Datasets:

dmariko

/

init_data

like 0

Acknowledgements

None.

PMC10061705

Authors’ contributions

& L.C.

All authors made substantial contributions to the study. Study concept and design: Y.S. & L.C.; literature search: Y.S. W.L. & Z.Z.; acquisition of data: W.L., Z.Z., S.M.L., Y.Y.C.& L.Y.; statistical analyses: Y.S. & L.C.; drafting of the manuscript: Y.S. & L.C.. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

PMC10061705

Funding

None.

PMC10061705

Availability of data and materials

The research data has been uploaded to the database. The public access to the database is open. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

PMC10061705

Declarations

PMC10061705

Ethics approval and consent to participate

Ethics approval has been granted by the Institutional Review Board of Shengjing Hospital of China Medical University, Shenyang, China (approval number: 2021PS511K, approval date:12/05/2021). Written informed consent was obtained from all subjects participating in the trial. The trial protocol followed the Declaration of Helsinki.

PMC10061705

Consent for publication

Not applicable.

PMC10061705

Competing interests

The authors declare that they have no competing interests.

PMC10061705

References

PMC10061705

1. Introduction

MSI, cancers, colorectal cancer, CRC, heterogeneous diseases, NB, deaths, Colorectal cancer

COLORECTAL CANCER, CANCERS, MICROSATELLITE INSTABILITY, MAY, REGRESSION, COLORECTAL CANCER

Academic Editor: Xing NiuChemotherapy is not recommended for patients with deficient mismatch repair (dMMR) in colorectal cancer (CRC); therefore, assessing the status of MMR is crucial for the selection of subsequent treatment. This study is aimed at building predictive models to accurately and rapidly identify dMMR. A retrospective analysis was performed at Wuhan Union Hospital between May 2017 and December 2019 based on the clinicopathological data of patients with CRC. The variables were subjected to collinearity, least absolute shrinkage and selection operator (LASSO) regression, and random forest (RF) feature screening analyses. Four sets of machine learning models (extreme gradient boosting (XGBoost), support vector machine (SVM), naive Bayes (NB), and RF) and a conventional logistic regression (LR) model were built for model training and testing. Receiver operating characteristic (ROC) curves were plotted to evaluate the predictive performance of the developed models. In total, 2279 patients were included in the study and were randomly divided into either the training or test group. Twelve clinicopathological features were incorporated into the development of the predictive models. The area under curve (AUC) values of the five predictive models were 0.8055 for XGBoost, 0.8174 for SVM, 0.7424 for NB, 8584 for RF, and 0.7835 for LR (Delong test, Colorectal cancer (CRC) is one of the most common cancers worldwide and the second leading cause of cancer-related deaths [Microsatellite instability (MSI) is the abnormal shortening or lengthening of 1–6 repeat base pair units of DNA caused by inactivation of the MMR system [Recently, artificial intelligence has become a research hotspot in medicine due to the potential to achieve high-precision automated diagnosis of heterogeneous diseases. Skrede et al. [Based on simple clinicopathological indicators and with reference to previous studies, four machine learning models and a logistic regression model were developed in this study to predict CRC lacking DNA MMR, aid clinicians in identifying MMR status, and provide a reference for a precise treatment plan for patients.

PMC9969972

2. Materials and Methods

PMC9969972

2.1. Study Population

CRC

MAY, COMPLICATIONS

Retrospective analysis of 2279 patients of CRC with confirmed diagnosis at Wuhan Union Hospital from May 2017 to December 2019 was done. Patients with the following conditions were excluded from the study: (i) no MMR status outcome, (ii) no complete clinical data, and (iii) history of radiotherapy and chemotherapy prior to MMR status identification. A total of 2279 patients were enrolled in our study and randomly assigned to train and test sets in a 7-to-3 ratio. The consensus criteria for dMMR protein diagnosis were to select CRC patients who met the Revised Bethesda Guidelines (RBG) and then underwent MSI testing and/or immunohistochemical staining for MMR protein. This study was approved by the Ethics Committee of Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (No. 2018-S377). All patients signed an informed consent form stating that they understood the procedure and its potential complications and agreed to participate in this study.

PMC9969972

2.2. Data Collection

squamous cell carcinoma antigen, tumor

TUMOR

Baseline clinicopathological information on the patients obtained from the hospital's medical records included the following serum tumor markers: carcinoembryonic antigen (CEA), glycoantigen 19-9 (CA19-9), glycoantigen 12-5 (CA12-5), glycoantigen 72-4 (CA72-4), glycoantigen 15-3 (CA15-3), alpha-fetoprotein (AFP), serum squamous cell carcinoma antigen (SCC), ferritin (FERR), cytokeratin 19 fragment cyfra21-1 (CYFRA21-1), serum neuron-specific enolase (NSE), pathological type, histological type, age, sex, location, diameters, number of sampled lymph nodes (LNs), number of positive LNs, T-stage, N-stage, M-stage, perineural invasion, and vascular invasion. MMR status was assessed by immunohistochemistry (IHC) and was determined by MSH2, MSH6, MLH1, and PMS2 markers. We defined dMMR as a lacking expression of one or more MMR proteins, while tumor with intact MMR proteins was categorized as pMMR.

PMC9969972

2.3. Four Machine Learning Classifiers and a Conventional Logistic Regression Model

NB

REGRESSION

In this study, we built four machine learning models (extreme gradient boosting (XGBoost), support vector machine (SVM), naive Bayes (NB), and random forest (RF)) and a conventional logistic regression (LR) model using the caret package for R language (version 6.0-90) to diagnose dMMR discriminatively. An analysis of colinearity was performed on the initial 23 variables to exclude significantly correlated variables. Subsequently, least absolute shrinkage and selection operator (LASSO) regression and RF were used for variable selection. LASSO regression is known to be able to remove unimportant variables via the regression coefficients penalizing the size of the parameters. Applying the LASSO regression method, feature selection and predictive signature building were done. LASSO regression shrinks the coefficient estimates toward zero, with the degree of shrinkage dependent on an additional parameter,

PMC9969972

2.4. Data Analysis

Continuous variables between the dMMR and pMMR groups were analyzed using the Student

PMC9969972

3. Results

PMC9969972

3.1. Patient Characteristics

tumor-related

We screened 3566 patients with CRC, and 2279 eligible patients were enrolled in our study. All eligible patients were recruited from Wuhan Union Medical College Hospital. In a ratio of 7 : 3, 1595 patients were allocated to the training group and 684 to the testing group. The detailed screening process is shown in The demographic, clinical, and tumor-related characteristics of the patients included in the study are summarized in

