Identification of preclinical dementia according to ATN classification for stratified trial recruitment: A machine learning approach.

INTRODUCTION: The Amyloid/Tau/Neurodegeneration (ATN) framework was proposed to identify the preclinical biological state of Alzheimer's disease (AD). We investigated whether ATN phenotype can be predicted using routinely collected research cohort data. METHODS: 927 EPAD LCS cohort participants free of dementia or Mild Cognitive Impairment were separated into 5 ATN categories. We used machine learning (ML) methods to identify a set of significant features separating each neurodegeneration-related group from controls (A-T-(N)-). Random Forest and linear-kernel SVM with stratified 5-fold cross validations were used to optimize model whose performance was then tested in the ADNI database. RESULTS: Our optimal results outperformed ATN cross-validated logistic regression models by between 2.2% and 8.3%. The optimal feature sets were not consistent across the 4 models with the AD pathologic change vs controls set differing the most from the rest. Because of that we have identified a subset of 10 features that yield results very close or identical to the optimal. DISCUSSION: Our study demonstrates the gains offered by ML in generating ATN risk prediction over logistic regression models among pre-dementia individuals.

Relative Risk of Death in Bulgarian Cancer Patients during the Initial Waves of the COVID-19 Pandemic.

BACKGROUND: The COVID-19 pandemic has led to millions of documented deaths worldwide, with diverse distribution among countries. Surprisingly, Bulgaria, a middle-income European Union member state, ranked highest in COVID-19 mortality. This study aims to assess whether Bulgarian cancer patients experienced a higher relative risk (RR) of death compared to the general Bulgarian population during the pandemic. MATERIALS AND METHODS: Data from the Bulgarian National Statistical Institute and the Bulgarian National Cancer Registry were analyzed to estimate monthly RR of death in cancer patients compared to the general population before and during the first two years of the pandemic. The impact of the COVID-19 waves and predominant SARS-CoV-2 variants on RR was evaluated on various cancer types and age groups using a multiple linear regression approach. RESULTS: During the COVID-19 waves, both the general population and cancer patients experienced a significant increase in mortality rates. Surprisingly, the RR of death in cancer patients was lower during pandemic waves. The results from the statistical modeling revealed a significant association between the COVID-19 waves and reduced RR for all cancer patients. Notably, the effect was more pronounced during waves associated with the Alpha and Delta variants. The results also showed varying impacts of the COVID-19 waves on RR when we analyzed subsamples of data grouped depending on the cancer type, age and sex. CONCLUSIONS: Despite increased overall mortality in Bulgarian cancer patients during the pandemic, the RR of death was lower compared to the Bulgarian general population, indicating that protective measures were relatively effective in this vulnerable group. This study underscores the importance of implementing and encouraging preventive measures, especially in cancer patients, to mitigate the impact of future viral pandemics and reduce excess mortality.

Detection of Alzheimer's Disease Using Logistic Regression and Clock Drawing Errors.

Alzheimer's disease is an incurable disorder that accounts for up to 70% of all dementia cases. While the prevalence of Alzheimer's disease and other types of dementia has increased by more than 160% in the last 30 years, the rates of undetected cases remain critically high. The present work aims to address the underdetection of Alzheimer's disease by proposing four logistic regression models that can be used as a foundation for community-based screening tools that do not require the participation of medical professionals. Our models make use of individual clock drawing errors as well as complementary patient data that is highly available and easily collectible. All models were controlled for age, education, and gender. The discriminative ability of the models was evaluated by area under the receiver operating characteristic curve (AUC), the Hosmer-Lemeshow test, and calibration plots were used to assess calibration. Finally, decision curve analysis was used to quantify clinical utility. We found that among 10 possible CDT errors, only 3 were informative for the detection of Alzheimer's disease. Our base regression model, containing only control variables and clock drawing errors, produced an AUC of 0.825. The other three models were built as extensions of the base model with the step-wise addition of three groups of complementary data, namely cognitive features (semantic fluency score), genetic predisposition (family history of dementia), and cardio-vascular features (BMI, blood pressure). The addition of verbal fluency scores significantly improved the AUC compared to the base model (0.91 AUC). However, further additions did not make a notable difference in discriminatory power. All models showed good calibration. In terms of clinical utility, the derived models scored similarly and greatly outperformed the base model. Our results suggest that the combination of clock symmetry and clock time errors plus verbal fluency scores may be a suitable candidate for developing accessible screening tools for Alzheimer's disease. However, future work should validate our findings in larger and more diverse datasets.

Approximate Bayesian computation for the natural history of breast cancer, with application to data from a Milan cohort study.

We explore models for the natural history of breast cancer, where the main events of interest are the start of asymptomatic detectability of the disease (through screening) and the time of symptomatic detection (through symptoms). We develop several parametric specifications based on a cure rate structure, and present the results of the analysis of data collected as part of a motivating study from Milan. Participants in the study were part of a regional breast cancer screening program, and their ten-year trajectories were obtained from administrative data available from the Italian national health care system. We first present a tractable model for which we develop the likelihood contributions of the observed trajectories and perform maximum likelihood inference on the latent process. Likelihood based inference is not feasible for more flexible models, and we implement approximate Bayesian computation (ABC) for inference. Issues that arise from the use of ABC for model choice and parameter estimation are discussed, including the problem of choosing appropriate summary statistics. The estimated parameters of the underlying disease process allow for the study of the effect of different examination schedules (age range and frequency of screening examinations) on a population of asymptomatic subjects.

