In Vitro Activities of Terbinafine against Aspergillus Species in Comparison with Those of Itraconazole and Amphotericin B

Abstract

Compared with the in vitro activities of itraconazole (geometric mean MIC [GM], 0.56 μg/ml) and amphotericin B (GM, 0.66 μg/ml), the in vitro activity of terbinafine was inferior against Aspergillus fumigatus (GM, 19.03 μg/ml) (P < 0.05) and superior against A. flavus (GM, 0.10 μg/ml), A. terreus (GM, 0.16 μg/ml), and A. niger (GM, 0.19 μg/ml). Clinical correlation is required, as trailing endpoints are problematic.

Aspergillus fumigatus remains the most common airborne fungal pathogen worldwide (8). Not surprisingly, the ubiquity of this organism coupled with the substantial increase in immunocompromised patients has resulted in a dramatic rise in the incidence of invasive aspergillosis over the last 2 decades. In a recent review of more than 1,000 autopsies in a university hospital, invasive aspergillosis was diagnosed as the cause of death in 4% of cases (17). Estimates of prevalence vary according to the underlying condition, ranging from 1% in patients with systemic lupus erythematosus to as high as 25 to 40% in patients with chronic granulomatous disease (4).

Terbinafine (TERB) is an allylamine antifungal agent available in both topical and oral preparations. It is widely utilized in the treatment of dermatophyte infections. It is also reported to have good activity in vitro against Cryptococcus (12), some species of Candida (12), Penicillium marneffei (9), Aspergillus (5, 14–16), and other filamentous fungi (5). In vivo activity against Pneumocystis carinii has been described (2), and clinical activity against pulmonary aspergillosis has been reported (13).

In this study, we determined the in vitro antifungal activities of TERB against six different species of Aspergillus, with a disproportionate number resistant to itraconazole (ITZ). We also examined the fungicidal activity of TERB. Furthermore, we directly compared both inhibitory and fungicidal results with those of ITZ and amphotericin B.

A total of 100 clinical Aspergillus isolates comprising 60 A. fumigatus isolates, 13 isolates each of A. flavus and A. niger, 12 isolates of A. terreus, and 1 isolate each of A. nidulans and A. oryzae were used. Ten isolates were resistant to ITZ, 9 being A. fumigatus. Isolates for which the MICs of ITZ and amphotericin B were known were included to ensure quality control (3, 6). All cultures were cultivated from frozen stock on Sabouraud dextrose agar (Oxoid, Basingstoke, United Kingdom) for 3 to 4 days at 30°C.

TERB (Novartis Pharma, Basel, Switzerland) was provided as the standard hydrochloride salt, and ITZ (Janssen Pharmaceuticals, Beerse, Belgium) was provided in pure powder form. Both were supplied by their respective manufacturers. Amphotericin B with desoxycholate (AMB) was obtained from E. R. Squibb & Sons Limited, Middlesex, United Kingdom.

Stock solutions (3,200 μg/ml) of all drugs were prepared using appropriate solvents—TERB (dimethyl sulfoxide containing 5% Tween 80), ITZ (1:1 acetone and 0.2 M HCl), and AMB (sterile distilled water)—and adjusted for potency when necessary. Each stock was then dispensed into aliquots and stored in glass vials, protected from the light, at −20°C until required.

TERB was tested using two different media: RPMI 1640 medium (Sigma, Poole, United Kingdom) containing 2% glucose, buffered with morpholinepropanesulfonic acid (MOPS; Sigma), and adjusted to pH 7.0 (RPMI) and Casitone (pH 7.0) (Difco, Detroit, Mich.) supplemented with 2% glucose (CAS). ITZ and AMB were also tested using both media for comparison purposes. MICs were determined using a microtiter method, with final drug concentrations ranging from 0.03 to 16 μg/ml for all antifungal agents. The method using RPMI has previously been used to substantiate ITZ resistance in A. fumigatus (3). All drugs were tested concurrently, and the entire study, including reproducibility testing, was performed using the same batch of each medium.

The inoculum was prepared by suspending conidia in phosphate-buffered saline containing 0.05% Tween 80, counted using a hemocytometer, and then diluted to yield a final concentration of 5 × 10⁵ conidia/ml. A positive control (drug-free) was included for each isolate. Microtiter trays were incubated in moist chambers at 37°C for 48 h. The MICs were read visually and were defined using a no-growth endpoint.

Minimum fungicidal concentrations (MFCs) were also determined for all drugs. For each isolate, 100 μl was removed from all wells without visible growth. Each aliquot was spot inoculated onto a blood agar plate, and the liquid was allowed to soak into the agar. When dry, the plate was streaked to separate any conidia and to remove them from the drug source. The plates were incubated at 37°C for 48 h. The MFC was defined as the lowest drug concentration that allowed the growth of five colonies or fewer (≥99.99% kill).

