Objective: to investigate the antitumor effects of various forms of vitamin B12 in combination with various synergistic vitamins and evaluate the prospects for clinical applications.

Material and methods. Cell lines BT-474 (breast ductal carcinoma) and A549 (lung carcinoma) were used as an in vitro cell model, and transplantable epidermoid Lewis lung carcinoma (LLC) was used as an in vivo animal tumor model. Animal studies of LLC were carried out on 25 male F₁ hybrid mice (age 2.5–3 months, body weight 23–26 g). In silico research was conducted as a systematic computer analysis of 9,326 scientific sources.

Results. In vitro studies on cultures of two human tumor cell lines (BT-474 and A549) confirmed the cytotoxic effect of vitamin B12 (aquacobalamin). It has been shown that vitamin B12 has weak cytotoxic properties in the concentration range of 3.125–200 μg/L (IC50>200 nM), and its hydrophobic derivative (heptamethyl cyanoquacobyric acid ester) significantly reduces the survival of tumor lines. BT-474 and A549 cells at high concentrations (100–200 µg/l, IC50~100 nM). Experimental animals with an in vivo LLС model easily tolerated a drug based on vitamin B12. Exposure to the drug up to the 21^st day of LLС development was accompanied by an increasing tendency to inhibit tumor growth by 10–20% (р=0.059). The results of a systematic in silico review of the literature show that clinical data confirmed the significant antitumor effect of vitamin B12.

Conclusion. The cellular model indicated the antitumor properties of vitamin B12 and its hydrophobic derivative. With subchronic intragastric administration of B12 to tumor-bearing animals, a steady tendency to inhibit the LLС growth was observed. Analysis of clinical data confirmed the feasibility of the antitumor use of vitamin B12 individually and in combination with synergistic vitamins.

Background. The search for promising nonsteroidal anti-inflammatory drugs (NSAIDs) is aimed, in particular, at identifying molecules with multitargeted anti-inflammatory and analgesic effects (including through central mechanisms).

Objective: To study the interactions of a candidate NSAID molecule (SV-1010) with opioid receptors and compare them with the effects of known agonist molecules (butorphanol and U-50488) using chemoreactomic analysis and docking.

Material and methods. Chemoreactomic analysis of NSAID mechanisms of action was conducted in three stages: data sampling, establishment of lists of molecules with known properties, and calculation of Kd binding constants and EC50 activation constants. Docking of kappa opioid receptors was performed using MarvinSketch, MOPAC2012, and AutoDock Vina. A comparison of the results of chemoreactomic modeling and docking was performed.

Results. Chemoreactomic analysis of the interactions of the studied molecules with opioid receptors showed that the median and average values ​​of the binding constants Kd of the SV-1010 compound are comparable with the estimates of the constants obtained for butorphanol and U-50488 (75–98 nM for delta receptors, 62–81 nM for kappa receptors, 198–244 nM for mu receptors). Among the studied opioid receptor subtypes, the lowest Kd values ​​were established for SV-1010 for kappa receptors (64.8±46.3 nM; delta and mu receptors: 79.9±77.6 and 243.8±246.9 nM, respectively). No significant difference in the binding of SV-1010 molecules to kappa-1 and kappa-2 opioid receptors was detected (Kd in the range of 23.7–54.5 nM). Docking of the studied molecules into the structure of the human kappa receptor allowed us to obtain Kd values ​​and formulate the mechanism of binding of SV-1010 to the kappa-opioid receptor site (potentially, the key binding amino acids of the kappa-opioid receptor site are ILE730, VAL667, MET579, ILE726, TRP723, ILE460 and TYR464). A comparison of the results of chemoreactomic modeling and docking made it possible to find a correlation expressed by the equation “35.8x – 4790” with a correlation coefficient close to unity. The results of chemoreactome modeling of EC50 constants confirmed the results of the Kd binding constant analysis, including the finding that SV-1010 exhibits greater affinity for kappa receptors than for mu receptors.

Conclusion. Chemoreactomic and docking modeling of the SV-1010 molecule's effects support the hypothesis that this compound may be a kappa-opioid receptor agonist, indicating the potential for experimental and other studies of SV-1010 with a focus on kappa-opioid receptors.

Background. Underreporting of adverse drug reactions is a pressing public health concern in the Russian Federation. Addressing this requires evaluating the pharmacovigilance system performance, analyzing stakeholder involvement (patients, healthcare and pharmaceutical professionals, and industry representatives), and refining existing approaches through the enhancement of interagency cooperation and digital solutions.

Objective: To assess the role and level of involvement of pharmaceutical professionals in the spontaneous reporting system of the Russian Federation, identify the primary barriers hindering their participation, and determine areas for improving pharmacovigilance.

Materials and methods. The study analyzes public statistical data from regulatory agencies in the Russian Federation (Roszdravnadzor), the United States (FAERS), and Germany (BfArM) to compare the reporting profiles of key groups. Due to the lack of aggregated national data for the Russian Federation, the authors conducted a survey in 2025 via the Yandex Forms platform, involving 101 respondents (consumers, healthcare professionals, and pharmaceutical professionals).

