Impact of a non‐fluoridated microcrystalline hydroxyapatite dentifrice on enamel caries progression in highly caries‐susceptible orthodontic patients: A randomized, controlled 6‐month trial

Abstract Aim The aim of the present randomized, controlled trial was to compare the impact of the regular use of a fluoride‐free microcrystalline hydroxyapatite (HAP) dentifrice and a 1400 ppm fluoride control dentifrice on caries progression in 150 highly caries‐active orthodontic patients. Methods The primary outcome was the occurrence of lesions with International Caries Detection and Assessment System (ICDAS) ≥code 1 on the vestibular surfaces of teeth 15‐25 within 168 days after fixation of orthodontic brackets. Secondary outcomes were lesion development ICDAS ≥code 2, the plaque index, and the gingival index. Results In total, 147 patients were included in the intent‐to‐treat (ITT) analysis; 133 finished the study per protocol (PP). An increase in enamel caries ICDAS ≥code 1 was observed in 56.8% (ITT) and 54.7% (PP) of the HAP group participants compared with 60.9% (ITT) and 61.6% (PP) of the fluoride control group. Non‐inferiority testing (ITT and PP) demonstrated the absence of a significant difference between the groups. No significant differences in secondary outcomes were observed between the groups. Conclusion In highly caries‐active patients, the impact of the regular use of a microcrystalline HAP dentifrice on caries progression is not significantly different from the use of a 1400 ppm fluoride toothpaste (ClinicalTrials.gov: NCT02705456).


| INTRODUC TI ON
Recent findings, mostly derived from in vitro studies, have suggested that microcrystalline hydroxyapatite (HAP) particles might be suitable candidates for the prevention of demineralization and the stimulation of remineralization processes on enamel and dentine surfaces. [1][2][3] Furthermore, in an in situ study, the use of (a pure HAP was used) HAP microcluster-containing mouthrinse significantly reduced bacterial colonization on bovine enamel slabs worn intraorally by healthy volunteers. 4 Hannig and Hannig put these in situ and in vitro findings into a more comprehensive perspective by stating that established physiological tooth wear constantly releases HAP particles into the oral environment, which might subsequently interfere with demineralization and remineralization processes, as well as with the metabolism of the oral microbiota at the tooth-bacterial biofilm interface. 5 The impact of microcrystalline HAP as an ingredient in dentifrices has been positively evaluated in controlled clinical trials on dentinal hypersensitivity, [6][7][8][9] and parameters of periodontal health. 10 To date, however, comparable data regarding the caries-inhibiting properties of HAP toothpastes are lacking. As orthodontic therapy with fixed appliances is known to be associated with an increased incidence of the overgrowth of a caries-promoting microbiota 11 and the development of white spot enamel caries lesions, [12][13][14] the aim of the of the present study was to assess the caries-inhibiting impact of the regular use of a fluoride-free HAP dentifrice in this particular group of patients with caries risk. Due to the abundant evidence for the caries preventive efficacy of fluorides, 15,16 clinical caries studies might no longer involve a true negative control for obvious ethical reasons.
Thus, a non-inferiority trial was conducted. The study hypothesis to be tested was whether or not the regular use of the HAP test dentifrice was inferior to the regular use of a fluoridated control in terms of caries prevention.

| MATERIAL S AND ME THODS
The investigation was designed as a multicenter, prospective, parallel-group, two-arm, double-blinded, randomized, clinical non-inferiority trial to be performed at the German study centres Wuerzburg (leading study center), Regensburg, Munich, Dresden, and Frankfurt. The study protocol was prepared in accordance with the Declaration of Helsinki and met the good clinical practice criteria. It was approved by the ethics committee of the University of Wuerzburg (file no. 184/13) and was registered at ClinicialTrials.gov (identifier no.: NCT02705456).

