We conducted a single-centre, randomised (to order in which toothpastes were used), oral/dental examiner- and specimen analyst-blind, three-period, three-treatment, in-situ crossover study at the Indiana University Oral Health Research Institute. The Indiana University Institutional Review Board (IRB #1709160589) approved the protocol; the study was designed according to CONSORT guidelines and was conducted in accordance with the Declaration of Helsinki, the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use and local laws and regulations. As the study involved human participants, it was registered at
clinicaltrials.gov (#NCT03296072). There was one administrative change to the protocol prior to study start that did not affect study flow or outcomes. Anonymised individual participant data and study documents can be requested for further research from
www.clinicalstudydatarequest.com.
Participants
Participants were recruited from an existing panel of individuals who had been pre-fitted with a maxillary palatal appliance capable of housing prepared enamel specimens. All participants were from the Indianapolis, IN area, where community water contains approximately 0.75 ppm fluoride [
19]. Eligible participants were aged 18 to 65 years, in good general and oral health and had unstimulated/stimulated salivary flow rates of ≥0.2 mL/min and ≥ 0.8 mL/min, respectively. Exclusion criteria included: presence of cavitated carious lesions (determined by visual assessment only), moderate or severe periodontal conditions, or severe tooth wear; wearing an oral appliance or orthodontia (except permanent lower retainers); any condition that might have influenced the study or impacted participant safety and wellbeing; pregnancy; breastfeeding; use of any medication that could interfere significantly with salivary flow; an intolerance/hypersensitivity to study materials; use of any investigational products or participation in another clinical trial within 30 days of screening.
Enamel specimen preparation
Bovine enamel specimens (1488 in total), taken from the central area of the buccal surface of lower incisors (up to two specimens per tooth), were ground-polished flat until the enamel surfaces had a minimum 3 mm × 3 mm facet in the centre. Specimens were then serially polished with grit papers of descending coarseness, finished with a polishing cloth wetted with a 1 μm diamond suspension then sonicated to remove any adherent polishing particles. The resulting specimens had a thickness range of 1.7–2.2 mm.
Changes in the mineral content of the enamel specimens during the experiment were evaluated using indentation to determine surface microhardness (SMH) [
8,
20,
21]. At baseline, five indentations 100 μm apart were made in the centre of each enamel specimen using a Knoop diamond under a 50 g load, applied for 11 s. Analysis of indent lengths was performed on a specimen level and then averaged, with a participant-level average value used for analysis.
For the first in vitro erosive challenge, enamel specimens were demineralised by immersion in 35 ± 1 mL grapefruit juice (100% juice, Kroger Co, Cincinnati, OH, USA; pH range 2.8–3.1 over the different study days) with no stirring for 25 min, then thoroughly rinsed with deionised water. This time period was chosen as previous studies have shown this to be an adequate amount of time in demineralising conditions for the surface microhardness test to detect changes without significant surface loss (in the absence of agitation) [
22]. Indentation lengths were then determined as before.
Following sterilisation with ethylene oxide, demineralised enamel specimens were secured onto plastic holders and attached to the palatal appliance, four on each side. Following the in situ challenge period (details below), after removal from the appliance, the specimens underwent a second in vitro erosive challenge as above (Fig.
1).
In situ procedure
Participants were required to complete four study visits: screening (Visit 1), then three treatment visits (Visits 2, 3, and 4) at which each of the three treatments were evaluated in a crossover manner. Each treatment visit was separated by a washout period of at least 3 days that included at least 1 day’s use of the participant’s regular toothpaste and 2 days' use of a non-fluoridated (non-marketed) toothpaste immediately prior to the visit using a provided toothbrush (Oral-B® Sensi Soft Manual; Procter & Gamble Company, Cincinnati, OH, USA). At Visit 1, participants gave written informed consent to take part in the study and their demographics, medical history and prior medications were recorded. Oral hard tissue (OHT) and oral soft tissue (OST) examinations were performed, followed by saliva flow rate assessment. Eligible participants had their palatal appliance assessed for comfort, with new appliances made for those whose appliance no longer fitted adequately.
At Visits 2, 3, and 4, participants underwent a pre-treatment OST examination, then the study investigator (the oral/dental examiner) placed the palatal appliance holding the eight pre-demineralised bovine tooth enamel specimens in the participant’s mouth. An equilibration period of at least 5 min was given to allow for development of a salivary pellicle on the enamel blocks.
Participants received the study treatments detailed in Table
1 in a predetermined order according to a randomisation schedule generated by a contracted statistical analysis organisation using a Williams Square design balanced for first period carryover. Study numbers were allocated in ascending order as each participant was entered into the trial.
