The exponential growth of biomedical data necessitates advanced tools for efficient information extraction (IE) to support clinical decision-making and research. Large language models (LLMs) have emerged as transformative solutions, yet their application in healthcare raises critical trade-offs between open-source (OSS) and proprietary models. This review evaluates IE workflows such as named entity recognition, relation extraction, and terminology normalization, through five axes: performance (including schema fidelity), reproducibility, cost, transparency & auditability, and patient-centric governance. While proprietary models excel in schema compliance and complex reasoning, OSS models offer advantages in auditability, local control, and cost-effectiveness. Challenges such as schema fidelity, reproducibility, and ethical considerations like algorithmic fairness and data sovereignty are emphasized. The analysis highlights that OSS models, though requiring domain-specific adaptation, enable greater transparency and customization for privacy-sensitive tasks, whereas proprietary systems face limitations in bias mitigation and regulatory alignment. By addressing technical, ethical, and operational challenges, this work underscores the importance of context-aware model selection to balance innovation with accountability in clinical AI deployment. The findings advocate for hybrid approaches that integrate OSS flexibility with proprietary capabilities, ensuring equitable, reliable, and compliant healthcare solutions.

The integration of biological tissues with conducting polymers is crucial for the advancement of bioelectronics. Although the potential toxicity of some conducting-ink components is an acknowledged constraint for in vivo use, a more fundamental limitation is the lack of molecular understanding of how conducting polymers interact with biological barriers. This study addresses some of the limitations by presenting a novel method for direct polymerization of pyrrole on excised skin in vitro and investigates the evolving skin/conducting polymer interface using electrical impedance spectroscopy (EIS), X-ray diffraction, vapor sorption, and complementary analyses. Direct polymerization of pyrrole on skin is performed by applying an aqueous pyrrole solution to the dermal side and an oxidant (ferric ion solution) on the stratum corneum (SC) side of the skin membrane fixed in a Franz cell. EIS measurements show a marked reduction in skin barrier resistance after PPy formation, whereas control experiments produced the opposite trend, confirming that the resistance drop arises specifically from polymerization. Our complementary analyses indicate that polymerization is primarily localized at the solution-SC interface but also support the formation of an electrically coupled pathway across the barrier. These results provide molecular-level insight into polymer-SC interactions, imply the formation of a coherent conductive junction that extends through the SC layers, and inform the design of future skin-integrated bioelectronic materials.

Niacinamide (NIA) is a widely used skincare ingredient with established benefits for skin barrier support, inflammation reduction, and dermal health. However, the mechanisms governing its transdermal delivery remain insufficiently understood, particularly regarding how formulation pH influences its permeation through the stratum corneum (SC). This study investigates how donor phase pH (5.0 vs. 7.4) modulates NIA skin permeation and how these effects relate to pH induced changes in SC electrical properties. Franz cell diffusion experiments were combined with electrical impedance spectroscopy (EIS) using full‑thickness human skin and 3D reconstructed epidermal tissue models. Permeation was quantified over 24 h and in pH switch experiments, while EIS characterized pH dependent changes in membrane resistance (Rmem) and effective capacitance (Ceff). Additional analyses assessed microbial conversion of NIA to nicotinic acid during prolonged exposure. Neutral donor pH (7.4) increased NIA permeation by roughly twofold compared with acidic pH (5.0) in both membrane types. Correspondingly, pH 7.4 decreased Rmem and increased Ceff, indicating pH driven changes in SC lipid organization and dielectric behavior. These effects were reversible and likely stem from alterations in SC lipid domains, including pH dependent partial deprotonation of free fatty acids that modifies the continuous lipid regions and introduce localized structural microdefects. Such changes enhance NIA and ion permeability and increase SC dielectric properties at neutral pH. Although microbial conversion of NIA to nicotinic acid was negligible within the first 24 h, it became clearly detectable upon prolonged experiments. In conclusion, donor phase pH is a critical determinant of NIA skin permeation, primarily through reversible modulation of SC lipid structure and transport pathways. These findings highlight the importance of pH control in topical formulations and underscore the need to consider microbiota‑mediated transformations when evaluating the efficacy and safety of skin care products containing NIA.

Replicating the complex physiology of human skin in vitro remains a challenge for cutaneous oncology. Although several three-dimensional (3D) skin and melanoma models have been developed, reproducible skin equivalent systems that enable direct and standardized comparisons between healthy and melanoma-containing skin remain valuable experimental tools. This study aimed to optimize and characterize skin equivalents containing either melanocytes or melanoma cells and to evaluate their utility as platforms for investigating melanoma-associated skin biology and UVB-induced responses in vitro. To achieve this, 3D skin equivalents containing fibroblasts, keratinocytes, and either melanocytes or melanoma cells were reconstructed on a polystyrene scaffold to generate healthy (mc) and melanoma (mm) models. Morphological characteristics were compared with clinically verified human tissue sections, and functional responses to UVB irradiation were assessed with a focus on the kynurenine pathway. The models demonstrated a distinct dermal-epidermal architecture and distinguishable melanoma-associated features, including epidermis-confined melanoma cell clusters, in mm constructs. UVB exposure induced differential responses between models, including significant differences in kynurenine pathway regulation. These findings suggest that the developed skin equivalents provide a reproducible in vitro platform for studying melanoma-associated skin biology and treatment-related metabolic responses and may support future mechanistic studies of skin cancer progression.