PMC9969972

3.2. Construction of Predictive Models

Twenty-three variables were initially included based on the simple clinicopathological data of the patients. The collinearity between variables was excluded before modelling. The results of the variable correlation analysis (

PMC9969972

3.3. Performance of Models

CRC

The 2279 patients with CRC were randomly divided into training and test sets in a 7 : 3 ratio. The receiver operating characteristic (ROC) curve was used to evaluate the performance of the four machine learning models and LR model. As shown in

PMC9969972

3.4. Variable Importance Analysis

REGRESSION

We performed feature importance analysis for the variables selected using LASSO regression and random forest, respectively, and the results are displayed in Figures

PMC9969972

4. Discussion

MSI, tumor, colorectal cancer, CRC

REGRESSION, TUMOR, COLORECTAL CANCER

CRC remains a major healthcare burden with a high mortality rate worldwide [In previous studies, [Few studies have built machine learning models based on simple clinicopathological indicators to predict dMMR or MSI status in patients with CRC. Notably, several studies have been reported to use deep learning methods based on pathologically stained images to identify the MMR/MSI status in patients with CRC. Echle et al. [The development of scoring systems has also been effective in predicting the dMMR or MSI status in patients with CRC. Jenkins et al. [Our previously established scoring model incorporated nine clinicopathological features: age, location, tumor diameter, degree of differentiation, number of sample lymph nodes, PNI, number of positive lymph nodes, CA72-4, and CEA. The machine learning model built in this study excluded the CEA feature and included the three features NSE, N staging, and TNM staging, which were the results of a series of feature screenings. Compared with similar studies, we included a relatively large number of features, but the predictive performance of the model did not correlate with the number of indicators included. Of these, tumor location and size contributed most to the model. The comparison revealed that four characteristics—age, tumor location, Crohn's-like reaction, and TIL—were frequently included in the scoring model and had high coefficients. The difference is that although in the current study the tumor location is divided into rectum and colon, it was not additionally classified into proximal, distal, or left and right; nor were the two features Crohn's-like reaction and TIL included.We analyzed the strengths and weaknesses of each model. Support vector machines have the advantage of being able to perform linear and nonlinear classification and regression but struggle to deal effectively with complex and large data. RF and gradient boosting (e.g., XGBoost) have the advantage of being able to understand the importance of each feature for prediction, explain how decisions are made, and are easier to train and tune. The disadvantage is that these models are unsuitable for regression [Current research on the prediction of MMR status in colorectal cancer has two main approaches. One is to build a deep learning model to predict the MMR status by identifying the pathologically stained sections. The second is to build scoring models, filter the final incorporated features of the models through univariate and multivariate analyses, and then predict the MMR status. The novelty of this study is the combination of artificial intelligence methods and simple clinicopathological indicators to predict the status of MMR in patients with CRC.

PMC9969972

5. Conclusions

tumor, CRC

REGRESSION, TUMOR

In this study, we built four sets of machine learning models and a conventional logistic regression model to predict the lack of DNA MMR in patients with CRC based on simple clinicopathological indicators. Our results show that machine learning models can be incorporated with accurate and consistent predictive behavior. In fact, machine learning models show better performance at identifying dMMR than the conventional logistic regression methods. To the best of our knowledge, this study is the first to propose a machine learning approach to analyze and model the MMR status of patients based on simple clinicopathology and tumor markers. In addition, our single-center sample was sufficiently large to draw conclusions with some reference values. In future studies, we aim to incorporate TIL and Crohn's-like response features into the prediction model, refine some of the features such as the location of the tumor, and add an external validation group to improve the predictive power of the model.

PMC9969972

6. Limitations

tumor

TUMOR, INFILTRATION

Our study has some limitations. First, the population in our cohort comprised of persons from one region of China (Wuhan), which may limit the generalizability of the predictive models and require further validation in patients from different geographic regions. Second, this was a nonrandomized retrospective analysis. Therefore, potentially biased comparisons such as in the inclusion of patients or sample selection bias could have occurred. Third, the included indicators need to be refined; for example, the location of the tumor needs to be subdivided into left or right and proximal or distal colon. Crohn's-like reaction and tumor lymphocyte infiltration also need to be included in the model. Finally, we only performed internal validation in this study, and further external validation groups are needed to verify the predictive effect.

PMC9969972

Acknowledgments

This study was supported by the National Natural Science Foundation of China (grant number 82170678), Hubei Province Key Research and Development Program of China (Science and Technology Innovation Special Project) (grant number 2021BAA04 4), and Wuhan Strong Magnetic Field Interdisciplinary Fund (grant number WHMF202113). We thank Lizhao Yan for the support of statistical analysis.

PMC9969972

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding authors on reasonable request.

PMC9969972

Ethical Approval

This study was performed in line with the principles of the Declaration of Helsinki. Studies involving human participants were reviewed and approved by the Ethics Committee and the Institutional Review Committee of Wuhan Union Medical College (No.2018-S377).

PMC9969972

Consent

Informed consent was obtained from all individual participants included in the study. The recruited volunteers were requested to sign an informed consent form. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

PMC9969972

Disclosure

Some part of our manuscript was previously published as a preprint as per the following link:

PMC9969972

Conflicts of Interest

The authors declare that they have no competing interests.

PMC9969972

Authors' Contributions

Conceptualisation was done by Zhenxing Jiang, Yinghao Cao, Lizhao Yan, and Shenghe Deng. Acquisition of data, analysis, and interpretation of data were done by Junnan Gu, Le Qin, Fuwei Mao, Yifan Xue, Fumei Shang, and Wentai Cai. Writing—original draft—was done by Zhenxing Jiang, Shenghe Deng, Junnan Gu, and Le Qin. Writing—review and editing—was done by all authors. Supervision was done by Ke Wu, Kailin Cai, Xiu Nie, Hongli Liu, Kaixiong Tao, and Jiliang Wang. Zhenxing Jiang, Lizhao Yan, and Shenghe Deng contributed equally to this work.