Results. The study analyzed the causes of underreporting across all target groups, evaluating awareness levels, reporting tool availability, motivation, and other contributing factors. A pronounced imbalance was identified in the Russian reporting system: unlike the multi-channel models of the United States and Germany, the majority of reports in the Russian Federation are submitted by healthcare professionals. The conducted survey revealed that pharmaceutical professionals account for only 8% of reports, which is inconsistent with their legal obligations. Healthcare professionals were found to be the primary contributors (61%), followed by pharmaceutical companies (17%) and patients or their relatives (14%). Notably, despite high pharmacovigilance awareness (88.9%), only 40.7% of pharmaceutical professionals reported encountering consumer complaints. Their preferred method of report submission is direct communication with pharmaceutical companies (51.9%), with online reporting forms being the most convenient format (51.9%). Analysis of comments identified key barriers, including insufficient feedback, procedural uncertainty, and a lack of confidence regarding the effectiveness of reports.

Conclusion. The study identified key problem areas and proposed strategies for improving the pharmacovigilance system in the Russian Federation. Pharmaceutical professionals represent a critical but under-involved category of reporters within the pharmacovigilance system. Improving the completeness and quality of drug safety data requires a comprehensive approach, encompassing educational reform, implementation of digital tools to streamline reporting, and the launch of awareness campaigns. Furthermore, it is essential to foster an environment where reporting adverse reactions is perceived as an integral part of professional responsibility.

Background. Many pharmaceuticals, including antibiotics, diuretics, some antitumor agents, hormones, etc., can promote the depletion of magnesium (Mg), pyridoxine (vitamin B6, VB6), and other micronutrients (MNs) in the body. This process may lead to the development of hypomagnesemia and concomitant MN deficiencies, which are associated with a range of adverse effects, including neurotoxicity, cardiotoxicity, hepatotoxicity, etc. Moreover, the resulting micronutrient deficiency (MND) may paradoxically aggravate the underlying pathophysiological mechanisms of the diseases for which these drugs are prescribed, thereby potentially diminishing therapeutic efficacy and contributing to treatment-related complication.

Objective: Chemoreactomic assessment of anti-micronutrient (anti-MN) effects of all drugs included in the Anatomical Therapeutic Chemical (ATC) classification system.

Material and methods. Using modern data mining techniques, including mathematical approaches from topological data analysis, labeled graph theory (chemographs), and related method, this study performed a systematic computer-based analysis of databases describing the Mg-depleting effects of drugs; original algorithms for numerically predicting the Mg- and VB6-removing effects of drugs. Original algorithms were developed for the numerical prediction of Mg- and VB6-depleting properties of drugs, as well as for the assessment of other anti-MN effects. These algorithms were subsequently applied in a chemoreactomic screening of 2,527 drugs classified within the ATC system.

Results. A database describing anti-MN properties of drugs was created for 24 MN balance indicators for 18 MNs. Algorithms for predicting the anti-MN properties of drugs were developed with a classification accuracy of 92±10% in cross-validation (the accuracy of predicting VB6 MND – 88%, Mg MND – 94-98%). On average, each drug from the ATC group accounts for 8.5±6.5 anti-MN effects. Only 100 out of 2527 (4%) drugs did not exhibit a negative impact on MN, primarily amino acids, MNs themselves, and choline drugs. The most pronounced negative impact of the drugs under study was related to the metabolism of vitamin D3 (505 ATC categories), VB6 (475 ATC categories), iron (419 ATC categories), vitamin B1 (386 ATC categories), and Mg (375 ATC categories). VB6 MND was caused by 1701 drugs, Mg MND – by 1064 drugs. Antibiotics for systemic use (ATC code J01), psycholeptics (N05) and psychoanaleptics (N06), antineoplastic agents (L01), sex hormones and modulators of the reproductive system (G03), analgesics (N02), antidepressants (N06A), diuretics (C03), antihistamines for systemic use (R06A), anti-inflammatory and antirheumatic agents (M01), direct-acting antivirals (J05A), and antiepileptic agents (N03A) were found to affect adversely the homeostasis of both Mg and VB6. A detailed description of the anti-Mg and anti-VB6 properties of these drug classes was provided. The data obtained via chemoreactomic analysis were compared with that obtained by experimental and clinical studies of Mg and VB6 preparations.

Conclusion. The conducted chemoreactomic analysis provides a substantiated basis for supporting pharmacotherapy with selected medicinal preparations based on organic salts of Mg and VB6.

Objective: To develop and validate a method for evaluating the regional economic efficiency of integrating artificial intelligence (AI) into target disease (TD) detection compared to conventional diagnostics.

Material and methods. Medical data from 381 patients with skin neoplasms (291 benign and 90 malignant cases) were analyzed to develop and validate a method of economic efficiency evaluation. Two diagnostic routing scenarios were simulated: AI-assisted routing (62% threshold) and conventional three-stage routing without AI. The assessment involved calculating financial costs per each identified TD case and for all its cases in the region.

