Test and reference products
The pediculicidal test product (PB790) used for both the in vitro studies and the clinical trial, is a dimeticone-based medical device from the NYDA® product family. All components of the NYDA® formulations have a long history of use in cosmetics and pharmaceutics and are regarded as safe. The dimeticones represent the active component by blocking the respiratory spiracles, thereby inducing death of both lice and eggs by suffocation [23]. The test product was provided by G. Pohl-Boskamp GmbH & Co. KG (Hohenlockstedt, Germany).
A formulation containing 1% permethrin (Nix®, CVS Pharmacy, USA) was used as internal control for the assessment of permethrin resistance of the louse strain in the in vitro bioassay. The active ingredient of the reference product used in the clinical trial is 0.5% permethrin (cis−/trans-relation 25:75) in alcoholic solution (InfectoPedicul®; InfectoPharm Arzneimittel und Consilium GmbH, Germany). Permethrin is a broad-spectrum synthetic pyrethroid and has been widely used for the treatment of head louse infestations in children. InfectoPedicul® and NYDA® are recommended by the Federal Office of Consumer Protection and Food Safety in Germany (Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, 20.10.2015) as disinfestation agents for the treatment of head louse infestation and are thus recommended pursuant to Section 18 of the German Infection Protection Act (IfSG).
Assessment of pediculicidal activity in vitro
Adult body lice (Pediculus humanus humanus) fed on rabbits were kept at 30.0 ± 1.0 °C and 73.8 ± 3.8% relative humidity and used within 24 h after the last feeding. The procedures were applied as described by Oliveira et al. and Sonnberg et al. [27, 28]. In brief, 40–50 ml of the test product were placed into porcelain bowls, and pre-warmed. For every test run, 30 male and female lice were transferred into a plastic sieve. The sieve was placed into the porcelain bowl, and lice were completely immersed in the product for 10 min. Upon immersion, lice were carefully separated from each other using a plastic spatula. As a negative control, identical procedures were followed with demineralized water.
Exposure was stopped by washing the test product off with a 1:4 solution of shampoo (pH Eucerin® Dermo capillaire, Beiersdorf AG, Hamburg) in tap water. The sieve was then dabbed on tissue paper, rinsed with lukewarm tap water for 1 minute and finally again dabbed on a tissue paper. Afterwards, the lice were individually transferred into labelled glass vials containing some gerbil (Meriones unguiculatus) hairs as a walking substrate. The vials were left open to observe the lice with a binocular microscope. The time points of observation were: 10, 20, and 30 min, 1, 2, 3, 8 and 24 h after treatment. During monitoring, the viability of each louse was categorized as follows:
-
L: Alive and walking
-
M: Moribund, reflexes and small movements observed, not walking
-
G: Without reflexes, only gut movements are observed
-
D: Dead, no movements are observed
Based on published criteria for the assessment of mortality and in compliance to the CRO’s standard protocol, lice in categories M, G and D were defined as non-viable [27, 29, 30]. Tests were repeated three times to obtain a total of 90 lice in each group.
Assessment of ovicidal activity in vitro
To evidence ovicidal activity of the test product, we performed laboratory tests. Two in vitro hair tuft bioassays were performed. First, lice were immersed in the test and control products, respectively. To simulate more closely the exposure conditions as they exist for in vivo application of the test product, a spray protocol was developed in a second bioassay.
Eggs from permethrin-resistant head lice (Pediculus humanus capitis, BR-HL strain), originally collected from infested children in Bristol/UK and maintained on an in vitro rearing system at the University of Massachusetts at Amherst/USA were used in the hair tuft bioassays, similar to the method developed by Strycharz et al. (2012) [31].
For the immersion bioassays, eggs attached to human hair tufts (at least 30 eggs/hair tuft of different developmental stages) were treated with the dimeticone test product or control following an immersing-swirling method for 10 min. A tuft with attached eggs was saturated with the test product by immersing it into 0.5 ml for 30 s with swirling on a small glass slide to ensure saturation and complete egg coverage. As an internal control for permethrin resistance, tufts with attached eggs were saturated for 30 s with 0.5 ml of the 1% permethrin product as above. For a negative control, a dry tuft with attached eggs was saturated with 0.5 ml distilled deionized water (ddH2O) as above. After treatments, the tufts were transferred to a clean Petri dish and placed in an incubator (31 °C, 70–80% relative humidity) for 10 min.
