Daily Cosmetic Research Analysis
A multicenter randomized trial shows poly-L-lactic acid (PLLA) outperforms hyaluronic acid for midfacial volume restoration with durable 12-month efficacy and comparable safety. A biomaterial study presents decellularized Wharton’s jelly scaffolds that enhance cartilage regeneration while inhibiting angiogenesis. A multimodal deep learning model accurately predicts acute dermal toxicity, advancing 3Rs-aligned safety assessment for cosmetics and chemicals.
Summary
A multicenter randomized trial shows poly-L-lactic acid (PLLA) outperforms hyaluronic acid for midfacial volume restoration with durable 12-month efficacy and comparable safety. A biomaterial study presents decellularized Wharton’s jelly scaffolds that enhance cartilage regeneration while inhibiting angiogenesis. A multimodal deep learning model accurately predicts acute dermal toxicity, advancing 3Rs-aligned safety assessment for cosmetics and chemicals.
Research Themes
- Evidence for aesthetic injectables (PLLA vs hyaluronic acid)
- Biomaterials for cartilage regeneration with anti-angiogenic properties
- AI-driven, multimodal toxicity prediction reducing animal testing
Selected Articles
1. Efficacy and Safety of Poly-l-Lactic Acid for Correction of Midfacial Volume Loss and Contour Defects: A Prospective, Multicenter, Randomized, Parallel-Controlled, Evaluator-Blinded, Superiority Trial.
In a multicenter randomized, assessor-blinded superiority trial (n=331), PLLA outperformed hyaluronic acid for midfacial volume restoration, with higher MMVS and GAIS responses at 6 and 12 months. Safety was comparable, with only slightly increased transient injection-site reactions in the PLLA group.
Impact: This RCT provides robust head-to-head evidence supporting PLLA over hyaluronic acid for durable midface augmentation with acceptable safety. It can directly inform product selection and patient counseling in aesthetic practice.
Clinical Implications: PLLA can be considered a first-line filler for midfacial volume loss when durability is prioritized, with counseling on mild transient injection site reactions. Follow-up beyond 12 months and comparative data with multiple HA products remain needed.
Key Findings
- Primary efficacy: PLLA achieved higher MMVS efficacy than HA (90.57% vs 51.01%; difference 39.56%, 95% CI 30.34%–48.78%).
- Durability: At 6 months, MMVS 93.04% vs 69.33% and GAIS (investigator) 99.37% vs 86.67%; at 12 months, MMVS 84.91% vs 46.98% and GAIS 94.34% vs 74.50% (all p<0.05).
- Patient-reported outcomes: Satisfaction was higher in the PLLA group (p<0.05).
- Safety: No significant differences in vitals, labs, or overall adverse events; slightly higher injection-site reactions with PLLA resolved in 1–3 days.
Methodological Strengths
- Prospective, multicenter, randomized, assessor-blinded superiority design.
- Consistency across worst-case and per-protocol analyses with validated scales (MMVS, GAIS).
Limitations
- Assessor-blinded but not fully double-blinded; potential performance/placebo effects cannot be excluded.
- Comparator limited to one HA control; follow-up limited to 12 months.
Future Directions: Longer-term comparative studies versus multiple HA formulations and other biostimulatory fillers, standardized injection protocols, and imaging-based volumetric outcomes will strengthen generalizability and mechanistic understanding.
