Bristol-Myers Squibb (Sweden)

Automated microbial detection systems are becoming increasingly essential in pharmaceutical manufacturing to enhance efficiency, accuracy, and compliance with Good Manufacturing Practice (GMP) standards. The APAS Independence System, developed by Clever Culture Systems, was evaluated at Bristol Myers Squibb (BMS) to assess its suitability for environmental monitoring applications. The APAS system is capable of automated plate counting of both 55 mm and 90 mm plates, following an offline incubation at routine EM parameters. The study compared APAS automated colony counts to traditional manual counts for compendial strains and environmental monitoring samples across multiple weeks of EM sampling. Performance metrics for both plate sizes were analyzed, such as accuracy, precision, false negative and false positive rates, and robustness. The APAS system demonstrated high accuracy (≥70%) and reliability for most tested organisms. Therefore, the system is being implemented for GMP use, with a validation approach that combines vendor validation of the Artificial Intelligence with site specific instrument qualification. This evaluation provides insight into the capabilities and limitations of automated colony detection, informing future digitalization strategies in microbiological quality control.

Predictive biomarkers for patients with early rheumatoid arthritis (RA) are needed. This exploratory post hoc analysis investigated inflammatory biomarkers associated with baseline disease activity and biomarkers predictive of treatment response in patients who were seropositive with early RA from a phase 3 study. The AVERT-2 study (NCT02504268) included patients who were disease-modifying antirheumatic drug naive and anti-citrullinated protein antibody positive randomized to abatacept and methotrexate (MTX) or placebo and MTX for 56 weeks. Correlations were assessed among biomarkers and disease activity, pharmacodynamic (PD) changes on disease-associated biomarkers in response to treatment, and baseline biomarkers to predict treatment response at week 52. In the analysis, 446 patients received abatacept and MTX and 300 received placebo and MTX. Of 103 biomarkers, 47 demonstrated a significant reduction in PD changes with abatacept and MTX versus placebo and MTX, with 18 out of 47 biomarkers showing correlations with baseline disease activity. High baseline interleukin-10 (IL-10) levels were associated with greater probability of achieving efficacy measures by week 52 in patients receiving abatacept and MTX versus placebo and MTX, consistent with a significant reduction in disease activity with abatacept and MTX (P < 0.03 to P < 0.0007). Higher baseline IL-10 levels were associated with lower bone erosive development with abatacept and MTX compared with placebo and MTX. Treatment with abatacept and MTX resulted in significantly greater reduction of biomarkers relevant to RA, and high baseline IL-10 levels predicted better treatment response with abatacept and MTX versus placebo and MTX across clinical outcomes. Findings suggest a well-suited mechanism of action for abatacept and MTX in patients with early RA and high baseline IL-10 levels.

The discovery of KRAS G12C inactive state inhibitors represented a significant advancement in the field of precision oncology. While inactive state inhibition shows promise in patients, Switch II (SWII)-binding inhibitors targeting both inactive and active states remain an underdeveloped therapeutic modality. Here, we describe the discovery of such KRAS G12C dual inhibitors that bind the SWII allosteric site using a chemically differentiated warhead to covalently modify boththe KRAS G12C inactive and active states. Co-crystal structures reveal that these inhibitors perturb a key water-mediated hydrogen bonding network and trigger allosteric remodeling of the GTP-bound protein surface and SWI that prevents binding to downstream effectors. Consistent with simultaneous targeting of the active and inactive states, dual inhibitors provide rapid covalent target engagement and suppression of MAPK signaling. However, they demonstrate similar efficacy in cellular and in vivo modelswhen compared to inactive state-selective ones despite faster target inactivation. Furthermore, both inhibitor classes show similar cellular efficacy in the presence of growth factors that drive KRAS, wt NRAS, and wt HRAS to the active state. These data provide the first detailed account of targeting both the active and inactive states of KRAS G12C and highlight the absence of a mechanistic advantage in contexts dependent on prolonged target inhibition.

S ummary There is an overwhelmingly large body of literature and numerous algorithms already available on “multimodal AI”, based on different modeling and fusion paradigms. As multimodal AI becomes fully integrated into healthcare applications, the ability to say “I’m not sure” or “I don’t know” when uncertain is a necessary capability for safe clinical deployment, requiring the augmentation of current multimodal AI techniques. Our goal in this work is not to add yet another method to the growing toolbox of multimodal AI, but rather to (i) clarify how existing supervised multimodal AI methods can be understood through the lens of uncertainty quantification, and (ii) develop a unified framework that provides valid and interpretable uncertainty estimates for multimodal AI practitioners. We introduce Coracle, a conformalized framework for multimodal AI that seamlessly adapts to early, late, and intermediate fusion. Coracle provides theoretical marginal confidence guarantees and achieves valid finite-sample coverage without relying on distributional assumptions. Through extensive simulations and analyses of diverse biomedical multiview datasets, we demonstrate that Coracle consistently outperforms existing non-conformalized methods in both uncertainty estimation and calibration. By equipping published multiview models with valid, interpretable confidence measures, Coracle advances the development of trustworthy, uncertainty-aware statistical tools for biomedical decision support. Coracle is publicly available as an open-source R package at https://github.com/himelmallick/Coracle .

