Authors: Vanessa R Barrs, Stefan Hobi, Angeline Wong, Jeanine Sandy, Lisa F Shubitz, Paweł M Bęczkowski
Categories: Review, Antifungal triazoles, amphotericin B, dimorphic fungal infections, echinocandins, lagenidiosis, moulds, pythiosis, terbinafine
Source: Journal of Feline Medicine and Surgery
Authors: Vanessa R Barrs, Stefan Hobi, Angeline Wong, Jeanine Sandy, Lisa F Shubitz, Paweł M Bęczkowski
Invasive fungal infections (IFIs) and oomycoses (hereafter termed invasive fungal-like infections [IFLIs]) are characterised by penetration of tissues by fungal elements. The environment is the most common reservoir of infection. IFIs and IFLIs can be frustrating to treat because long treatment times are usually required and, even after attaining clinical cure, there may be a risk of relapse. Owner compliance with medication administration and recheck examinations can also decline over time. In addition, some antifungal drugs are expensive, have variable interpatient pharmacokinetic properties, can only be administered parenterally and/or have common adverse effects (AEs). Despite these limitations, treatment can be very rewarding, especially when an otherwise progressive and fatal disease is cured.
In the second of a two-part article series, the spectrum of activity, mechanisms of action, pharmacokinetic and pharmacodynamic properties, and AEs of antifungal drugs are reviewed, and the treatment and prognosis of specific IFIs/IFLIs - dermatophytic pseudomycetoma, cryptococcosis, sino-orbital aspergillosis, coccidioidomycosis, histoplasmosis, sporotrichosis, phaeohyphomycosis, mucormycosis and oomycosis - are discussed. Part 1 reviewed the diagnostic approach to IFIs and IFLIs.
Information on antifungal drugs is drawn from pharmacokinetic studies in cats. Where such studies have not been performed, data from ‘preclinical’ animals (non-human studies) and human studies are reviewed. The review also draws on the wider published evidence and the authors’ combined expertise in feline medicine, mycology, dermatology, clinical pathology and anatomical pathology.
AMB (amphotericin B); FC (flucytosine); FCZ (fluconazole); ISA (isavuconazole); ITZ (itraconazole); KCZ (ketoconazole); PCZ (posaconazole); TRB (terbinafine); VCZ (voriconazole).
For invasive fungal infections (IFIs) and oomycoses (hereafter termed invasive fungal-like infections [IFLIs]) in cats, antifungal susceptibility testing should be performed when possible, as outlined in Part 1.^ 1 ^ Targeted antifungal therapy based on the susceptibility profile of the pathogen isolated is an important factor in determining treatment success.

Penetration of antifungal drugs at the site of infection to achieve minimum inhibitory concentrations (MICs) against the target pathogen is essential for efficacy. In general, only free drug is considered biologically active; thus, drug concentrations in interstitial fluids (eg, cerebrospinal fluid [CSF]) are important determinants of antifungal efficacy.^ 2 ^ Small polar compounds with low plasma binding (eg, fluconazole [FCZ], isavuconazole [ISA], flucytosine [FC]) have volumes of distribution similar to total body water, penetrate well into aqueous sites (eg, ocular fluid, CSF) and have body plasma concentration ratios of ~1. In contrast, highly lipophilic drugs (eg, itraconazole [ITZ], posaconazole [PCZ]) that are also highly protein-bound have much higher volumes of distribution, penetrate best into tissues with high lipid content and have plasma concentration ratios of >1, but may not penetrate aqueous sites.^ 2 ^

For azoles, FC and echinocandins, the ratio between the area under the plasma concentration-time curve (AUC) and the MIC (AUC:MIC) for the fungal pathogen best correlates with antifungal efficacy. For amphotericin B (AMB), which is a concentration-dependent antifungal with a post-antifungal effect, the ratio between peak concentration (Cmax) and the MIC (Cmax:MIC) correlates best with efficacy.^
3
^

AMB is a polyene macrolide antibiotic.
✜ Activity AMB is fungicidal against many moulds, yeasts, thermally dimorphic fungi and Zygomycetes. Some species of Aspergillus (eg, Aspergillus terreus) have innate resistance.
✜ Mechanism of action AMB is fungicidal, with concentration-dependent activity.^ 7 ^ Molecules of the drug bind to ergosterol molecules in the fungal cell membrane, forming channels that assemble into a transmembrane pore; subsequent leakage of intracellular ions leads to cell death by osmotic lysis (Figure 1). AMB also causes free radical formation, further increasing fungal membrane permeability.^ 8 ^
✜ Mechanism of resistance Acquired resistance is uncommon and the underlying mechanisms are incompletely understood.
✜ Pharmacokinetics AMB is generally available in formulations for intravenous (IV) use (see box) owing to negligible oral bioavailability. However, orally administered nanofor-mulations are likely to become commercially available.^ 9 ^ After administration, AMB is highly protein-bound and also binds to cholesterol in cell membranes. It is distributed in the liver, spleen, bone marrow, kidneys and lungs. It has low water solubility and exhibits poor penetration of CSF and ocular fluid, except in the presence of inflammation, and thus has clinical efficacy for treatment of fungal central nervous system (CNS) and ocular infections. It is cleared slowly from the plasma and is eliminated unchanged in bile and urine.