PMC9969972

Supplementary Materials

tumor, NB

REGRESSION, COLORECTAL CANCER, TUMOR

Analytical diagram of the percentage of the situation within each variable. We performed a statistical analysis of the individual signs of the included patients, which is presented in the form of a bar chart that clearly shows the proportion of the number of patients for each variable.Click here for additional data file.Display of variable coefficients in logistic regression model. For the logistic regression model, the coefficients assigned to each feature were recorded.Click here for additional data file.LASSO regression feature filtering. LASSO (least absolute shrinkage and selection operator) regression based on five times tenfold cross-validation was used for feature selection.Click here for additional data file.Patient screening process. The detailed process of patient selection.Colinearity analysis. Variables exhibiting colinearity were excluded from variate analysis. The darker blue color indicates higher colinearity.LASSO regression feature filtering. LASSO (least absolute shrinkage and selection operator) regression based on five times tenfold cross-validation was used for feature selection.Random forest feature filtering. Random forest based on five times tenfold cross-validation was used to perform feature selection.Receiver operating characteristic (ROC) curves of predictive models. Diagnostic abilities of predictive models for the differential diagnosis of dMMR and pMMR in the test set. ROC curves of predictive model created by LR, NB, RF, SVM, and XGBoost.Significance analysis by LASSO regression. For the LASSO (least absolute shrinkage and selection operator) regression, we give the normalized regression coefficients for each feature.Significance analysis by Random forest. We used the machine learning technique and random forest, to determine feature importance.Variable importance analysis. The merged variables were performed for feature importance analysis by random forest.Clinical characteristics of the patients with colorectal cancer.Risk factors for deficient MMR in colorectal cancer. Five predictive models based on simplified clinicopathological features and serum tumor biomarkers. Performance of different predictive models to identify dMMR. Accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of five predictive models in the test set.

PMC9969972

Subject terms

Sheehan Disability, post-traumatic stress disorder, PTSD, DSM-5

SECONDARY

This multi-site, randomized, double-blind, confirmatory phase 3 study evaluated the efficacy and safety of 3,4-methylenedioxymethamphetamine-assisted therapy (MDMA-AT) versus placebo with identical therapy in participants with moderate to severe post-traumatic stress disorder (PTSD). Changes in Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) total severity score (primary endpoint) and Sheehan Disability Scale (SDS) functional impairment score (key secondary endpoint) were assessed by blinded independent assessors. Participants were randomized to MDMA-AT (Results from the phase 3 placebo-controlled MAPP2 trial show that MDMA-assisted therapy reduces post-traumatic stress disorder (PTSD) symptoms and functional impairment in a diverse population with moderate to severe PTSD.

PMC10579091

Main

trauma, PTSD, chronic sexual abuse, stress disorder

Post-traumatic stress disorder (PTSD) is a serious neuropsychiatric condition affecting approximately 5% of the US population each yearMounting evidence supports substituted phenethylamine 3,4-methylenedioxymethamphetamine-assisted therapy (MDMA-AT) as a treatment for PTSDDue to disparities in trauma exposure, gender-diverse and transgender individuals, ethnoracial minorities, first responders, military personnel, veterans and victims of chronic sexual abuse have a disproportionately higher risk of developing PTSD

PMC10579091

Results

PMC10579091

Demographics and baseline characteristics

MAY

Participants were recruited from 21 August 2020 to 18 May 2022 (last participant visit on 2 November 2022). Overall, 324 individuals were screened, and 121 were enrolled. Of these, 17 individuals did not meet enrollment confirmation after initiation of preparation therapy, and 104 were confirmed for randomization: 53 were assigned to MDMA-AT and 51 to placebo with therapy (Fig.

PMC10579091

Primary outcomes

PTSD

MDMA-AT significantly attenuated PTSD symptomology versus placebo with therapy, as measured by a reduction in CAPS-5 total severity score from baseline to 18 weeks. Mixed models for repeated measures (MMRM) analysis of the de jure estimand showed a least squares (LS) mean (95% confidence interval (CI)) change of −23.7 (−26.94, −20.44) for MDMA-AT versus −14.8 (−18.28, −11.28) for placebo with therapy (treatment difference: −8.9 (−13.70, −4.12),

PMC10579091

Secondary outcomes

Sheehan Disability

MDMA-AT significantly mitigated clinician-rated functional impairment, as measured by a reduction in the Sheehan Disability Scale (SDS) from baseline. MMRM analysis of the de jure estimand revealed that the LS mean change (95% CI) in SDS total scores was −3.3 (−4.03, −2.60) with MDMA-AT versus −2.1 (−2.89, −1.33) with placebo with therapy (treatment difference: −1.20 (−2.26, −0.14);

PMC10579091

Exploratory outcomes

In the MDMA-AT group, 45 of 52 (86.5%) participants were responders with a clinically meaningful improvement at 18 weeks after baseline, defined as a ≥10-point reduction in CAPS-5 total severity score, versus 29 of 42 (69.0%) in the placebo with therapy group (Fig.

PMC10579091

Discussion

psychiatric, PTSD, comorbid personality disorders

CARDIOVASCULAR DISEASE

In this confirmatory phase 3 study of participants with moderate to severe PTSD, MDMA-AT significantly improved PTSD symptoms and functional impairment, as assessed by CAPS-5 and SDS, respectively, compared to placebo with therapy over 18 weeks. Notably, 45 of 52 (86.5%) participants treated with MDMA-AT achieved a clinically meaningful benefit, and 37 of 52 (71.2%) participants no longer met criteria for PTSD by study end. In a historic first, to our knowledge, for psychedelic treatment studies, participants who identified as ethnically or racially diverse encompassed approximately half of the study sample. These findings confirm and extend the results observed in MAPP1 (ref. Given the diverse population and degree of participant complexity, the replication of efficacy is particularly notable. In our study, 26.9% (28/104) of participants expressed moderate PTSD, whereas, in MAPP1, all participants expressed severe PTSDMDMA simultaneously induces prosocial feelings and softens responses to emotionally challenging and fearful stimuliConsistent with MAPP1, no new major safety issues were reported. Common TEAEs were similar to previous studies and consistent with expected effects of MDMAAlthough the sample sizes of the MAPP1 and MAPP2 phase 3 studies had 90% statistical power and were developed with guidance from the FDA to ensure adequate, rigorous testing of outcomes, these evaluations did not extend further than 2 months after therapy and were intended to support an acute treatment course. To support these studies, data from the ongoing follow-up of participants from phase 2 and 3 studies (ClinicalTrials.gov Identifier: Several limitations may impact the integration of MDMA-AT into clinical care, including the exclusion of participants with high suicide risk, comorbid personality disorders and underlying cardiovascular disease. Observed effect sizes for MDMA-AT (between-group, The notable effect seen in the placebo with therapy arm could suggest the standalone value of the manualized inner-directed therapy that was developed for use with MDMA. Additional head-to-head studies will need to be conducted to evaluate whether this form of manualized therapy provides greater value in the treatment of PTSD than the current first-line cognitive behavioral therapy and prolonged exposure therapy treatmentsAlthough treatment expectancy, per se, was not measured in this study, prospective treatment expectancy would likely have been high in both study arms, with random assignment expected to distribute this equally between groups. Although expectancy effects are a well-known issue in psychiatric clinical trials and are intertwined with the observation of treatment benefit during a trialThe therapists who participated in this study were required to complete the sponsor’s training program (see This confirmatory phase 3 trial showed consistent benefits of MDMA-AT in an ethnoracially diverse group of individuals with longstanding moderate to severe PTSD and numerous comorbidities. The dropout rate was low, and treatment was generally well tolerated. These findings represent the culmination of over two decades of research

