Alarming India‐wide phenomenon of antifungal resistance in dermatophytes: A multicentre study

An alarming increase in recalcitrant dermatophytosis has been witnessed in India over the past decade. Drug resistance may play a major role in this scenario.


| INTRODUC TI ON
Superficial dermatophytosis is an infection of the skin or nail that affects 20%-25% of the world's population. 1 Patients suffer from severe itching, sleep disturbance, stigmatisation and depression. 2,3 Over the past few years, healthcare professionals in India have witnessed a significant increase in the number of patients presenting with dermatophytosis, as well as in the number of difficult-to-treat and recalcitrant cases. 4 The latter has been attributed to multiple causes including an abuse of irrational fixed drug combination (FDC) creams containing potent steroids, an altered immune response of the host, and microbiological (in vitro) resistance of the causative fungi. 5,6 In India, the most common fungi responsible for dermatophytosis belong to the Trichophyton (T.) mentagrophytes/interdigitale species complex. These are characterised by a particular genotype called T mentagrophytes internal transcribed spacer (ITS) Type VIII, which is endemic to India and is the cause of an epidemic of tinea cruris and tinea corporis resistant to the widely used drug terbinafine (TRB). [7][8][9] In vitro resistance to TRB has been reported in 17% and 32% of such isolates. 10,11 Several single point mutations in the fungal squalene epoxidase gene (SQLE), which encodes the target for TRB, have been recorded in T rubrum and T mentagrophytes/interdigitale. 11,12 These mutations lead to substitutions at one of the four amino acid positions Leu393, Phe397, Phe415 and His440 and were associated with TRB resistance. Recently, more substitutions were discovered, including in other countries such as Denmark. 13,14 The objectives of our study were (a) to determine the species pattern and the prevalence of dermatophytes resistant to TRB, itraconazole (ITC) and voriconazole (VRC) on a large panel of patients from different sites in India, (b) to compare the results between these sites, and (c) to determine whether all cases resistant to TRB can be attributed to mutations in the fungal SQLE gene.

| Patients
A total of 402 patients with clinically suspected dermatophytosis from eight different locations in India participated in this study. Figure 1 shows the number and distribution of participants in each centre. From each patient, one skin scraping from the periphery of the infected lesion was collected between 2017 and 2019. This includes 199 patients whose species detection was performed in a previously published multicentric study. 7

| Fungal identification
Each sample was transferred onto two culture media: Sabouraud's 4% dextrose agar (SDA) containing the antibiotic chloramphenicol, and SDA containing chloramphenicol as well as cycloheximide in order to eliminate moulds. Cultures were incubated at 28°C for at least 4 weeks and checked visually every 3 days. Criteria for the detection of dermatophytes were based on colour, texture and growth speed. Cultures fulfilling the listed criteria were subject to light microscopy using lactophenol cotton blue dye. In order to detect dermatophyte DNA, samples underwent a polymerase chain reaction using an enzyme-linked immunosorbent assay (PCR-ELISA). DNA was extracted using QIAamp ® DNA Mini QIAcube Kit (Qiagen). The following probes for PCR were used: T rubrum, T interdigitale/T mentagrophytes, Microsporum canis and T benhamiae. Identification was further confirmed by Sanger sequencing of the ITS region (mainly partial 18.S rRNA, ITS1, 5.8S rRNA, ITS2 and partial 28.S rRNA) of the rDNA and the translation elongation factor (TEF)-1α gene. 15 For ITS amplification, the primers V9G (5′-TTACGTCCCTGCCCTTTGTA-3′) and LSU266 (5′-GCATTCCCAAACAACTCGACTC-3′) were used. 16 The TEF-1α gene was amplified using the primers EF-DermF (5′-CACATTAACTTGGTCGTTATCG-3′) and EF-DermR (5′-CATCCTTGGAGATACCAGC-3′). 15 We compared the sequences with the GenBank database (https://blast.ncbi.nlm.nih.gov/Blast.cgi) using nucleotide BLAST searches.   20 The map in Figure 1 was obtained by using the tmap package 21 within the software R and edited in Microsoft

| Statistical analyses and map generation
Word.

| Terbinafine resistance rates in India
A total of 297 dermatophyte isolates were tested for growth ability on SDA containing 0.2 µg/mL TRB. Terbinafine MICs were also determined by broth microdilution method (Table 1). All isolates with F I G U R E 2 Counts and rates of detected dermatophytes in different regions of India. Detection was established by cultivation, light microscopy, PCR-ELISA, and/or sequencing of the ITS region of the rDNA or the TEF-1α gene. 'Trichophyton mentagrophytes/interdigitale' declare strains which showed a positive PCR-ELISA result, but did not yield an ITS or TEF1-α sequence due to insufficient amounts of DNA

| Amino acid substitutions in squalene epoxidase
In order to discover reasons for in vitro TRB resistance in these

