PHARMACOLOGY AND THERAPEUTICS
SIMPLIFIED EXPERIMENTAL HUMAN DERMATOPHYTOSIS MODEL RAZA ALY, PH.D., HOWARD I. MAIBACH, M.D., IRWIN HO, M.D.. AND BEATRICE B. ABRAMS, P H . D .
Abstract The authors have improved and simplified previous methods for producing localized mycotic skin infections in an attempt to compare topical antifungal agents for their bioequivalency. Healthy human volunteers who had negative results for commercial, purified Trichophyton antigen (Trichophytin) were inoculated with Trichophyton mentagrophytes on two sites on each forearm in a randomized study designed to compare the antifungal activities of two ciclopirox olamine formulations. The lesions, easily induced by the authors' method, were localized and did not spread under the occlusive dressings. Infections established at the four sites on 26 subjects were treated twice daily for 14 days with the two active drug formulations and their vehicles. There were no significant differences in culture-documented cure rates or alleviation of clinical signs and symptoms between ciclopirox olamine lotion and cream. Each drug was significantly better than its vehicle. The authors' method seems to be effective and suitable for therapeutic studies.
insufficient and delay the developmental process. Effective human screens should be of short duration and limited variability; they also should be amenable to use in relatively small numbers of subjects and predictive of results encountered in highly variable, costly, and timeconsuming comparative clinical trials. Several investigators have attempted to study the pathogenesis, immunology, and treatment of dermatophytosis by using a human experimental model to produce infections. Reinhart et al. devised a reliable and reproducible method for producing infection on the glabrous skin in humans;' however, their occlusive procedures were cumbersome and uncomfortable for the participants. We have improved and simplified the previous methods to induce experimental dermatophytosis in humans. This article describes the utility of an experimental human dermatophytosis model in comparing the therapeutic (antifungal) activity of two formulations of a standard antifungal agent. Materials and Methods
The development of new topical antifungal agents can be divided into three phases: (1) in vitro or animal (preclinical) studies to identify effective therapeutic agents and potentially useful vehicles; (2) initial human testing to evaluate the safety and efficacy of the drug and, if necessary, to refine the delivery system; and (3) largescale clinical trials to demonstrate the safety and efficacy of the final formulation in a specific indication. In ideal situations, in vitro and animal models permit expeditious, effective development; unfortunately, satisfactory models of fungal infections are lacking. Because of the difficulties in extrapolating from preclinical to clinical results, human screening procedures often are
The preparation of conidia and the procedure of inoculation were performed according to the methods of Reinhardt et al.' A 2-week-old culture of the granular variety of Trichophyton mentagrophytes subcultured from ATCC 18748 was harvested in sterile saline with 0.01% Tween 40, homogenized by shaking with glass beads, and filtered through a column of glass wool. A hemocytometer was used to quantify the individual conidia found in the top portion of a suspension that had been mixed and had settled. To determine conidial viability, the original conidial suspension was diluted serially in saline and plated on potato dextrose agar. More than 90% of the conidia were found to be viable. The conidial preparation was further diluted to 300conidia/0.1 mL The subjects were at least 18 years of age, selected without regard to race or sex, and in good general health as confirmed by a brief history and physical examination. All subjects signed a statement of informed consent. Before entering the study, women of childbearing age were tested for pregnancy. Those who were pregnant or nursing were excluded from participation. The rest of these women used adequate contraceptive measures during the trial. We also excluded subjects requiring or using any other medication that would affect the evaluation of the test materials and subjects who had taken any antimicrobial drug 7 days before enrollment.
Erom the Department of Dermatology, University of Galifornia Medical Genter, San Erancisco, Galifornia, and the Department of Glinical Research, Hoechst-Roussel Pharmaceuticals, Inc., Somerville, New Jersey. Supported by a grant from Hoechst-Roussel Pharmaceuticals, Inc., Somerville, New Jersey. Address for correspondence: Raza Aly, Ph.D., Department of Dermatology, Box 0536, University of Galifornia at San Erancisco Medical Genter, San Erancisco, GA 94143-0536. 122
Experimental Human Dermatophytosis Aly et al.