Results. With the use of the Derma Onko Check AI program, the proportion of unreasonable referrals decreased from 40.6% to 6.9% for dermatologists/venereologists and from 22% to 7.6% for oncologists. Calculations performed using the developed method show that unreasonable financial costs per TD case (C43 skin melanoma) in the Moscow Region amounted to 282,268.98 rubles with the use of AI compared to 579,069.26 rubles with conventional diagnostics. The ratio of reasonable to unreasonable costs was 1.7 with the use of the AI (indicating a predominance of reasonable costs) and 0.31 without AI (where unreasonable costs exceed reasonable costs). When extrapolated to the regional level (Moscow, 1470 cases of skin melanoma in 2024), the potential reduction in unreasonable costs amounts to 436,296,411.6 rubles.

Conclusion. Using skin melanoma as an example, the developed method demonstrated the high economic efficiency of integrating AI into TD diagnostics. This technology provides a means to optimize patient routing, reduce the financial burden on the healthcare system, and ensure earlier detection of socially significant diseases. This method can be adapted to evaluate the economic efficiency of AI integration in diagnosis of other pathologies.

Objective: To analyze the legal and ethical aspects of regulating artificial intelligence (AI) in medicine in key jurisdictions (United States, European Union, China, Russia), to identify regulatory gaps, ethical dilemmas and prospects for harmonization of standards.

Material and methods. National and international regulatory documents (GDPR, AI Act, FDA, NMPA), scientific publications, clinical cases and regulatory initiatives (IMDRF, WHO) were reviewed. Methods for comparative legal analysis and systematization of ethical and legal norms were used.

Results. Considerable differences in approaches to AI regulation were identified, including flexibility in the US, the ethical centricity in the EU, centralization in China and an emerging framework in Russia. Key issues were emphasized, such as algorithmic bias, AI transparency, responsibility, and the conflict between innovation and security.

Conclusion. The harmonization of international standards, the introduction of dynamic regulation and the strengthening of interdisciplinary cooperation should be pursued to achieve a balance between innovation and the protection of patients' rights.

Objective: To develop and validate a method for evaluating the economic efficiency of target disease (TD) diagnostics performed via artificial intelligence (AI)-assisted multi-stage patient routing.

Material and methods. The evaluation method was developed through a simulation of two diagnostic routing scenarios (with and without AI program output) based on data from 381 patients with malignant and benign skin neoplasms. This approach was validated using output of the Derma Onko Check AI program, employing previously proposed diagnostic algorithms for melanocytic skin tumors (n=230) at a 62% routing threshold. Formulas were derived to calculate the financial cost (FC) ratio, the cost of identifying one TD case, and coefficients for avoidable and potential avoidable costs to enable mapping within a quadrant matrix. The evaluation method factors in not only the avoidable costs of medical interventions but also the potential avoidable costs (losses) resulting from delayed TD detection.

Results. The implementation of diagnostic algorithms based on the output of the Derma Onko Check AI program demonstrated high economic efficiency. The FC ratio of 0.49 indicates a 51% reduction in the total FCs for melanocytic skin tumors compared to conventional diagnostics. The analysis of avoidable and potential avoidable costs revealed a 59.0% decrease in avoidable costs (R_TC_AC=0.41) and a 51.0% decrease in potential avoidable costs (R_TC_PAC=0.49). These results fall within the optimal efficiency zone of the quadrant matrix.

Conclusion. The obtained results validate factoring missed-case treatment costs into both parts of the R_TC formula. Thisensures accurate comparability of diagnostic approaches with different FC structures and clinical outcomes.

Background. No systematic data are currently available on long-term trends in the product range, prices, and supply structure of antithrombotic drugs in the Republic of Uzbekistan, which hinders pharmaceutical supply planning.

Objective:  To conduct a comprehensive analysis of the antithrombotic drug market in the Republic of Uzbekistan covering the period 2010–2024.

Materials and methods. This study presents the first multi-parametric analysis of the antithrombotic drug market in the Republic of Uzbekistan over a 15-year period. The analysis evaluated registration trends, supply volumes, price ranges, manufacturing countries, and the pharmacoeconomic characteristics of antithrombotic agents. Data were obtained from the State Register of Medicines, Medical Devices, and Medical Equipment, published by the Ministry of Health of the Republic of Uzbekistan, as well as the Drugs Audit system. Methodological approaches included product range analysis, trend analysis, price analysis, calculation of the compound annual growth rate (CAGR), and correlation analysis (Pearson correlation coefficient, p<0.05).

Results. The assortment of antithrombotic drugs expanded from 53 to 141 trade names over the period under review, with an average of approximately 50% entering active commercial circulation. Supply volumes increased 6.6-fold, peaking at 12.19 million packages (2020), followed by a decline (CAGR: −8.03%). Although imported products account for 83% of the product range, the share of domestic manufacturers has increased since 2017. The leading International Nonproprietary Names in the market include acetylsalicylic acid, enoxaparin, and clopidogrel; the market share of rivaroxaban is growing.

Conclusion. The antithrombotic drug market in the Republic of Uzbekistan has become more diversified, yet it remains import-dependent. The findings of this study can be used to inform public policy and procurement planning.

Background. The introduction of artificial intelligence (AI)-driven computer vision programs (CVPs) into medical diagnostics has imposed stricter requirements on the quality of input photographic images. Imaging conditions vary significantly across medical fields, necessitating the establishment of field-specific reference ranges for photometric and textural parameters to ensure that AI models maintain reproducible diagnostic accuracy. A systematic analysis of the causes of erroneous classifications by CVPs is essential for their clinical application and further improvement.