Besides immersion, the protocol of the spray bioassays followed the procedures described above, including a 10 min exposure time. Hair tufts with attached eggs were saturated with the dimeticone test product or with ddH2O by spraying the formulation until complete egg coverage was achieved by visual observation. The pre-optimized spraying technique using 12 pumps was employed to establish consistent applications. Due to the texture of the internal permethrin control product, the immersion protocol was used for these groups.
At the end of the exposure time, the tufts from both bioassays were shampoo-washed by applying 0.5 ml baby shampoo (Johnson and Johnson, New Brunswick, USA). The shampoo-treated hair tufts were sequentially washed in three separate water baths, each placed on a magnetic stirrer for 40 s per wash and air dried on filter paper for 5 min at room temperature. Dried tufts with treated eggs were placed into covered sterile glass Petri dishes and moved to an incubator at 31 °C, 70–80% relative humidity. Egg viability was recorded daily by examining individual eggs for proper site, shape, and color. The number of lice that hatched from eggs was recorded and used to determine the relative ovicidal activity of the treatments. Underdeveloped eggs and stillborn lice were recorded as dead. All treatments were performed three times.
Clinical trial
Study design/regulatory background
This monocentric, randomized, controlled investigator-blinded trial with a 1:1 allocation ratio was approved by the Ethical Review Board of the Landesärztekammer Thüringen/Germany and by the German Federal Institute for Drugs and Medical Devices (BfArM) by the department for medical devices for the test product, and the department for medicinal products for the reference product, respectively. The design of this study and all assessments performed followed international agreements regarding clinical trials with pediculicides [32]. The primary objective of the RCT was to show that the cure rate, corrected for re-infestation, of the dimeticone treatment is superior to 70% (literature-based lowest acceptance rate). The main secondary objectives were to show that the cure rate of the test product was superior or non-inferior to the reference product, and to assess safety and tolerability of the pediculicidal products tested.
The study protocol followed the ethical principles of the Declaration of Helsinki, ICH-GCP guidelines (International Conference on Harmonization of Technical Requirements for registration of Pharmaceutical for Human Use – Good Clinical Practice), requirements of the German Drug Law including the GCP regulation, and the German Medical Device Law. The trial was registered at the EU Clinical Trials Register (EudraCT2016–004635-20). The study adheres to CONSORT guidelines for reporting clinical trials.
Eligibility criteria
Patients of both sexes with active head louse infestations were eligible. Children were included, if they were 2 years of age or older. The reference product was administered according to the most current SPC that was available in Germany before start of the study. For children aged 2 months to 3 years, a maximum dose of 25 ml must not be exceeded, which was strictly followed for children < 3 years. Active infestation was defined as the presence of at least 5 live lice, confirmed by diagnostic combing using combs with a tooth gap of 0.2 mm and tips with blunt parallel-sided teeth.
Patients or their guardians had to be capable of understanding written informed consent form and to give written informed consent after being informed about benefits and potential risks of the trial, as well as details on the insurance covering the subjects participating in the study. Patients had to agree not to use any other anti-lice treatment for the duration of the study. Female patients of childbearing potential had to be tested negative for pregnancy and had to agree to use a reliable method of birth control or remain abstinent during the study.
Patients were excluded if they had used any head lice treatment within the last 30 days prior to the screening visit; used systemic or topical drugs or medications, including systemic antibiotics, which in the opinion of the investigative personnel may interfere with the study results; in the case of allergies or hypersensitivities against any of the active ingredients or the constituents of the products used, skin allergies, multiple drug allergies or multiple allergies to cosmetic products, severe acute scalp disorders, hair longer than mid-back, high probability or known not to follow instructions, or previous participation in this study or in any other investigational trial within the preceding 30 days. Pregnant and breast-feeding women were also excluded.
We did not include patients unable to understand the written and verbal instructions given by the study personnel, in particular regarding the risks and inconveniences. Personnel directly affiliated with this study and/or their immediate families and G. Pohl-Boskamp employees or employees of third-party organizations involved in the study were also excluded.