Sensitivity to dermatophyte antigens was another criterion for selection. Skin tests were performed by intradermal injection of 0.1 ml of the commercial, purified Trichophyton antigen, Trichophytin. The test site was examined immediately after injection and after 72 hours. Subjects who had any skin reaction to the antigen at the 72-hour evaluation were classified as "positive reactors" and were excluded from the study. Those who had no skin reaction were classified as "negative reactors" and remained in the study. For inoculation, two sites were located on the volar aspect of each arm, at least 2.5 cm apart (periphery to periphery). A layer of transparent dressing (Tegaderm, 3M Corp., St. Paul, MN) was applied firmly to each arm (Fig. 1). This dressing had two rectangles cut into it, and the openings were placed over the two test sites. Approximately 300 conidia were delivered to each skin site (a total of four sites) exposed through the openings in the transparent dressing, and a sterile 2 x 2.5 cm gauze pad was placed over each inoculated site. The pad then was saturated with sterile distilled water and occluded with another, uncut, piece of transparent dressing (Fig. 2). Without removal of the dressing, additional water was added to the gauze pads after 48 hours, with the use of a 22-gauge needle. After 4 days, the dressings were removed. Three days after removal of the occlusive dressings, application of the test materials began. The products were assigned randomly to test sites; however, to avoid accidental cross-contamination, sites treated with the active drug were not on the same arm as sites treated with vehicle. Test materials were applied to each site twice daily for 14 consecutive days. After application, each site was covered with a semi-occlusive dressing (gauze covering secured with Micropore tape; 3M Corporation). The test sites were evaluated clinically on days 1 (before application of test materials), 4, 9, and 14 of treatment by rat-
Figure 2. Establishment of infections at multiple sites on the arms of volunteers. Inoculated sites were covered with a sterile gauze pad that had been saturated with sterile water and then covered with an occlusive dressing. The dressings were removed after 4 days. ing hyperpigmentation, scaling, erythema, edema, vesicles, and spreading of the lesion. The following arbitrary scale was used for judgment: 0 = none; 1 = mild; and 2 = notable effect. Scrapings were obtained from each site for mycologic evaluations on day 1 of treatment (before application of test materials) and 1 day after the last day of treatment (day 15). Before test sites were scraped on day 15, the lesions were cleansed gently with soap and water. The scrapings were inoculated on Sabouraud culture medium and incubated at room temperature for at least 5 days before they were read. Plates were scored as showing positive or negative results for growth. We assessed the safety of the materials and procedures by observing and interviewing the subjects throughout the trial. Sign tests^ evaluated the significance of the differences between cream and lotion, cream and cream vehicle, and lotion and lotion vehicle in culture results and sign severity scores (the clinical evaluation). A test was deemed significant if the P value was less than 0.05. A two-tailed test was used to evaluate the differences between cream and lotion formulations, and a one-tailed test was used for the other comparisons. The ability^ to detect a 20% difference in negative cultures from the reference formulation was calculated for the cream versus lotion comparison for which the difference was not statistically significant. This analysis involved the normal approximation procedure with correction for continuity, and a = 0.05.'' Ciclopirox olamine cream was the reference formulation. The exact confidence intervals for culture results on day 16 were calculated according to the method of Wilkes.^
RESULTS
:'
Thirty subjects who met the criteria participated in the infection study. Four did not complete the investigation. Of these, one subject removed the occlusive device prematurely, and the remaining three did not have lesions on both arms at the test sites. Only subjects with lesions on both arms were allowed to continue in the study. Subjects who were not selected were treated
Figure 1. Establishment of infections with a subculture of T. mentagrophytes ATCC 18748 on the arms of volunteers. A transparent dressing (Tegaderm, 3M Gorp., St. Paul, MN) was used to contain the inoculum at the site of application. Each site was inoculated with approximately 300 spores per 0.1 mL. 123
International Journal of Dermatology Vol. 31, No. 2, February 1992
immediately with conventional topical creams. The ages of the 26 subjects who did complete the study, and whose data were used in the statistical analyses, ranged from 22 to 40 years. Twenty-one were women and 5 were men. Twenty-two of the subjects were white, 3 were black, and 1 was Hispanic. All lesions remained localized within the cut-out areas of the transparent dressings (Fig. 3), and, at baseline evaluation, cultures at all four test sites of tbe 26 treated subjects had positive findings. On day 15 of the study (after 14 days of treatment), 62% (16 of 26) of the cream-treated sites were culture negative, compared with 85% (22 of 26) of the lotion-treated sites (Fig. 4). There was no statistically significant difference, however, between culture results for the cream and those for the lotion. In addition, 38% (10 of 26) of the cream vehicle-treated sites and 19% (5 of 26) of the lotion vehicle-treated sites were culture negative. The differences in the numbers of culture-negative sites treated with the cream versus cream vehicle and lotion versus lotion vehicle were significant (P < 0.05). Two statistical procedures were used to test the reliability of tbe assumption that tbe cream and lotion products were "bioequivalent." With the use of the normal approximation procedure, the ability (statistical power) to detect a 20% difference in data for the two formulations with regard to negative cultures was 0.5. Exact 95% confidence intervals are shown in Table 1.
