Efficacy of Antiviral Drugs against Feline Immunodeficiency Virus

Feline immunodeficiency virus (FIV) is one of the most common infectious agents affecting cats worldwide .FIV and human immunodeficiency virus (HIV) share many properties: both are lifelong persistent lentiviruses that are similar genetically and morphologically and both viruses propagate in T-lymphocytes, macrophages, and neural cells. Experimentally infected cats have measurable immune suppression, which sometimes progresses to an acquired immunodeficiency syndrome. A transient initial state of infection is followed by a long latent stage with low virus replication and absence of clinical signs. In the terminal stage, both viruses can cause severe immunosuppression. Thus, FIV infection in cats has become an important natural model for studying HIV infection in humans, especially for evaluation of antiviral compounds. Of particular importance for chemotherapeutic studies is the close similarity between the reverse transcriptase (RT) of FIV and HIV, which results in high in vitro susceptibility of FIV to many RT-targeted antiviral compounds used in the treatment of HIV-infected patients. Thus, the aim of this article is to provide an up-to-date review of studies on antiviral treatment of FIV, focusing on commercially available compounds for human or animal use.

DNA or RNA strands, but because of small differences in the molecular structure, chain termination results or nonfunctional nucleic acids are produced [5,8,9]. NARTIs are accepted as false substrates by viral enzymes as well as by cellular enzymes, which is the main reason for their toxicity [10].

Zidovudine
Zidovudine (3′-azido-2′,3′-dideoxythymidine, AZT) was first synthesized in the 1960s [11] as a potential anticancer drug. In 1985 it was shown to be effective against HIV [12] and became the first drug approved for treatment of HIV infection [13].
The anti-FIV activity of zidovudine has been assessed in numerous in vitro studies in different cell systems [14,[15][16][17][18][19][20][21][22][23][24][25][26]. The first in vitro study was carried out in 1989, when North and coworkers showed that zidovudine inhibited FIV replication in Crandell-Rees feline kidney (CRFK) cells. The susceptibility of FIV to zidovudine was similar to that of HIV [27]. There is evidence that FIV can become resistant to nucleoside analogues, as is the case in HIV. Zidovudine-resistant FIV mutants can arise after only six months of use, and a single-point mutation in the FIV gene is responsible for resistance [10].
In vivo, zidovudine can reduce plasma viral load, improve the immunologic and clinical status of FIV-infected cats, increase quality of life, and prolong life expectancy [16]. In placebo-controlled trials, zidovudine improved stomatitis and increased the CD4/CD8 ratio in naturally FIV-infected cats. In some cats with FIV-associated neurologic signs, marked improvement was reported within the first days of therapy [28,29].
Zidovudine not only inhibits RT, but also cellular polymerases, and this can lead to bone marrow suppression. Regular blood cell counts are necessary during zidovudine treatment because non-regenerative anemia is a common side effect [28]. Cats with bone marrow suppression should not be treated with zidovudine. Most FIV-infected cats treated with zidovudine for as long as two years tolerated the drug well. The hematocrit can decline within three weeks of initiating treatment to approximately 50% of baseline but increases afterwards in most cases, even without discontinuation of treatment. If the hematocrit drops below 20%, discontinuation of treatment is recommended, and anemia usually resolves within a few days. Other side effects in cats, including vomiting or anorexia, are rare [28].
Stavudine is active against FIV in vitro [18][19][20]23,26,35,36]. Mutants of FIV that are resistant to stavudine and cross-resistant to several other antivirals, including zidovudine, have been detected. Resistance is caused by a single-point mutation in the RT-encoding region of the pol gene [26]. No in vivo data in FIV-infected cats have been published.

Didanosine
Didanosine (2′,3′-dideoxyinosine, ddl) was shown to be active against HIV in 1986 [37]. In the United States, it was the second drug to be approved for treatment of HIV and has been on the market since 1991 [5].
Lamivudine is active against FIV in vitro [3,20,21,23,38,41]. A combination of zidovudine and lamivudine had synergistic anti-FIV activities in cell cultures [41]. FIV mutants resistant to lamivudine and containing a point mutation in the RT gene were selected in vitro and showed cross resistance to zidovudine [23].
In one in vivo study, experimentally FIV-infected cats were treated with a high-dose zidovudine/ lamivudine combination, which protected some cats from infection when treatment was started before virus inoculation. However, zidovudine/lamivudine treatment showed no anti-FIV activity in chronically infected cats. Severe side effects, including fever, anorexia, and marked hematologic changes, were observed in some of the cats with this high-dose dual-drug treatment [41]. Thus, high-dose lamivudine treatment alone, or in combination with zidovudine, is not recommended in naturally FIV-infected cats.