Objective: To develop a generalized methodology for analyzing the causes of classification errors in photographic images by processed by AI-driven CVPs. The proposed methodology enables the establishment of reference ranges for photometric and textural parameters for specific medical imaging applications, as well as the development of criteria for excluding images with anomalous values when during preprocessing in medical software systems.

Material and methods. The methodology includes eight sequential stages. A dataset of photographic images verified by histological and dermatoscopic examinations was compiled, classified using the CVPs, and assigned to either of the four standard categories (true positive, true negative, false positive, and false negative). For each photographic image, thirteen photometric and textural quality metrics were calculated, including: brightness, contrast, sharpness, entropy, high-frequency saturation, proportions of overexposed and underexposed pixels, and mean values ​​and standard deviations of the color channels. Systematic between-group differences were identified using one-way ANalysis Of VAriance (ANOVA), Welch's test, and Spearman's rank correlation analysis. The image regions that determine the neural network decision were localized using explainable AI techniques. Reference ranges were established from the characteristics of correctly classified photographs (true positive and true negative categories), defined as intervals of [mean − 2 std; mean + 2 std]. The effectiveness of parameter normalization was assessed by the improvement in accuracy, sensitivity, and specificity.

Results. The proposed methodology was tested using the Derma Onko Check and Melanoma Check CVPs as an example. Its application allowed statistically significant intergroup differences in photometric and textural parameters to be identified (F=13.50–39.31, p<0.001 for the main metrics of the one-way ANOVA; F=5.41–72.29, p<0.001 for the conclusion category by the two-way ANOVA). The analysis confirmed that the observed patterns were independent of the specific CVP used (p=0.39–0.96 for the program factor; p=0.15–0.92 for the interaction effect). Multivariate analysis further demonstrated significant differences among classification outcome groups based on the combined set of image quality metrics (Wilks' lambda 0.639; F=10.37; p<0.001) and established key independent predictors of classification errors through logistic regression (Fast Fourier Transform blur: odds ratio (OR)  3.08; sharpness: OR 0.31; proportion of overexposed pixels: OR 1.64). Reference ranges were established for brightness (0.467–0.942), contrast (0.066–0.333), entropy (3.626–5.590), and high-frequency saturation (23.82–56.48), along with critical thresholds for image exclusion from inference (proportion of overexposed or darkened pixels greater than 55%). The use of a targeted preprocessing module for normalization of deviating parameters falling outside the reference ranges ensured an increase in diagnostic accuracy by +0.014–0.017 in absolute values across all studied CVP configurations, with a predominant increase in specificity (+0.015–0.019).

Conclusion. The proposed methodology for analyzing the causes of erroneous classification of photographic images by AI-driven CVPs was tested on the example of Derma Onko Check and Melanoma Check using a limited dataset (460 photographic images representing two morphological subgroups of melanocytic skin tumors). The extension of the methodology to other areas of medical imaging (ophthalmology, histology, or ultrasound diagnostics), where image acquisition conditions differ substantially, will require validation on representative multicenter datasets with recalculation of parameter reference ranges to reflect the imaging specifics of each domain. The integration of modules for detecting and excluding images with abnormal metric values constitutes a natural practical implication of the proposed methodology and ensures a reproducible increase in the clinical accuracy of AI-driven CVP systems.

Background. The rapid development of digital technologies and the increasing recognition of real‑world evidence (RWE) underscore the need for its broader integration in comprehensive drug assessment, including the development of national restrictive lists of drugs). In this context, an analysis of the availability of local (Russian) real‑world studies (RWS) and their use in comprehensive drug assessment is highly relevant.

Objective: To analyze the availability of RWE and the frequency of its use in comprehensive drug assessment.

Material and methods. The availability of RWS was assessed for 172 drug proposals for inclusion in national restrictive lists between 2020 and 2024. The study was conducted by calculating the proportion of proposals, for which publications reporting the results of local RWS were available at the time of submission. RWS were identified through a systematic search. The frequency of RWE use in comprehensive drug assessment was evaluated through a content analysis of 28 drug dossiers submitted for inclusion in 2022. The frequency of RWE use was also determined in published cost-effectiveness analyses (CEAs) and budget impact analyses (BIAs) of drugs proposed for inclusion in 2018–2024 (116 publications).

Results. Over the period from 2020 to 2024, local RWS were published for 34% of drugs proposed for inclusion in the national restrictive lists. In 2022, 86% of drug dossiers included RWS. Among the published CEAs and BIAs submitted in 2018–2024, RWE was used in 42% and 63% of cases, respectively.

Conclusions. Russian RWS can already be regarded as a promising source of evidence for comprehensive drug assessment. At the same time, the relatively low publication rate of local RWS indicates the need to further accelerate the generation of such evidence. The high demand for RWE in comprehensive drug assessment underscores the importance of its integration in the development of national restrictive drug lists.

Background. In Russia, one of the key mechanisms for ensuring access to modern drug therapy for patients with severe chronic and rare diseases is the federal drug provision program for 14 High-Cost Nosologies (HCN).