| Study design
The design of the study is schematically depicted in Figure 1. At visit 1 (−4 to −28 days prior to baseline), patients scheduled for orthodontic therapy were screened for study eligibility. Those meeting the eligibility criteria were asked to participate, and after providing informed consent, they were scheduled for the baseline visit 2 (day 0).At visit 2, the plaque index (PI) and the gingival index (GI) scores were recorded from the vestibular surfaces of teeth 15-25, followed by professional tooth cleaning and the subsequent assessment of the vestibular enamel surfaces of teeth 15-25 according to International Caries Detection and Assessment System (ICDAS) II criteria. Orthodontic brackets were then adhesively mounted to the vestibular surfaces.
No sealants, fluoride varnishes, or any other caries-preventive layers surrounding the brackets were applied. Using a randomization list, a supply of either the test dentifrice or the control dentifrice, calculated to be adequate for 4 weeks of 2× daily repeated toothbrushing, as well as a standardized electric toothbrush (Oral-B Pulsar 35; Procter & Gamble GmbH, Schwalbach, Germany) to be used for the duration of the study, were given to the study patients. Practical training was provided for the dosing of the assigned dentifrice (2× daily a streak of approximately 1 g) and the use of the electric toothbrush, and the patients were instructed to return all the assigned toothpaste tubes at the next scheduled visit. At day 28, the sequence of recording the PI, GI, and ICDAS II scores was repeated, as described for visit 2. As F I G U R E 1 Schematic study design. CHX, chlorhexidine; GI, gingival index; ICDAS, International Caries Detection and Assessment System; PI, plaque index an additional caries-preventive measure, teeth 15-25 were disinfected with a topically-applied 1% chlorhexidine gel. Toothpaste tubes supplied at visit 2 were collected, and a new supply was provided for the next 4 weeks. At day 56 (visit 4), oral hygiene reinstruction was provided, as well as cleaning/disinfection procedures and return/handing over of the toothpaste supply, as described earlier. At day 84 (visit 5), the recording of the PI, GI, and ICDAS II scores and cleaning and disinfection were repeated, as described earlier. In addition to a new supply of toothpaste, a new electric toothbrush was also provided.
At day 112 (visit 6) and day 140 (visit 7), the performed procedures were identical to those at day 56 (visit 4). At day 168 (visit 8), the final assessment of the PI, GI, and ICDAS II scores and the return of the study dentifrices were conducted, as described before. Furthermore, at each study visit, the patients were asked about the occurrence of important problems or unintended effects related or unrelated to the use of the study dentifrices.

| Study population
The trial was performed in healthy adolescents and young adults who were scheduled for orthodontic therapy with fixed appliances.

| Exclusion criteria
The inclusion criteria were: (a) untreated caries lesions of ICDAS code 3-6 on any tooth; (b) treated carious lesions of ICDAS code 3-6 on the vestibular surfaces of teeth 15-25; (c) diseases or conditions or the regular use of related medications that interfere with salivary flow; (d) antibiotic therapy within the past 6 weeks before study participation or the necessity for antibiotic prophylaxis during dental interventions; and (e) known allergies to ingredients in the experimental dentifrices.

| Control dentifrice
A commercially-available fluoridated toothpaste (meridol Zahnpasta; CP GABA GmbH, Hamburg, Germany) was used as a positive control.

| Primary outcome
The primary study outcome was the percentage of participants in each experimental group with a new occurrence of at least one enamel caries lesion ICDAS ≥code 1 on the vestibular surfaces of teeth 15-25 during the observation period of 168 days.

| Caries assessment
The occurrence of caries was evaluated visually on the vestibular surfaces of teeth 15-25 according to ICDAS-II criteria. 18 The examination was performed at baseline, prior to the fixation of the orthodontic brackets, and was repeated after 28 days, 84 days,and 168 days. All teeth were professionally cleaned before each assessment. The development of a caries lesion ICDAS >code 3 during the course of the study on any tooth and observed at any visit was defined as an immediate study exit criterion.