DRUG
VEHICLE
CREAM LOTION Figure 4. Percent of treated sites from which cultured samples were negative for the d e r m a t o p h y t e after 14 days of treatment (for each treatment, n = 26). Sites were cleansed with soap and water before scraping. Sabouraud medium was used for cultures. Plates were scored either positive or negative after a minimum incubation time of 5 days.
Similarities in the therapeutic efficacy of ciclopirox olamine cream and lotion were apparent from frequency distributions of the major sign severity scores (erythema, edema, vesicles, scaling); those for the visit on day 14 are shown in Figure 5. After 2 weeks of treatment, the signs at most sites treated either with ciclopirox olamine cream or lotion were scored as 1 (mild) or 0 (none). The signs at vehicle-treated sites generally were scored as 1 (mild) or 2 (notable). Again, there were no apparent differences in the distributions for ciclopirox olamine cream and ciclopirox olamine lotion; however, tbe distributions for the active drugs differed from those of their respective vehicles. No statistical analyses were done on these distributions. Subjects whose infections had not cleared by the end of the study were treated with ciclopirox olamine cream 1% until the lesions resolved.
DISCUSSION
We have provided an improved and simplified method for inducing experimental ringworm infections in humans. Our model provides a reliable method for investigators who are interested in the biologic characteristics of human dermatophyte infections. Besides being a useful tool for evaluating antifungal agents in a standardized population, the model also can support studies of immune responses to dermatophytes and their pathogenesis. The four important factors that must be considered in providing standard infections are concentration of the spore inoculum, hydration of the skin test site, duration of occlusion, and the immune status of the subject. The gauze covering the inoculated site should remain damp throughout the occlusion period. The
Figure 3. Established lesions on the arm of a volunteer, the window made in the occlusive underlying adhesive dressing prevented the spread of the experimentally induced infections beyond the defined areas. .
124
Experimental Human Dermatophytosis Aly et al.
Table 1. Culture Results for Day 15 Formulation Point Estimate Exact 95% Confidence Interval Cream Cream vehicle Lotion Lotion vehicle
0.62 0.38 0.85 0.19
and lotion formulations is evident in the results of clinical evaluations and mycologic tests. The mycologic cure rates for the cream and cream vehicle (62% and 38%, respectively) are similar to the rates obtained with the cream and its vehicle in a parallel group, multicentered, clinical trial,'' which involved tinea corporis and tinea cruris infections of mixed dermatophyte origin. Culture conversions in this large study were observed at day 14 in 76% of the subjects treated with ciclopirox olamine cream and 45% of the subjects treated with the cream vehicle. Our experimental dermatophytosis model, which provides well-localized and controlled infections, should prove useful in the early assessment of promising test formulations and possibly could serve as a model for bioequivalence determinations for topical antifungal drugs.
0.41-0.80 0.20-0.59 0.64-0.96 0.07-0.39
CO Lin COCrm Un Veh CrmVeh
25 ERYTHEMA 20
o
a 10
DRUG NAMES 0 1 2 SIGN SEVERITY SCORE
0
ciclopirox olamine lotion: Loprox lotion ciclopirox olamine cream: Loprox cream
1 2 SIGN SEVERITY SCORE
25 VESICLES
.
