Publications

The development of new therapeutics and the validation of pathogenetic cancer mechanisms require representative laboratory models. However, existing collections represent only a fraction of the diversity observed in human cancer. Recent technologies have enabled efficient in vitro model derivation (for example, tumour organoids). However, whether these maintain essential properties of patient tumours during long-term expansion has not been systematically investigated. Here we present results of a large-scale international programme—the Human Cancer Models Initiative—which involved the generation of a resource of 665 next-generation models from 2,780 donors with 25 cancer types and integrated tumour–model whole genome, exome, methylome and transcriptome analyses. The resource provides 522 models with comprehensive clinical data, 153 models of rare cancers and 71 models from participants with non-European ancestry. Analyses of 421 matched tumour–model pairs reveal high genetic (97.8%) and epigenetic (95%) concordance and define correlates of model discordance. Single-nucleus RNA sequencing of tumour–model pairs reveals subsets of models in which culture conditions significantly influence cell states. Finally, we characterize model preservation of extrachromosomal DNA and post-treatment mutational signatures to provide opportunities to study therapeutic resistance. This model repository is being made available to the community—including multimodal molecular profiling, clinical information and integrative software tools—thus providing a valuable resource for preclinical investigation of cancer pathogenesis and treatment response.

Since the clinical trials for the first COVID-19 vaccines in 2020, interest in RNA-based therapeutics has grown rapidly, with promising applications in vaccines, oncology, and gene therapy. This surge has created a strong demand for scalable, cost-effective, and robust manufacturing platforms for messenger RNA. However, current mRNA purification largely relies on batch-wise chromatography and tangential flow filtration, which face limitations in scalability, cost, and compatibility with continuous production. Chromatographic techniques often require harsh conditions, such as high pH, salt, or organic solvents, that may compromise mRNA stability. Additionally, extensive sample conditioning (e.g., dilution, heating) is typically required to reduce aggregation and facilitate column loading, further hindering continuous operation. To address these challenges, a fully continuous precipitation-based method for mRNA purification is developed. The process consists of an optimized precipitation step using PEG6000 and NaCl in a tubular reactor, followed by two continuous TFF stages for washing and buffer exchange. The overall process achieves yields of 92% and purities of 95%, with no detectable double-stranded RNA formation, residual proteins, fragmentation, or aggregates. Compared to traditional approaches, this method achieves higher yields and purities while offering enhanced process robustness and integration potential. The final mRNA product can be directly encapsulated into lipid nanoparticles without further conditioning, with no observed degradation or aggregation. This platform offers a scalable, flexible alternative to chromatography, suitable for integration into end-to-end continuous mRNA manufacturing.

Lipid nanoparticles (LNPs) are the leading platform for delivering nucleic acid therapeutics, produced by rapidly mixing lipids in ethanol with nucleic acid cargo in an aqueous buffer. LNP production is often approached with a mixing-focused mindset that reduces the entire self-assembly process to a single step, obscuring the relationship between the process inputs and LNP properties. Here, we present a method for producing mRNA-loaded LNPs, with independent and predictive control over both the size and morphology and without compromising other quality attributes. By decoupling particle design from mixing and formulation changes, this method enables the rational engineering of LNPs with defined properties. The method leverages mixing under high fusogenicity conditions, achieved by modulating the solvent composition, followed by timed postinjection of an aqueous buffer to kinetically arrest LNPs at the desired properties. We demonstrate the method using benchmark LNP formulations in an impinging jet mixer, a state-of-the-art technology for LNP manufacturing. The resulting LNPs exhibit up to an 8-fold increase in in vitro transfection efficacy compared to those produced by the conventional method. In addition, the method facilitates quality control and supports predictive modeling and rational process translation.

The increasing demand for mRNA-based therapeutics requires scalable and cost-effective purification methods. Chromatography-based approaches, such as Oligo-dT affinity chromatography, require multiple processing steps, including buffer exchange and high-temperature treatments, which can lead to mRNA degradation and increased costs. Although precipitation is commonly used at the lab scale, its application in large-scale mRNA purification remains underexplored. In this study, we developed a rapid and efficient precipitation-based method for mRNA purification. The effects of different precipitation conditions, including salt type, co-precipitating agents, and time, were evaluated in terms of recovery yield and purity. Precipitation experiments were conducted for two different in vitro transcription crude materials containing mRNAs of varying lengths and concentrations. The optimal conditions, identified as a combination of NaCl and PEG 6000, achieved recovery yields of 80-93% and purities of 80-83%, with no detectable double-stranded RNA formation or fragmentation. The precipitated mRNAs were successfully transfected into 293T and A549 cells, showing protein expression comparable to commercial and chromatography-purified mRNAs. These findings position PEG/NaCl precipitation as a rapid, scalable, and cost-effective alternative to traditional chromatography methods, significantly reducing process complexity and time for large-scale mRNA production.