Nucleotide Analogue Reverse Transcriptase Inhibitors
Similar to NARTIs, NtARTIs also interact with the catalytic site of RT and are incorporated into the elongating proviral DNA strand, causing chain termination [5,42]. They compete with natural nucleotides and therefore function as competitive substrate inhibitors. However, in contrast to NARTIs, NtARTIs already contain one phosphate group and thus need only two intracellular phosphorylation steps for conversion to their active forms because the first and often rate-limiting phosphorylation step is unnecessary [5,42,43].

Adefovir
Adefovir (2-(6-amino-9H-purin-9-yl)-ethoxy-methyl-phosphonic acid, PMEA) is active against herpesviruses, hepadnaviruses (hepatitis B), and retroviruses [44]. Adefovir is not licensed as an HIV drug, but is currently available as an oral formulation (bis-POM PMEA) approved for the treatment of chronic hepatitis B. Adefovir belongs to the acyclic nucleoside phosphonates, in which the alkyl side chain of purines and pyrimidines is linked to a modified phosphate moiety and a C-P phosphonate linkage replaces the normal O5′-P phosphate linkage [43,45]. This phosphonate bond is not hydrolysable, which makes it more difficult to cleave off these compounds once they have been incorporated at the 3′-terminal end of the elongating proviral DNA strand [5]. Adefovir inhibits FIV replication in vitro [46].
Several studies have investigated the efficacy of adefovir in either experimentally and naturally FIV-infected cats [47,[48][49][50][51][52][53]. A few of those studies showed some efficacy, but also reported severe side effects, mainly non-regenerative anemia. In a recent study, adefovir was administered to FIV-infected cats in a six-week placebo-controlled, double-blinded clinical trial; ten cats received adefovir (10 mg/kg SC twice weekly) and ten cats received placebo. There was no decrease in proviral or viral loads in treated cats, and treated cats developed a progressive, sometimes life-threatening anemia, which is a common adverse effect of NtARTIs [53]. This shows that results obtained in experimental studies cannot always be applied to a field situation and emphasizes the importance of controlled clinical field trials. Based on the lack of efficacy in the recent placebo-controlled field trial and the side effects, adefovir cannot be recommended for treatment of FIV-infected cats.
Tenofovir is effective against FIV in vitro [25,45], and there is some evidence that tenovovir might have greater anti-FIV efficacy with less cytotoxicity than other antiretroviral compounds, including adefovir [45,54]. However, in vivo studies are lacking and should be a focus of future research.

Non-Nucleoside Reverse Transcriptase Inhibitors
Most of the NNRTIs are highly specific for HIV-1 and are not active against other retroviruses, including HIV-2 and FIV [7,42]. Unlike NARTIs and NtARTIs, which bind to the catalytic site of RT, non-nucleoside RT inhibitors interact with an allosteric site of the enzyme [5] and are not incorporated into the proviral DNA strand [42]. They are classified as non-competitive inhibitors of RT and do not require intracellular activation for inhibition of the enzyme [8,42]. NNRTIs are a group of structurally diverse compounds that all bind a single site of the RT [55]. The interaction with the allosteric site which is located in close proximity to the catalytic site, leads to a number of conformational changes of the RT [55,56]. Among other effects, these changes cause a reduction in the interaction between the DNA primer and the polymerase domain of the enzyme and thus, inhibit virus replication [55,56].
Three of the FDA-approved NNRTIs (nevirapine, delavirdine, efavirenz) have been shown not to be effective against FIV in vitro [40,57]. In vivo studies have not been performed, presumably because of the lack of in vitro efficacy [7,57]. Only one old NNRTI, suramin, with a broad antiviral spectrum has been used in veterinary medicine. A major breakthrough in the treatment of FIV would be the discovery of more NNRTIs with activity against FIV RT.