Twenty percent of the isolates (20 of 100) were randomly selected and retested against each drug to assess the reproducibility of the method.

The differences between species were analyzed using the Kruskal-Wallis test for multiple comparisons (SPSS for Windows).

Summaries of the in vitro susceptibility values of the 100 isolates tested are shown in Table 1. For TERB, it was found that RPMI gave a narrower range of MICs, particularly for A. fumigatus, than did CAS. Therefore, the CAS values for TERB have been used for all subsequent analyses. However, in fitting with a validated method (3), RPMI values have been used for ITZ and AMB. Isolates of A. fumigatus (n = 60) were found to be significantly less susceptible (P < 0.05) to TERB than isolates of A. flavus (n = 13), A. niger (n = 13), and A. terreus (n = 12). Perhaps surprisingly, given the mode of action of TERB, the TERB geometric mean (GM) MIC (in micrograms per milliliter) for the ITZ-resistant A. fumigatus isolates (14.8, n = 9) appeared slightly lower than that of the ITZ-susceptible A. fumigatus isolates (19.9, n = 51).

TABLE 1.

In vitro susceptibilities of 100 Aspergillus isolates to TERB, ITZ, and AMB

Species (no. of isolates)	Antifungal agent	Medium	MIC (μg/ml)a
			GMb	Range	50%	90%
A. fumigatus (60)	TERB	RPMI	30.55	4–>16	>16	>16
	TERB	CAS	19.03	2–>16	>16	>16
	ITZ	RPMI	0.65	0.125–>16	0.25	>16
	ITZ	CAS	0.63	0.06–>16	0.25	>16
	AMB	RPMI	0.40	0.125–0.5	0.5	0.5
	AMB	CAS	0.94	0.5–2	1	1
A. flavus (13)	TERB	RPMI	0.14	≤0.03–>16	0.06	0.5
	TERB	CAS	0.10	≤0.03–>16	0.06	0.25
	ITZ	RPMI	0.43	0.25–>16	0.25	0.5
	ITZ	CAS	0.50	0.125–>16	0.25	4
	AMB	RPMI	2.35	0.5–>16	2	>16
	AMB	CAS	2.11	1–>16	2	4
A. niger (13)	TERB	RPMI	0.29	0.125–1	0.25	1
	TERB	CAS	0.19	0.06–1	0.125	0.5
	ITZ	RPMI	0.85	0.5–4	1	1
	ITZ	CAS	6.46	1–>16	8	>16
	AMB	RPMI	0.34	0.125–1	0.5	1
	AMB	CAS	0.53	0.25–1	0.5	1
A. terreus (12)	TERB	RPMI	0.24	0.125–1	0.25	0.25
	TERB	CAS	0.16	0.125–1	0.125	0.25
	ITZ	RPMI	0.31	0.25–0.5	0.25	0.5
	ITZ	CAS	0.94	0.25–8	0.5	4
	AMB	RPMI	3.56	2–8	4	4
	AMB	CAS	4.76	2–16	4	16
A. nidulans (1)	TERB	RPMI	—	0.06	—	—
	TERB	CAS	—	0.03	—	—
	ITZ	RPMI	—	0.06	—	—
	ITZ	CAS	—	0.06	—	—
	AMB	RPMI	—	1	—	—
	AMB	CAS	—	4	—	—
A. oryzae (1)	TERB	RPMI	—	0.03	—	—
	TERB	CAS	—	0.03	—	—
	ITZ	RPMI	—	0.125	—	—
	ITZ	CAS	—	0.125	—	—
	AMB	RPMI	—	1	—	—
	AMB	CAS	—	1	—	—
All isolates (100)	TERB	RPMI	4.07	≤0.03–>16	>16	>16
	TERB	CAS	2.60	≤0.03–>16	4	>16
	ITZ	RPMI	0.56	0.06–>16	0.5	4
	ITZ	CAS	0.83	0.06–>16	0.25	>16
	AMB	RPMI	0.66	0.125–>16	0.5	4
	AMB	CAS	1.20	0.25–>16	1	4

^a50% and 90%, MICs at which 50 and 90% of isolates are inhibited, respectively. —, not calculated. 

^bIn calculation of the GM values, MICs of 0.03 or >16 μg/ml were classed as 0.03, or 32 μg/ml, respectively. 

For all isolates, TERB, ITZ, and AMB were fungicidal in 35, 18, and 71% of instances, respectively, in the drug range tested. TERB, however, was fungicidal for 87.5% of non-A. fumigatus aspergilli. MFC GMs and ranges (in micrograms per milliliter) for all 100 isolates were 5.91 and ≤0.03 to >16 for TERB, 22.62 and 0.06 to >16 for ITZ, and 8.00 and 0.5 to >16 for AMB. Species differences were apparent. For more than half (58%) of all isolates of A. terreus tested, the TERB MFC was over 10-fold higher than the corresponding MIC, and they therefore were defined as tolerant. The only other isolate for which this was the case was the isolate of A. oryzae.