Objective: To evaluate the current performance of the 14 HCN program from the perspective of healthcare professionals directly involved in its implementation and to identify priority areas for program improvement.

Material and methods. A survey was administered to specialists responsible for drug provision within the 14 HCN program. The questionnaire included nine questions using interval and ordinal response scales. Of 87 questionnaires received, 28 valid responses were included in the final analysis. Instrument reliability was assessed using Cronbach’s alpha, and agreement among expert judgments was evaluated using Kendall’s coefficient of concordance and Pearson’s chi-square test.

Results. The questionnaire demonstrated satisfactory internal consistency (α=0.799). According to respondents, the most critical challenges of the program were insufficient funding (mean score 4.29), the need to improve the drug provision system (4.21), and the need to automate the submission and review of applications for medicinal products (4.18). Moderate agreement among experts was observed only in ranking nosologies by priority for implementing improvement measures (W=0.3089). Hemophilia, multiple sclerosis, and malignant neoplasms of lymphoid, hematopoietic, and related tissues were identified as the highest-priority conditions.

Conclusion. The findings confirm need to strengthen the 14 HCN program and may inform the development of organizational and pharmacoeconomic optimization strategies.

Background. Digital media monitoring is a requirement of Good Pharmacovigilance Practices (GVP) of the Eurasian Economic Union (EAEU). However, standardized approaches for processing unstructured social media data on adverse reactions (ARs) associated with the medical use of drugs are currently lacking.

Objective. To develop and test an organizational model for digital media monitoring aimed at identifying AR reports, adapted to the requirements of EAEU GVP, as well as to assess its feasibility on the example of the Russian-language segment of the VKontakte social network.

Materials and methods. A retrospective analysis of content published in seven VKontakte communities for patients with rheumatoid arthritis (with a combined audience of more than 54,000 subscribers) during the period from January to December 2025 was conducted. A total of 58,884 posts and comments were analyzed. Searches were performed using Russian keywords corresponding to “side effects” in English. The developed data-processing protocol includes assessment of the minimum validity criteria of posts and comments (presence of an identifiable reporter, patient, suspected drug, and description of an AR); coding of identified ARs using the Medical Dictionary for Regulatory Activities (MedDRA, version 27.0) by two independent experts; comparison with the approved prescribing information for drug use; and calculation of inter-rater agreement using Cohen’s kappa coefficient.

Results. A total of 1,463 reports mentioning ARs associated with drug use were identified. Of these, 108 reports (7.4%) met the established validity criteria. Coding of the 108 valid reports yielded 76 unique preferred terms for ARs consistent with MedDRA terminology. Comparison of these terms with the approved prescribing information showed that 59 ARs were classified as listed (expected) ARs, whereas 17 were not included in the prescribing information and were, therefore, classified as unlisted (unexpected) ARs. These data served as a basis for generating hypotheses for subsequent validation using signal management. Inter-rater agreement between the two independent experts regarding the presence of ARs was high (Cohen’s kappa = 0.87). The proportion of publications that could not be coded because of incomplete or nonspecific descriptions was 92.6% of those mentioning ARs, confirming the high level of informational noise in social media.

Conclusion. The proposed organizational model enables formalization of the social media monitoring process, ensuring reproducibility and compliance with EAEU GVP regulatory requirements. The identification of ARs not described in the approved prescribing information demonstrates the potential of social networks as an additional source of drug safety hypotheses and highlights the need to integrate such monitoring into the routine pharmacovigilance activities of pharmaceutical organizations.

Objective: To calculate direct retail pharmacotherapy and monitoring costs for type 2 diabetes mellitus (T2DM) and obesity in adults in the Russian Federation based on the current Ministry of Health clinical guidelines.

Material and methods. Using median retail prices from the apteka.ru and uteka.ru pharmacy aggregators in April 2026, costs for 14 T2DM and 5 obesity pharmacotherapy regimens were calculated. The dataset included 254 price items for T2DM drugs and 44 items for obesity drugs. The analysis was conducted from the perspective of the patient as a retail market consumer. Direct non-drug medical monitoring costs, including laboratory tests, physician visits, and self-monitoring of blood glucose, were additionally assessed.

Results. T2DM pharmacotherapy costs ranged from 175 to 14,716 rubles per month, corresponding to an approximately 84-fold difference between the least and most expensive regimens. Three cost strata were identified: (I) low-cost (<500 rubles per month); (II) medium-cost (1,000–7,000 rubles per month); and (III) high-cost (>12,000 rubles per month) strata. Low-cost regimens included metformin and its combinations with sulfonylureas; medium-cost regimens included dipeptidyl peptidase-4 inhibitors, sodium-glucose transport-2 inhibitors, insulin and injectable semaglutide; high-cost regimens included liraglutide, tirzepatide, and oral semaglutide. Obesity pharmacotherapy costs ranged from 2,958 rubles per month for sibutramine to 29,568 rubles per month for semaglutide (2.4 mg once a week). The total annual recommended monitoring costs including self-monitoring of blood glucose reached 32,900–63,200 rubles for patients receiving oral therapy and 51,150–99,700 rubles for patients receiving insulin studies.