PMC10579091

Methods

PMC10579091

Study design and oversight

PTSD

This multi-site, randomized, double-blind, placebo-controlled study assessed the efficacy and safety of MDMA-AT versus placebo with therapy in participants diagnosed with moderate or severe PTSD (

PMC10579091

Participants

PTSD, DSM-5

After written informed consent, participants were screened for eligibility. Adults (≥18 years of age) meeting the full DSM-5 criteria for current PTSD per CAPS-5 assessment

PMC10579091

Randomization and masking

Participants were randomized in a 1:1 allocation and in a blinded fashion to the MDMA-AT and placebo with therapy groups, stratified by clinical site. Randomization was managed via an interactive web randomization system (IWRS) (IT Clinical version 11.0.1) based on a centralized randomization schedule developed by an independent third-party vendor to maintain blinding.A central pool of blinded independent assessors was used to mitigate the risk of functional unblindingTo ensure that all site and sponsor staff were shielded from study outcomes, the blinded independent assessor pool collected and stored outcome measures in a dedicated database that was separate from the blinded clinical database. A blinding survey was conducted at study termination (visit 20) to assess if participants thought that they received MDMA or placebo.

PMC10579091

Procedures

Trial procedures were consistent with MAPP1 (ref. Within the MDMA-AT group, three participants did not undergo dose escalation in experimental sessions 2 and 3, and two participants experienced dose administration timing errors (Supplementary Table

PMC10579091

Outcomes

ideation, death, congenital anomaly, birth defect, TEAEs, PTSD, impairment or damage, disability or incapacity, Suicidality

ADVERSE EVENTS, EVENT, SECONDARY, ADVERSE EVENT, EVENTS

Independent assessors conducted CAPS-5 and SDS outcome assessments at baseline, after experimental sessions 1 and 2 and 6–8 weeks after experimental session 3 (18 weeks after baseline) via video interviews. Primary and secondary objectives were mean change in CAPS-5 total severity and SDS scores, respectively, for MDMA-AT versus placebo with therapy from baseline to 18 weeks after baseline.Exploratory outcome measurements included characterization of the treatment response and differences between the treatment groups by demographics and characteristics. Responder analyses were based on categorical diagnostic assessment data and the CAPS-5 total severity score assessment. PTSD severity was defined using the CAPS-5 total severity score as follows: asymptomatic (0–10), mild (11–22), moderate (23–34), severe (35–46) and extreme (47+) (ref. Safety objectives included assessment of differences between groups in severity, incidence and frequency of TEAEs, serious TEAEs, TEAESIs, suicidal ideation and behavior and vital signs. TEAEs were defined as any adverse event that occurred during the treatment period from the first experimental session to the last integration session. The severity of TEAEs was determined by the site physician as mild (no limitation in normal daily activity), moderate (some limitation in normal daily activity) or severe (unable to perform normal daily activity). A serious TEAE was defined as any unforeseen medical event at any dose of the drug that resulted in death; was life-threatening; required inpatient hospitalization; caused significant disability or incapacity; resulted in a congenital anomaly or birth defect; or required intervention to prevent permanent impairment or damage. Serious TEAEs also included any event, based on medical judgement, that jeopardized the participant or may have required intervention to prevent one of the events listed previously. With the exception of serious adverse event reporting, relatedness to study drug was not assessed by investigators, to preserve blinding. In an effort to identify common adverse events that may be most related to MDMA, TEAEs occurring with incidence >10% and at least twice the prevalence in the MDMA-AT group versus the placebo with therapy group are reported. Suicidality was tracked at each study visit using the C-SSRS (see the

PMC10579091

Statistical analysis

depression, PTSD, TEAEs

DISEASE, SECONDARY, DISORDERS, DISORDER

SAS version 9.4 (SAS Institute) was used for analyses. Sample size was calculated to achieve a power of 90% at an alpha of 0.0499.Efficacy was tested using an MMRM analysis comparing the change from baseline to 18 weeks after baseline in CAPS-5 and SDS scores between treatment groups in two-sided tests with alpha set at 0.0499. The alpha was adjusted to account for an administrative interim analysis for sample size re-estimation conducted after all participants were enrolled and 60% of primary endpoint data had been collected. Fixed effects were treatment, visit, treatment group by visit interaction and dissociative subtype; baseline CAPS-5 score was a covariate. Primary and secondary efficacy analyses used a de jure (related to initially randomized treatment) estimand and a supportive de facto (treatment policy) estimand of the modified intention-to-treat population, which required exposure to MDMA or placebo and at least one follow-up CAPS-5 assessment, as in MAPP1 (ref. In additional exploratory analyses, 13 covariates were assessed in the model, with alpha set at 0.0499: age, sex (self-reported), prior use of selective SSRIs, work disability, disease severity, PTSD duration, dissociative subtype, overnight site stay, site ID, moderate depression (as measured by the BDI-II), severe adverse childhood experiences and moderate alcohol and substance use disorder risk (as measured by the Drug Use Disorders Identification Test and the Alcohol Use Disorders Identification Test). Analyses of primary or secondary outcomes by gender were not planned a priori; some exploratory analyses included sex as a covariate (Safety analysis evaluated TEAEs at the participant level, including all participants who received MDMA or placebo. Causal association with MDMA was determined based on relative incidence of TEAEs with at least a two-fold difference between groups.

PMC10579091

Adverse events of special interest

ideation, sudden death, cardiac arrhythmias, overdose, syncope, seizures, MDMA abuse, self-harm

SUDDEN DEATH, CARDIAC ARRHYTHMIAS, VENTRICULAR TACHYCARDIA, ADVERSE EVENTS, TORSADE DE POINTES, VENTRICULAR FIBRILLATION AND FLUTTER, VENTRICULAR EXTRASYSTOLES

In accordance with FDA guidance, special attention was paid to a subset of adverse events, TEAESIs, relating to cardiac function, suicide risk and MDMA abuse, misuse or diversion. TEAESIs involving cardiac function that could be indicative of QT prolongation or cardiac arrhythmias were collected, including torsade de pointes, sudden death, ventricular extrasystoles, ventricular tachycardia, ventricular fibrillation and flutter, non-postural syncope and seizures. TEAESIs involving suicide risk included suicide, suicide attempts, self-harm associated with suicidal ideation, suicide ideation assessed as a score of 4 or 5 on the C-SSRS and suicidal ideation judged by the investigator to be serious/severe. TEAESIs involving terms of MDMA abuse, misuse, drug diversion, dependence or overdose were also collected.