SCALING
Acknowledgments: Dr. Ray Chen helped with the statistical analysis. Charlene Bayles and Dr. Stan Miller provided technical assistance. Linda Setescak and Joanne Robinett gave administrative assistance during the study. Dr. Dehra Jan Bibel helped in the preparation of the manuscript.
20
15 a: 10
REFERENCES SIGN SEVERITY SCORE
1.
SIGN SEVERITY SCORE
Figure 5. Number of sites with specified sign severity score after 14 days of treatment (for each treatment, n = 26). Eor each sign: 0 = none; 1 = mild; and 2 = notable. CO Ltn, ciclopirox olamine lotion 1%; CO Crm, ciclopirox olamine cream 1%; Ltn Veh, lotion vehicle; Crm Veh, cream vehicle.
2. 3.
4.
immune status of the experimental subject is important because subjects with positive reactions to Trichophyton antigen in skin tests react differently to fungal infections than those whose results are negative.'' The results of this 2-week assay in a limited number of subjects suggest that the two formulations tested (ciclopirox olamine lotion and cream) are therapeutically equivalent in the treatment of experimentally induced superficial T. mentagrophytes infection. The equivalency of the cream
5. 6.
7.
125
Reinhardt JH, Allen AM, Gunnison D, et al. Experimental human Trichophyton mentagrophytes infections. J Invest Dermatol 1974; 63:419-422. Conover WJ. Practical nonparametric statistics. 2nd ed. New York: John Wiley and Sons, 1980:122. Lehmann EL, D'Ahrera JH. Nonparametric statistical methods based on ranks. San Erancisco: Holden-Day, 1975:156. Hsieh CC. Note on interval estimation of the difference between proportions from correlated series. Stat Med 1985; 4:23-27. Wilkes SS. Mathematical statistics. New York: John Wiley and Sons, 1962:368. Wood SR, Cruckshank CND. The relation between trichophytin sensitivity and fungal infections. Br J Dermatol 1962; 74:329-336. Bogaert H, Cordero C, Ollague W, et al. Multicentre double-blind clinical trials of ciclopirox olamine cream 1% in the treatment of tinea corporis and tinea cruris. J Int Med Res 1986; 14:210-216.
SIMPLIFIED EXPERIMENTAL HUMAN DERMATOPHYTOSIS MODEL RAZA ALY, PH.D., HOWARD I. MAIBACH, M.D., IRWIN HO, M.D.. AND BEATRICE B. ABRAMS, P H . D .
Abstract The authors have improved and simplified previous methods for producing localized mycotic skin infections in an attempt to compare topical antifungal agents for their bioequivalency. Healthy human volunteers who had negative results for commercial, purified Trichophyton antigen (Trichophytin) were inoculated with Trichophyton mentagrophytes on two sites on each forearm in a randomized study designed to compare the antifungal activities of two ciclopirox olamine formulations. The lesions, easily induced by the authors' method, were localized and did not spread under the occlusive dressings. Infections established at the four sites on 26 subjects were treated twice daily for 14 days with the two active drug formulations and their vehicles. There were no significant differences in culture-documented cure rates or alleviation of clinical signs and symptoms between ciclopirox olamine lotion and cream. Each drug was significantly better than its vehicle. The authors' method seems to be effective and suitable for therapeutic studies.