| Interexaminer reliability
To ensure interexaminer reliability, prior to the study onset all examiners were instructed to pass the ICDAS e-learning course at the icdas. org website and were subsequently trained in person by an experienced expert (K.H.K.) to perform ICDAS assessments in reference patients. Grading skills were retrained three times during the course of the study using another Internet-based ICDAS training tool. Interrater reliability analysis revealed a mean weighted ĸ = 0.75 for the first assessment run, which increased to ĸ = 0.80 for the final calibration, indicating "substantial agreement" among the different examiners throughout the study. 19

| Statistical analysis
The primary outcome measure was analyzed primarily for the per protocol (PP) population and repeated for sensitivity reasons for the intent-to-treat (ITT) population. The exact confidence limits (Clopper-Pearson) were computed to test non-inferiority. 22 For the primary outcome measure, non-inferiority was claimed if the upper limit of the one-sided 95% confidence for the corresponding difference between test and control dentifrice was less than Δ (difference) ≤20%.
In addition, two-sided Wilcoxon-Mann-Whitney tests were used for between-group comparisons, and Friedman tests for withingroup comparisons, for secondary outcomes.

| Sample size calculation
Based on a reported caries incidence rate of approximately 60% in a preceding caries trial assessing orthodontic patients with fixed braces who were not being preselected for particular caries-promoting risk factors, 13 the likelihood for the occurrence of an ICDAS code 2 lesion during the 168-day observation period in this cohort of cariesrisk individuals with elevated salivary numbers of caries-promoting mutans streptococci was extrapolated to be P = 80% for the control group using the fluoridated toothpaste. The difference between both experimental groups was not regarded to be clinically relevant and was set to Δ ≤ 20%. A sample size of 2 × 74 study patients was calculated to be sufficient to reject the null hypothesis, that the test dentifrice is inferior to the control dentifrice, using a non-inferiority margin of Δ = 20% for the primary outcome measure and one-sided, exact Fisher's test (α = 5%, power = 80%).

| Blinded change of the primary outcome
A blinded analysis of the ICDAS data at the end of the study revealed that the overall observed occurrence of ICDAS lesions ≥code 2 in the study population was 29.3%, and therefore considerably lower than the anticipated value (P = 80%) used for the sample size calculation. As the difference between the groups was not regarded to be clinically relevant and had been set in the study protocol to Δ ≤ 20%, a clinically meaningful verification of non-inferiority was no longer warranted. Thus, the primary endpoint was changed to the more frequent overall occurrence of ICDAS lesions ≥code 1 (59.2%).
We decided to keep the original primary endpoint as an additional secondary outcome in the statistical data analysis.
While it might have been debatable to keep the original noninferiority margin of Δ = 20% when switching the primary outcome of the trial, despite an overall incidence of the revised primary outcome (ICDAS lesion code 1) of only 60%, the subsequent analysis of the unblinded PP dataset revealed that the actual difference between both experimental groups was 6.2% in favor of the HAP test dentifrice with an exact upper one-sided 95% confidence limit of 8.3% (i.e substantially lower than the preset non-inferiority margin of Δ = 20%).

| Blinding and randomization
The trial was designed to blind study patients and examiners to the group assignment. Both study dentifrices (test/control) were filled into neutral plastic tubes of identical shape and color by an independent, good manufacturing practice-certified laboratory for cosmetics. Using block randomization with a block size of 4, a random list was generated to code label test and control tubes with consecutive unique identification numbers. The randomization of dentifrice assignment was stratified by the study center. Distribution of the experimental dentifrices to the study patients followed the sequence of the identification numbers and was performed by trained study nurses not involved in the examination of the study participants.
To maintain blinding of examiners and study patients, the study patients were instructed not to discuss toothpaste-related issues with the examiners, but with the study nurses only, who were also responsible for instructing the patients in efficacious oral hygiene and taking back the empty or unused dentifrice tubes at the subsequent visits. The number of study nurses varied between a minimum of one and a maximum of four per study center.

| Number and severity of International Caries Detection and Assessment System score increases
The number and severity of ICDAS score increases on the vestibular surfaces of teeth 15-25 over the course of the study are shown in Table 1

| Effect of study site on the primary outcome measure
The effect of study site on the primary outcome measure Δ ICDAS score ≥1 at day 168 was evaluated by logistic regression analysis.
It included the factor's study site, treatment group, and the interaction between the study site and treatment group. Due to small sample sizes, the data for the Dresden, Munich and Frankfurt study sites were pooled (N = 40 patients). The results revealed a significantly lower incidence of the primary outcome at day 168 (P < 0.001) at the combined smaller centres (Dresden, Munich, and Frankfurt) when compared to the study centres in Regensburg (N = 72 patients) and Wuerzburg (N = 35 patients). However, there was no significant interaction between the study site and treatment group, proving that the factor study site did not significantly affect efficacy differences between the treatment groups (Tables 2 and 3).

