carbon atom were more effective against HHV-6B than the CDV control.
In vitro studies testing a wide range of acyclic nucleoside analogs (S)-3-deaza-HPMPA also referred to as 9-(S)-[3-hydroxy-2-(phosphonomethoxy) propyl]-3-deazaadenine or 3-deaza-HPMPA has shown very promising results with an EC50 of 1.4 µM (Reymen 1995, Naesens 2006). Arylsulfone derivatives exhibit an indirect inhibition viral DNA synthesis, but do not affect the DNA polymerase. Since its method of action is unique and it is effective against HHV-6 it holds promise for future studies (Naesens 2006)
Table 2. Comparison of EC50 Values
Status | Drug | EC 50 | Target/Description | References |
In use for CMV | Foscarnet | 8.4 μM | Viral polymerase | Reymen 1995 |
In use for CMV | Cidofovir | 12-15μM | Viral polymerase | Reymen 1995 |
In use for CMV | Ganciclovir/valganciclovir | >25 μM | Viral polymerase | Reyman 1995 |
In use for malaria | Artesunate | 3.8 μM | Cellular pathways | Milbradt 2009 |
Phase III for CMV | Brincidofovir (CMX-001) | 3 nM
(HHV-6A) |
Viral polymerase | Williams Aziz 2005, Keith 2004 |
Phase II for CMV | Cyclopropravir (MBX-400) | 0.7 μM (HHV-6B) | Pronucleotide analog | Prichard 2013 |
Phase II for CMV | Cyclopropravir (MBX-400) | 7.8 μM (HHV-6A) | Pronucleotide analog | Prichard 2013 |
Phase II for VZV | 3-Deaza-HPMPA | 1.4 µM | Pronucleotide analog | Reyman 1995, Naesens 2006A |
Arylsufone derivitives | 2.2 µg/ml | Inhibit late viral gene expression | Naesens 2006C | |
Benzimidazole analog 4 | 2.8 µM
(HHV-6A) |
Nucleoside analog | Prichard 2011 | |
Benzimidazole analog 4 | 9.7 µM
(HHV-6B) |
Nucleoside analog | Prichard 2011 | |
CMV423 | 49-53 nM | Inhibits DNA replication | Naesens 2007/DeBolle 2004 |
Blood Brain Barrier
HHV-6 encephalitis is a condition with a high rate of morbidity and requires urgent treatment to prevent death or permanent sequelae [Zerr 2011, Hoshino 2011, Hill 2012]. Therefore, the ability of the drug to penetrate the blood-brain barrier (BBB) is of critical importance in these cases. Both foscarnet and ganciclovir have been found to successfully penetrate the BBB [AHFS Drug Information, Sjovall 1989]. Although the product literature for cidofovir states that CSF concentrations of cidofovir was undetectable at <.1 ug/mL, in a patient whose corresponding serum concentration was 8.7 ug/ML 15 minutes after an infusion at 5 mg/kg. However, cidofovir may be helpful in patients who have a compromised blood brain barrier, and there are isolated reports of successful cidofovir use in encephalitis cases Blick 1997. In rabbits, radiolabelled cidofovir crossed the blood brain barrier but were found at much lower levels than were observed in the kidney tissue (EMEA 2005).
Immunotherapy
Adoptive T cell therapy has been proven to be practical for HHV-6 with the successful expansion of HHV-6 peptide specific T cells that can kill virus infected target cells (Gerdemann 2013). A recent report of a small study suggests that this approach has potential for in vivo use as well (Papadopoulou 2014). A donor lymphocyte infusion was used successfully in a patient with HHV-6 encephalitis (Yoshihara 2004). Another immunotherapy agent, Active Hexose Correlated Compound (AHCC) was shown to reduce saliva DNA HHV-6 levels in patients undergoing chemotherapy in Japan by 24% (p<0.05) while improving quality of life measures, hematotoxicity and hepatotoxicity (Ito 2014).
