Machine learning-driven optimization of mRNA-lipid nanoparticle vaccine quality with XGBoost/Bayesian method and ensemble model approaches

mRNA vaccines for prostate cancer: A novel promising immunotherapy

The treatment of advanced prostate cancer (PCa) primarily based on androgen deprivation therapy (ADT); however, patients inevitably progress to the castration-resistant prostate cancer (CRPC) stage. Despite the recent advancements in CRPC treatment with novel endocrine drugs that further inhibit androgen receptor signaling, resistance ultimately develops, underscoring the urgent need for new effective therapeutic strategies. Therapeutic cancer vaccines, a form of immunotherapy, exert anti-cancer effects by activating the host's immune system. Over the past few decades, various conventional therapeutic PCa vaccines based on cells, microbes, proteins, peptides, or DNA have been developed and tested in patients with advanced PCa. These attempts have largely failed to improve survival, with the sole exception of sipuleucel-T, which extended the median overall survival of asymptomatic or minimally symptomatic metastatic CRPC (mCRPC) patients by four months. The rapid development and high efficacy of mRNA vaccines during the COVID-19 pandemic have garnered worldwide attention. Compared to conventional vaccines, mRNA vaccines offer several unique advantages, including high production efficiency, low cost, high safety, strong immune response induction, and high adaptability and precision. These attributes make mRNA vaccines a promising frontier in the treatment of advanced PCa.

A Versatile Strategy to Transform Cationic Polymers for Efficient and Organ-Selective mRNA Delivery

The progress of mRNA therapeutics underscores the imperative demand for the development of targeted delivery systems. While cationic polymers hold promise as genetic vectors, their poor in vivo efficacy and numerous variants highlight the urgent need for a universal functionalization strategy to bolster their delivery capabilities. Here, we present a versatile strategy to transform low-cost commercial cationic polymers into phospholipidated and alkylated polymers (PAPs), enabling efficient and organ-selective mRNA delivery in vivo. This straightforward post-functionalization method can be readily broadened to a diverse array of existing cationic polymers, enhancing their cellular uptake, endosomal escape, and mRNA release functionalities. Consequently, PAPs facilitate up to 30,500-fold higher mRNA expression compared to their unmodified counterparts in vivo. Notably, the one-component PAPs enable spleen-specific mRNA delivery, with their vaccine application validated in a mouse melanoma model following intravenous administration. Better still, PAPs can synergize with different helper lipids to formulate four-component lipid nanoparticles (LNPs), achieving respective lung- and liver-specific mRNA delivery. Noteworthy is that these organ-selective mRNA delivery systems significantly outperform previous polymer and LNP benchmarks. This transformation strategy for cationic polymers represents a generalized methodology to give highly effective mRNA carriers, highlighting substantial potential for clinical translation of mRNA therapies with organ-targeting requirements.

Enhancing quality-by-design through weighted goal programming: a case study on formulation of ultradeformable liposomes

INTRODUCTION

Optimization of pharmaceutical formulations requires advanced tools to ensure quality, safety, and efficacy. quality-by-design (QbD), introduced by the FDA, emphasizes understanding and controlling processes early in development. Advanced optimization methods, such as desirability, have surpassed traditional single-objective techniques. Others, such as weighted goal programming (WGP) offers unique advantages by integrating decision-maker preferences, enabling balanced solutions for complex drug delivery systems. This study applies WGP to optimize timolol (TM)-loaded nanoliposomes aligning with QbD principles.

METHODS

The optimization process followed six steps: identifying factors and responses, developing a Design of Experiments (DoE) plan, defining ideal and anti-ideal points, setting aspiration levels, assigning relative weights, and applying WGP compared to desirability function. Minimized and balanced deviations from aspiration levels served as criteria for selecting the most robust optimization results. Six responses were analyzed: vesicle size ( z 1 ) , polydispersity index ( z 2 ) , zeta potential ( z 3 ) , deformability index ( z 4 ) , phosphorus content ( z 5 ) , and drug entrapment efficiency ( z 6 ) .

RESULTS

WGP produced a more balanced formulation that simultaneously optimized multiple responses. By incorporating the importance of each response, the WGP approach improved control over size, colloidal stability, and drug entrapment, based on its mathematical formulation. Comparative analysis with the desirability function confirmed that WGP effectively addressed potential tradeoffs without oversimplifying conflicting objectives.

CONCLUSIONS

This case-study demonstrates WGP potential as an advanced multi-objective optimization tool for pharmaceutical applications, improving upon traditional methods in complex formulations. Its ability to harmonize multiple critical attributes in line with QbD highlights its value in developing high-quality pharmaceutical products.

Development and validation of a prediction model for myelosuppression in lung cancer patients after platinum-based doublet chemotherapy: a multifactorial analysis approach

OBJECTIVE

To develop an individualized prediction model for myelosuppression risk in lung cancer patients undergoing platinum-based doublet chemotherapy and validate its predictive efficacy.

METHODS

A retrospective analysis was conducted on the clinical data of 584 lung cancer patients who received platinum-based doublet chemotherapy at The Affiliated Hospital of Qingdao University between January 2016 and December 2020. Patients were randomly assigned to a training cohort (n=391) and a validation cohort (n=193). Myelosuppression occurred in 280 (71.6%) patients in the training cohort and 132 (68.4%) in the validation cohort. Univariate analysis and LASSO regression were used to identify independent risk factors for myelosuppression. Prediction models were developed using Support Vector Machine (SVM), Random Forest, Extreme Gradient Boosting (XGBoost), and Adaptive Boosting (Adaboost). Model performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and Decision Curve Analysis (DCA). The SHAP algorithm was employed to evaluate feature importance, and a nomogram was developed for individual risk prediction.