PMC10579091

Reporting summary

Further information on research design is available in the

PMC10579091

Online content

Any methods, additional references, Nature Portfolio reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-023-02565-4.

PMC10579091

Supplementary information

CONSORT checklist, MAPP2 study collaborators, Supplementary Methods, Supplementary Figs. 1 and 2, Supplementary Tables 1–9 and Supplementary References.Reporting Summary

PMC10579091

Supplementary information

The online version contains supplementary material available at 10.1038/s41591-023-02565-4.

PMC10579091

Acknowledgements

CARPENTER, PBC

The authors thank all of the participants and their support networks. See the This study was funded by Multidisciplinary Association for Psychedelic Studies (MAPS) with support from the Steven and Alexandra Cohen Foundation and organized by MAPS Public Benefit Corporation (PBC). MAPS PBC was responsible for overseeing the collection, analysis and interpretation of the data. Medical writing assistance was provided by J. Carpenter and M. Yochum of BOLDSCIENCE, funded by MAPS PBC.

PMC10579091

Author contributions

S.H., B.Y.-K., C.H. and A.d.B. contributed to data analysis; these authors and J.M.M. directly accessed and collectively verified the underlying data. All authors had full access to the trial data, contributed to the interpretation of the data and contributed to writing the manuscript. The authors attest to the accuracy and completeness of the reported data, accept responsibility to submit for publication and confirm that the trial conformed to the protocol and the statistical analysis plan (available via

PMC10579091

Peer review

PMC10579091

Data availability

The data that support the findings of this study are available from the sponsor beginning 1 year after completion of the trial. However, restrictions apply to the availability of these data, which were used under license for the current study and so are not publicly available. Data are, however, available from the authors upon reasonable request and with the permission of the sponsor. All requests for raw and analyzed data are promptly reviewed to verify if the request is subject to any confidentiality obligations. Participant-related data not included in the paper were generated as part of clinical trials and may be subject to participant confidentiality. Any data that can be shared will be released via a data use agreement. Proposals should be directed to

PMC10579091

Code availability

Commercially available software (SAS version 9.4 or higher, SAS Institute) was used for analyses, in keeping with the statistical analysis plan.

PMC10579091

Competing interests

PBC

J.M.M. has received research support from MAPS; grants/contracts from the Veterans Administration (Merit Award) and the FDA (Research Award); has received royalties/licenses from UCLA (for a patent licensed to UCSF for cell screening); has received payment/honoraria from Stanford (for lecturing to undergraduate students) and Johns Hopkins (for presenting grand rounds); has a patent licensed to UCSF for cell screening; has been a reviewer for NIAAA CTN; has been a member of CA DOJ RAP; and has been a grant reviewer for the Australian Medical Research Council. M.O.G.: Aguazul-Bluewater, Inc has received research support from MAPS PBC and payments from Cybin (training and consultation), from Horizons Conference and from Naropa University. B.v.d.K. has received royalties from Penguin Random House (book,

PMC10579091

References

PMC10579091

Supplementary Material

cirrhosis, end-stage liver disease

CIRRHOSIS

To explore the potential mechanisms underlying the effects of a probiotic in cirrhotic patients, we analyzed the blood metabolome using proton nuclear magnetic resonance (In a previous double-blinded, placebo-controlled, randomized clinical trial, we observed that a multistrain probiotic improved cognitive function and risk of falls, reinforced the intestinal barrier, and modulated the proinflammatory state in patients with cirrhosis.We analyzed serum metabolome in samples from 32 outpatients with cirrhosis included in the study mentioned above.Patients had a mean age of 64.5 years, 65.6% were women, the etiology of cirrhosis was alcohol in 50%, the mean model for end-stage liver disease (MELD) score was 9.3, 81.2% had previous decompensations, 84.3% had previous falls, and 28.1% presented a PHES <−4. There were no statistical differences between the 2 groups at baseline.The untargeted observation of the serum metabolome by (A) Volcano plots showing the change between baseline and 12 weeks in all metabolites identified in the probiotic group and in the placebo group. (B) Changes in glutamine/glutamate ratio in patients treated with the probiotic and patients treated with placebo. Results are expressed as mean (SD). Abbreviation: FDR, false discovery rate.Regarding metagenomic results, when we focused on the serum glutamine/glutamate ratio, analyzing all the 30 samples at the genus level, we observed this ratio was associated with an abundance of The increase in the serum glutamine/glutamate ratio observed in the present study in probiotic-treated patients could reflect a higher capacity of ammonia clearance because ammonia binds to glutamate to generate glutamine.Our findings are in line with a recent study in rats with bile duct ligation, which showed the multistrain probiotic used here ameliorated locomotor activity and improved the neurometabolic profile assessed by in vivo brain ultrahigh field The exact mechanisms underlying the increase in the glutamine/glutamate ratio cannot be elucidated from the present study. One possible mechanism is that the probiotic could induce a decrease in glutaminase activity or an increase in glutamine synthetase activity, as has been observed in experimental models with other treatments for HE, such as rifaximinRegarding gut microbiota composition, with the limitations of the small sample size, here we focused on the glutamine/glutamate ratio and found that higher values in blood were associated with lower abundance of In conclusion, metabolomic analysis in patients with cirrhosis treated with this multistrain probiotic showed an increase in the serum glutamine/glutamate ratio. This finding contributes to the understanding of the mechanisms involved in the effects of this probiotic and could help in the design of new approaches to improve patients’ outcomes and health-related quality of life.

PMC10079330

ACKNOWLEDGMENTS

The authors thank Carolyn Newey for English language revision.

PMC10079330

FUNDING INFORMATION

This study was partially funded by grant PI12/00629 from the Instituto de Salud Carlos III. Mendes S.A. (Lugano, Switzerland) also partially funded the study.