Test toothpaste | 0.254% w/w NaF (1150 ppm fluoride), 5% KNO3, PVM/MA copolymer, sodium lactate, pH 6.2 |
Placebo toothpaste | 5% KNO3, PVM/MA copolymer, sodium lactate (0 ppm fluoride) |
Reference toothpastea | 0.454% w/w SnF2 (1100 ppm fluoride), zinc citrate (‘SnF2-Zn’) |
The oral/dental examiner and the specimen analyst were not permitted in the room where study products were dispensed. The oral/dental examiner, specimen analyst, study statistician and any relevant study sponsor or research centre employees were blinded to treatment received. While treatment group was not revealed to the participant during the study and the study toothpastes were supplied to the study site in over-wrapped tubes to conceal product identity, participants could not be deemed fully blinded as, according to ICH guidelines, for a truly double blind study, the products would need to be identical in every way, including taste and texture, which they were not.
At each study visit, following the equilibration period, the participant was provided with a new toothbrush (as above) loaded with 1.5 g of the assigned toothpaste. The participant brushed the buccal surfaces of their natural teeth only for 25 timed seconds and then swished the resulting toothpaste slurry around their mouth for 95 s. After expectorating the slurry, the participant gently rinsed their mouth with 15 mL tap water for 10 s before expectorating again.
After completing the brushing/rinsing procedures, participants continued to wear their palatal appliance for a total of 4 h. Enamel specimens were removed in a predetermined order from the appliance at 2 and 4 h (four at each timepoint). Once the appliance was removed from the participant’s mouth after 4 h, a post-treatment OST examination was performed.
Safety
Adverse events (AEs) and any abnormalities in the OHT or OST examinations were recorded from the end of screening until 5 days after the last administration of study product. Clinical judgement was exercised by the oral/dental examiner to diagnose the AE and to assess the relationship between the study product and occurrence of each AE, with intensity graded as mild, moderate, or severe.
Specimen analysis
Indentation lengths (details above) were determined prior to (B) and after (E1) the first in vitro erosive challenge, after the treatment-induced in situ rehardening phase (R), and after the second in vitro erosive challenge (E2) (Fig.
1). The extent of rehardening was calculated as SMH recovery (SMHR) where %SMHR = {(E1-R)/(E1-B)}*100 [
23]. Overall resistance of treated enamel to the erosive challenge was calculated as relative erosion resistance (RER), where %RER = {(E1-E2)/(E1-B)}*100 [
24]. Acid resistance following intra-oral rehardening after treatment with the study toothpastes was calculated as the acid resistance ratio (ARR) where ARR = 1-{(E2-R)/(E1-B)} [
7].
EFU was assessed by microdrill enamel biopsy after the in situ rehardening period, before the second erosive challenge [
25]. Enamel specimens were drilled to a depth of 100 μm through the entire lesion, with four cores obtained per specimen. The enamel powder sample pooled from the four cores was dissolved in perchloric acid (20 μl of 0.5 M HClO
4) then buffered with a citrate/ethylenediaminetetraacetic acid solution prior to analysis via a pre-calibrated fluoride-specific electrode. The diameter of the drilled cores was measured via light microscopy and the EFU expressed as amount of fluoride divided by the combined area of the enamel cores (μg F/cm
2). These values were averaged across the four enamel specimens evaluated at each timepoint to produce the participant-wise mean EFU.
Statistical analysis
Sufficient participants were screened so that up to 66 could be randomised to treatment, aiming to ensure that 60 evaluable participants completed the study. This sample was calculated to have 90% power to detect a difference in mean %SMHR of 5.0 between study products at 4 h, assuming a standard deviation (SD) of differences of 11.92 (from a previous, unpublished study, data on file), using a paired t-test with a 0.05 two-sided significance level. This sample size was calculated to have 80% power to detect a difference in %RER of 7.4 between study products. While these specific calculations were based on an unpublished study, participant numbers are similar to or higher than a number of previous in situ erosion-remineralisation model studies [
6,
7,
11‐
13,
16,
20].
Efficacy analyses were conducted on a modified intent-to-treat (mITT) population, defined as all participants who were randomised into the study, received at least one dose of study product and had at least one post-baseline efficacy assessment. The safety population included all randomised participants who received at least one dose of study product.
The primary efficacy endpoint was the difference between the Test and Placebo toothpastes in %SMHR after 4 h of intraoral exposure. The difference was required to be statistically significant (p < 0.05) to meet the success criteria of the study. Secondary efficacy endpoints included the differences in %RER and EFU between the Test and Placebo toothpastes and in all measures between the Test and Reference toothpastes after 4 h. Exploratory endpoints included the difference in all measures for all paired efficacy comparisons after 2 h. Post-hoc analyses were comparison of ARR values between all treatments at both timepoints, and all other paired efficacy comparisons for each endpoint between the Placebo and Reference toothpastes at both timepoints.
Statistical analyses for all endpoints were performed using a mixed model analysis of variance (ANOVA) model that included fixed factors for study period and treatment and a random effect for participant. Statistical testing of all endpoints in this study was conducted at a two-sided significance level of 0.05. As a primary objective was defined prior to analysis, there was no adjustment for multiple comparisons. Adjusted means of all treatments and treatment differences were provided together with their standard error (SE), 95% confidence intervals (CIs) and p-values.