Antifungal Drugs

Classification

Polyenes: amphotericin B, nystatin(topical/oral non-absorbed). Azoles: fluconazole, itraconazole, voriconazole, posaconazole(triazoles) · ketoconazole, clotrimazole, miconazole(imidazoles, mostly topical now). Echinocandins: caspofungin, micafungin, anidulafungin. Antimetabolite: flucytosine. Squalene epoxidase inhibitor: terbinafine. Other: griseofulvin.

Organizing Concept: ERGOSTEROL

Fungal membrane analogue of human cholesterol — present in fungi, absent in humans → basis of selective toxicity. Adverse effects trace to IMPERFECT selectivity(partial cross-reactivity with cholesterol or human enzymes resembling fungal ergosterol-synthesis enzymes).

Mechanism

Polyenes(amphotericin B): bind ergosterol DIRECTLY → pore formation → membrane leakage → FUNGICIDAL. Selectivity = ergosterol-vs-cholesterol binding preference, REAL but IMPERFECT(residual cholesterol affinity, esp renal tubular cells) → nephrotoxicity(below). Liposomal formulations = ↓off-target renal binding, preserve efficacy.

Azoles: inhibit fungal CYP450 14α-demethylase(lanosterol demethylase) — blocks lanosterol→ergosterol conversion → depletes ergosterol+toxic intermediate accumulation. FUNGISTATIC generally. Fungal-vs-human CYP450 selectivity real but imperfect, WORST for KETOCONAZOLE(least selective) → endocrine SE(below) → largely topical-only today, displaced systemically by more selective triazoles.

Echinocandins: inhibit β-(1,3)-D-glucan synthase → blocks β-glucan(fungal cell wall component, ABSENT in humans entirely — no cell wall) → CLEANEST selectivity of any antifungal class(parallel to β-lactam absolute selectivity) → most favourable SE profile among systemics.

Flucytosine: fungal-selective PRODRUG — fungal permease uptake + fungal cytosine deaminase(humans LACK this) → converts to 5-FU → disrupts fungal RNA/DNA synthesis. Elegant selectivity(humans can’t activate at all) but resistance emerges FAST with monotherapy → always combination(classically +amphotericin B, cryptococcal meningitis).

Terbinafine: inhibits squalene epoxidase(EARLIER step than azoles) → depletes ergosterol + toxic SQUALENE accumulation(extra fungicidal mechanism beyond depletion alone). Dermatophyte infections(onychomycosis — excellent keratin penetration/accumulation).

Griseofulvin: deposits in NEW keratin(hair/skin/nails) + disrupts fungal microtubule/mitotic spindle → inhibits cell division. DISTINCT mechanism from ergosterol-pathway drugs. Dermatophyte infections of hair/nails(keratin-affinity delivers drug to infection site).

Adverse Effects

Amphotericin B(“amphoterrible”): infusion reactions(“shake and bake,” cytokine release, premedicate) + NEPHROTOXICITY(dose-limiting — imperfect cholesterol-binding selectivity, direct tubular damage+vasoconstriction; lipid formulations ↓but don’t eliminate). Tolerated given broad reliable fungicidal activity in severe systemic infection with few alternatives.

Azoles: hepatotoxicity(class-wide). Endocrine effects WORST with KETOCONAZOLE(inhibits human steroidogenic CYP450 — testosterone+cortisol synthesis) → gynaecomastia, ↓libido, adrenal insufficiency — DIRECT consequence of poor fungal-vs-human CYP450 selectivity(triazoles much less prominent). Significant CYP450 interactions(both substrate+inhibitor of CYP3A4) — statins, warfarin, tacrolimus/cyclosporine. Voriconazole: distinctive transient VISUAL disturbances(altered colour perception, blurred vision) with initial dosing, self-limiting.

Echinocandins: minimal SE(absolute target selectivity) — mild infusion reactions+hepatotoxicity, NO nephrotoxicity/endocrine burden.

Flucytosine: bone marrow suppression — small amount converts to 5-FU via GUT BACTERIAL cytosine deaminase(not fungal-selective route) → 5-FU itself NOT fungal-selective → same cytotoxic marrow effects as chemotherapeutic 5-FU. Shows how elegant target selectivity can still leak host toxicity via indirect route.