Purpose: To evade immunosurveillance many cancers convert tryptophan (Trp) into kynurenine (Kyn), which induces immunotolerance and suppresses immune responses. Elevated Kyn amounts have been found in blood from patients with cutaneous melanoma. This study aimed to investigate whether higher Kyn abundance and lower Trp abundance can be detected on the surface of cutaneous melanoma lesions compared with adjacent non-lesional skin.

Methods: Sixteen patients with suspected melanomas were enrolled in this study. All lesions were excised and histopathologically diagnosed: 7 lesions were diagnosed as invasive malignant melanomas (MM), 6 as melanomas in situ (MIS), and 3 as benign lesions (BL). Non-invasive metabolite sampling was performed by tape stripping of suspected skin lesions and adjacent healthy non-lesional (NL) skin. Trp, Kyn, tyrosine (Tyr), and phenylalanine (Phe) were quantified by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Electrical impedance spectroscopy (EIS) measurements were conducted to assess skin barrier integrity.

Results: Levels of all metabolites, Tyr (x6), Phe (x6), Trp (x5), and Kyn (x3), were significantly higher in MM lesions compared with adjacent NL skin, resulting in an elevated Trp/Kyn ratio. Trp levels increased less than Phe and Tyr levels, suggesting a potential increase in Trp depletion. Skin resistance in MM lesions was half that of NL skin. No differences were observed between MIS or BL and NL skin.

Conclusions: Non-invasive skin sampling revealed elevated Tyr, Phe, Trp and Kyn levels in MM skin, which is likely the result of compromised skin barrier at this stage of cutaneous melanoma.

Early detection of cancer via biomarkers is vital for improving patient survival rates. In the case of skin cancers, low-molecular-weight biomarkers can penetrate the skin barrier, enabling non-invasive sampling at an early stage. This study focuses on detecting tryptophan (Trp) and kynurenine (Kyn) on the surface of reconstructed 3D melanoma and melanocyte models. This is examined in connection with IDO-1 and IL-6 expression in response to IFN-γ or UVB stimulation, both crucial factors of the melanoma tumor microenvironment (TME). Using a polystyrene scaffold, full-thickness human skin equivalents containing fibroblasts, keratinocytes, and melanocytes or melanoma cells were developed. The samples were stimulated with IFN-γ or UVB, and Trp and Kyn secretion was measured using HPLC-PDA and HPLC-MS. The expression of IDO-1 and IL-6 was measured using RT-qPCR. Increased Trp catabolism to Kyn was observed in IFN-γ-stimulated melanoma and melanocyte models, along with higher IDO-1 expression. UVB exposure led to significant changes in Kyn levels but only in the melanoma model. This study demonstrates the potential of skin surface Trp and Kyn monitoring to capture TME metabolic changes. It also lays the groundwork for future in vivo studies, aiding in understanding and monitoring skin cancer progression.

Background and aim: Melanoma is a fatal form of skin cancer that carries a grave prognosis if the cancer cells spread and form metastases. The Kynurenine (Kyn) pathway is activated by the enzyme indoleamine 2,3-dioxygenase 1 (IDO-1) and has been shown to have a role in tumour progression. We have previously shown that interferon-γ (IFN-γ) acts as an inducer of tryptophan (Trp) degradation to Kyn in keratinocytes of the basal layer in a 3D epidermis model. Before extending our reconstructed human epidermis model to not only contain keratinocytes but also fibroblasts and melanocytes/melanoma cells, we have in this study set out to investigate possible differences between primary adult melanocytes and six melanoma cell lines regarding the expression of the immune checkpoint inhibitors IDO-1 and programmed death ligand 1 (PD-L1) together with Kyn production.

Methods: The melanocytes and melanoma cells were stimulated with 1–20 ng/ml of IFN-γ and the levels of Trp to Kyn degradation were monitored with high-performance liquid chromatography (HPLC). To analyze the viability of the cell types after IFN-γ treatment, an MTT assay was performed. mRNA quantity of IDO-1, PD-L1 and IFN-γ receptor (IFN-GR1) was analyzed with qPCR.