Suramin
Suramin (1-(3-benzamido-4-methylbenzamido)-naphthalene 4,6,8-trisulfonic acid sym-3′-urea sodium salt), a sulfated naphthylamine and trypan red derivative, is one of the oldest known antimicrobial agents. It has been used as an antitrypanosomal agent and for the treatment of some tumors, such as prostate cancer [58]. It also has an inhibitory effect on the RT activity of retroviruses and has also been used in humans with HIV infection [59]. Suramin inhibits RT by interacting with the template-primer binding site of the enzyme. Thus, it competitively binds to the primer binding site (without being a nucleoside analogue) and inhibits the template-primer binding that is necessary for DNA elongation. Suramin can therefore be classified as an NNRTI [60].
Suramin is effective against feline leukemia virus (FeLV) in vivo [61,62], and thus, could potentially be active against FIV, although this has not been investigated.
Suramin is associated with a significant number of severe side effects in humans, such as nausea and anaphylactic shock as immediate reactions during administration and peripheral neuritis leading to palmar-plantar hyperesthesia, photophobia, skin reactions, agranulocytosis, hemolytic anemia, and destruction of the adrenal cortex as later side effects [58,59,[63][64][65]. In cats with FeLV infection, the major adverse effects of suramin were transient vomiting and anorexia [61].

Nucleotide Synthesis Inhibitors
Nucleotide synthesis inhibitors prevent synthesis of nucleotides through various mechanisms. They have a broad spectrum of activity but are associated with marked toxicity mainly because they are non-selective and therefore also interfere with normal cellular nucleotide synthesis. Some, for instance foscarnet, interfere with the exchange of pyrophosphate from deoxynucleoside triphosphate during viral replication by binding to RT or DNA polymerase, thereby preventing nucleotide synthesis [66]. Others, such as ribavirin, inhibit inosine monophosphate dehydrogenase after intracellular phosphorylation, which in turn leads to inhibition of guanosine monophosphate.

Foscarnet
Foscarnet (phosphonoformic acid, PFA) has broad-spectrum antiviral activity against DNA and RNA viruses, including retroviruses. It is FDA-approved for the treatment of HIV-associated cytomegalo and herpes simplex virus infections in humans [67]. Foscarnet is usually administered intravenously by continuous intravenous infusion because of its short half-life, which has also been demonstrated in cats [68]. Oral administration of the drug is possible but can result in irritation of mucous membranes and oral bleeding. Foscarnet has many side effects, including nephrotoxicity and myelosuppression, in both humans and cats. It also is toxic to epithelial cells and mucous membranes, resulting in gastrointestinal side effects and genital epithelium ulceration. In addition, it chelates various cations, which can lead to hypocalcemia, hypomagnesemia, and hypokalemia [69,70].
In vitro, foscarnet has been shown to be active against FIV, but foscarnet-resistant FIV strains can develop [14]. No in vivo studies in FIV-infected cats have been carried out, likely because of the severe side effects and necessity for continuous intravenous administration of the drug.

Ribavirin
Ribavirin (1-β-D-ribofuranosyl-1 H-1,2,4-triazole-3-carboxamide, RTCA) has marked in vitro antiviral activity against a variety of DNA and RNA viruses [71]. Systemic administration of ribavirin is limited in cats because of side effects [72]. Sequestration of ribavirin within erythrocytes results in hemolysis, even when low doses of the drug are used [73,74]. In addition, there is a dose-related toxic effect on bone marrow, primarily on megakaryocytes, resulting in thrombocytopenia and hemorrhage. With prolonged ribavirin treatment or at higher doses, the production of erythrocytes and neutrophils also is suppressed. Ribavirin also can induce hepatic toxicity. An attempt to decrease the toxicity of ribavirin by incorporating it into lecithin-containing liposomes and administering it at lower doses was not successful [75].
Ribavirin is active against many viruses in vitro, including FIV [23,76]. Therapeutic concentrations are difficult to achieve in vivo because of toxicity [74]. To date, the efficacy of ribavirin has not been investigated in FIV-infected cats.