Conclusion. Transition to modern pharmacotherapy regimens for T2DM and obesity is associated with a substantial increase in direct drug expenditure. Injectable semaglutide belongs to the medium-cost stratum, whereas oral semaglutide remains in the high-cost stratum, which may reflect the technological complexity of the oral salcaprozate sodium formulation and limited competition in this segment. Accounting for monitoring costs shows that the true direct financial burden on patients substantially exceeds drug costs, particularly in the low- and medium-cost strata. These data provide a basis for subsequent pharmacoeconomic analyses.

Objective: To develop and systematize a methodology for constructing datasets for computer vision programs (CVPs) in clinical medicine; to propose and substantiate a multicriteria classification of CVPs and to characterize their clinical use at three organizational levels: the patient, the clinician and the medical organization.

Material and methods. The study systematizes the concepts of "dataset" and "database" in medical artificial intelligence, develops a multicriteria classification of CVPs, and , and characterizes their clinical use. The study employed terminological and comparative analyses, as well as a and classification approach.

Results. The concepts of "dataset" and "database" were clarified, and their key characteristics were formalized. A six-stage methodology for constructing a dataset from a clinical database was described, comprising extraction, de-identification, verification, structuring, balancing, and documentation. A four-criteria classification of CVPs was proposed and substantiated, based on input-data modality, number of models used, deployment setting, and integration with clinical equipment. The clinical application of CVPs was characterized at three organizational levels: the patient, the clinician, and the medical organization.

Conclusion. Proper dataset construction is a key determinant of the clinical validity of CVPs. The proposed classification provides a standardized description of these systems, enabling their valid comparison and regulatory evaluation. The identification of the levels at which CVP is used in clinical practice helps optimize its integration into real-world healthcare settings.

Background. Although clinical guidelines prioritize oral administration of nonsteroidal anti-inflammatory drugs (NSAIDs), the distribution of administration routes across Russian regions has not been quantitatively characterized. Market concentration indices and cost decomposition have scarcely been used in Russian studies of drug utilization.

Objective: To quantify the regional consumption of systemic NSAIDs in the retail sector of the pharmaceutical market in the Samara Region for 2023–2025 and to compare the structural position of ketorolac with published European pharmacoepidemiological data.

Material and methods. A retrospective regional study on the drug utilization was conducted using anonymized sales data on systemic NSAIDs from the pharmacy retail sector in the Samara Region for the years 2023–2025. The data source was the AlphaRM database. Consumption volume was calculated in defined daily doses (DDD) according to the WHO ATC/DDD index, accounting for administration route; consumption intensity was expressed as DDD per 1,000 inhabitants per day (DDD/1,000 inhabitants/day, DID). Additional analyses included the drug utilization 90% (DU90) segment, market concentration indicators (Herfindahl–Hirschman Index, HHI), cumulative share of the three leading molecules (concentration ratio of top-3 firms, CR3), and budget burden structure. The resulting indicators were compared against findings from European NSAID utilization studies.

Results. Total consumption of systemic NSAIDs in the Samara Region was 40.1 DID in 2023 and 43.1 DID in 2025, comparable to Nordic reference values. Ketorolac consistently ranked among the top three molecules in the DU90 segment throughout 2023–2025, accounting for an average of approximately 6.1 DID and 14.7% of total DDD volume. This structural position of ketorolac substantially exceeded values reported in European sources, with moderate market concentration (HHI<0.15, stable CR3 values).

Conclusion. While the overall consumption of systemic NSAIDs in the Samara Region is comparable to European countries, the structural position of ketorolac in the regional retail segment substantially exceeds the published reference values. Due to ketorolac's restricted regulatory status, these findings should be regarded as a population-level signal that requires further confirmation through an assessment of individual prescriptions and clinical outcomes. The methods employed in this study can be applied to other regions of the Russian Federation.

Background. Standard error matrices systematically underestimate the clinical value of diagnostic artificial intelligence (AI) tools by misclassifying cautionary conclusions regarding histologically benign, yet clinically suspicious neoplasms with clinical signs of malignancy as false-positive outcomes.

Objective: To develop and validate a methodology for assessing the clinical effectiveness and accuracy of diagnostic AI tools in dermato-oncology, incorporating clinical caution alertness as an independent metric.

Material and methods. A total of 342 skin lesions were evaluated at the Burdenko Main Military Clinical Hospital (2025–2026). While a standard error matrix was used or the formal accuracy assessment, the clinical evaluation relied on two newly developed: a four-category clinical effectiveness system (complete concordance, concordance in clinical caution, discordance in clinical caution, complete discordance), and a clinical accuracy formula derived from a clinical case matrix (Justified Conclusion, Unjustified Conclusion, Missed Justified Conclusion and Not Missed Justified Conclusion). Wilson confidence intervals (CIs) were applied.

Results. The formal evaluation yielded an accuracy was of 89.7% (100.0% sensitivity, 84.8% specificity). The analysis of clinical effectiveness revealed complete concordance in 82.2% of cases, concordance in clinical caution in 12.9%, discordance in clinical caution in 1.2%, and complete discordance in 3.8%. Clinical accuracy metrics were as follows: 98.5% sensitivity (95% CI 94.6–99.6), 88.1% specificity (95% CI 83.0–91.8), 92.1% accuracy (95% CI 88.8–94.5).