insufficient and delay the developmental process. Effective human screens should be of short duration and limited variability; they also should be amenable to use in relatively small numbers of subjects and predictive of results encountered in highly variable, costly, and timeconsuming comparative clinical trials. Several investigators have attempted to study the pathogenesis, immunology, and treatment of dermatophytosis by using a human experimental model to produce infections. Reinhart et al. devised a reliable and reproducible method for producing infection on the glabrous skin in humans;' however, their occlusive procedures were cumbersome and uncomfortable for the participants. We have improved and simplified the previous methods to induce experimental dermatophytosis in humans. This article describes the utility of an experimental human dermatophytosis model in comparing the therapeutic (antifungal) activity of two formulations of a standard antifungal agent. Materials and Methods
The development of new topical antifungal agents can be divided into three phases: (1) in vitro or animal (preclinical) studies to identify effective therapeutic agents and potentially useful vehicles; (2) initial human testing to evaluate the safety and efficacy of the drug and, if necessary, to refine the delivery system; and (3) largescale clinical trials to demonstrate the safety and efficacy of the final formulation in a specific indication. In ideal situations, in vitro and animal models permit expeditious, effective development; unfortunately, satisfactory models of fungal infections are lacking. Because of the difficulties in extrapolating from preclinical to clinical results, human screening procedures often are
The preparation of conidia and the procedure of inoculation were performed according to the methods of Reinhardt et al.' A 2-week-old culture of the granular variety of Trichophyton mentagrophytes subcultured from ATCC 18748 was harvested in sterile saline with 0.01% Tween 40, homogenized by shaking with glass beads, and filtered through a column of glass wool. A hemocytometer was used to quantify the individual conidia found in the top portion of a suspension that had been mixed and had settled. To determine conidial viability, the original conidial suspension was diluted serially in saline and plated on potato dextrose agar. More than 90% of the conidia were found to be viable. The conidial preparation was further diluted to 300conidia/0.1 mL The subjects were at least 18 years of age, selected without regard to race or sex, and in good general health as confirmed by a brief history and physical examination. All subjects signed a statement of informed consent. Before entering the study, women of childbearing age were tested for pregnancy. Those who were pregnant or nursing were excluded from participation. The rest of these women used adequate contraceptive measures during the trial. We also excluded subjects requiring or using any other medication that would affect the evaluation of the test materials and subjects who had taken any antimicrobial drug 7 days before enrollment.
Erom the Department of Dermatology, University of Galifornia Medical Genter, San Erancisco, Galifornia, and the Department of Glinical Research, Hoechst-Roussel Pharmaceuticals, Inc., Somerville, New Jersey. Supported by a grant from Hoechst-Roussel Pharmaceuticals, Inc., Somerville, New Jersey. Address for correspondence: Raza Aly, Ph.D., Department of Dermatology, Box 0536, University of Galifornia at San Erancisco Medical Genter, San Erancisco, GA 94143-0536. 122
Experimental Human Dermatophytosis Aly et al.
Sensitivity to dermatophyte antigens was another criterion for selection. Skin tests were performed by intradermal injection of 0.1 ml of the commercial, purified Trichophyton antigen, Trichophytin. The test site was examined immediately after injection and after 72 hours. Subjects who had any skin reaction to the antigen at the 72-hour evaluation were classified as "positive reactors" and were excluded from the study. Those who had no skin reaction were classified as "negative reactors" and remained in the study. For inoculation, two sites were located on the volar aspect of each arm, at least 2.5 cm apart (periphery to periphery). A layer of transparent dressing (Tegaderm, 3M Corp., St. Paul, MN) was applied firmly to each arm (Fig. 1). This dressing had two rectangles cut into it, and the openings were placed over the two test sites. Approximately 300 conidia were delivered to each skin site (a total of four sites) exposed through the openings in the transparent dressing, and a sterile 2 x 2.5 cm gauze pad was placed over each inoculated site. The pad then was saturated with sterile distilled water and occluded with another, uncut, piece of transparent dressing (Fig. 2). Without removal of the dressing, additional water was added to the gauze pads after 48 hours, with the use of a 22-gauge needle. After 4 days, the dressings were removed. Three days after removal of the occlusive dressings, application of the test materials began. The products were assigned randomly to test sites; however, to avoid accidental cross-contamination, sites treated with the active drug were not on the same arm as sites treated with vehicle. Test materials were applied to each site twice daily for 14 consecutive days. After application, each site was covered with a semi-occlusive dressing (gauze covering secured with Micropore tape; 3M Corporation). The test sites were evaluated clinically on days 1 (before application of test materials), 4, 9, and 14 of treatment by rat-