| Health status, age, and sex
All study patients were healthy.

| Occurrence of International Caries Detection and Assessment System lesions ≥code 1 and ≥code 2
The occurrence of ICDAS lesions ≥code 1 (primary outcome) and ICDAS lesions ≥code 2 (secondary outcome) is depicted in

| Non-inferiority analysis
The difference between both experimental groups in terms of the percentage of study participants experiencing a new occurrence of at least one ICDAS lesion ≥code 1 (primary outcome) or at least one ICDAS lesion ≥code 2 (secondary outcome), including the corresponding one-sided 95% confidence intervals, as displayed in Table 5. As the upper limits of the 95% confidence intervals for the primary outcome were well below the given noninferiority margin of Δ ≤ 20% for both analysis sets (PP: 8%, ITT: 9%), the HAP group was considered to be non-inferior to the fluoride control.
Regarding the secondary outcome (ICDAS lesion ≥code 2), the upper limits of the 95% confidence intervals were also substantially below the given non-inferiority margin of 20% for both analysis sets (PP: 3%, ITT: 7%), again indicating that the HAP test group was noninferior to the fluoride control.

| Plaque Index and Gingival Index
The results of the ITT analysis of the Pl and the GI data are shown in Table 6. The mean Pl and GI scores increased significantly (P < 0.0001) from baseline to day 168 in both groups, but they were not significantly different between the groups at any time point.

| Methods
Caries detection and grading in this trial followed the principles of ICDAS-II, 18 an internationally-established, state-of-the-art caries assessment method that is particularly suitable and appropriate for the differentiation and grading of incipient enamel caries. Due to repeated examiner calibrations, the mean weighted kappa for interrater reliability increased from initially 0.75 for the first to 0.80 for the final calibration assessment, demonstrating an overall in the upper range of the kappa reliability scores reported by other controlled clinical trials and indicative of "substantial" agreement. 19

| Data analysis
Whether the occurrence frequency of ICDAS code 1 enamel caries lesions used in the present study is the most suitable primary endpoint for a non-inferiority caries trial is subject to discussion.
However, the adjunctive analysis of the PP dataset regarding the frequency and severity of the occurrence of enamel caries lesions during the observation period, as depicted in Table 1, only endorsed the identified absence of relevant differences between the groups.
The data for the secondary outcomes (PI and GI) further confirmed the findings of preceding studies, reporting a significant increase in gingival inflammation and bacterial plaque mass after the onset of orthodontic therapy with fixed appliances. 13,24 Differences between the groups regarding PI and GI were not significant for any of the evaluated time points, which was also in good agreement with the results of a previous trial comparing the plaque-and gingivitisreducing properties of a fluoride-free HAP test dentifrice and a fluoridated amine fluoride (AmF)/SnF 2 control in a study cohort of patients suffering from mild to moderate periodontitis. 10

| Outlook
While the safety of fluoride-based caries prevention has been firmly established by numerous studies, 16 dosage and toxicity aspects must always be considered. A caries-inhibiting increase in the applied fluoride dosage in caries-risk patients, as described by Sonesson et al., 25 might thus not be feasible in infants and children up to the age of 8 years due to the associated risk for the development of dental fluorosis. Although not verified by clinical studies thus far, an increase in the dosing or application frequency of HAP toothpaste might also potentially boost the caries-inhibition efficacy in caries-active patients, as HAP is a potent buffer under acidic conditions that is able to neutralize organic acids. Unlike fluorides, a HAP dosage increase is not affected by any toxicity issues, even in infants and children, as HAP is the major mineral phase of all human hard tissues. 5

| Conclusions
The data of this 6-month, clinical non-inferiority trial demonstrate, that in highly caries-active orthodontic patients, the impact of the regular use of a fluoride-free, microcrystalline HAP dentifrice on caries progression is not significantly different from the use of a