| Triazole susceptibility
We investigated antifungal susceptibility to ITC and VRC in 297 Trichophyton isolates. Table 2 provides descriptive data on the drug's MICs. Significant regional differences were not observed (Table S2).  Isolates with low sensitivity to VRC had a low sensitivity to ITC. Isolates resistant to both ITC and VRC were more prevalent in TRB-sensitive isolates (42%) than in TRB-resistant isolates (18%), assuming a MIC cut-off of 0.25 µg/mL for azoles (Table 3). Conversely, isolates resistant to TRB were more common in the azole-sensitive group (77%) than in the azole-resistant group (51%). Fisher's exact test proved these differences to be extremely statistically significant (two-tailed P < .0001).
Surprisingly, triazole susceptibility seemed to depend on the SQLE genotype. Figure 6 illustrates that the MIC means (derived from distribution curves) were higher in isolates showing an  Note: Triazole resistance was more frequent in TRB-sensitive isolates, while TRB resistance was more frequent in triazole-sensitive isolates (P < .0001). The reason for previous divergent findings regarding the commonest species in dermatophytosis (T mentagrophytes vs T interdigitale) might lie in the process of species determination itself and differences in the used nomenclature. 28,29 It is of utmost importance for researchers to adjust submitted sequence names according to the latest taxonomic agreements, which is unfortunately not yet common practice. Failing this, confusion in species identification will continue unabated.

| Terbinafine resistance rates
The results obtained by the broth microdilution method for the identification of TRB resistance of Trichophyton strains were consistent with those obtained by the agar dilution method. The agar dilution method may serve as a practical alternative to the broth microdilution method as it is faster and much more cost-effective.

| Associations of in vitro terbinafine resistance and amino acid substitutions in the squalene epoxidase
All isolates of T mentagrophytes ITS Type VIII with Phe397Leu or Leu393Phe substitution in the SQLE protein were shown to be resistant in agar cultures and had high MICs of ≥ 8 µg/mL. The value of 8 µg/mL was recently proposed as the epidemiological cut-off value (ECV) for TRB resistance. 31 Isolates with a non-mutated SQLE or carrying a single Ala448Thr substitution continuously featured low MICs of ≤0.125 µg/mL and did not grow on SDA containing 0.2 µg/ mL. The SQLE of strains with an MIC between 0.25 and 1 µg/mL revealed other missense mutations. Of important note, a strain isolated from a patient for which TRB treatment was unsuccessful revealed a MIC of 1.0 µg/mL and a Gln408Leu substitution in SQLE. 14 The same value was obtained for a strain analysed in the present study. Therefore, strains with MIC ≥ 1 µg/mL can be clinically resistant to TRB.
The MICs of sensitive strains of T mentagrophytes isolates (0.125 µg/mL) were found to be higher than that of T interdigitale isolates from tinea unguium and tinea pedis in Europe. 13 These isolates had MICs of 0.008-0.03 µg/mL. The MIC of 0.125 µg/ mL in the present study corresponds to the minimal value given by other authors for sensitive isolates in India. 9,11

| Various prevalence of mutations in the SQLE gene
The high prevalence of isolates with a Phe397Leu mutation is strik-

| Trends in triazole sensitivity among isolates
Triazole MICs in the present study from isolates originating in Northern India were equal to, lower, [9][10][11]22 and some higher 23,24 when compared with other studies.
A strong, positive, linear correlation between MICs of ITC and VRC was found. Isolates with reduced susceptibility to ITC tended to be less susceptible to VRC, and vice versa. These findings may implicate that triazole resistance in dermatophytes is not drug specific but is mediated by a shared mechanism of resistance. To the best of Triazole resistance was more frequent in TRB-sensitive isolates, while TRB resistance was more frequent in triazole-sensitive isolates. This may be because patients were treated with either TRB or azole, but not with both classes of drugs. As data on pre-treatment were only available for a fraction of patients in this study, we can neither prove nor reject this hypothesis. Investigations correlating the exposure of patients to antifungals with in vitro resistance are necessary.
We observed that reduced susceptibility to triazoles in  The novel squalene epoxidase substitution Ala448Thr was associated with higher MICs of ITR and VRC and could have predictive value of triazole insensitivity. TRB-resistant isolates were only encountered if a substitution was located between amino acids

| CON CLUS IONS
Leu393 and Ser443 while all strains which lacked a substitution in the squalene epoxidase were sensitive to TRB in the present study.
Therefore, checking specifically the C-terminus of SQLE allows the identification of TRB resistance in T rubrum or T mentagrophytes/interdigitale. The accuracy and feasibility of treatment decisions based on SQLE mutations should be evaluated in the future-especially since TRB resistance is marching across India and beyond. [43][44][45]

ACK N OWLED G M ENTS
We thank Marina Fratti for excellent technical assistance. This research received no specific grant from any funding agency in the public, commercial or not-for-profit-sectors.

CO N FLI C T O F I NTE R E S T
There are no conflicts of interest to disclose.