Reproducibility studies showed that for TERB, only 10 of 20 (50%) isolates retested produced a result within 1 twofold dilution. Most isolates (95%) were within a 3-twofold-dilution range. However, all discrepant results were due to trailing endpoints. In contrast, 19 of 20 (95%) and 18 of 20 (90%) isolates retested for ITZ and AMB, respectively, gave a result within 1 twofold dilution.

Our data show that TERB has potent in vitro activity against A. flavus, A. niger, and A. terreus at lower concentrations than both ITZ and AMB. Interestingly, however, limited activity was observed against A. fumigatus. Other recent studies have shown good activity (MICs ≤ 1 μg/ml) against Aspergillus species, including A. fumigatus (5, 15, 16). To date, there is no standardized method available for testing TERB, since it is currently not included within the NCCLS M-38P reference method for conidium-forming filamentous fungi (11). These recent studies have all used RPMI 1640 as the test medium, with various inoculum concentrations (2 × 10³ to 5 × 10³ to ∼1 × 10⁶ CFU/ml), temperature (30 to 35°C), and duration of incubation (48 to 72 h). In addition, different endpoints have been used (80 to 100% inhibition). Our method using RPMI 1640 medium has previously been validated in vivo for ITZ against A. fumigatus (3). In our hands, RPMI 1640 medium proved to have trailing endpoints when TERB was being tested; therefore, we felt that it was an inappropriate choice of medium for our validated method, which uses a no-growth endpoint. In the present study, CAS medium also suffered, to a lesser extent, from trailing endpoints, and more so with A. fumigatus than any of the other species. If trace growth had been ignored when reading the MICs, for 60% of the isolates of A. fumigatus the MIC would have been reduced by at least 2 dilutions and the GM MIC would have become 5.1 μg/ml. This may suggest that a no-growth endpoint is too stringent when testing TERB. Interestingly, if the ITZ-susceptible and -resistant A. fumigatus isolates were reanalyzed ignoring trace growth, the situation would be reversed, with the GM MIC of TERB for the ITZ-resistant isolates (10.1 μg/ml, n = 9) being somewhat higher than that for the ITZ-susceptible isolates (4.5 μg/ml, n = 51). Another vital factor may be inoculum size. Subsequent studies in our laboratory have shown that MICs may be reduced by up to 3 dilutions when a final inoculum of 5 × 10⁴ CFU/ml is used. Trailing endpoints are also diminished (J. Mosquera, unpublished data). However, the TERB MICs for isolates of A. fumigatus are still consistently higher than the TERB MICs for other species. Extensive in vivo work will be required to establish which method gives the best correlation with clinical outcome.

We found that TERB showed primary fungicidal activity (MFC within 2 dilutions of the MIC) against A. niger and A. flavus but not usually against A. terreus. Other studies have reported the fungicidal nature of TERB against Aspergillus, but A. terreus was not tested (14, 16). This may present additional problems clinically in terms of duration of treatment and potential relapse.

Our finding that TERB appears less active against A. fumigatus than other pathogenic species of Aspergillus may present two problems in clinical terms. Firstly, A. fumigatus accounts for around 90% of human infections (8), secondly, treatment is very often commenced without positive cultures. In addition, our MICs for A. fumigatus were higher than levels achievable in blood, since peak concentrations in plasma of ∼1.7 μg/ml are typically seen within 1.5 h of an oral 250-mg dose (7). No intravenous preparation is available. Combination therapy may, therefore, hold advantages in this situation. In vitro synergy between TERB and either AMB or ITZ has been reported for Candida (1; A. W. Fothergill, I. Leitner, J. G. Meingassner, N. S. Ryder, and M. G. Rinaldi, Abstr. 36th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E53, 1996), Cryptococcus (Fothergill et al., 36th ICAAC), and Scedosporium prolificans (10). Synergistic in vitro activity with ITZ or AMB against Aspergillus has also been described (L. Rodero, R. Vitale, S. Cordoba, E. H. Reinoso, and G. Davel, Abstr. 38th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E-63b, 1998; N. S. Ryder and I. Leitner, Abstr. 36th Intersci. Conf. Antimicrob. Agents Chemother., abstr. E54, 1996), but this has yet to be confirmed in vivo (W. R. Kirkpatrick, R. K. Mcatee, N. S. Ryder, and T. F. Patterson, Abstr. 38th Intersci. Conf. Antimicrob. Agents Chemother., abstr. J-77, 1998).

Further in vivo and clinical studies, with both single- and combination-drug therapy, are necessary to fully determine the potential of this new application of TERB.