Griseofulvin: hepatotoxicity + INDUCES hepatic CYP450(↓OCP efficacy — same interaction category as rifampicin) + disulfiram-like reaction+alcohol.

Clinical Relevance

Ergosterol-centred framework explains BOTH efficacy and toxicity from same idea: drugs targeting ergosterol/its pathway work because fungi uniquely depend on this sterol, but the CLOSER a drug’s target resembles a human counterpart(cholesterol for amphotericin B; human CYP450 esp for ketoconazole), the MORE imperfect selectivity and MORE prominent host adverse effects. Echinocandins(clean profile) + flucytosine(elegant prodrug selectivity) = the 2 genuine exceptions where fungal target has NO meaningful human analogue.

Classification (by Mechanism)

Polyenes: amphotericin B, nystatin (topical/oral non-absorbed only)
Azoles: fluconazole, itraconazole, voriconazole, posaconazole (triazoles); ketoconazole, clotrimazole, miconazole (imidazoles, mostly topical now)
Echinocandins: caspofungin, micafungin, anidulafungin
Antimetabolite: flucytosine
Squalene epoxidase inhibitor: terbinafine
Other: griseofulvin

The Central Organizing Concept: Ergosterol

Nearly every systemic antifungal drug class targets ergosterol, the fungal cell membrane’s analogue of cholesterol in mammalian membranes — this single molecule (present in fungi, absent/replaced by cholesterol in humans) is both the basis of antifungal selective toxicity and the reason so many antifungal adverse effects stem from imperfect selectivity (partial cross-reactivity with the structurally similar mammalian sterol, cholesterol, or with mammalian steroid synthesis enzymes that resemble the fungal ergosterol-synthesis enzymes targeted).

Mechanism of Action

Polyenes (amphotericin B): bind ergosterol directly in the fungal membrane, forming pores that disrupt membrane integrity and cause leakage of intracellular contents — fungicidal. Because amphotericin B binds ergosterol directly (not an enzyme in its synthesis), its selectivity depends entirely on ergosterol-versus-cholesterol binding preference, which is real but imperfect — amphotericin B has some residual affinity for cholesterol in human cell membranes, particularly renal tubular cell membranes, the direct mechanistic basis for its dose-limiting nephrotoxicity (below), and the reason newer lipid-formulation preparations (liposomal amphotericin B) were developed specifically to reduce this off-target renal binding while preserving antifungal efficacy.

Azoles: inhibit fungal cytochrome P450 14α-demethylase (also called lanosterol demethylase), an enzyme required for converting lanosterol to ergosterol in the fungal sterol synthesis pathway — blocking this step depletes ergosterol and causes accumulation of toxic sterol intermediates, fungistatic (generally) rather than fungicidal. Fungal CYP450 selectivity over human CYP450 enzymes is again real but imperfect, particularly for ketoconazole, which shows the least selectivity among azoles for the fungal versus human enzyme — the direct mechanistic basis for ketoconazole’s endocrine adverse effects (below) and for its largely topical-only role today, systemic use having been substantially displaced by more selective triazoles.

Echinocandins: inhibit β-(1,3)-D-glucan synthase, blocking synthesis of β-glucan, a structural fungal cell wall component entirely absent in human cells (human cells have no cell wall at all, echinocandins’ target simply doesn’t exist to cross-react with) — this gives echinocandins the cleanest selective-toxicity profile of any antifungal class, mechanistically parallel to β-lactams’ absolute selectivity for a bacterial-cell-wall target humans entirely lack, and correspondingly the most favourable overall adverse-effect profile among systemic antifungals.

Flucytosine: a fungal-selective prodrug — taken up by a fungal-specific permease and converted intracellularly by fungal cytosine deaminase (an enzyme humans lack) to 5-fluorouracil, which then disrupts fungal RNA/DNA synthesis — genuinely elegant selective toxicity (humans cannot activate the prodrug at all), but used only in combination (classically with amphotericin B for cryptococcal meningitis) since resistance emerges rapidly with monotherapy.