PMC10079330

CONFLICTS OF INTEREST

The authors have no conflicts to report. Luca Laghi and Eva Román are co-first authors.Luca Laghi and German Soriano are co-corresponding authors.Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal’s,

PMC10079330

REFERENCES

PMC10079330

Background

caries

HYDROXYAPATITE, CARIES

Edited by: Dominic Augustine, M. S. Ramaiah University of Applied Sciences, IndiaReviewed by: Ralitsa Raycheva, Plovdiv Medical University, Bulgaria; Rodolfo Reda, Sapienza University of Rome, ItalyDental caries is a worldwide challenge for public health. The aim of this 18-month double-blinded, randomized, clinical trial was to compare the caries-preventing effect of a fluoride-free, hydroxyapatite toothpaste (test) and a toothpaste with sodium fluoride (1450 ppm fluoride; positive control) in adults.

PMC10393266

Methods

The primary endpoint was the percentage of subjects showing no increase in overall Decayed Missing Filled Surfaces (DMFS) index. The study was designed as non-inferiority trial. Non-inferiority was claimed if the upper limit of the exact one-sided 95% confidence interval for the difference of the primary endpoint DMFS between test and control toothpaste was less than the predefined margin of non-inferiority (Δ ≤ 20%).

PMC10393266

Results

HYDROXYAPATITE

In total, 189 adults were included in the intention-to-treat (ITT) analysis; 171 subjects finished the study per protocol (PP). According to the PP analysis, no increase in DMFS index was observed in 89.3% of subjects of the hydroxyapatite group and 87.4% of the subjects of the fluoride group. The hydroxyapatite toothpaste was not statistically inferior to a fluoride toothpaste with regard to the primary endpoint.

PMC10393266

Conclusion

Hydroxyapatite was proven to be a safe and efficient anticaries agent in oral care.

PMC10393266

Clinical trial registration

NCT04756557.

PMC10393266

1. Introduction

caries

CAVITY, HYDROXYAPATITE, CARIES

Hydroxyapatite, CaThe modes of action of hydroxyapatite in the oral cavity are based on physical, biochemical, and biological principles (The use of fluoride-free hydroxyapatite toothpastes has in recent years been shown to be a clinically proven approach to caries prevention (The efficacy of hydroxyapatite-based toothpastes in preventing caries has been demonstrated in adolescents (

PMC10393266

2. Materials and methods

PMC10393266

2.1. Objectives

caries

HYDROXYAPATITE, CARIES

The study objective was the investigation of the caries-preventing effect in adults aged 18–45 regularly using a fluoride-free, hydroxyapatite toothpaste (test toothpaste) or the fluoridated control toothpaste. The aim of the study was to compare the caries preventive effect of the test and control toothpaste to prove the non-inferiority of the test toothpaste compared with the control toothpaste.

PMC10393266

2.2. Study design

POLAND

The study was performed as a two-centered, double-blinded, randomized, and active-controlled parallel-group study with two arms (control and test). The study was performed at two study centers, i.e., Universities of Medical Sciences, Poznan and Bialystok, Poland.

PMC10393266

2.3. Inclusion and exclusion criteria

Systemic disorders, allergy, caries, DMFS, tooth

SEVERE PERIODONTITIS, ALLERGY, CARIES

The following inclusion and exclusion criteria were applied:A) Inclusion criteria• Provision of written informed consent.• Age 18-45 years (both men and women).• A minimum of 10 caries-free (DMFS index=0) molars and premolars.• Willing to use an electric (powered) toothbrush.B) Exclusion criteria • Untreated caries (subjects with untreated caries in need of restoration can become eligible after restorative therapy).• Severe periodontitis at the baseline visit (pocket depth on at least one tooth ≥ 5.5 mm).• Undergoing orthodontic treatment.• Known allergy to one of the ingredients of the study toothpastes.• Systemic disorders interfering with salivary function or flow rate.• Regular medication intake interfering with salivary function or flow rate. • Participation in any other clinical study within the past 3 months or ongoing.• Lack of intellectual or physical ability to conduct the study per protocol.• Any other reason that, in the opinion of the investigator, disqualifies the subject from participating in the study.

PMC10393266

2.4. Primary endpoint

Percentage of subjects showing no increase in overall Decayed Missing Filled Surfaces (DMFS) index (DMFS

PMC10393266

2.5. Secondary endpoints

PLAQUE, PLAQUE

A) Percentage of subjects experiencing no change in mineral density (as analyzed by laser diode near-infrared light transillumination of dental tissues with the use of DIAGNOcam) during the observation period of 18 months.B) Changes in the coverage of all teeth with bacterial plaque according to the criteria of the Plaque Control Record (PCR) during the observation period of 18 months (

PMC10393266

2.6. Toothpastes (blinded)

HYDROXYAPATITE

The test toothpaste and the active control toothpaste were provided in neutral plastic tubes all having the same shape; thus, both the subjects and the investigators were blinded. It is pertinent to mention that the subjects were requested not to discuss anything related to the toothpastes with any of the study team member who was involved in the clinical examinations. The neutral plastic tubes were labeled with a random subject number for each subject. The toothpastes were produced by an external certified laboratory. The composition of the toothpastes for both groups was identical (except for the main active ingredients, i.e., hydroxyapatite and fluoride). The compositions of the study toothpastes were as follows.

PMC10393266

2.6.1. Test toothpaste—hydroxyapatite toothpaste (fluoride-free)

HYDROXYAPATITE

Aqua, hydrated silica, glycerin, hydrogenated starch hydrolysate, hydroxyapatite (10%), xylitol, silica, cellulose gum, sodium methyl cocoyl taurate, sodium sulfate, 1,2-hexanediol, aroma, caprylyl glycol, sodium cocoyl glycinate.

PMC10393266

2.6.2. Active control toothpaste—fluoride toothpaste

HYDROXYAPATITE

Aqua, hydrated silica, glycerin, hydrogenated starch hydrolysate, xylitol, silica, cellulose gum, sodium methyl cocoyl taurate, sodium sulfate, 1,2-hexanediol, aroma, caprylyl glycol, sodium fluoride (1,450 ppm fluoride), sodium cocoyl glycinate.The composition of the hydroxyapatite test toothpaste was comparable to commercially available Karex toothpastes (Dr. Kurt Wolff GmbH & Co. KG, Bielefeld, Germany).

PMC10393266

2.7. Application

Tooth

Tooth brushing was performed twice a day (morning and evening; after the meals) for 3 min per episode. No other fluoride- and/or hydroxyapatite-containing oral care products, professional or home use, such as mouthwashes, gels, or fluoridate supplements, were allowed during the trial.

PMC10393266

2.8. Toothbrushes

Electric (powered) toothbrushes (Oral-B; P&G, Schwalbach, Germany) were used by the subjects, and the brushing heads were changed every 2 months.