Catalytic reactions involving multiple cycles are common in many key organic transformations. We report a detailed mechanistic analysis of one such network of "nested" cycles, the Ni-catalyzed cross-coupling of alkyl sulfonyl hydrazides with (hetero)aryl halides, employing a battery of techniques, including Temperature Scanning Reaction (TSR) calorimetry coupled with Reaction Progress Kinetic Analysis (RPKA), quantitative NMR time-course analysis, kinetic modeling, linear free energy analysis, and density functional theory (DFT) quantum chemical calculations. TSR/RPKA studies yield the simultaneous determination of the thermodynamic enthalpy of the reaction, the kinetic rate law, and the activation parameters (ΔH‡, ΔS‡, ΔG‡), for the rate-determining step. Expanding the analysis to the coupling of an array of electron-poor and electron-rich sulfonyl hydrazides (each readily accessible on the decagram scale) with a range of (hetero)aryl halide coupling partners introduces the concept of "kinetics matching," and demonstrates that optimal yields may be obtained when the reactivities of the separate cycles are tuned to work in unison. These requirements for reaction networks involving nested catalytic cycles are showcased by several real-world examples. Our work highlights the predictive power of this kinetics-focused mechanistic approach, which may be extended to other nested catalytic networks to select appropriate catalyst and substrate combinations for optimal outcomes.

<p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d618428e81"> <b>Objectives:</b> To recommend a nivolumab subcutaneous (nivo SC) pediatric dosing regimen without clinical data using a model informed drug delivery (MIDD) approach. The objective of this approach was to predict and compare nivo exposures across nivo SC vs. the approved nivo intravenous (IV) dosing regimens in pediatric patients (12 to &lt; 18 years) with unresectable or metastatic melanoma, adjuvant treatment of melanoma, and microsatellite instability high or mismatch repair deficient metastatic colorectal cancer. The recommended nivo SC pediatric dosing regimen was needed to achieve PK non-inferiority (geometric mean ratio [GMR] &gt;1) compared to the approved nivo IV regimen exposures in pediatric patients. <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d618428e86"> <b>Methods:</b> Previously established nivo population PK (popPK) models for the different tumor types were utilized. Stochastic simulations were performed using a pediatric population created from the NHANES database (2020-2023). To ensure sufficient subjects were included across a wide body-weight range (30 kg to ≥110 kg ), 800 subjects aged 12 to &lt; 18 years were randomly sampled from the database. Height and creatinine clearance from the NHANES database were utilized in the Schwartz equation to calculate estimated glomerular filtration rate (eGFR) for each sampled subject. Other covariates utilized from NHANES included, body weight, lean body mass, sex, and race. Another population of adult subjects (N = 500) was created by random sampling from the included adult subjects in each popPK dataset. Sensitivity analyses were performed to evaluate the impact of variations in SC bioavailability (F) and absorption rates (ka) on PK exposures. <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d618428e91"> <b>Results:</b> The nivo SC dosing regimens recommended for pediatric patients included nivo 1200 mg SC Q4W and nivo 600 mg SC Q2W for patients weighing ≥ 40 kg, and nivo 600 mg SC Q4W and nivo 300 mg SC Q2W for patients weighing &lt; 40 kg. The nivo 1200 mg SC Q4W and nivo 600 mg SC Q2W dosing regimens are the approved adult dosing regimens across a range of tumor types. In the assessments conducted, the GMRs were above 1.0 for all evaluated indications indicating that noninferiority was met. All pediatric SC median exposures fell within the 5th and 95th percentile of adult SC exposures at the proposed dosing recommendations. The simulated median exposures (Cavgd28, Cavgss, Cminss, Cmaxss), including subjects across the entire body weight range for all SC dosing regimens were below the median adult exposures for nivo 10 mg/kg IV Q2W, a dosing regimen considered safe and tolerable. In addition, no meaningful differences in exposures were observed with the sensitivity analyses for ka and F. <p xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" dir="auto" id="d618428e96"> <b>Conclusions:</b> Based on the totality of the data, the exposures of the proposed dosing regimens for nivo SC in pediatric patients (12 to &lt; 18 years) are expected to result in a comparable benefit-risk profile as in adults. Utilizing MIDD approaches to support dose selection in pediatric patients is a powerful tool to support dose recommendations in patient populations with limited clinical data.