Setting
Recruitment, treatment and diagnostic assessment of patients took place from 18th April, 2017 (first patient in) to 16th March, 2018 (last patient out) at a clinical trial center in Erfurt, Germany, specialized in conducting trials with children and experienced in diagnosis and treatment of head louse infestations. Patients/guardians/caretakers were informed about this study via advertisements, flyers, letters to primary schools and preschools, and internet on-screen displays. All texts used for recruitment had been approved by the responsible ethics committee. Patients/ guardians/caretakers had to contact the study center to arrange an appointment, and after initial assessment of infestation and informed consent procedure, patients were included and allocated to treatment groups.
Intervention
Two topical applications, as recommended for treatment of head louse infestations, were performed [32, 33]. The first treatment was applied on day 0 (V1), the second on day 7 or 8 (V3).
Treatments required individual amounts of the investigational products to completely cover hair and scalp of the patient. Both products were applied as recommended by the manufacturers. The test product was evenly sprayed until the hair was completely wetted with the solution and then massaged into the dry (not washed) hair over its full length, with special diligence on the base of the hair near the scalp and the ear region. Curly, long and thick hair was treated in strands. After 10 min without covering, the hair was combed carefully with the nit-comb to remove suffocated lice and eggs. Afterwards the product was washed out with a commercially available shampoo and the hair was rinsed thoroughly.
For application of the reference product, the hair was washed with a commercially-available shampoo and towel-dried. Afterwards, the product was evenly dispensed onto the hair over its full length, again, with special diligence on the base of the hair near the scalp and the ear region. Curly, long and thick hair was treated in strands. After 45 min without covering the hair, the product was washed out with warm water without shampoo. Afterwards, the hair was again carefully towel-dried. Before the hair was completely dry, combing with a nit-comb was performed according to the manufacturer’s instruction for use, to eliminate the eggs. The participants dried their hair by themselves or with the help from their guardians, using a hair dryer.
Metal combs were used, with a gap of 0.2 mm and tips with blunt parallel-sided teeth. The same comb types were used for both groups.
According to the instructions for the reference product, participants were instructed not to wash their hair with shampoo within the next 3 days, to possibly increase the ovicidal efficacy of permethrin. To keep both treatments as similar as possible, the instruction of not washing the hair within 3 days after application was also given for the test group.
Informed consent, outcome measures and assessments
The primary outcome was defined as the cure rate at the end of day 10 (V4), corrected for re-infestation. Assessments for head louse infestations were performed on days 1 (V2), 7 or 8 (V3) and 10 (V4), by blinded study staff. Definition of re-infestation was based on current literature: no adult lice or third stage nymphs following first treatment at day 1 (V2), and no more than two adult lice or third stage nymphs found by combing following second treatment on day 10 (V4) [34].
Secondary outcome measures included: global tolerability and local tolerability, rated by the patients and the blinded investigator via 4-point VRS (Tables 4 and 5); skin irritation and eye irritation, assessed by the blinded investigator using a 4-point VRS. The esthetical effect of the products was evaluated (look of hair; feeling of hair; sensation on scalp; categories: “strongly agree”, “agree”, “disagree”, “strongly disagree”) by structured questionnaires. Date, number and type of adverse events were documented.
A total of four assessments were performed (V1-V4). At day 0, during the screening visit, patients were registered and enrolled. After informed consent procedure and signing of the informed consent form (ICF), confirmation of the diagnosis, and assessment for exclusion and inclusion criteria, participants were randomized to one of the two intervention groups. Baseline data (gender, age, hair length, hair type, medical history, concomitant medication) were assessed before treatment. Within 1 hour before first application and 1 hour after application of the respective product, the following assessments were done: assessment of skin irritation, assessment of eye irritation and assessment of adverse events (AE). Additionally, 1 hour after application on V1 and V3 and also during V2 and V4, assessment of global tolerability and local tolerability were performed. These assessments were repeated on days 7 (V3, before and after treatment) and on 10 (V4, final assessments). Application of the questionnaire on esthetical effects was done only after treatment on day 0 (V1) and day 7 (V3).
At all visits after treatment (V2-V4), AEs were recorded for start and end dates and times, seriousness, expectedness, severity, causal relationship to investigational product, and causal relationship to study procedure and evaluated for events per subject and study group. All deteriorations were documented as AEs.
Upon recommendations of the Robert Koch Institute [33] and in line with current state-of-the-art clinical trials [24], patients were allowed to use a nit-comb between study visits. At every visit, patients or their guardians (in the case of minors) were asked if they used the nit-comb provided and answers were documented in the electronic case report form (eCRF).
All assessments of primary and secondary outcome measures were performed by investigators blinded to the treatment applied.