CYP2C19 and CYP2D6 Genotypes and Metabolizer Status Distribution in a Bulgarian Psychiatric Cohort.

CYP2D6 and CYP2C19 are enzymes of essential significance for the pharmacokinetics of a multitude of commonly used antidepressants, antipsychotics, antiemetics, ß-blockers, opioids, antiestrogen, antacids, etc. Polymorphisms in the respective genes are well established as resulting in functional differences, which in turn can impact safety and efficacy. Importantly, the prevalence of genetic CYP2D6 and CYP2C19 variability differs drastically between populations. Drawing on the limited information concerning genotype frequencies in Bulgaria, we here analyzed 742 Bulgarian psychiatric patients predominantly diagnosed with depression and/or anxiety. Specifically, we analyzed frequencies of CYPC19*2, *4 and *17, as well as of CYP2D6*2, *3, *4, *5, *6, *10 and *41. In total, 571 out of 742 patients (77%) carried at least one variant which impacts metabolizer status. Overall, 48.6% of the studied individuals were classified as non-normal metabolizers of CYP2D6 with most exhibiting reduced function (38.2% intermediate metabolizers and 6.6% poor metabolizers). In contrast, for CYP2C19, the majority of non-normal metabolizers showed increased functionality (28.9% rapid and 5.5% ultrarapid metabolizers), while reduced activity metabolizer status accounted for 25.6% (23.8% intermediate and 1.8% poor metabolizers). These results provide an important resource to assess the genetically encoded functional variability of CYP2D6 and CYP2C19 which may have significant implications for precision medicine in Bulgarian psychiatry practice.

Myopia in Bulgarian school children: prevalence, risk factors, and health care coverage.

BACKGROUND: The prevalence of myopia has increased in recent years, with changes being dynamic and uneven in different regions. The purpose of this study is to evaluate the prevalence of visual impairment caused by myopia in Bulgarian school children, associated risk factors, and health care coverage. METHODS: A cross-sectional study among 1401 children (mean age 10.38, standard deviation 2.70) is conducted in three locations in Bulgaria from 2016 to 2020. Refractive error is measured with an auto-refractor in the absence of cycloplegia, the visual acuity is assessed without refractive error correction. A paper-based preliminary questionnaire is used to collect data on previous eye examinations, prescribed optical vision correction, regularity of wearing corrective glasses and risk factors. RESULTS: Children with myopic objective refraction ≤ -0.75 D and decimal visual acuity ≤ 0.8 on at least one of the eyes are 236 out of 1401 or 16.85%. The prevalence of myopia varies depending on age, geographical location, and school profile. The rate of myopic children in age group 6-10 is 14.2% compared to 19.9% in age group 11-15. The prevalence of myopic children in the urban populations is 31.4% (capital) and 19.9% (medium-sized town) respectively, and only 8.4% in the rural population. Our results show 53% increase of the prevalence of myopia in the age group 11-15 compared to a 2009 report. The analysis of data associated with health care coverage factors of all myopic pupils shows that 71.6% had a previous eye examination, 43.2% have prescription for corrective glasses, 27.5% wear their glasses regularly. Risk factors for higher odds of myopia are gender (female), age (adolescence), and parents with impaired vision. Residence in a small town and daily sport activities correspond to lower odds for myopia. The screen time (time in front of the screen calculated in hours per day) is self-reported and is not associated with increased odds of myopia when accounted for the other risk factors. CONCLUSIONS: The prevalence of myopia in this study is higher compared to previous studies in Bulgaria. Additional studies would be helpful in planning adequate prevention and vision care.

Exploring pharmacogenetic variation in a Bulgarian psychiatric cohort.

INTRODUCTION: Pharmacogenetics in psychiatry is currently gaining momentum. The efficiency of antipsychotic therapy is often limited by the lack of response and the presence of side effects. Pharmacogenetic variation is probably one of the causative factors for the observed interindividual differences in the response to and the side effects of antipsychotics, which could be addressed and whose negative effects could be avoided or mitigated. AIM: The present study aimed to conduct a comprehensive analysis of the frequency of DRD2 rs1799732, COMT rs4680, MC4R rs489693, and HTR2C rs3813929 in Bulgarian psychiatric patients. MATERIALS AND METHODS: The frequency of genotypes and the alleles of variants DRD2 rs1799732, COMT rs4680, MC4R rs489693, and HTR2C rs3813929 were studied in a cohort of 515 Bulgarian psychiatric patients using the polymerase chain reaction (PCR) method. RESULTS: We found no significant difference between our cohort and the dataset of the 1000 Genomes Project. Moreover, we found that 433 out of 515 patients carried at least one, and 191 out of 515 carried at least two variants which, based on multiple scientific sources with consistent findings, could potentially alter the expected response rate, time to respond and/or risk of side effects to antipsychotic medications. CONCLUSIONS: Considering the consistent data about the frequency of these pharmacogenetic variants, testing these genetic variants may prove useful in clinical practice. Further studies regarding the clinical interpretation and frequency distribution in larger cohorts and different populations are warranted.