RESULTS

LASSO regression identified 10 independent risk factors for myelosuppression: age, body mass index (BMI), white blood cell count, neutrophil count, platelet count, total protein, gender, treatment regimen, targeted therapy, and first chemotherapy cycle. In the training cohort, the XGBoost model exhibited the best performance, with an area under the curve (AUC) of 0.855 (95% CI: 0.813-0.897), while the AUC in the validation cohort was 0.793. SHAP analysis identified white blood cell count, platelet count, neutrophil count, BMI, and age as the most influential predictors. The SHAP analysis based on the XGBoost model demonstrated substantial value.

CONCLUSION

This study successfully developed an individualized prediction model for myelosuppression risk in lung cancer patients following platinum-based doublet chemotherapy, with the XGBoost model achieving high predictive accuracy and clinical utility. The model provides a valuable tool for guiding precision medicine.

Computational biology and artificial intelligence in mRNA vaccine design for cancer immunotherapy

Messenger RNA (mRNA) vaccines offer an adaptable and scalable platform for cancer immunotherapy, requiring optimal design to elicit a robust and targeted immune response. Recent advancements in bioinformatics and artificial intelligence (AI) have significantly enhanced the design, prediction, and optimization of mRNA vaccines. This paper reviews technologies that streamline mRNA vaccine development, from genomic sequencing to lipid nanoparticle (LNP) formulation. We discuss how accurate predictions of neoantigen structures guide the design of mRNA sequences that effectively target immune and cancer cells. Furthermore, we examine AI-driven approaches that optimize mRNA-LNP formulations, enhancing delivery and stability. These technological innovations not only improve vaccine design but also enhance pharmacokinetics and pharmacodynamics, offering promising avenues for personalized cancer immunotherapy.

Interaction design in mRNA delivery systems

Following the coronavirus disease 2019 (COVID-19) pandemic, mRNA technology has made significant breakthroughs, emerging as a potential universal platform for combating various diseases. To address the challenges associated with mRNA delivery, such as instability and limited delivery efficacy, continuous advancements in genetic engineering and nanotechnology have led to the exploration and refinement of various mRNA structural modifications and delivery platforms. These achievements have significantly broadened the clinical applications of mRNA therapies. Despite the progress, the understanding of the interactions in mRNA delivery systems remains limited. These interactions are complex and multi-dimensional, occurring between mRNA and vehicles as well as delivery materials and helper ingredients. Resultantly, stability of the mRNA delivery systems and their delivery efficiency can be both significantly affected. This review outlines the current state of mRNA delivery strategies and summarizes the interactions in mRNA delivery systems. The interactions include the electrostatic interactions, hydrophobic interactions, hydrogen bonding, π-π stacking, coordination interactions, and so on. This interaction understanding provides guideline for future design of next-generation mRNA delivery systems, thereby offering new perspectives and strategies for developing diverse mRNA therapeutics.

'Applications of machine learning in liposomal formulation and development'

Machine learning (ML) has emerged as a transformative tool in drug delivery, particularly in the design and optimization of liposomal formulations. This review focuses on the intersection of ML and liposomal technology, highlighting how advanced algorithms are accelerating formulation processes, predicting key parameters, and enabling personalized therapies. ML-driven approaches are restructuring formulation development by optimizing liposome size, stability, and encapsulation efficiency while refining drug release profiles. Additionally, the integration of ML enhances therapeutic outcomes by enabling precision-targeted delivery and minimizing side effects. This review presents current breakthroughs, challenges, and future opportunities in applying ML to liposomal systems, aiming to improve therapeutic efficacy and patient outcomes in various disease treatments.

Paracyclophane-based ionizable lipids for efficient mRNA delivery in vivo

mRNA therapeutics utilizing lipid nanoparticle (LNP) delivery technology represent a medical innovation for the treatment of various diseases. Amine-derived ionizable cationic lipids have been regarded as the pivotal component of LNPs, which often utilize commercially available small amine molecules as their cores. Given that even minor changes in the structure of ionizable lipids can result in significant differences in the delivery performance, there is a growing need to redesign the lipid amine-cores to optimize mRNA therapy. Here, we rationally design and synthesize a library of 198 paracyclophane-based ionizable lipids (PILs), which are then formulated into LNPs for mRNA delivery in vitro and in vivo. The resulting PIL LNPs display favorable characteristics, including appropriate particle sizes, zeta potentials, mRNA binding capability, efficacious endosomal escape, and robust mRNA delivery in vitro. Tailoring the PIL structures further enables mRNA expression specifically in the liver or simultaneously across multi-organs in vivo. Notably, the optimized PIL LNPs demonstrate superior efficacy compared to the U.S. Food and Drug Administration (FDA) approved DLin-MC3-DMA LNPs following intravenous administration. Additionally, when administered intramuscularly, our PIL LNPs exhibit higher efficacy than the SM-102 and ALC-0315 LNPs that are employed in the coronavirus disease 2019 (COVID-19) mRNA vaccines. These findings demonstrate the potential of paracyclophane-based ionizable lipids in advancing mRNA therapeutics, particularly for liver-targeted drugs and vaccines.