✜ Adverse effects AMB is nephrotoxic. It causes renal oxidative injury from induced expression of pro-inflammatory cytokines (especially interleukin 6 and tumour necrosis factor alpha); it inhibits sodium-potassium ATPases and proton exchange, contributing to renal tubular acidosis and increased tubule permeability; and it causes afferent glomerular arteriolar vasoconstriction.^ 8 ^ The resultant decrease in renal blood flow and distal tubular ischaemia cause a reversible decrease in the glomerular filtration rate (GFR). Nephrotoxicity is usually reversible weeks to months after AMB discontinuation, provided the recommended total cumulative dose is not exceeded.
✜ Clinical use See Table 1 and also the later section on ‘Management of invasive fungal and fungal-like infections in cats’ for information regarding the use of AMB in specific infections.
Flucytosine (FC), also known as 5-flucytosine, is a pyrimidine analogue.
✜ Activity FC is fungistatic, with activity only against Cryptococcus and Candida species.


✜ Mechanism of action FC is a prodrug that is converted to 5-flurouracil (5-FU) by cytosine deaminase inside the fungal cell. 5-FU is either incorporated into RNA, disrupting protein synthesis, or converted to a compound that inhibits DNA synthesis.
✜ Mechanism of resistance Mutations in fungal cytosine permease (which internalises FC into the fungal cell) or fungal cytosine deaminase confer resistance. Resistance arises rapidly during treatment if FC is used as monotherapy.
✜ Pharmacokinetics FC is available in oral and IV formulations. The oral formulation has been used in cats but is expensive and can be difficult to source for veterinary use. In humans, the drug has high oral bioavailability, high water solubility, minimal protein binding and penetrates well into most body sites. It is 90% eliminated by glomerular filtration.
✜ Adverse effects Mammalian cells lack cytosine deaminase, resulting in selective fungal toxicity. However, gastrointestinal microbes may convert FC to 5-FU, leading to bone marrow toxicity (anaemia, leukopenia, thrombocytopenia). In patients with renal impairment, the most common adverse effects (AEs) are nausea, vomiting, diarrhoea and hepatotoxicity.^ 8 ^
✜ Clinical use FC has synergistic activity when combined with AMB against cryptococcal isolates and effectively penetrates the blood-brain barrier.^12,13^ In combination with AMB, FC is indicated for the treatment of CNS and ocular cryptococcosis, or for cases of cryptococcosis that are refractory to azole monotherapy. The drug should be avoided, or a lower dose used, if renal function is impaired.
✜ Mechanism of action Azole antifungals, including imidazoles (eg, ketoconazole [KCZ], clotrimazole) and triazoles (eg, FCZ), contain a heterocyclic five-member ring with either two nitrogen atoms (imidazoles) or three nitrogen atoms (triazoles). All imida-zoles are topical formulations, except for KCZ, which also has a tablet formulation. Oral KCZ has been withdrawn from the market in most jurisdictions due to hepatotoxicity in humans.
Azole antifungals inhibit 14a-demethylase, an enzyme belonging to the cytochrome P450 family (Figure l), which is also known as CYP51 in moulds and ERG11 in yeasts. It is encoded by CYP51A/CYP51B and ERG11 for moulds and yeasts, respectively.^ 3 ^ Azoles bind to the haem group of 14a-demethylase to prevent conversion of lanosterol to ergosterol, an essential fungal cell membrane component. Ergosterol depletion and the accumulation of methylated sterol precursors results in inhibition of fungal cell growth and/or fungal cell death.
✜ Mechanism of resistance Some fungal species have innate resistance to azoles. Resistance can also be acquired after therapeutic or environmental azole exposure. Widespread use of azole antifungals in agriculture has been increasingly associated with azole resistance in human clinical isolates of Aspergillus fumigatus.^ 14 ^ Although resistance is uncommon so far in veterinary clinical isolates, ongoing surveillance via antifungal susceptibility testing would be prudent.^15,16^
✜ Adverse effects AEs and drug-drug interactions are mostly associated with cross-inhibition of mammalian cytochrome P450 enzymes, resulting in inhibition of synthesis of cholesterol, cortisol and sex steroids, as well as inhibition of cytochrome P450-dependent drugs such as ciclosporin. This occurs most commonly with KCZ, followed by ITZ, VCZ and FCZ. Azoles also inhibit P-glycoprotein efflux pumps and, together with cytochrome P450 inhibition, this can result in increased oral absorption, tissue distribution and concentration of drugs. Azoles should not be used in pregnant animals as they are potentially teratogenic.