Figure 2. Establishment of infections at multiple sites on the arms of volunteers. Inoculated sites were covered with a sterile gauze pad that had been saturated with sterile water and then covered with an occlusive dressing. The dressings were removed after 4 days. ing hyperpigmentation, scaling, erythema, edema, vesicles, and spreading of the lesion. The following arbitrary scale was used for judgment: 0 = none; 1 = mild; and 2 = notable effect. Scrapings were obtained from each site for mycologic evaluations on day 1 of treatment (before application of test materials) and 1 day after the last day of treatment (day 15). Before test sites were scraped on day 15, the lesions were cleansed gently with soap and water. The scrapings were inoculated on Sabouraud culture medium and incubated at room temperature for at least 5 days before they were read. Plates were scored as showing positive or negative results for growth. We assessed the safety of the materials and procedures by observing and interviewing the subjects throughout the trial. Sign tests^ evaluated the significance of the differences between cream and lotion, cream and cream vehicle, and lotion and lotion vehicle in culture results and sign severity scores (the clinical evaluation). A test was deemed significant if the P value was less than 0.05. A two-tailed test was used to evaluate the differences between cream and lotion formulations, and a one-tailed test was used for the other comparisons. The ability^ to detect a 20% difference in negative cultures from the reference formulation was calculated for the cream versus lotion comparison for which the difference was not statistically significant. This analysis involved the normal approximation procedure with correction for continuity, and a = 0.05.'' Ciclopirox olamine cream was the reference formulation. The exact confidence intervals for culture results on day 16 were calculated according to the method of Wilkes.^
RESULTS
:'
Thirty subjects who met the criteria participated in the infection study. Four did not complete the investigation. Of these, one subject removed the occlusive device prematurely, and the remaining three did not have lesions on both arms at the test sites. Only subjects with lesions on both arms were allowed to continue in the study. Subjects who were not selected were treated
Figure 1. Establishment of infections with a subculture of T. mentagrophytes ATCC 18748 on the arms of volunteers. A transparent dressing (Tegaderm, 3M Gorp., St. Paul, MN) was used to contain the inoculum at the site of application. Each site was inoculated with approximately 300 spores per 0.1 mL. 123
International Journal of Dermatology Vol. 31, No. 2, February 1992
immediately with conventional topical creams. The ages of the 26 subjects who did complete the study, and whose data were used in the statistical analyses, ranged from 22 to 40 years. Twenty-one were women and 5 were men. Twenty-two of the subjects were white, 3 were black, and 1 was Hispanic. All lesions remained localized within the cut-out areas of the transparent dressings (Fig. 3), and, at baseline evaluation, cultures at all four test sites of tbe 26 treated subjects had positive findings. On day 15 of the study (after 14 days of treatment), 62% (16 of 26) of the cream-treated sites were culture negative, compared with 85% (22 of 26) of the lotion-treated sites (Fig. 4). There was no statistically significant difference, however, between culture results for the cream and those for the lotion. In addition, 38% (10 of 26) of the cream vehicle-treated sites and 19% (5 of 26) of the lotion vehicle-treated sites were culture negative. The differences in the numbers of culture-negative sites treated with the cream versus cream vehicle and lotion versus lotion vehicle were significant (P < 0.05). Two statistical procedures were used to test the reliability of tbe assumption that tbe cream and lotion products were "bioequivalent." With the use of the normal approximation procedure, the ability (statistical power) to detect a 20% difference in data for the two formulations with regard to negative cultures was 0.5. Exact 95% confidence intervals are shown in Table 1.
DRUG
VEHICLE
CREAM LOTION Figure 4. Percent of treated sites from which cultured samples were negative for the d e r m a t o p h y t e after 14 days of treatment (for each treatment, n = 26). Sites were cleansed with soap and water before scraping. Sabouraud medium was used for cultures. Plates were scored either positive or negative after a minimum incubation time of 5 days.
Similarities in the therapeutic efficacy of ciclopirox olamine cream and lotion were apparent from frequency distributions of the major sign severity scores (erythema, edema, vesicles, scaling); those for the visit on day 14 are shown in Figure 5. After 2 weeks of treatment, the signs at most sites treated either with ciclopirox olamine cream or lotion were scored as 1 (mild) or 0 (none). The signs at vehicle-treated sites generally were scored as 1 (mild) or 2 (notable). Again, there were no apparent differences in the distributions for ciclopirox olamine cream and ciclopirox olamine lotion; however, tbe distributions for the active drugs differed from those of their respective vehicles. No statistical analyses were done on these distributions. Subjects whose infections had not cleared by the end of the study were treated with ciclopirox olamine cream 1% until the lesions resolved.