Receptor Homologues/Antagonists
Receptor homologues/antagonists bind to the virus or to the cellular receptor, leading to inhibition of viral cell-surface binding. Most of the receptor homologues/antagonists are highly selective for HIV and not useful in veterinary medicine. An exception is the class of antiviral compounds called bicyclams, which have been used in cats with FIV infection. Bicyclams act as potent and selective CXC chemokine receptor 4 (CXCR4) antagonists [77,78]. Chemokine receptors belong to the group of seven transmembrane-proteins that enable signal transmission through rapid influx of calcium into the cell. They are essential co-receptors for HIV as well as for FIV during infection of CD4+ lymphocytes [79,80]. By binding to CXCR4, bicyclams prevent interaction of CXCR4 with other ligands, thereby inhibiting the entry of HIV or FIV into the cell [81][82][83].
Plerixafor is active against FIV in vitro [82]. In a placebo-controlled double-blinded clinical trial, treatment of naturally FIV-infected cats with plerixafor resulted in a significant decrease in proviral load in treated cats when compared to the placebo group. There was a concomitant decrease in serum magnesium levels, which did not produce any clinical consequences. Development of resistance of FIV isolates to plerixafor did not occur during treatment [53]. In cats, plerixafor is administered at a dosage of 0.5 mg/kg every 12 h. Monitoring of magnesium and calcium levels should be performed at regular intervals during treatment [53]. Further studies investigating the potential of this promising drug are needed.

Protease Inhibitors
Protease inhibitors (PI) specifically bind to the active site of the protease and therefore prevent viral replication. Several PIs have been used for successful treatment of HIV. Nevertheless, side effects and development of viral resistance were found during treatment, and therefore additional compounds that bind to sites other than the active site of the protease have been developed [85][86][87].

Integrase Inhibitors
The enzyme integrase catalyzes strand transfer (3′-end joining), which inserts both viral DNA ends into a host cell chromosome during proviral DNA integration [5,88]. Once integrated, the provirus persists in the host cell genome and functions as a template for replication of the viral genome, leading to the formation of new viruses [89]. The high degree of conservation of integrase-active sites across many retroviruses suggests that FIV might also be sensitive to integrase inhibitors [90]. Integrase inhibitors act through inhibition of integration of the proviral DNA that is produced by reverse transcription of the viral RNA genome [91].

Raltegravir
Raltegravir is used as an anti-HIV compound. The drug was shown to be active against FIV in vitro [92], but FIV was less susceptible to raltegravir than HIV [92].
No studies in FIV-infected cats exist so far. Although there are no in vivo studies on the efficacy of raltegravir in FIV-infected cats, the drug recently was shown to be effective against FeLV and was safe in cats [93].

Interferons
Interferons (IFNs) are polypeptide molecules with various biological functions [94]. They play an important role in mediating antiviral and antigrowth responses and in immune response modulation [95]. They can be divided into type I and type II IFNs, both of which have antiviral properties. Type I IFNs, including IFN-α, IFN-β, and IFN-ω, are produced by virus-infected cells [94,96], whereas type II IFN, consisting of only IFN-γ, is produced by activated T lymphocytes and natural killer cells in response to recognition of virus-infected cells [97]. IFNs act in an autocrine or paracrine fashion [98] inducing an anti-viral state in non-infected cells. IFNs bind to specific cell surface receptors and result in the transcription of IFN-stimulated genes. The products of these genes are proteins with potent anti-viral properties that interfere with various stages of viral replication [98]. Several studies suggest that retroviral protein synthesis is not affected by IFNs and therefore conclude that the antiviral activity of IFNs is mainly related to interference with later stages of the viral replication cycle such as virion assembly and release [94,99]. Interferons also trigger virus-infected cells to undergo apoptosis by activating gene expression for apoptosis [97,99], which prevents the spread of virus from infected cells and aids in the clearance of virus infection [97]. Human IFNs have been manufactured by recombinant DNA technology and are available commercially. Recombinant feline IFN-ω is on the market in Japan, Australia, and many European countries and is licensed for use in cats and dogs.
Human IFN-α becomes ineffective after three to seven weeks of parenteral use in cats because of the production of neutralizing antibodies [100]. Anti-IFN-α antibody production does not occur with oral administration of IFN-α and therefore this route allows for a longer period of treatment. IFN-α is inactivated by gastric acid and destroyed by trypsin and other proteolytic enzymes in the duodenum [101], which means that direct antiviral effects are unlikely after oral application. However, oral IFN-α appears to have immuno-modulatory activity, because it can stimulate local lymphoid tissue. The release of cytokines by lymphatic cells in the oropharyngeal area triggers a cascade of immunologic responses with systemic effects [102][103][104].
RHuIFN-α has been shown to be active against FIV in vitro [105]. Although frequently used in the field for treating FIV-infected cats, controlled studies evaluating the effect of parenteral administration of rHuIFN-α in FIV-infected cats have not been conducted.
Use of oral rHuIFN-α in 24 ill, naturally FIV-infected cats (50 U/kg applied to the oral mucosa daily for seven days on alternating weeks for six months, followed by a two-month break, and then repetition of the six-month treatment) resulted in improvement of clinical signs (e.g., fever, lymphadenopathy, opportunistic infections) in a placebo-controlled, double-blinded study [106]. However, proviral and viral loads were not monitored during thiat study and therefore it is impossible to conclude whether treatment with rHuIFN-α had indeed an effect on FIV, or rather on secondary infections.