✜ Activity FCZ has activity
FCZ has no or limited activity against moulds (including Aspergillus species), dema-tiaceous fungi and Zygomycetes.^ 17 ^
✜ Pharmacokinetics Food intake and gastric pH have no effect on absorption (Tables 1 and 2).^18,19,24^ FCZ has low protein binding and, unlike other azoles, crosses the blood-brain barrier in the absence of inflammation. It exhibits excellent penetration into most tissues, including the CNS, aqueous humour and lungs.^ 18 ^ Excretion is >90% renal (~75% is excreted unchanged in urine), with a small amount excreted in faeces.^ 26 ^
✜ Adverse effects Overall, FCZ has the most favourable safety profile of all the azoles. The principal AEs are gastrointestinal signs, particularly inappetence, but also vomiting and diarrhoea. Hepatotoxicity, resulting in elevations of serum alanine aminotransferase (ALT), occurs less commonly than with ITZ.
✜ Clinical use FCZ is the first-line treatment for cryptococcosis and fungal urinary tract infections caused by yeasts or thermally dimorphic fungi. Renal function should be assessed before use (eg, urine specific gravity, serum urea, creatinine and symmetric dimethyl arginine). Dose reductions and extended dosing intervals are necessary if renal function is impaired. Owing to its low efficacy against many moulds, FCZ should not be used for the treatment of mould infections unless specifically indicated by the results of antifungal susceptibility testing.
ITZ is the most common antifungal drug used in feline practice and has replaced KCZ for the treatment of IFIs because of its higher potency, broader spectrum of activity and improved safety profile.
✜ Activity ITZ has activity
ITZ has no or limited activity against cryptic Aspergillus section Fumigati species that cause sino-orbital aspergillosis (SOA) in cats, Fusarium species and Zygomycetes (eg, Mucor and Rhizopus species).^ 27 ^
✜ Pharmacokinetics Oral bioavailability and absorption is variable (Table 2) as ITZ is highly lipophilic and almost insoluble in water. The oral solution contains ITZ complexed with hydroxypropyl-β-cyclodextrin to maintain solubility and has three times higher bioavail-ability than proprietary capsules.^ 20 ^ Proprietary capsules contain small sugar spheres coated with a spray-dried dispersion of ITZ, a technique that improves oral bioavailability of drugs with low water solubility.^ 28 ^ Generic formulations are now available, at significantly lower cost. Blood concentrations of ITZ in dogs receiving proprietary vs generic capsules have been found to be similar.^ 29 ^ The same study also evaluated cats, but there were too few for meaningful comparison. Capsules should be given with food (Table 1) since gastric acid secretion increases drug solubility and absorption. Concurrent administration of gastric acid suppressants (eg, proton pump inhibitors) may decrease absorption. Compounded solutions and capsules should not be used -these are not prepared using an appropriate excipient or spray-dried dispersion and are, therefore, not absorbed.^ 22 ^
Owing to high protein binding, ITZ concentrations in CSF, ocular fluid and urine are negligible, but in the presence of inflammation ITZ can penetrate the blood-brain barrier. ITZ is distributed extensively in tissues throughout the body where active drug concentrations accumulate and persist far longer than in plasma, especially in skin and claws.^ 30 ^ ITZ is metabolised in the liver. Excretion is primarily in bile; <1% of active drug is excreted in urine.
✜ Adverse effects AEs are generally dose-related and are relatively common, but resolve after temporary drug discontinuation. Treatment can usually be resumed at a lower dose. Therapeutic drug monitoring (see earlier) would be advantageous to determine if plasma concentrations are adequate. If not available, use of a less hepatotoxic azole with activity against the fungal agent is indicated (eg, FCZ or PCZ; Table 1). Gastrointestinal signs, especially anorexia and vomiting, and hepatotoxicity (anorexia, vomiting, jaundice) are most common (Table 1).^ 21 ^
✜ Clinical use Baseline measurement of serum ALT should be performed and ITZ should not be prescribed if there is severe concurrent liver disease. See ‘Management of invasive fungal and fungal-like infections in cats’ for more information regarding the use of ITZ for specific infections.
✜ Activity PCZ has potent activity against yeasts, thermally dimorphic fungi and moulds, including dermatophytes, Fusarium species and dematiaceous fungi.^ 31 ^ It is fungi-cidal against Cryptococcus species (at high concentrations), Aspergillus species and some Zygomycetes, including Mucor species.^27,32^
✜ Pharmacokinetics This second-generation triazole is structurally similar to ITZ and thus almost insoluble in water. Pharmacokinetic data in cats are available for the proprietary oral suspension (40 mg/ml; Table 2) and IV preparation.^ 22 ^ A delayed-release tablet (100 mg) has high bioavailability in humans and dogs, but cannot be divided, crushed or chewed, and so smaller doses cannot be given. A proprietary delayed-release oral suspension (Noxafil PowderMix 300 mg, 30 mg/ml; Merck) is available for use in children. The powder mix must be prepared and administered using special notched syringes to prevent aggregation of the suspension and ensure delivery of the correct dose. Once mixed, the suspension can only be used once. A pharmacokinetic study is needed to determine whether this formulation would be useful in cats.
