Ensuring drug safety using AI models for adverse drug reaction prediction

An AI model developed to predict adverse drug reactions could potentially support early-stage drug safety assessment before clinical trials.

Adverse drug reactions (ADRs) are a significant cause of hospital admissions and treatment discontinuation worldwide. Conventional approaches often fail to detect rare or delayed effects of medicinal products. In order to improve early detection, a research team from the Medical University of Sofia developed a deep learning model to predict the likelihood of ADRs based solely on a drug’s chemical structure.

The model was built using a neural network trained using reference pharmacovigilance data. Input features were derived from SMILES codes – a standard format representing molecular structure. Predictions were generated for six major ADRs: hepatotoxicity, nephrotoxicity, cardiotoxicity, neurotoxicity, hypertension, and photosensitivity.

A flowchart illustrating chemical transformation, featuring molecular structures, fragment analysis, and decomposition stages.
Visual representation of SMILES and the process of molecular deconstruction. Adapted from Wu JN, Wang T, Chen Y, Tang LJ, Wu HL, Yu RQ. t-SMILES: a fragment-based molecular representation framework for de novo ligand design. Nat Commun. 2024 Jun 11;15(1): 4993. https://doi.org/10.1038/s41467-024-49388-6.

“We could conclude that it successfully identified many expected reactions while producing relatively few false positives,” the researchers write in their paper published in the journal Pharmacia, concluding it “demonstrates acceptable accuracy in predicting ADRs.”

Infographic detailing an AI model predicting adverse drug reactions for various compounds.

Testing of the model with well-characterized drugs resulted in predictions consistent with known side-effect profiles. For example, it estimated a 94.06% probability of hepatotoxicity for erythromycin, 88.44% for nephrotoxicity and 75.8% for hypertension in cisplatin. Additionally, 22% photosensitivity was predicted for cisplatin, while 64.8% photosensitivity was estimated for the experimental compound ezeprogind. For enadoline, a novel molecule, the model returned low probability scores across all ADRs, suggesting minimal risk.

Notably, these results demonstrate the model’s potential as a decision-support tool in early-phase drug discovery and regulatory safety monitoring. The authors acknowledge that performance of the infrastructure could be further enhanced by incorporating factors such as dose levels and patient-specific parameters.

Research article:

Ruseva V, Dobrev S, Getova-Kolarova V, Peneva A, Getov I, Dimitrova M, Petkova V (2025) In situ development of an artificial intelligence (AI) model for early detection of adverse drug reactions (ADRs) to ensure drug safety. Pharmacia 72: 1–8. https://doi.org/10.3897/pharmacia.72.e160997

Probiotic supplementation may help reduce chemotherapy side effects in breast cancer

A new study published in the journal Pharmacia showed that multi-strain probiotics can have impact on chemotherapy-related side effects.

Chemotherapy is one of the most popular ways to treat breast cancer. Even though it has proven to be effective, it also has downsides, such as a higher risk of side effects, as it doesn’t only attack the cancer cells but can also affect normal cells. Chemotherapy-related side effects such as fatigue, weakness, and proneness to infections can influence treatment adherence        and undermine the effectiveness of the therapy and the patients’ quality of life.

Probiotics are among the emerging alternative supplements being studied for their potential benefit in cancer treatment. A new study published in the open-access journal Pharmacia explored the effects of multi-strain probiotics on chemotherapy-related side effects using Karnofsky performance score, complete blood count, and blood biochemistry.

The findings show that probiotics with 7 strains have a potential effect in patients with breast cancer, especially in the improvement of common chemotherapy-related side effects, such as fatigue and nausea. Another potential effect of multi-strain probiotics is improving blood urea nitrogen levels. Significant improvement was observed in the ability to perform activities of daily living, which means patients can go about their lives more independently and reduce their reliance on others. This might also be related to the improvement of nausea and fatigue symptoms.