Feline Interferon-ω
Recombinant feline interferon-ω (rFeIFN-ω), the corresponding feline interferon, is licensed for use in veterinary medicine in Japan, Australia, and some European countries. It can be used in cats for long periods without antibody development, and no major severe side effects have been reported [107].
IFN-ω inhibits FIV replication in vitro [105]. One placebo-controlled, multicenter study that investigated the effect of parenteral rFeIFN-ω against FIV infection in 62 naturally FIV-infected cats (treated with 10 6 U/kg SC q 24 h on five consecutive days) did not find a difference in the survival rate in treated cats. However, some improvement in clinical scores, including eight categories of clinical signs (rectal temperature, behavior, appetite, thirst, dehydration, mucous membrane appearance, stomatitis, and death) as well as improvement in laboratory abnormalities (leukopenia, leukocytosis, and anemia) occured [107]. In another study, which evaluated naturally FIV-infected cats housed in a shelter, some clinical improvement was observed after parenteral rFeIFN-ω (10 6 U/kg SC q 24 h on FIVe consecutive days for three cycles), but this study lacked a placebo control. In that same study, hematologic values remained within reference intervals, and there were no biochemical abnormalities associated with rFeIFN-ω treatment [96].
A recent study evaluated the use of oral administration of rFeIFN-ω for the treatment of eleven client-owned, naturally FIV-infected cats with clinical signs [108]. The treatment protocol was 10 5 U/cat PO q 24 h for 90 consecutive days, administered by the cats' owners. A historical retrospective group was used as a control for comparison (10 6 U/kg SC q 24 h on five consecutive days for three cycles), but a placebo group was not included. Treatment with oral rFeIFN-ω resulted in a significant improvement in clinical scores (e.g., oral lesions, coat appearance, body condition score, and ocular discharge) after treatment. In addition, there was no significant difference between the SC historical control group and the PO group, suggesting that oral administration of rFeIFN-ω might be a viable and less expensive alternative [109]. In a recently published study that assessed viremia, provirus load, and blood cytokine profile in naturally FIV-infected cats treated with oral rFeIFN-ω (10 5 U/cat PO q 24 h for 90 days) or with subcutaneous rFeIFN-ω (10 6 U/cat SC q 24 h for 5 consecutive days in three courses), no change in the level of viremia or in most cytokine levels was found; a placebo control group was not included [109]. The fact that virus load remained unchanged but some clinical improvement was observed in earlier studies suggests that rFeIFN-ω has an effect on secondary infections rather than on FIV itself [94]. As there are major differences in outcomes of the different studies on feline IFN-ω in FIV-infected cats. Thus, a definitive conclusion cannot be drawn without additional randomized, placebo-controlled, and double-blinded studies that include a sufficiently high number of naturally FIV-infected cats.

Conclusions
Unfortunately, the efficacy of antiviral compounds for the treatment of FIV in cats has been generally poor. The duration of treatment in many clinical trials was relatively short and might have been inadequate for infections with a long clinical course. In addition, it is difficult to compare treatment results of cats infected experimentally and kept under laboratory conditions and pet cats infected with field strains of FIV. Therefore, further well-designed double-blinded, placebo-controlled trials using antiviral drugs in naturally FIV-infected cats are needed to determine the efficacy and side effects of different antiviral compounds. Table A1. Treatment options (antiviral drugs) for FIV-infected cats (including EBM grades for judgment of the available efficacy data; EBM grades used according to the European Advisory Board of Cat Diseases (ABCD).

Non-Nucleoside Reverse Transcriptase Inhibitor
Suramin no nd likely too toxic IV

Nucleotide Synthesis Inhibitors
Foscarnet (PFA) yes [14] nd effective in vitro, but too toxic IV Ribavirin yes [23,76] nd possibly effective, but too toxic in cats IV

Receptor Homologues/Antagonists
Plerixafor yes [82] yes [53] some effect in a study in privately-owened cats (thus, can be considered as treatment) I