PMC10393266

2.9. Clinical examinations

carious lesion, caries, tooth

PLAQUE, CARIES, PLAQUE

DMFS Index calculation (• When a carious lesion or both a carious lesion and a restoration were present, the surface was listed as a D.• When a tooth was extracted due to caries, it was listed as an M.• When a permanent filling was present, or when a filling was defective but not decayed, this surface was counted as an F. Surfaces restored for reasons other than caries were not counted as an F. The total count was 128 surfaces for a maximum of 28 teeth. DIAGNOcam was used according to the instructions of the manufacturer (KaVo Dental, Biberach, Germany). The following classification was used (0 = Light transmission unchanged1 = Shadow visible in enamel2 = Shadow visible in dentin The Plaque Control Record (PCR) is an established method of recording the presence of the plaque on individual tooth surfaces (

PMC10393266

2.10. Course of the study

caries, tooth

CARIES

Potential subjects were informed by the investigator about the nature, significance, and scope of the clinical trial according to the requirements described in the written subject information. Before study enrollment, the willingness of the subjects to properly follow the study protocol throughout the 18 months of the study was assessed. Subjects were enrolled as study participants only when they had given their written informed consent. Subjects had to meet all inclusion criteria and no exclusion criteria. Subjects disqualified due to untreated caries became eligible after restorative therapy.Once informed consent had been given, an initial examination took place that covered the following aspects:Screening subjects for study eligibility (inclusion and exclusion criteria).Collection of demographic data.Baseline data collection performed in the following sequence: PCR, DMFS, DIAGNOcam.After the analysis of the PCR record using a plaque-staining solution, a professional tooth cleaning was performed. The professional tooth cleaning was performed at visits 1 and 4 only, and there was no fluoride application after the cleaning.Finally, the study subjects received an electric toothbrush with three brushing heads (replaced every 2 months) and the allocated toothpaste (test or control) by a trained study nurse or dentist not involved in clinical study examinations.The proper use of the assigned electric toothbrush and the issued toothpaste were also instructed by this study nurse or a dentist not involved in the clinical study examinations. One-hundred eighty-two days after the baseline visit, the PCR and DMFS were reassessed as described for the baseline visit. Subsequently, a study nurse or a dentist who has not involved in clinical study examinations handed out three new brushing heads for the electric toothbrush and a new supply of the assigned experimental toothpaste (test or control) for the next 182 days. Finally, subjects received a new appointment for visit 3. Three hundred sixty-four days after the baseline visit, the procedure performed on Visit 2 was repeated, after which the subjects received a new appointment for Visit 4. Five hundred forty-six days after the baseline visit, the following assessments were repeated: PCR, DMFS after professional cleaning, and DIAGNOcam as described from the baseline visit.An overview of the study course in detail is presented in Study flowchart.

PMC10393266

2.11. Methods of determining safety

ADVERSE EVENTS, CAVITY

Safety assessments were performed in all subjects at every visit. The clinical examiner visually examined each subject's oral cavity and perioral area as well as questioned the subjects on any adverse events. All observed or voluntarily reported adverse events (AEs) and serious adverse events (SAEs) regardless of the experimental group were recorded.

PMC10393266

2.12. Monitoring

RECRUITMENT

The monitor (Dr. Egmont Zieseniß, Inpharm Consulting, Germany) reviewed the case report forms with the investigator and his staff at regular intervals:First monitoring visit after 30% recruitment.Second monitoring visit after 12 months.Third monitoring visit at study close-out.During the study, the monitor reviewed the case report forms at the above intervals to verify completeness, plausibility and consistency of the data, protocol adherence, and the progress of enrolment, and to ensure that study supplies are being stored, dispensed, and accounted for according to specifications.

PMC10393266

2.13. Determination of sample size and statistical analysis

caries lesions

REGRESSION, SECONDARY

It was assumed that the primary endpoint (i.e., no increase in DMFS index, i.e., ΔDMFS = Visit 4 - Visit 1 ≤ 0) will occur in about 60% of study subjects. A sample size of 77 study subjects per arm (two arms) was calculated to be sufficient to reject the null hypothesis that the test toothpaste is inferior to the control toothpaste, using a non-inferiority margin of Δ = 20% for the primary endpoint one-sided chi-square test (α = 5%, power = 80%) as described in a previous study (A non-inferiority testing was performed. The non-inferiority margin Δ was set to 20% between the groups. The primary endpoint was analyzed for the per-protocol (PP) population and, in addition, repeated for sensitivity reasons for the ITT population. The confidence interval approach was used to test non-inferiority (The confidence interval approach was also performed to test non-inferiority of the secondary endpoint's percentage of subjects experiencing no increase in the overall number of caries lesions (as examined using DIAGNOcam) as explorative statistical evaluation. Moreover, a logistic regression analysis was performed for this secondary endpoint as a dependent variable and toothpaste, center, gender, and age as independent variables (covariates).A two-sided independent

PMC10393266

2.14. Ethical committee and study registration

POLAND

The study for both centers was approved by the ethical commission of the Poznan University of Medical Sciences, Poznan, Poland (No. 691/20; November 04, 2020). The study was registered at

PMC10393266

3. Results

PMC10393266

3.1. Subjects

HYDROXYAPATITE

Overall, 194 subjects were included: 97 (50%) applied the hydroxyapatite toothpaste (test toothpaste) and 97 (50%) applied the fluoride toothpaste (control toothpaste) as randomized. The study was prematurely terminated in 20 (10.3%) subjects for several reasons (test toothpaste: All subjects who performed at least one follow-up visit (V2, V3, or V4) were included in the intention-to-treat (ITT) population. The study was terminated in five subjects at visit V1 (baseline); i.e., these subjects were not included in the ITT population. In total, 189 (97.4%) subjects were included in the ITT population. Finally, 174 (89.7%) subjects completed visit V4 (final visit): hydroxyapatite toothpaste: The primary endpoint was analyzed for the PP population and, in addition, repeated for sensitivity reasons for the ITT population.

PMC10393266

3.2. Patient flowchart and analysis sets

The patient flow chart according to the CONSORT statement is shown in Patient flow chart.In the following, the results are reported for the PP population (“analyzed subjects”) if not mentioned otherwise.

PMC10393266

3.3. Demographic data

The demographic data of the PP population (Summary of the demographic data of the PP population.Age, weight, height, and sex did not differ significantly between groups (two-sided p-values > 0.10; t-test or chi-square test).