Conclusion. Relying solely on the formal diagnostic accuracy of AI tools underestimates their clinical value. The proposed clinical accuracy and effectiveness metrics ensure an objective assessment of AI tools in relation to the real-world tasks in dermatooncological patient routing.

The prevalence of obesity and type 2 diabetes mellitus continues to rise, imposing a substantial medical, social, and economic burden on healthcare systems worldwide. Glucagon-like peptide-1 receptor agonists have demonstrated robust efficacy in glycemic control, weight management, and cardiovascular risk mitigation, which has led to their widespread integration into clinical practice. However, the high cost of originator products, recurrent supply shortages, and inequitable access to therapy remain significant barriers, particularly within constrained healthcare budgets. This review analyzes current data on the clinical comparability of biosimilars of glucagon-like peptide-1 receptor agonists, regulatory requirements for their approval, and international experience with their clinical uptake. Particular attention is given to economic considerations, including the impact of price competition on treatment costs, out-of-pocket expenditures, and potential budgetary implications. Real-world evidence suggests that the financial burden may adversely affect treatment initiation, adherence, and timely achievement of therapeutic targets. Noteworthy is that the introduction of biosimilars does not automatically translate into substantial price reductions; this requires supportive regulatory and reimbursement policies, transparent pricing mechanisms, effective pharmacovigilance, and educational initiatives for both clinicians and patients. In the Russian context, the integration of biosimilars may enhance improved access to incretin-based therapy, provided that standards of quality, safety, and evidence-based evaluation are maintained. Achieving a balance between clinical effectiveness and economic sustainability is essential for ensuring long-term accessibility of glucagon-like peptide-1 receptor agonist therapy amid growing demand.

Objective: To demonstrate the financialandclinicalvalue of ABC/VENanalysisforevaluating the rationalityofresource allocation underbudgetary constraints.

Material and methods. Data from the inventory and expenditure records of a psychiatric facility for the period 2022–2024 were analyzed. Pharmaceutical expenditures were calculated (ABC-analysis), and rational medicine use was assessed (VEN-analysis). The categories and structure of expenditures were evaluated using ABC/VEN matrix analysis. Spearman’s correlation coefficient and an optimization coefficient were calculated to determine the relationships between variables and evaluate the rationality of resource allocation in terms of the alignment of expenditures with clinical significance.

Results. During the analyzed period, the share of Category I expenditures increased both in terms of the number of subcategories (from 57.9% to 86.66%) and costs (from 84.7% to 96.98%). Conversely, the share of Category II decreased both in terms of the number of items (from 38.6% to 10.34%) and costs (from 14.6% to 3.02%). Category III expenditures (low-priority drugs) were gradually eliminated from the procurement structure, dropping from 3.5% to complete absence in 2024. An increase was observed in the optimization coefficient for Subcategory AV (up to 0.47 in 2023–2024), alongside a growth in the share of expenditures for Group V medicines from 48% to 97%. Subcategories AN and AE were completely eliminated, and the correlation between the shares of medicine items and expenditures became stronger.

Conclusion. In 2022–2024, the procurement structure realigned toward Category I, which is consistent with the principles of evidence-based medicine and the strategic objectives of expenditure optimization while maintaining a high level of therapeutic efficacy.

Background. Obesity represents one of the most critical chronic non-communicable diseases worldwide, driving an an increased risk of cardiovascular disease, type 2 diabetes mellitus, and premature mortality. Recent years have seen considerable attention given to the development of new pharmacological approaches for obesity management. Among these, glucagon-like peptide-1 receptor agonists have attracted particular interest due to their effects on appetite regulation and metabolic control. Semaglutide, one of the most extensively studied agents in this class, demonstrates substantial weight-reducing effects alongside favorable cardiometabolic outcomes.

Objective: To analyze current evidence on the clinical effectiveness, safety, and pharmacoeconomic aspects of semaglutide use in the treatment of obesity.

Material and methods. An analytical review of publications indexed in international (PubMed/MEDLINE) and Russian (eLIBRARY.RU) scientific databases was conducted, covering studies published up to December 2025. The search strategy was based on combinations of keywords, such as “semaglutide”, “GLP-1 receptor agonist”, “obesity treatment”, “anti-obesity medications”, “weight loss”, “efficacy”, “safety”, “cost-effectiveness”, and “health economics”, alongside their Russian equivalents. The analysis included randomized controlled trials, meta-analyses, systematic reviews, and observational real-world studies. Additionally, pharmacoeconomic studies assessing cost-effectiveness, cost-utility, and budget impact of therapy were considered.

Results. Although semaglutide therapy leads to clinically meaningful weight reduction, improved metabolic parameters, and reduced cardiovascular risk, its clinical use requires careful consideration of the safety profile, treatment duration, and economic implications. Pharmacoeconomic analyses suggest that despite relatively high acquisition costs, the use of semaglutide may contribute to a reduction in long-term healthcare expenditures by decreasing the incidence of obesity-related complications.

Conclusion. While semaglutide is a promising pharmacological option in comprehensive obesity management, further clinical and economic studies are required to evaluate its long-term effectiveness and impact on healthcare systems.

