Solid tumors, with their intricate cellular architecture and genetic heterogeneity, have long posed therapeutic challenges. The advent of the CRISPR genome editing system offers a promising, precise genetic intervention. However, the journey from bench to bedside is fraught with hurdles, chief among them being the efficient delivery of CRISPR components to tumor cells. Lipid nanoparticles (LNPs) have emerged as a potential solution. This biocompatible nanomaterial can encapsulate the CRISPR/Cas9 system, ensuring targeted delivery while mitigating off-target effects. Pre-clinical investigations underscore the efficacy of LNP-mediated CRISPR delivery, with marked disruption of oncogenic pathways and subsequent tumor regression. Overall, CRISPR/Cas9 technology, when combined with LNPs, presents a groundbreaking approach to cancer therapy, offering precision, efficacy, and potential solutions to current limitations. While further research and clinical testing are required, the future of personalized cancer treatment based on CRISPR/Cas9 holds immense promise.

Preventable medication harm in oncology is often driven by drug-related adverse events (AEs) that trigger order changes such as holds, dose reductions, delays, rechallenges, and enhanced monitoring. Much of the evidence needed to make these decisions lives in unstructured clinical texts, where large language models (LLMs), a type of artificial intelligence (AI), now offer extraction and reasoning capabilities. In this narrative review, we synthesize empirical studies evaluating LLMs and related NLP systems applied to clinical text for oncology AEs, focusing on three decision-linked tasks: (i) AE detection from clinical documentation, (ii) Common Terminology Criteria for Adverse Events (CTCAE) grade assignment, and (iii) grade-aligned actions. We also consider how these findings can inform pharmacist-facing recommendations for order-level safety. We conducted a narrative review of English-language studies indexed in PubMed, Ovid MEDLINE, and Embase. Eligible studies used LLMs on clinical narratives and/or authoritative guidance as model inputs or reference standards; non-text modalities and non-empirical articles were excluded. Nineteen studies met inclusion criteria. LLMs showed the potential to detect oncology AEs from routine notes and often outperformed diagnosis codes for surveillance and cohort construction. CTCAE grading was feasible but less stable than detection; performance improved when outputs were constrained to CTCAE terms/grades, temporally anchored, and aggregated at the patient level. Direct evaluation of grade-aligned actions was uncommon; most studies reported proxies (e.g., steroid initiation or drug discontinuation) rather than formal grade-to-action correctness. While prospective, real-world impact reporting remained sparse, several studies quantified scale advantages and time savings, supporting an initial role as high-recall triage with pharmacist adjudication. Overall, the evidence supports near-term, pharmacist-in-the-loop use of AI for AE surveillance and review, with CTCAE-structured, citation-backed outputs delivered into the pharmacist’s electronic health record order-verification workspace as reviewable artifacts. Future work must standardize reporting and CTCAE/version usage, and measure grade-to-action correctness prospectively, to advance toward order-level decision support.

Opioid-related iatrogenic harms, including opioid use disorder, overdose, and opioid-induced respiratory depression, constitute a major patient safety challenge. Although clinicians document key safety signals in unstructured clinical narratives, many of these indicators are not readily captured by conventional surveillance approaches that rely on structured administrative data. This clinically focused narrative review synthesizes 47 empirical studies published between 2009 and 2025 that applied artificial intelligence (AI) methods to identify opioid-related harms from clinical text and to address the resulting ascertainment gap. Across studies, administrative coding systems, including ICD-10, often under-ascertain opioid-related events, whereas text-based AI can identify additional cases and contextual details often documented primarily in narrative records, such as fluctuating mental status, suspected drug causality, and responses to naloxone. Methodologically, the literature has progressed from interpretable rule-based lexicons to machine learning and deep learning models and, more recently, to transformer-based approaches, including large language models (LLMs) for classification and schema-driven extraction. Rule-based systems established the feasibility of transparent surveillance and frequently recovered clinically documented cases missed by billing codes. Subsequent supervised and deep learning approaches expanded scalability and, in a smaller subset of studies, were integrated into electronic health record workflows with operational metrics reported. More recent transformer- and LLM-based studies emphasize richer extraction schemas and benchmark development, including characterization of overdose context and intentionality and identification of potential prodromal neurocognitive signals, although external validation, calibration, and prospective outcome evaluation remain inconsistently reported. Given that the evidence base is predominantly retrospective and that clinical workflow studies remain comparatively few, a pragmatic near-term clinical role is to provide detection-to-triage decision support rather than autonomous diagnosis, in which systems surface candidate cases with reviewable evidence for clinician adjudication. Future progress will require greater standardization of phenotype definitions, routine equity auditing and subgroup reporting, stronger external validation and calibration at operational thresholds, and a shift from retrospective discrimination metrics toward prospective assessments of the clinical workflow impact, clinical utility, and patient-centered outcomes.