Other new agents in development
Two novel agents have been reported with good activity against CMV and would be expected to be effective for HHV-6 as well. Digoxin analogues have been shown to inhibit antiviral activity at nanomolar concentrations (Cai 2014). Also an artemesinin derived dimer, diphenylo phosphate, was shown to have highly selective activity against CMV, reducing intermediate early as well as late viral activity (He 2013).
Discussion
Foscarnet is considered the preferential treatment option for patients with anemia—as the administration of ganciclovir poses an additional risk of dose-limiting hematological toxicity [Wagstaff 1994]. Risks associated with foscarnet include renal toxicity [Wagstaff 1994] as well as complications from catheter-related deep vein thrombosis and infection. Unlike cidofovir, foscarnet cannot be administered in a peripheral vein. Of the three drugs, the only one available in oral form is Valcyte, a prodrug of ganciclovir.
GCV, CDV, and PFA inhibit CMV much more selectively than HHV-6, and all three drugs have poor bioavailability with a heightened risk for toxicity in the clinical setting [Williams-Aziz 2005, Williams 2003].
Other drugs in various stages of development have shown effectiveness against HHV-6 in vitro (Table 3). Brincidofovir (CMX001), currently in stage III clinical trials for CMV disease, combines CDV with a lipid-ester derivative which allows it to enter the cell more easily, and gives it a 100-fold advantage over CDV in the in vitro inhibition of HHV-6, allowing it to be effective at sub-cytotoxic concentrations [Painter 2011]. There is also some evidence that CMX001 can successfully penetrate the BBB, at least in animals [Ruiz 2011, Quenelle 2010]. Artesunate, a drug used internationally for the treatment of malaria—and in the USA for the treatment of severe malaria—has shown efficacy against HHV-6 in vitro, readily crosses the BBB [Milbradt 2009], and has been used in several centers for ganciclovir resistant CMV infections [Shapira 2008].
There are many good candidates with promising in vitro lab results. There is a tremendous need for the development of novel antivirals to improve treatment options for acute HHV-6 disease, especially HHV-6 encephalitis.
Ganciclovir Resistance
Manichahn et al. 2001 described a strain of HHV-6B acquiring ganciclovir resistance following serial passages in-vitro in the presence of the drug. Sequencing revealed two single nucleotide substitutions that resulted in amino acid substitutions (M318V and A961V) of U69 protein kinase and U38 DNA polymerase, two proteins involved in replication of the virus. The same amino acid substitution (A961V) of U69 protein kinase was found in HHV-6 infected PBMCs of a ganciclovir resistant AIDS patient, confirming its existence in humans (Manichahn 2001).
Another group (Safronetz 2003) induced ganciclovir resistance with mutations in U69 using a recombinant baculovirus system. By comparing baculoviruses expressing either a WT or mutant HHV-6 U69 protein kinase, they were able to identify three amino acid substitutions (M318V, C448G and C463Y) in the protein that could independently confer ganciclovir resistance. Isegawa et al. 2009 also reported a case of ganciclovir resistant HHV-6 in a human host. The strain of HHV-6, which was 100 fold more resistant to ganciclovir than WT, was isolated from blood following 2-3 months of ganciclovir therapy of a patient who underwent allogeneic stem cell transplantation. A mutation in U38 DNA polymerase (P462S) was found which corresponds to a mutation in the UL54 gene of ganciclovir resistant CMV. Another group described the development of a denaturing high-performance liquid chromatography assay (Nakano 2009) and a single nucleotide polymorphism detection assay (Isegawa 2010) for rapid detection of mutations in U69 protein kinase previously shown to induce ganciclovir resistance.
Cidofovir and Foscarnet Resistance
A strain of HHV-6B resistant to foscarnet was selected for in-vitro and then characterized (Bonnafous 2007). Four amino acid substitutions (T435R, H507Y, C525S and F292S) located in U38 DNA polymerase were identified to be in association with foscarnet resistance. The same group later selected for cidofovir-resistant HHV-6B isolates (Bonnafous 2008). Again, the mutation accounting for drug resistance was found in U38, resulting in an amino acid substitution, R798I, in a conserved domain of viral DNA polymerase.