Background. Hemophilia is a rare disease covered by Russia’s federal program for 14 High-Cost Nosologies (HCN), which ensures nationwide access to expensive drug therapies funded from the federal budget. The program currently includes 11 hemophilia drug products; however, the absence of transparent, clinically grounded criteria for selecting among these options in specific clinical scenarios complicates therapeutic decision-making.

Objective: To conduct a systematic review of the clinical effectiveness for hemophilia drugs included in the 14 HCN program and to develop an approach for establishing treatment selection criteria for specific patient groups.

Material amd methods. A literature search was performed in PubMed/MEDLINE and the Cochrane Library. Comparative clinical trials, systematic reviews, and meta-analyses were included. In addition, clinical guidelines for hemophilia and prescribing information for the relevant drugs were analyzed.

Results. The available evidence base is heterogeneous and does not permit a comprehensive comparison of all drugs included in the program. Nevertheless, several patterns in clinical effectiveness and important limitations of the evidence relevant to treatment selection were identified. Based on clinical trial data, clinical guidelines, and prescribing information, a methodological approach for developing drug selection criteria was proposed. This approach incorporates patient models, typical clinical scenarios, and an automated treatment selection algorithm.

Conclusion. The proposed approach may support the standardization of prescribing in hemophilia and can be adapted for other diseases included in the 14 HCN program.

Background. Post-COVID syndrome (PCS) presents with significant clinical heterogeneity and is associated with reduced quality of life, impaired status, and diminished work capacity. As no causal therapy for PCS has been approved, it is essential to distinguish the role of early antiviral treatment during acute COVID-19 (which may mitigate the risk of long-term sequelae) from the use of antiviral agents in patients with established PCS.

Objective: To review the clinical, pharmacological, technological, and pharmacoeconomic factors underlying the use of contemporary antiviral strategies for COVID-19 and PCS.

Material and methods. A narrative review was conducted in accordance with the SANRA scale criteria and the CHEERS 2022 statement for reporting economic evaluations. The literature search was conducted in PubMed/MEDLINE, ScienceDirect, SpringerLink, Google Scholar, eLibrary, and CyberLeninka. Eligible publications included randomized and observational clinical trials, systematic reviews, pharmacoeconomic evaluations, and research on antiviral agents, biological therapeutics, and emerging drug delivery technologies.

Results. Strong evidence supports early initiation of antiviral therapy in high-risk patients during acute COVID-19. Nirmatrelvir/ritonavir demonstrates the greatest clinical benefit in reducing hospitalization and mortality rates, although its effect on PCS risk remains uncertain. Remdesivir use is limited by its intravenous administration and the associated healthcare resource utilization, while molnupiravir is generally considered alternative. The economic value of monoclonal antibodies depends on the susceptibility of circulating SARS-CoV-2 variants. In the STOP-PASC trial, a 15-day course of nirmatrelvir/ritonavir showed no meaningful improvement in symptoms among patients with established PCS. Small interfering RNAs, nanoformulations, prodrugs, and intranasal delivery systems are supported primarily by preclinical or early-phase clinical data. Pharmacoeconomic assessments should distinguish between preventing PCS and treating established PCS, while accounting for disease phenotype, quality of life, work capacity, and rehabilitation needs.

Conclusion. Routine use of antiviral agents in established PCS is not supported by current evidence. Further investigation may be warranted in subgroups with indications of viral or antigenic persistence. Early antiviral therapy for acute COVID-19 in high-risk patients remains justified, although its potential to prevent PCS requires further evaluation.

Background. Artificial intelligence (AI) plays a central role in contemporary medicine. The rapid increase in the number of developed approved software products, the deployment of clinical decision support systems, and the growing body of published research indicate that AI technologies are transitioning from the experimental stage to routine clinical use. However, the field still lacks unified terminology and a systematic classification of AI-based medical software.

Objective: To develop a coherent terminology and classification framework for AI in medicine; to compare major AI architectures; and to define several new concepts needed to describe contemporary AI software architectures.

Material and methods. A systematic search of PubMed/MEDLINE, Scopus, Web of Science, and regulatory databases (FDA, EMA) was conducted. Eligible sources included peer-reviewed publications and regulatory guidance documents in English and Russian (1950–2025). Terminology systematization and classification were performed using content analysis, and internationally established terms were adapted to Russian-language usage through expert consensus.

Results. The article examines the development of AI in clinical medicine and healthcare, tracing its progression from the expert systems of the 1970s to contemporary large language models and multimodal systems. It systematizes the major neural network architectures used in medical AI, including convolutional neural networks (CNN), vision transformers (ViT), hybrid CNN+ViT architectures, recurrent neural network (RNN) models, long short-term memory (LSTM) networks, and transformer-based large language model (LLM) (such as bidirectional encoder representations from transformer, BERT), and generative pre-trained transformer (GPT). This analysis provides a comprehensive classification of modern neural network architectures and AI systems according to modality, number of models, and deployment setting, and formulates practical recommendations to support method selection. In addition, the study systematizes and adapts the concepts of “inference”, “single-model/multi-model AI”, “computer vision program”, and recommends their standardized use in Russian-language medical terminology.

Conclusion. The development of a unified terminology and standardized approaches to the classification and validation of AI programs is essential for their safe and effective clinical application.