Close-up photo of a Black doctor showing pills and pill bottle at the camera

In conclusion, multi-strain probiotics have been shown to have potential benefits in chemotherapy-related side effects for patients with breast cancer undergoing chemotherapy. Further, larger study is needed to explore these effects and to confirm the benefits of multi-strain probiotics before they are used in clinical practice.

Research article:
Kirtishanti A, Wijono H, Kok T, Setiawan E, Tanggo VVCM, Zahara GS, Davina W, Presley B (2025) Effect of multi-strain probiotics supplementation on chemotherapy-related side effects among patients with breast cancer: A pilot trial. Pharmacia 72: 1-9. https://doi.org/10.3897/pharmacia.72.e144998

Anti-melanoma activity found in green-produced nanosilver-chlorhexidine complex

A trustful concentration range was established in which the active agent was effective against the cancer cells but remained non-toxic to the normal cells.

Melanoma is a fast-progressing skin cancer characterized by a high mortality rate after metastasis. Local chemotherapy could be considered a therapeutic approach only in stage 0 of progression (in situ melanoma) and in the postoperative phase after surgical removal of suspected skin lesions. For this purpose, drugs such as Imiquimod, 5-Fluorouracil, Dacarbazine, and Doxorubicin have been tested and shown positive effects. Recently, metal nanoparticles as separate therapeutic units or drug carriers have also fallen into the research focus.

Silver nanoparticles (AgNPs), in particular, are widely recognized as multifunctional tools in nanomedicine, drug delivery, and theranostics. They exert wide-spectrum antimicrobial and anti-tumor properties but their pharmacological effects are in a tight bond with the so-called “surface functionality”. For example, negatively charged and spherical particles are proven less toxic than positively charged particles, especially irregularly shaped (e.g., rods, wires, etc.). Still, because of this, the former are less potent anti-cancer agents.

Contemporary pharmaceutical development relies very much on eco-friendly technologies (often referred to as “green” technologies) that avoid the use of toxic solvents and reagents. This research has implemented one such method based on Camellia sinensis (green tea leaves) as a natural reducer of silver ions. To achieve enhanced antimicrobial and anti-tumor activity, the thus obtained AgNPs were further conjugated to chlorhexidine (Cx+) — a broad-spectrum antimicrobial agent and a cationic surfactant. Indeed, the AgNP-Cx+ complexes have shown highly increased antimicrobial properties, about 18-fold stronger anti-melanoma activity, and 3 times better tumor selectivity compared to the non-functionalized AgNPs.

Anti-tumor activity of silver nanoparticles (AgNPs) and chlorhexidine-silver nanoparticles conjugates (AgNP-Cx+) against melanoma. Legend: ζ-zeta potential – characterizes the charge of the silver nanoparticles; dH hydrodynamic diameter – the size that the particles acquire in an aqueous dispersion; HaCaT – normal cells, human keratinocytes; SH-4 – tumor cells, human melanoma; IC50 – the concentration that inhibits cell proliferation at 50%; SD – standard deviation of the values based on six repetitions of the experiment; ppm – part per million; Selectivity Index – an index that is calculated as the quotient of IC50 on normal cells and IC50 on tumor cells; the greater its value, the more selective the agent against the cancer cells; p-value – statistical indicator; the lower the value the greater the statistical significance between the results obtained with the different samples; a borderline for significance is normally set at p<0.05; n/a – not applicable.

The most valuable outcome of this new study, published in the journal Pharmacia, was the establishment of an adhesive patch prototype as a topical dosage form for the AgNP-Cx+ complex. The used polymers, Hydroxypropyl methylcellulose and Eudragit® RS, demonstrated a lack of negative interference with the antiproliferative action of the active agent but also ensured twice as high activity and even better selectivity against the tumor cells.

Original source:

Ivanova NA (2025) Anti-melanoma activity of green-produced nanosilver-chlorhexidine complex. Pharmacia 72: 1-7. https://doi.org/10.3897/pharmacia.72.e143419