Synthetic RNA has become an essential modality in therapeutic development. Linear mRNA is already clinically validated, which demonstrated that in vitro-transcribed (IVT) RNA can achieve robust protein expression in humans and can be manufactured at a large scale. Circular RNA (circRNA) represents a more recent format characterized by a covalently closed backbone that confers enhanced resistance to exonucleases and supports sustained translation when paired with appropriate regulatory elements. Although both formats are produced through cell-free synthesis, their manufacturing pathways are distinct. Linear mRNA synthesis requires transcription, capping, polyadenylation, and stringent removal of double-stranded RNA contaminants. circRNA production generally proceeds through transcription of a linear precursor followed by enzymatic or ribozyme-mediated circularization, with emerging strategies such as permuted intron-exon designs improving efficiency and reducing extraneous sequence content. This review summarizes the principal methods used to generate linear and circRNA and identifies the technical barriers that must be overcome during the manufacturing process.

Multistate survival models (MSMs) are widely used in the medical field of clinical studies. For example, in type 2 diabetes mellitus (T2D), these models can be applied to describe progression in T2D by predefining several T2D states based on available biometric measurements such as hemoglobin A1 C (HbA1c). In most cases, MSMs come with an assumption that the examination process is independent of disease progression. However, in practice, complete independence between disease progression and examination processes is unrealistic, as the frequency at which a patient accesses healthcare may vary based on treatment and/or control of the health condition. We built a joint model of a 4-state transition process of T2D with informative examination scheme (i.e., the patterns of examination times are not random). Risk factors including age, sex, race, and socioeconomic disadvantage were included in a log-linear model examining T2D transition intensities and healthcare visit frequencies. Parameters of the joint model are estimated under the framework of likelihood function by the expectation–maximization (EM) algorithm. The joint model demonstrated that people living in neighborhoods with greater socioeconomic disadvantage had a lower healthcare visit frequency under all 4 defined T2D statuses. Evaluation of race/ethnicity revealed that comparing to non-Hispanic White patients, Black patients had higher risk for progressing from Normal to Prediabetes, T2D, and Uncontrolled T2D states. Our joint model offers a framework for analyzing multistate survival processes while accounting for the dependence between disease progression and examination frequency. Unlike traditional MSMs that estimate only transition intensities, our model captures variations in healthcare visit frequencies across different disease states, providing a more comprehensive understanding of disease dynamics and healthcare access patterns.

Multistate transition models (MSTMs) are valuable tools depicting disease progression. However, due to the complexity of MSTMs, larger sample size and longer follow-up time in real-world data, the computation of statistical estimation and inference for MSTMs becomes challenging. A bounded Taylor series in Newton–Raphson procedure is proposed which leverages the uniformization technique to derive maximum likelihood estimates and corresponding covariance matrix. The proposed method, namely uniformization Taylor-bounded Newton–Raphson, is validated in three simulation studies, which demonstrate the accuracy in parameter estimation, the efficiency in computation time and robustness in terms of different situations. This method is also illustrated using a large electronic medical record data related to statin-induced side effects and discontinuation.