Although absorption of PCZ is not affected by gastric acid, administration with food increases bioavailability of the oral adult suspension (up to four-fold in humans). PCZ is highly protein-bound and is widely distributed into tissues, especially the liver, kidneys, lungs and myocardium.^3,33^ Similar to ITZ, concentrations in CSF are low or negligible, but PCZ can penetrate the blood-brain barrier in the presence of inflammation. In humans, ~65% is excreted unchanged in faeces, while ~15% is metabolised in the liver by uridine diphosphate glucuronidation, with inactive metabolites being eliminated in faeces and urine.^ 34 ^ Only trace amounts of PCZ are excreted unchanged in urine.^ 35 ^ Cats are deficient in some uridine diphosphate glucuronosyltransferases compared with humans and dogs,^ 36 ^ which may contribute to the relatively longer half-life of PCZ in cats (Table 2). PCZ inhibits cytochrome P450 enzymes to a lesser extent than ITZ.^ 37 ^
✜ Adverse effects Hepatotoxicity and drug-drug interactions can occur, but are considerably less common than experienced with ITZ treatment.
✜ Clinical use PCZ should be reserved for fungal infections that are not susceptible to ITZ or FCZ, or where ITZ is not tolerated due to hepatotoxicity. Owing to the low bioavail-ability and variable absorption of PCZ, therapeutic drug monitoring should be performed to check that plasma drug concentrations are reaching target levels. Generally, PCZ is well tolerated in cats for periods of many months.^6,11,38^
✜ Activity VCZ is fungistatic against yeasts and thermally dimorphic fungi, and fungici-dal against moulds including some Aspergillus species, Scedosporium species and Fusarium species. It has no activity against Sporothrix species and Zygomycetes.
✜ Pharmacokinetics This second-generation triazole is structurally similar to FCZ. In cats, pharmacokinetic data are available for the IV solution, oral suspension and tablet formula-tions.^ 10 ^ Oral bioavailability of the tablet is high but the presence of food in the stomach reduces and delays absorption (Table 1).^25,39^ The oral suspension is unsuitable for use in cats as it causes profuse ptyalism after administration, which may prevent adequate drug absorption.^ 10 ^
Similar to FCZ, VCZ has very good penetration into tissues, including CSF, ocular fluid and lungs.^ 40 ^ In contrast to FCZ, elimination of VCZ is predominantly by metabolism rather than by renal clearance of unchanged drug. Saturation of metabolic clearance results in non-linear dose-dependent pharmacokinetic parameters, such that a greater than proportional increase in exposure is seen with increasing VCZ dose. Metabolites, which are mostly inactive, are excreted in urine (~80% in dogs and humans) and to a lesser extent in faeces (~20%).
In humans, the major metabolic pathways include fluoropyrimidine N-oxidation by hepatic cytochrome P450 enzymes, followed by glucuronide conjugation, as well as fluoropyrimidine hydroxylation and methyl hydroxylation.^ 25 ^ Multiple oral doses of VCZ in rats and dogs result in lower systemic drug concentrations due to autoinduction of VCZ metabolism, with liver cytochrome P450 enzymes leading to increased drug clear-ance.^ 25 ^ By contrast, repeated oral doses of VCZ in cats (12.5 mg q48h after a loading dose of 25 mg) resulted in increasing plasma drug concentrations over time; even by day 14, when dosing was stopped, steady state had not been reached.^ 10 ^
✜ Adverse effects VCZ has been associated with severe AEs in cats, including anorexia, ataxia, pelvic limb paresis, photophobia and blindness.^11,41,42^
✜ Clinical use Further pharmacokinetic studies are required to determine the best dosing regimen for cats and to confirm that AEs are dose-related. Owing to the high frequency of AEs, VCZ is not recommended for first- or second-line therapy in cats. Its use should be based on antifungal susceptibility data and therapeutic drug monitoring should be performed to track plasma drug concentrations.
✜ Activity ISA has broader antifungal activity compared with other triazoles, by virtue of a side-arm of the ISA molecule that orients it to engage the triazole ring to the binding pocket of CYP51.^ 43 ^ It has broad-spectrum and predominantly fungistatic activity against yeasts, thermally dimorphic fungi, moulds and some Zygomycetes, including Mucor species. The MIC90 (concentration that inhibits growth of 90% of isolates) against Cryptococcus neoformans and Cryptococcus gattii species complex isolates is similar to that of other triazole antifungals.^ 44 ^
Similar to VCZ, ISA has high MICs ( activity) against clinical isolates of cryptic Aspergillus species that cause SOA in cats.^ 45 ^
✜ Pharmacokinetics This second-generation triazole is available in IV and oral formulations that contain the soluble prodrug isavu-conazonium; unlike VCZ and ICZ, ISA does not require a potentially nephrotoxic cyclodextrin vehicle.^ 46 ^ After administration, the prodrug is cleaved by plasma esterases to form ISA and a prodrug cleavage product BAL8728. ISA undergoes oxidative metabolism in the liver by cytochrome P450 enzymes and, in humans, <1% of the dose is excreted in urine. Thus, while not preferred for fungal urinary tract infections, it can be safely administered without dose adjustment to patients with renal impairment.^ 8 ^
In contrast to VCZ, ISA has linear dose-dependent pharmacokinetics, and improved safety and tolerability.