PMC10393266

3.4. Primary efficacy parameter—DMFS index

PMC10393266

3.4.2. Confirmative statistical test (primary analysis) and additional statistical tests

DMFS

REGRESSION

The primary efficacy endpoint was the percentage of subjects showing no increase in DMFS Index (ΔDMFS = DMFS Percentage of subjects without an increase of DMFS Index in the PP and ITT population.ΔDMFS%With ΔDMFS%ΔDMFS%The upper limit of the one-sided 95% confidence interval for the difference in percentage of subjects without an increase of DMFS (ΔDMFS%In addition, a logistic regression analysis was performed with the primary endpoint (increase in DMFS vs. no increase in DMFS) as dependent variable and toothpaste, center, sex, and age as independent variables (covariates).The results for the PP population confirmed that the “risk” of an increase in DMFS did not differ significantly between control and test toothpastes.Furthermore, the risk of an increase in DMFS did not differ significantly between centers, was not dependent on age, but was significantly higher in men compared to women. The corresponding odds ratio revealed that the “risk” of an increase in DMFS was 4.7-fold higher in men: An increase in DMFS was observed in 13 of 57 (22.8%) men and only in 7 of 114 (6.1%) women of the PP population.

PMC10393266

3.5. Secondary efficacy parameters

caries lesions

PLAQUE

Secondary efficacy parameters were (A) the overall number of caries lesions and (B) the coverage of all teeth with bacterial plaque according to the criteria of the PCR index (

PMC10393266

3.5.2. NCL—explorative statistical tests

NCL, caries

REGRESSION, SECONDARY, CARIES

The secondary efficacy endpoint for NCL is the proportion of subjects showing no increase in NCL% (ΔNCL = NCL% ΔNCL%With ΔNCL%ΔNCL%Percentage of subjects without increase of NCL% (percentage of caries lesions) in the PP set.The upper limit of the one-sided 95% confidence interval for the difference in proportion of subjects without an increase of NCL (ΔNCL%In addition, a logistic regression analysis was performed for the secondary endpoint NCL (increase in NCL vs. no increase in NCL) as dependent variable and toothpaste, center, sex, and age as independent variables (covariates). The results for the PP population confirmed that the “risk” of increase in NCL did not differ significantly between control and test toothpaste. Furthermore, the risk did not differ significantly between centers and was not dependent on age and sex.

PMC10393266

3.5.4. PCR—explorative statistical tests

A two-sided Mann–Whitney–Wilcoxon test was applied (instead of

PMC10393266

3.6. Safety

All randomized subjects (

PMC10393266

4. Discussion

caries lesions, caries, diet behaviors, tooth, NCL, orthodontic

SECONDARY, HYDROXYAPATITE, CARIES, DENTIN HYPERSENSITIVITY

In the present clinical trial, the effectiveness of hydroxyapatite toothpaste in preventing caries development was compared with that of fluoride (1,450 ppm) toothpaste, to determine the non-inferiority of the hydroxyapatite toothpaste to the fluoride toothpaste. Analysis of both the primary and the secondary outcome measures demonstrated non-inferiority of the two products to each other. This result is in agreement with two previously published clinical trials (The present study used a toothpaste with sodium fluoride (1,450 ppm fluoride) as control toothpaste. Sodium fluoride is one of the most frequently used fluoride compounds in fluoridated toothpastes worldwide (Approximately almost 90% of the subjects remained caries-free (no increase in DMFS) in the present study, representing the caries-preventing effect of both toothpastes within the study duration. This is higher than in data from Rao et al. (The primary parameter in the present study was the DMFS index. This index was established in clinical studies that investigated the caries-preventing effect of toothpastes (When comparing the mean DMFS increase on the tooth surface level during the 18-month study duration with the previously published studies (Comparison on the mean increase in DMFS between this study and the published studies on adults using fluoridated toothpastes.In addition to DMFS, the overall number of caries lesions was determined with DIAGNOcam. DIAGNOcam is a modern device for laser diode near-infrared light transillumination of dental tissues. The appliance combines fiber optic transillumination with near-infrared (NIR) light and a digital camera to non-ionizing imaging of dental tissues. DIAGNOcam enables the visualization of caries lesions on occlusal and approximal surfaces and measures their severity in real time. It may be used without any limitation to monitor the caries process. It is a non-invasive and non-destructive tool (The difference in DMFS (visual) and NCL based on DIAGNOcam (near-infrared light transillumination) can be explained by the sensitivity of this method. While decayed surfaces can only be detected on the outer area of the tooth visually by an experienced clinician, transillumination methods can also detect lesions underneath the surface (A notable trend was also observed in the PCR index of the hydroxyapatite group: Unlike the fluoride group, the PCR scores in the hydroxyapatite group showed a tendency to a lowering of scores during the course of the study (A strength of the present study is that the same toothpaste composition was used for the hydroxyapatite toothpaste and the fluoride toothpaste except that they differed in their main active ingredients (i.e., hydroxyapatite vs. fluoride). All other ingredients remained the same. While the non-inferiority of hydroxyapatite to fluoride toothpastes was shown in caries risk groups [children (Secondly, as we tested the influence of both toothpastes on caries progression, the patients were not monitored according to dietary habits, and they kept their sugar intake under real-life conditions. There were no specific inclusion or exclusion criteria concerning the diet. However, the subjects had comparable diet behaviors due to living in urban areas in both study centers (A limitation of this study is that the study population was relatively homogeneous, consisting of healthy, predominantly young adults without significant oral health problems. However, previously published RCTs on hydroxyapatite toothpastes were performed in caries risk groups, i.e., orthodontic patients (There are several reasons why hydroxyapatite toothpastes represent an alternative to fluoride toothpastes. Hydroxyapatite is a safe active ingredient (Hydroxyapatite has been shown to be not only efficient in caries prevention, but also in other fields of oral prevention, such as reduction of dentin hypersensitivity (In summary, the three clinical caries studies [Paszynska et al. (

PMC10393266

5. Conclusion

HYDROXYAPATITE

The results of this long-term double-blinded, randomized clinical trial in adults clearly show the non-inferiority of the fluoride-free hydroxyapatite toothpaste to the toothpaste with 1,450 ppm fluoride with regard to the primary endpoint DMFS index. According to the per-protocol analysis, no increase in DMFS index was observed in 89.3% of subjects of the hydroxyapatite group and 87.4% of subjects of the fluoride group. In conclusion, hydroxyapatite was proven to be a safe and efficient anticaries agent in oral care.

PMC10393266

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

PMC10393266

Ethics statement

POLAND

The studies involving human participants were reviewed and approved by the Poznan University of Medical Sciences, Poznan, Poland (No. 691/20; November 4, 2020). The patients/participants provided their written informed consent to participate in this study.

PMC10393266