Sample size
According to previous preclinical studies with the dimeticone test product and to preclinical and clinical studies with products from the NYDA® family, a 90% cure rate was expected for the test product. A pre-defined limit of 70% was determined from reported cure rates for permethrin-containing products (range from 34.8 to 98.0%, mean 66.6%, approximated to a 70.0% cure rate) and was defined to be the minimal acceptable cure rate [35,36,37,38]. A sample size of 42 was required for a one group χ2-test comparing cure rate of 90% with a fixed limit of 70% (two-sided test; alpha-level of 0.05; power = 90%; software nQuery advisor 7.0; power = 80%: sample size = 34). Assuming a 10% drop out rate and 5% re-infestation rate, 49 cases would be required in each group, which was rounded up to 50.
Randomization
Participants were randomized to one of the two head louse treatments by a computer-generated code using randomly mixed blocks of 10, with a final and random 1:1 allocation. Randomization, enrollment of participants, and assignment of specific participants to one of the two interventions were performed by an investigator not involved with assessment of outcome measures.
Blinding
The study was observer-blinded. As both products differ substantially from each other in terms of packaging, smell, application method and exposure time, double-blinding was not possible.
All staff members involved in assessments of primary and secondary outcome measures (investigators and study staff performing the assessments of hair and scalp, eyes, as well the efficacy and safety evaluations) were blinded to treatment assignment. The assessors were not involved in handling, storage, and use of the products, and did not have access to the eCRF entries regarding application of investigational products, such as used amount or start and stop time of exposure.
Statistical analyses
Pediculicidal activity in vitro
At each point of time of observation, the relative frequencies of non-viable lice (categories M + G + D) were calculated for each test run, and the arithmetic mean ± standard deviation (SD)was calculated. Relative frequencies of lice treated with the test product and the water control were compared using Fisher’s exact test using the Statistica® v.7.1 software (StatSoft, Tulsa, USA). For determination of pediculicidal activity, endpoint mortality of lice (mortality determined 24 h after treatment) was compared between test and control.
Ovicidal activity in vitro
The mean percent ovicidal activity (± SD) was determined and statistically analyzed using Shapiro-Wilk Test for determination of normal distribution, 2-factorial ANOVA and test for homogeneity of variance (Levenes). Number of replicates was 3 (immersion protocol) and 4 (spray protocol), respectively.
Clinical trial
Statistical analysis of data obtained from the RCT followed the predefined statistical and analytical plan (SAP). Descriptive statistics (continuous variables: mean ± SD; ordinal variables: median, min, max, Q1, Q3; categorical variables: counts and relative frequencies) was performed by visit and treatment.
For the primary objective, the following null hypothesis was tested: H0prim: pT = 70%. If pT > 70% and if the null hypothesis was rejected by a two-sided, one sample χ2 test at 0.05 level, superiority would be concluded. The primary objective was analyzed in the full analysis set (FAS). The secondary objectives were analyzed in FAS and per-protocol population (PP).
Due to no observed differences for the cure rates after correction for re-infestation (100% efficacy for both treatments), non-inferiority testing was performed in the PP population, with a predefined non-inferiority margin of 7.5%. The following null-hypothesis was tested at an α-level of 0.025. H0,NI: pT-pR < δ, whereby δ = − 7.5%. The lower, one sided 97.5% confidence interval of difference pT-pR was used for the test. With a confidence limit of ≥ − 7.5, hypothesis H0,NI was rejected and non-inferiority was concluded. The differences of cure rates pT-pR as well the cure rates pT and pR are presented with two-sided 95% confidence interval. NCSS 12.0.2 software was used to calculate the 95% confidence interval for the difference of zero (Miettinen-Nurminen Score).
Cure rates (pT and pR) and difference between cure rates (pDiff = pT - pR) with their two-sided 95% CI including results of two-sided, two sample Fisher’s exact test were calculated. Effect of hair characteristics on efficacy of the products was evaluated via Kruskal-Wallis-Test.
The use of a nit-comb during clinical trials is controversial, as combing may increase efficacy [32]. To estimate the effect of combing, in a post hoc analysis, the association between the occurrence of treatment failures and patients’ implementation of combing was evaluated. For this purpose, treatment failures and combing were evaluated for independence of categorical variables by chi-squared test and by Fisher’s exact test (for small expected frequencies).