✜ Adverse effects Vomiting was observed 6 and 8 h after dosing in two of four cats administered a single 100 mg ISA capsule.^ 23 ^ In humans, AEs are infrequent and usually limited to mild nausea, vomiting and diarrhoea; also, hepatotoxicity and drug-drug interactions are uncommon compared with other triazole antifungals.^ 47 ^
✜ Clinical use Only single-dose IV and oral pharmacokinetic data (Tables 1 and 2) are available for cats.^ 23 ^ Reports of use of ISA in the field are not yet available. Based on in vitro data and human studies, ISA would not be indicated for empirical therapy of SOA in cats.
✜ Activity Terbinafine (TRB) is highly fungi-cidal against dermatophytes, moulds, thermally dimorphic fungi and some yeasts.
✜ Mechanism of action A synthetic allyl-amine, TRB inhibits squalene epoxidase, blocking conversion of squalene to lanosterol (Figure 1). This results in depletion of ergosterol and accumulation of toxic intra-cellular squalene, which damages fungal cell membranes.
✜ Mechanism of resistance Point mutations in the fungal squalene epoxidase gene (SQLE) confer resistance.^ 48 ^
✜ Pharmacokinetics TRB is available as IV and oral (tablet) formulations. It is highly protein-bound and lipophilic, has moderate oral bioavailability in cats (31%) and is rapidly absorbed.^ 49 ^ The drug distributes extensively to peripheral body fluids and tissues, especially fat, stratum corneum, hair follicles, sebum and claws. TRB is extensively metabolised by cyto-chrome P450 enzymes in the liver before elimination in faeces and urine. No active drug is excreted in urine.^ 50 ^
✜ Adverse effects Oral TRB is mostly well tolerated in cats; however, it can cause anorexia, vomiting and/or diarrhoea in some.^51-53^ Administration with food may prevent vomiting.^ 52 ^ Hepatotoxicity, characterised by elevations in liver enzymes and anorexia/vomiting, usually resolves with drug discontinuation. In humans, TRB occasionally causes severe irreversible acute hepatotoxicity.^ 54 ^ In cats, oral TRB can cause intense facial pruritus with excoriations from self-trauma (Figure 2), which may be followed by urticaria and a maculopapular skin eruption on the face, thorax and abdomen.^ 53 ^
✜ Clinical use In cats, TRB is most commonly used as monotherapy for dermatophytosis. When TRB is combined with an azole, there is a synergistic effect in vitro against some fungi, such that the MIC of the agents in combination is lower than for either agent alone, indicating the potential for lower drug doses to be used, with lower risks of toxicity.
A synergistic effect has been reported in vitro for TRB combined with ITZ or FCZ against isolates of dermatophytes, Tricho -phyton species, Aspergillus species, dematia-ceous fungi, C neoformans/C gattii species complexes and Pythium species.^48,55,56^ Synergism of VCZ and TRB combination therapy was shown in vivo for Aspergillus species in an invertebrate model of infection.^ 57 ^ A randomised controlled clinical trial found that combination therapy with ITZ and TRB was superior to use of either drug alone for the treatment of dermatophytosis in humans.^ 58 ^ Patients treated with TRB plus VCZ for Lomentospora (formerly Scedo sporium) prolifi-cans had better survival rates than those treated with other antifungal regimens.^ 59 ^
Representative drugs in this class include caspofungin acetate, micafungin, anidulafun-gin and rezafungin. Caspofungin is the only echinocandin for which pharmacokinetic data are available in cats.^ 60 ^
✜ Activity Echinocandins are fungicidal against Candida species, and fungistatic against Aspergillus, Penicillium and Purpureocillium species.^ 61 ^ They have some activity against Sporothrix, Bipolaris, Scedosporium, Pseud al -lescheria and Fonsecaea species.^ 62 ^

Echinocandins show no activity against fungal genera that have p-(1,6)-D-glucan as the most abundant cell wall component, including Cryptococcus, Mucorales and Tricho -sporon species.^61,63^ Activity is limited against Histo plasma, Blastomyces and Coccidioides species.
✜ Mechanism of action Drugs in this class are semisynthetic cyclic lipopeptides derived from echinocandin metabolites of moulds such as pneumocandin A0 and B0. They have a hexacyclic peptide core that is anchored to the fungal cell membrane by a hydrophobic fatty acid side chain.^
64
^ Echinocandins inhibit p-(1,3)-D-glucan synthase, a transmembrane glycosyltransferase in the fungal cell wall with at least two subunits, Fks1p1 and Rho1p. Enzyme activity is inhibited by non-competitive binding to the Fks1p1 subunit encoded by three related FKS genes (FKS1, FKS2 and FKS3). This results in inhibition of fungal cell growth, or loss of the osmotic integrity of the fungal cell wall and subsequent cell lysis (Figure 1).
✜ Mechanism of resistance Mutations in the FKS1 gene are among the factors conferring resistance.
✜ Pharmacokinetics Owing to poor oral bio-availability, only IV formulations are available.