Terbinafine: inhibits squalene epoxidase, an earlier step in the ergosterol synthesis pathway than the azoles’ target — blocking this step both depletes ergosterol and causes toxic squalene accumulation within the fungal cell, contributing an additional fungicidal mechanism beyond simple ergosterol depletion. Used predominantly for dermatophyte infections (particularly onychomycosis, given excellent keratin penetration and accumulation in nail/skin).

Griseofulvin: deposits in newly forming keratin (hair, skin, nails) and disrupts fungal microtubule function/mitotic spindle formation, inhibiting fungal cell division — a genuinely distinct mechanism from the ergosterol-pathway drugs above, used specifically for dermatophyte infections of hair/nails where its keratin-affinity delivers the drug directly to the site of infection.

Adverse Effects

Amphotericin B: infusion-related reactions (fever, chills, rigors — sometimes called “shake and bake,” related to cytokine release, generally managed with premedication rather than requiring discontinuation) and nephrotoxicity (the dose-limiting toxicity, from the imperfect cholesterol-binding selectivity above — causes both direct tubular damage and renal vasoconstriction, and is a major driver of the shift toward lipid formulations, which substantially reduce but don’t entirely eliminate this risk) — the mnemonic “amphoterrible” reflects this genuinely demanding toxicity profile, still tolerated given amphotericin B’s broad, reliable fungicidal activity against severe/life-threatening systemic fungal infections where few alternatives are as broadly effective.

Azoles: hepatotoxicity (class-wide, monitored during therapy) and endocrine effects — most pronounced with ketoconazole, which at higher doses meaningfully inhibits human steroidogenic CYP450 enzymes (particularly those involved in testosterone and cortisol synthesis), causing gynaecomastia, decreased libido, and adrenal insufficiency — a direct consequence of ketoconazole’s comparatively poor fungal-versus-human CYP450 selectivity, in contrast to the far more selective triazoles (fluconazole, itraconazole, voriconazole), which show these endocrine effects much less prominently. Significant CYP450 drug-interaction potential (azoles are both substrates and inhibitors of human CYP3A4) is a shared, clinically important liability across the class, given how many other drugs (statins, warfarin, certain immunosuppressants like tacrolimus/cyclosporine) share this metabolic pathway. Voriconazole carries a distinctive additional adverse effect: transient visual disturbances (altered colour perception, blurred vision) with initial dosing, generally self-limiting.

Echinocandins: comparatively minimal adverse effects (per the absolute target-selectivity above) — mild infusion reactions and hepatotoxicity are the main concerns, without the nephrotoxicity or endocrine burden of the other systemic classes.

Flucytosine: bone marrow suppression — worth noting that flucytosine’s selectivity depends entirely on human cells lacking cytosine deaminase, but a small amount of conversion to 5-FU can occur via gut bacterial cytosine deaminase, and the resulting 5-FU is not itself selective for fungal cells, producing the same cytotoxic bone marrow effects 5-FU causes when used directly as a chemotherapeutic agent — a specific, mechanistically clear explanation for how a drug with elegant target-based selective toxicity can still cause meaningful host toxicity through an indirect route.

Griseofulvin: hepatotoxicity, and a specific, examined interaction — it induces hepatic CYP450 enzymes, reducing efficacy of co-administered oral contraceptives (the same category of interaction already seen with rifampicin), and disulfiram-like reaction with alcohol.

Clinical Relevance

The ergosterol-centred organizing framework explains both antifungal efficacy and much of antifungal toxicity from the same underlying idea: drugs targeting ergosterol or its synthesis pathway are effective specifically because fungi (unlike bacteria, and unlike human cells) depend on this one distinctive membrane sterol, but the closer a given drug’s target structurally resembles a human counterpart (cholesterol for amphotericin B, human CYP450 enzymes for azoles, particularly ketoconazole), the more imperfect its selective toxicity and the more prominent its host adverse effects — echinocandins’ clean safety profile and flucytosine’s elegant prodrug selectivity both make sense as the two genuine exceptions where the fungal target has no meaningful human analogue at all.

Ergosterol synthesis pathway (lanosterol → ergosterol) with azole and terbinafine blocking points marked at their respective enzymes.
Selective-toxicity spectrum from echinocandins (no human target at all) to ketoconazole (most human CYP450 cross-reactivity), explaining the adverse-effect gradient.