BACKGROUND: Poly-l-lactic acid (PLLA) is widely used in esthetic medicine due to its excellent biocompatibility and biodegradability. This study aimed to evaluate the effectiveness and safety of PLLA facial filler in correcting midfacial volume loss and/or contour defects. METHODS: In this prospective, multicenter, randomized, assessor-blinded, superiority clinical trial, 331 subjects were randomly assigned to receive either PLLA (experimental group) or hyaluronic acid (HA, control group). Efficacy was assessed using the Midfacial Volume Scale (MMVS) and Global Aesthetic Improvement Scale (GAIS). Safety was evaluated based on adverse events. RESULTS: The PLLA group demonstrated significantly higher efficacy in MMVS scores compared to the HA group (90.57% vs. 51.01%, difference of 39.56%, 95% CI: 30.34%-48.78%). These results were consistent across worst-case and per-protocol analyses. Secondary outcomes revealed higher MMVS and GAIS scores in the PLLA group at 6 months (MMVS: 93.04% vs. 69.33%, GAIS investigator rating: 99.37% vs. 86.67%) and 12 months (MMVS: 84.91% vs. 46.98%, GAIS investigator rating: 94.34% vs. 74.50%) (p < 0.05). Participant satisfaction surveys showed higher satisfaction in the PLLA group (p < 0.05). Safety analysis revealed no significant differences in vital signs, laboratory tests, or adverse events (p > 0.05), with a slightly higher incidence of injection site reactions in the PLLA group, which resolved within 1-3 days. CONCLUSION: The comprehensive analysis of results indicates that PLLA is a safe and effective treatment for the correction of midfacial volume loss and midfacial contour defects.
2. Decellularized cartilage scaffolds derived from wharton's jelly facilitate cartilage regeneration and inhibit angiogenesis.
Two decellularization strategies for Wharton’s jelly scaffolds were compared. NFT-dWJ preserved ECM better and showed superior mechanical properties, enhanced BMSC chondrogenesis, accelerated in vivo cartilage repair, and inhibited endothelial angiogenesis, indicating promise for cartilage regeneration while maintaining avascularity.
Impact: The scaffold simultaneously promotes chondrogenesis and inhibits angiogenesis—two often competing goals—offering a strategically designed matrix for cartilage regeneration relevant to reconstructive and aesthetic surgery.
Clinical Implications: Although preclinical, NFT-dWJ scaffolds could evolve into off-the-shelf matrices for articular or reconstructive cartilage repair (e.g., nasal/auricular), potentially improving outcomes by preserving avascular cartilage physiology.
Key Findings
- NFT-dWJ preserved native ECM structure and key components (collagen, glycosaminoglycans) better than TFT-dWJ.
- TFT-dWJ had higher pore size/porosity/swelling but lower Young’s modulus; degradation rates were similar between scaffolds.
- Both scaffolds were biocompatible; NFT-dWJ enhanced BMSC chondrogenesis in vitro and accelerated in vivo cartilage repair.
- dWJ scaffolds inhibited localized angiogenesis in HUVEC assays, supporting avascular cartilage regeneration.
Methodological Strengths
- Head-to-head comparison of two decellularization protocols with comprehensive physicochemical and biological characterization.
- Demonstrated efficacy across in vitro chondrogenesis assays and in vivo cartilage repair models.
Limitations
- Preclinical study; long-term durability, immunogenicity, and functional integration remain untested in large animals/humans.
- No direct comparison with clinically used scaffolds or cell-based constructs.
Future Directions: Evaluate long-term integration and mechanical performance in large animal models, compare with clinical gold-standard scaffolds, and assess manufacturability/sterilization and regulatory pathways.
The avascular nature of articular cartilage severely limits its ability to self-repair after injury, which poses a challenge for clinical treatment, and tissue engineering aims to address this issue with scaffold-based strategies. However, the defining characteristics of an optimal scaffold remain controversial. In this study, we prepared two types of decellularized wharton's jelly (dWJ) scaffolds by trypsin combined with repeated freeze-thawing (TFT) and nuclease combined with repeated freeze-thawing (NFT), respectively. The scaffolds were tested with general characterization, decellularization effect, extracellular matrix (ECM) composition and structure retention, mechanical properties, biocompatibility, in vivo and in vitro chondrogenic effects, and in vitro anti-angiogenic effects. The results showed that the TFT-dWJ scaffolds possessed higher pore size, porosity, and swelling rate, but their Young's modulus was lower than that of the NFT-dWJ scaffolds. Both scaffolds were generally similar in terms of degradation rates. In comparison, the native ECM structure and the major components of collagen and glycosaminoglycans were better preserved in NFT-dWJ scaffolds. Importantly, dWJ scaffolds showed favorable biocompatibility and markedly promoted the chondrogenic differentiation of bone marrow mesenchymal stem cells (BMSCs) in vitro, and accelerated cartilage damage repair in vivo. This was particularly evident with NFT-dWJ. Secondly, the dWJ scaffolds exhibited the capability to inhibit localized angiogenesis in human umbilical vein endothelial cells (HUVECs), a property that could be advantageous for preserving avascularity throughout the cartilage regeneration process. This study presents an ECM-derived scaffold fabrication strategy that optimally preserves matrix composition and microstructure, offering a promising solution for cartilage regeneration.