✜ Pharmacokinetics Protein binding has not been assessed in cats, but (as for other mammals) is likely to be very high, allowing for good distribution of caspofungin into the lungs, liver and spleen.^ 65 ^ However, this level of protein binding, together with the drug’s high molecular weight, limits CNS and ocular penetration. Caspofungin is rapidly distributed to tissues after administration, but elimination is slow due to delayed release from tissues. Elimination routes are primarily hepatic and renal, with excretion of predominantly inactive metabolites and only small proportions of parent drug.^ 66 ^ Echinocandins are metabolised in the liver to inactive metabolites by non-enzymatic hydrolysis and N-acetylation.^ 67 ^
✜ Adverse effects AEs are uncommon due to the targeted activity of caspofungin against the fungal cell wall. Drug-drug interactions and toxicity are minimal compared with triazole antifungals. AEs most commonly associated with altered drug metabolism in sick human patients include elevations in liver enzymes, fever, nausea or vomiting, and infusion reactions such as rash, pruritis and anaphylaxis.^ 68 ^ A pharmacokinetic study, in which caspofun-gin was administered to six cats for 7 days, reported transient fever in one cat and transient diarrhoea in another as the only AEs.^ 60 ^
✜ Clinical use Caspofungin has been used to treat SOA in cats.
✜ First-line therapy TRB and ITZ combination therapy (see ‘case notes’ later).
✜ Treatment considerations Prospective randomised controlled trials are not available to guide treatment. Successfully treated cases have mostly received antifungal monotherapy (ITZ or TRB) alone or combined with surgical excision.^69-71^ The authors suggest administering TRB and ITZ in combination until 3 months past clinical resolution (Table 1).^72,73^ Intra-abdominal dermatophytic pseudomyce-toma should be resected, or debrided if full resection is not possible. Surgical excision of cutaneous nodules can be performed with wide margins if only one to a few such nodules are present.^ 71 ^ Surgical excision of dermatophytic pseudomycetoma is not recommended as sole therapy as recurrence is likely.^69-71 ,74-78^
✜ Prognosis Recurrent disease - whether involving relapse of original lesion(s) or reinfection - is common.
✜ First-line therapy FCZ (rhinitis); AMB and FC combination therapy (CNS/severe disease).
✜ Treatment considerations FCZ is the drug of choice for rhinitis and localised cutaneous disease. The endpoint of therapy is complete resolution of clinical signs and a serum latex cryptococcal antigen test (LCAT) titre of zero. Treatment periods of 2-12 months (median 4 months) are typical for C gattii (molecular type VGI) and C neoformans (VNI).^ 79 ^ For other species in the C neoformans/C gattii species complexes, or when LCAT titres remain high, longer treatment times may be needed and clinical recrudescence may occur.
If LCAT titres remain persistently elevated or fail to decline after 6 weeks of FCZ administration, switching to an alternative azole is indicated. ITZ is the next best choice, with a median treatment duration of 9 months.^ 79 ^ Isolates exhibiting high MICs for FCZ or ITZ may be susceptible to other azoles, such as KCZ,^ 80 ^ VCZ or PCZ.
Surgical excision or debridement of large cryptococcal granulomas may be considered for some patients, if deemed safe, since antifun-gal drugs penetrate poorly into hypoperfused tissues. However, there is currently no evidence that this approach decreases treatment times or is superior to medical therapy alone.
First-line therapy for CNS (including ocular) or disseminated disease is AMB combined with FC, or with FCZ if FC is unavailable.^79,81^ Critical patients with increased intracranial pressure may benefit from adjunctive judicious glucocorticoid (eg, dexamethasone) administration for the first 48-72 h of treat-ment.^ 82 ^ Clinical improvement is expected within 7-10 days, and the AMB/FC combination is typically continued for 4-8 weeks. Treatment is then de-escalated to oral FCZ (or ITZ), which is administered until LCAT titres decrease to zero; prolonged treatment (≥2 years) may be needed to achieve this. Periodic follow-up monitoring of LCAT titres (eg, every 3-6 months) should be performed to detect recrudescence, which can occur in up to 30% of cases.^ 79 ^
In some cases of cryptococcosis (any form), LCAT titres may not decline to zero despite switching azoles (eg, from FCZ to ITZ). In such refractory cases, combination therapy with an azole and TRB (Table 1), or TRB monotherapy, may be effective.^ 83 ^ If LCAT titres decline four- to five-fold and then remain static despite these approaches, treatment response may be deemed adequate if accompanied by clinical remission.^ 83 ^ Serum LCAT titres should be monitored and treatment re-instituted if they increase.
✜ Prognosis The prognosis is generally favourable, with successful outcomes reported in 60-75% of cases.^79,84^ CNS involvement is a negative prognostic factor.^79,85^ However, in one study, cats with CNS, disseminated or refractory disease treated with AMB-containing protocols had a similar rate of successful outcomes as cats with less severe disease treated with azole monotherapy.^ 79 ^ Further research is needed to determine whether outcome is associated with the infecting species within the C neoformans/C gattii species complexes.
✜ First-line therapy Combination PCZ and TRB; or caspofungin, PCZ and TRB.
✜ Second-line therapy Combination therapy with AMB, PCZ and TRB.