Results: After 24 h, only the metastatic cell line WM-266-4 was affected by all concentrations of IFN-γ, whereas at 48 h, the higher IFN-γ concentrations gave a more pronounced effect on the viability in all cell types. Trp was detected at various levels in the culture medium from all cell types before and after IFN-γ treatment. The degradation to Kyn was detected in primary melanocytes, Mel Juso, and Mel Ho cell lines after 24 h of treatment and low levels of IFN-γ. However, the higher concentration of IFN-γ, 20 ng/ml, induced Kyn to various degrees in all cell types after 24 h. The change in mRNA quantity of IDO-1 and PD-L1 was similar in all cell types.

Conclusion: To conclude, no significant difference in upregulation of the immune checkpoint inhibitors PD-L1 and IDO-1 was seen between primary tumour and metastatic melanoma. IFN-γ stimulation of melanocytes and different stages of melanoma cell lines resulted in an increased Kyn/Trp ratio in the more aggressive melanoma cells when a high concentration was used (20 ng/ml) but when a lower concentration of IFN-γ (5 ng/ml) was used an increased Kyn/Trp ratio were detected in media from primary melanocytes and early-stage melanoma.

Introduction: To minimize the risk of developing foot-ulcers, persons with diabetes are given the advice to daily inspect their feet and to apply skincare formulations. However, commercially available skincare products have rarely been developed and evaluated for diabetes foot care specifically. The primary aim of this randomized controlled trial (RCT) is to evaluate the effects in reducing foot xerosis in persons with diabetes without footulcers using two skincare creams containing different humectants (interventions) against a cream base nonhumectant (comparator). Secondary outcomes are to evaluate differences on skin barrier integrity, lowmolecular weight biomarkers and skin microbiota, microcirculation including transcutaneous oxygen pressure, degree of neuropathy, and HbA1c between intervention-comparator creams. Methods: Two-armed double-blind RCT, registered in ClinicalTrials.gov Identifier: NCT06427889. With 80 % power, two-tailed significance of 2.5 % in each arm, 39 study persons is needed in each arm, total 78 persons, 98 including dropouts, to be able to prove a reduction of at least one category in the Xerosis Severity Scale with the intervention creams compared to the comparator. In one arm, each participant will treat one foot with one of the intervention creams (Oviderm (R) or Canoderm (R)), while the opposite foot will be treated with the comparator cream (Decubal (R) lipid cream), twice a day. If needed, participants are enrolled after a wash-out period of two weeks. The participants will undergo examinations at baseline, day 14 and day 28. Discussion: This RCT evaluate the potential effects of humectants in skin creams against foot xerosis in persons with diabetes.

The molecular composition of human skin is altered due to diseases, which can be utilized for non-invasive sampling of biomarkers and disease diagnostics. For this to succeed, it is crucial to identify a sampling formulation with high extraction efficiency and reproducibility. Highly hydrated skin is expected to be optimal for increased diffusion of low-molecular-weight biomarkers, enabling efficient extraction as well as enhanced reproducibility as full hydration represents a well-defined endpoint. Here, the aim was to explore water-based formulations with high water activities, ensuring satisfactory skin hydration, for non-invasive sampling of four analytes that may serve as potential biomarkers, namely tryptophan, tyrosine, phenylalanine, and kynurenine. The included formulations consisted of two hydrogels (chitosan and agarose) and two different liquid crystalline cubic phases based on the polar lipid glycerol monooleate, which were all topically applied for 2 h on 35 healthy subjects in vivo. The skin status of all sampling sites was assessed by electrical impedance spectroscopy and transepidermal water loss, enabling explorative correlations between biophysical properties and analyte abundancies. Taken together, all formulations resulted in the successful and reproducible collection of the investigated biomarkers. Still, the cubic phases had an extraction capacity that was approximately two times higher compared to the hydrogels.

Monitoring of low-molecular weight cancer biomarkers, such as tryptophan (Trp) and its derivative kynurenine (Kyn), might be advantageous to non-invasive skin cancer detection. Thus, we assessed several approaches of topical sampling of Trp and Kyn, in relation to phenylalanine (Phe) and tyrosine (Tyr), on the volar forearm of six healthy volunteers. The sampling was performed with three hydrogels (made of agarose or/and chitosan), hydrated starch films, cotton swabs, and tape stripping. The biomarkers were successfully sampled by all approaches, but the amount of collected Kyn was low, 20 ± 10 pmol/cm2. Kyn quantification was below LOQ, and thus, it was detected only in 20% of topical samples. To mitigate variability problems of absolute amounts of sampled amino acids, Tyr/Trp, Phe/Trp, and Phe/Tyr ratios were assessed, proving reduced inter-individual variation from 79 to 45% and intra-individual variation from 42 to 21%. Strong positive correlation was found between Phe and Trp, pointing to the Phe/Trp ratio (being in the 1.0–2.0 range, at 95% confidence) being least dependent on sampling materials, approaches, and sweating. This study leads to conclusion that due to the difficulty in quantifying less abundant Kyn, and thus the Trp/Kyn ratio, the Phe/Trp ratio might be a possible, alternative biomarker for detecting skin cancers.