DISCUSSION
We have provided an improved and simplified method for inducing experimental ringworm infections in humans. Our model provides a reliable method for investigators who are interested in the biologic characteristics of human dermatophyte infections. Besides being a useful tool for evaluating antifungal agents in a standardized population, the model also can support studies of immune responses to dermatophytes and their pathogenesis. The four important factors that must be considered in providing standard infections are concentration of the spore inoculum, hydration of the skin test site, duration of occlusion, and the immune status of the subject. The gauze covering the inoculated site should remain damp throughout the occlusion period. The
Figure 3. Established lesions on the arm of a volunteer, the window made in the occlusive underlying adhesive dressing prevented the spread of the experimentally induced infections beyond the defined areas. .
124
Experimental Human Dermatophytosis Aly et al.
Table 1. Culture Results for Day 15 Formulation Point Estimate Exact 95% Confidence Interval Cream Cream vehicle Lotion Lotion vehicle
0.62 0.38 0.85 0.19
and lotion formulations is evident in the results of clinical evaluations and mycologic tests. The mycologic cure rates for the cream and cream vehicle (62% and 38%, respectively) are similar to the rates obtained with the cream and its vehicle in a parallel group, multicentered, clinical trial,'' which involved tinea corporis and tinea cruris infections of mixed dermatophyte origin. Culture conversions in this large study were observed at day 14 in 76% of the subjects treated with ciclopirox olamine cream and 45% of the subjects treated with the cream vehicle. Our experimental dermatophytosis model, which provides well-localized and controlled infections, should prove useful in the early assessment of promising test formulations and possibly could serve as a model for bioequivalence determinations for topical antifungal drugs.
0.41-0.80 0.20-0.59 0.64-0.96 0.07-0.39
CO Lin COCrm Un Veh CrmVeh
25 ERYTHEMA 20
o
a 10
DRUG NAMES 0 1 2 SIGN SEVERITY SCORE
0
ciclopirox olamine lotion: Loprox lotion ciclopirox olamine cream: Loprox cream
1 2 SIGN SEVERITY SCORE
25 VESICLES
.
SCALING
Acknowledgments: Dr. Ray Chen helped with the statistical analysis. Charlene Bayles and Dr. Stan Miller provided technical assistance. Linda Setescak and Joanne Robinett gave administrative assistance during the study. Dr. Dehra Jan Bibel helped in the preparation of the manuscript.
20
15 a: 10
REFERENCES SIGN SEVERITY SCORE
1.
SIGN SEVERITY SCORE
Figure 5. Number of sites with specified sign severity score after 14 days of treatment (for each treatment, n = 26). Eor each sign: 0 = none; 1 = mild; and 2 = notable. CO Ltn, ciclopirox olamine lotion 1%; CO Crm, ciclopirox olamine cream 1%; Ltn Veh, lotion vehicle; Crm Veh, cream vehicle.
2. 3.
4.
immune status of the experimental subject is important because subjects with positive reactions to Trichophyton antigen in skin tests react differently to fungal infections than those whose results are negative.'' The results of this 2-week assay in a limited number of subjects suggest that the two formulations tested (ciclopirox olamine lotion and cream) are therapeutically equivalent in the treatment of experimentally induced superficial T. mentagrophytes infection. The equivalency of the cream
5. 6.
7.
125
Reinhardt JH, Allen AM, Gunnison D, et al. Experimental human Trichophyton mentagrophytes infections. J Invest Dermatol 1974; 63:419-422. Conover WJ. Practical nonparametric statistics. 2nd ed. New York: John Wiley and Sons, 1980:122. Lehmann EL, D'Ahrera JH. Nonparametric statistical methods based on ranks. San Erancisco: Holden-Day, 1975:156. Hsieh CC. Note on interval estimation of the difference between proportions from correlated series. Stat Med 1985; 4:23-27. Wilkes SS. Mathematical statistics. New York: John Wiley and Sons, 1962:368. Wood SR, Cruckshank CND. The relation between trichophytin sensitivity and fungal infections. Br J Dermatol 1962; 74:329-336. Bogaert H, Cordero C, Ollague W, et al. Multicentre double-blind clinical trials of ciclopirox olamine cream 1% in the treatment of tinea corporis and tinea cruris. J Int Med Res 1986; 14:210-216.