3. Deep Learning-Based Multimodal Fusion Approach for Predicting Acute Dermal Toxicity.
TriModalToxNet fuses 2D molecular images, SMILES embeddings, and fingerprints to predict acute dermal toxicity, achieving AUROC 95% and sensitivity 91.2% on 10-fold cross-validation with external validation support. Multimodal fusion outperformed bi-modal and single-modality baselines, aligning with 3Rs and regulatory-ready screening.
Impact: By integrating heterogeneous molecular representations, this model markedly improves dermal toxicity prediction, offering a scalable alternative to animal testing in cosmetics and chemical safety pipelines.
Clinical Implications: While not clinical, accurate in silico dermal toxicity prediction can streamline ingredient selection and preclinical safety assessment, potentially accelerating safe cosmetic and topical drug development while reducing animal use.
Key Findings
- TriModalToxNet achieved AUROC 95% and sensitivity 91.2% via stratified 10-fold cross-validation.
- External validation demonstrated robustness and generalizability beyond cross-validation.
- Multimodal fusion (2D images, SMILES, fingerprints) outperformed bi-modal and single-modality approaches.
Methodological Strengths
- Curated dataset of 3845 compounds with standardized evaluation (AUROC, sensitivity, MCC, accuracy).
- Model comparison including bi-modal baseline; both cross-validation and external validation performed.
Limitations
- Labels derived from rat/rabbit toxicity; human dermal translation requires further validation.
- Model interpretability and potential dataset biases are not fully addressed; code/data availability not specified.
Future Directions: Prospective benchmarking on human-relevant in vitro skin models, explainability analyses, and regulatory validation (e.g., OECD contexts) will be key for deployment.
Acute dermal toxicity testing is essential for assessing the safety of chemicals used in pharmaceuticals, pesticides, cosmetics, and industrial chemicals. Conventional toxicity testing methods rely significantly on animal tests, which are resource-intensive and time-consuming and raise ethical issues. To address these issues and support the 3Rs principle (replacement, reduction, and refinement) in animal testing, this study investigates whether a multimodal deep learning framework based on the fusion of heterogeneous molecular representations can yield a reliable and accurate model for the prediction of acute dermal toxicity. This study proposes TriModalToxNet, a novel architecture that extracts features from three distinct molecular representations: 2D molecular images through a 2D convolutional neural network, SMILES embeddings via a 1D convolutional neural network, and molecular fingerprints via a fully connected neural network. These extracted features are then concatenated and passed into a deep neural network for classification. For comparative purposes, this study also evaluates BiModalToxNet, a baseline model using only 2D molecular images and fingerprints. The models are trained and tested on a curated data set consisting of 3845 compounds derived from experimental rat and rabbit acute dermal toxicity studies. The proposed model is evaluated using multiple standard performance metrics such as area under the receiver operating characteristic curve, sensitivity, Matthews correlation coefficient, and accuracy derived from stratified 10-fold cross-validation and external validation. TriModalToxNet achieved an area under the receiver operating characteristic curve of 95% and a sensitivity of 91.2% in cross-validation. External validation was also conducted to further demonstrate the robustness and generalizability of the model. These results show that multimodal methods can attain better predictive performance than traditional single-modality methods. This TriModalToxNet framework highlights the potential for integration into regulatory frameworks, pharmaceutical screening pipelines, and advancing the field toward more ethical and efficient chemical safety assessment.