Advancing algorithmic drug product development: Recommendations for machine learning approaches in drug formulation.

Artificial intelligence is a rapidly expanding area of research, with the disruptive potential to transform traditional approaches in the pharmaceutical industry, from drug discovery and development to clinical practice. Machine learning, a subfield of artificial intelligence, has fundamentally transformed in silico modelling and has the capacity to streamline clinical translation. This paper reviews data-driven modelling methodologies with a focus on drug formulation development. Despite recent advances, there is limited modelling guidance specific to drug product development and a trend towards suboptimal modelling practices, resulting in models that may not give reliable predictions in practice. There is an overwhelming focus on benchtop experimental outcomes obtained for a specific modelling aim, leaving the capabilities of data scraping or the use of combined modelling approaches yet to be fully explored. Moreover, the preference for high accuracy can lead to a reliance on black box methods over interpretable models. This further limits the widespread adoption of machine learning as black boxes yield models that cannot be easily understood for the purposes of enhancing product performance. In this review, recommendations for conducting machine learning research for drug product development to ensure trustworthiness, transparency, and reliability of the models produced are presented. Finally, possible future directions on how research in this area might develop are discussed to aim for models that provide useful and robust guidance to formulators.

Exploring the role of ex vivo metabolism on blood and plasma measurements of oxytocin among women in the third stage of labour: A post hoc study.

AIMS: To examine the role of ex vivo oxytocin metabolism in post-dose peptide measurements. METHODS: The stability of oxytocin (Study 1) and oxytocinase activity (Study 2) in late-stage pregnancy blood was quantified using liquid-chromatography tandem mass-spectrometry (LC-MS/MS) and a fluorogenic assay, respectively. Analyses were conducted using blood from pregnant women (>36 weeks gestation) evaluated in lithium heparin (LH), ethylenediaminetetraacetic acid (EDTA) and BD P100 blood collection tubes with or without protease inhibitors. In addition, plasma oxytocin concentrations following administration of oxytocin 240 IU inhaled, 5 IU intravenous or 10 IU intramuscular in women in third stage of labour (TSL) were analysed using enzyme-linked immunosorbent assay (ELISA) and LC-MS/MS to understand how quantified peptide concentrations differ between these analytical methods (Study 3). RESULTS: Study 1: Oxytocin was stable in blood collected into EDTA tubes with or without protease inhibitors but not in LH tubes. Study 2: Blood collected into all EDTA-containing collection tubes led to near-complete inhibition of oxytocinase (≤100 min). In plasma, a 35% reduction in oxytocinase activity was observed in LH tubes with EDTA added. In plasma from late-stage pregnancy compared to nonpregnant participants, the oxytocinase activity was approximately 11-fold higher. Study 3: Plasma oxytocin concentrations from nonpregnant or women in TSL following exogenous oxytocin administration were ≤33 times higher when analysed using ELISA vs. LC-MS/MS methods. CONCLUSIONS: Collection of blood from late-stage pregnant women into tubes containing EDTA inhibits oxytocinase effectively stabilizing oxytocin, suggesting low concentrations of oxytocin after dose administration reflect rapid in vivo metabolism.

Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study.

In response to the increasing application of machine learning (ML) across many facets of pharmaceutical development, this pilot study investigated if ML, using artificial neural networks (ANNs), could predict the apparent degree of supersaturation (aDS) from two supersaturated LBFs (sLBFs). Accuracy was compared to partial least squares (PLS) regression models. Equilibrium solubility in Capmul MCM and Maisine CC was obtained for 21 poorly water-soluble drugs at ambient temperature and 60 °C to calculate the aDS ratio. These aDS ratios and drug descriptors were used to train the ML models. When compared, the ANNs outperformed PLS for both sLBFCapmulMC (r2 0.90 vs. 0.56) and sLBFMaisineLC (r2 0.83 vs. 0.62), displaying smaller root mean square errors (RMSEs) and residuals upon training and testing. Across all the models, the descriptors involving reactivity and electron density were most important for prediction. This pilot study showed that ML can be employed to predict the propensity for supersaturation in LBFs, but even larger datasets need to be evaluated to draw final conclusions.

BRAF p.Val600Glu (V600E) mutation detection in thyroid fine needle aspiration cell block samples: a feasibility study.

Assessing BRAF mutation status in thyroid fine needle aspiration (FNA) cytology samples by both immunohistochemistry (IHC) and molecular methods has been documented in recent literature. We aim to highlight issues relating to quality and quantity of cellular material and DNA extracted from cell block samples.BRAF mutation status was assessed by both molecular and IHC methods in cell block material from thyroid FNA samples over a range of diagnostic categories, and correlated with available follow-up resection specimens.Of 39 samples there were 14 cases with 'inconclusive' cytology (Bethesda diagnostic categories 3, 4 or 5) and 25 cases with malignant cytology. Follow-up information was available in 38 of 39 cases and resection material for comparison in 18 of 39 case. Detection of BRAF mutation in cell block samples by combined molecular and IHC methods showed 100% specificity and 71.4% sensitivity compared to subsequent histologically confirmed BRAF mutated papillary thyroid carcinoma. IHC detected BRAF mutation in two (8.2%) cases which were negative by molecular methods and confirmed mutation positive by IHC and molecular methods on subsequent histology. Low extracted DNA concentration did not appear to preclude detection of BRAF mutation, although cell blocks with lower tumour cell content were over-represented in cases that were wild-type on FNA material and BRAF mutant on subsequent histology.BRAF mutation detection in cell block material is feasible and highly specific for papillary thyroid carcinoma. Best results are obtained by a combination of molecular and IHC methods.