✜ Treatment considerations ITZ is not suitable for empirical treatment of SOA since most infections are caused by Aspergillus species with intrinsic resistance.^37,86,87^ Typically, isolates have high MICs for ITZ and VCZ, low MICs for PCZ, caspofungin and TRB, and variable MICs for AMB.^45,87^ Thus, PCZ and TRB are used for first-line therapy. AMB is not recommended for empirical therapy and its use should be guided by the results of anti-fungal susceptibility testing.
The reliably low MICs of caspofungin against infecting species make it attractive for adjunctive initial treatment, although this requires hospitalisation for daily IV infusions for 7-21 days. Placement of a central venous line enables daily infusions to be given for up to 3 weeks.^11,38^ Pharmacokinetic modeling data support the use of caspofungin in SOA, with effective target drug concentrations likely achieved when used as recommended in Table 1.^ 60 ^ The combination of caspofungin and PCZ has also been used to treat invasive aspergillosis in humans caused by the same agents as feline SOA (eg, Aspergillus udagawae).^ 88 ^ Some feline cases have shown improvement with the addition of orbital exenteration or debridement of orbital granu-lomas to the treatment regimen;^42,89^ overall, however, surgery has not been shown to improve outcomes compared with antifungal therapy alone.^ 11 ^
In cats treated successfully, oral antifungal therapy was administered for 4-8 months in some cases,^ 90 ^ and up to several years in others based on evidence of residual disease on serial head CT scans.^11,38^ In the light of the treatment duration, the authors strongly recommend therapeutic drug monitoring to enable individualised dose adjustments if PCZ levels are not optimal. In addition, monitoring for hepatotoxicity is recommended.
✜ Prognosis The prognosis is generally guarded to poor, but good outcomes can be achieved in some cases using combination antifungal therapy guided by antifungal susceptibility testing.
✜ First-line therapy ITZ; or ITZ and AMB combination therapy for severe disease/CNS involvement.
✜ Second-line therapy PCZ or FCZ, if ITZ is not tolerated. Therapeutic drug monitoring is recommended for azoles if the clinical response is inadequate.
✜ Prognosis Overall, the prognosis is good, but is poorer for cats with CNS involve-ment.^91,92^ Treatment periods of approximately 6 months are required to achieve cure, and relapses can occur.^ 91 ^
✜ First-line therapy FCZ or ITZ monothera-py; FCZ or ITZ combined with AMB for severe disease.
✜ Second-line therapy PCZ has a low MIC for Coccidioides species and could be considered for second-line therapy.^ 93 ^
✜ Treatment considerations FCZ is most frequently prescribed because of its low cost, excellent oral bioavailability and tissue penetration, and low AE profile. However, high MICs for FCZ are not uncommon and, in humans with coccidioidomycosis, high dose rates are often required to achieve a clinical response. ITZ should be considered if the response to FCZ is suboptimal. For severe or refractory disease, combination therapy with the addition of AMB is recommended.^ 93 ^
✜ Prognosis The prognosis is generally favourable, with most cats responding to FCZ or combination therapy, although long treatment periods (>1 year) are typical and recurrence occurs in ~25% of cats; these cases will usually respond to resumption of therapy.^ 94 ^
✜ First-line therapy ITZ monotherapy; or ITZ and AMB combination therapy for severe disease.^ 95 ^
✜ Treatment considerations FCZ can be used as an alternative to ITZ, although FCZ resistance has been reported in cats and humans on chronic therapy and MICs for FCZ are more variable than for ITZ.^96-98^ A third of Histoplasma species isolates from human patients that relapse on FCZ therapy have reduced FCZ susceptibility, whereas MICs for ISA usually remain low.^ 99 ^ PCZ and ISA both have good activity and either may be considered for cats that fail to respond to ITZ or FCZ therapy.
Median treatment durations of 6 months are typical, but longer treatment is often necessary to effect a cure, especially for disseminated disease.^100-102^ Low anti-inflammatory doses of short-acting glucocorticoids have been advocated by some as additional therapy for cats with severely compromised airways.^ 95 ^ In the same study, the use of glucocorticoids was also a negative prognostic indicator, as was the use of supplemental oxygen, which may have reflected the selection bias for severe disease.^ 95 ^
✜ Prognosis Overall, the prognosis is good, with approximately two-thirds of cases achieving sustained clinical remission. However, recrudescence is reported in up to 40% of cases that achieve clinical remission, after median disease-free intervals of approximately 1-3 years.^101,102^ Poor prognostic factors include severe respiratory, neurological, hepatic or haematological involvement.^ 95 ^
✜ First-line therapy ITZ for Sporothrix brasi-liensis infection; ITZ and TRB combination therapy for S schenckii infection (see ‘Treatment considerations’).
✜ Second-line therapy Refractory cases in Brazil, where S brasiliensis is endemic, have been cured with ITZ in combination with potassium iodide capsules.^103,104^ See ‘Treatment considerations’ for further discussion and alternative options for second-line therapy.
✜ Treatment considerations S brasiliensis usually has a low MIC for ITZ. Median duration of therapy is 4-6 months. Treatment should continue for 1 month after clinical resolution, extending to 2 months for lesions involving the nasal region or cases where there is respiratory involvement.^ 105 ^
AEs have occurred in 50% of cats receiving the second-line therapy of ITZ plus potassium iodide, including anorexia, lethargy, vomiting, diarrhoea and hepatotoxicity.^103,106^ The anti-fungal mechanism of action of potassium iodide - which is used for first-line therapy of cutaneous sporotrichosis in humans^ 107 ^ - is incompletely understood.
Although clinical data are lacking, based on antifungal susceptibility testing and clinical efficacy in other hosts, the authors suggest the following alternative options for second-line therapy of S brasiliensis
In contrast to S brasiliensis, isolates of S schenckii from Malaysia and Thailand have much higher MICs for azoles, TRB and AMB, and may be refractory to treatment.^110,113^ The combination of ITZ and TRB is synergistic, resulting in low MICs; hence it is recommended by the authors for first-line therapy.^ 111 ^

✜ First-line therapy Surgical excision and ITZ, PCZ or TRB monotherapy; or surgical excision and ITZ or PCZ combination therapy with TRB.
✜ Treatment considerations Solitary lesions (eg, cutaneous, pulmonary) have been cured after radical surgical excision (wide margins) combined with oral ITZ or PCZ for several months to prevent recurrence at the surgical site.^31,114^ For solitary digital lesions, amputation of the affected phalanx is suggested, together with a similar medical approach.
Given that >100 species of pigmented fungi cause phaeohyphomycosis in humans and animals, antifungal susceptibility testing is recommended to guide therapy. If this is not possible, molecular identification of the infecting agent can usually be achieved by panfun-gal PCR of formalin-fixed biopsy tissues (see Part 1^ 1 ^). This can help inform the best empirical therapy since antifungal susceptibility profiles vary between different fungal genera and species. Of 126 clinical and environmental isolates of Exophiala species, PCZ (0.063 μg/ml) and ITZ (0.125 μg/ml) had much lower MIC90 values than VCZ (1 μg/ml) and ISA (2 μg/ml), suggesting better suitability of ITZ or PCZ for first-line therapy.^ 115 ^ In one case series, recurrence of cutaneous lesions caused by Alternaria species was common after surgical excision and ITZ therapy for 2-4 months.^ 116 ^
✜ First-line therapy PCZ or ISA monotherapy; or PCZ or ISA and AMB combination therapy.
✜ Treatment considerations Few cases have been reported and treated in cats.^32,117-120^ In human infections, surgical resection or debride-ment of affected tissues, when possible, is recommended, before dissemination.^ 121 ^ PCZ and ISA have good activity but ITZ is not suitable for empirical therapy because of low activity against Zygomycetes in the subphylum Mucoromycotina, which are most common in cats. AMB also has excellent activity and could be used together with PCZ or ISA (Table 1). One case of nasal mucormycosis in a cat resolved after treatment with PCZ for 5 months.^ 32 ^
✜ First-line therapy Resection of accessible lesions with wide margins (eg, amputation of an affected digit, resection of a small intestinal mass with 5 cm margins) plus adjunctive medical
✜ Treatment considerations Theoretically, since oomycetes lack ergosterol in their cell membranes, azoles and TRB should not have efficacy. However, while MICs for each antifungal are high, they are low for these drugs in combination for Pythium insidiosum.^122,123^ Also, dogs with gastrointestinal pythiosis have been cured using a combination of ITZ, TRB and a low anti-inflammatory dose of prednisolone.^ 124 ^ Another study has shown that some antibiotics, including minocycline, doxycycline and clarithromycin, have much lower MICs than antifungal drugs for P insidiosum isolates.^ 125 ^ Synergistic effects of tetracyclines and macrolides also occur.^ 125 ^ Clinical cure or a significantly decreased organism burden was observed after azithromycin and minocycline treatment in rabbits with experimentally induced cutaneous pythiosis.^ 126 ^ The authors have successfully treated several cases of subcutaneous pythiosis in cats using a combination of surgical resection/debridement with either PCZ, TRB, azithromycin and doxycycline, or PCZ, minocycline and clarithromycin.

In dogs, an agricultural fungicide, mefen-oxam, used to control oomycete infections in plants through inhibition of RNA polymeras-es, has had some success as part of the treatment protocol for gastrointestinal and cutaneous pythiosis; it was variously used after surgical resection in combination with ITZ and TRB, or with minocycline.^127-129^ Mefenoxam has potent in vitro activity against Pythium and Lagenidium species, but pharma-cokinetic studies have not been performed in animals; nor has it been used in cats.^ 130 ^
Adjuvant immunotherapy using commercially formulated P insidiosum antigens administered as a subcutaneous injection has been successful in treating pythiosis in horses and humans. Its efficacy in dogs appears to be poor.^ 131 ^ One cat with extensive ulcerated cutaneous pythiosis showed dramatic shrinkage of the lesion after immunotherapy, but the cat was euthanased because of severe pleural effusion of unknown cause.^ 132 ^ Immuno-therapy has been used as adjunctive therapy in four successfully treated cats with pythiosis in Hong Kong, and was well tolerated (authors’ unpublished observations); however, its contribution to treatment success in these cases is unknown.