Search

Precision dosing, the tailoring of drug doses to optimize therapeutic benefits and minimize risks in each patient, is essential for drugs with a narrow therapeutic window and severe adverse effects. Adaptive dosing strategies extend the precision dosing concept to time‐varying treatments which require sequential dose adjustments based on evolving patient conditions. Reinforcement learning (RL) naturally fits this paradigm: it perfectly mimics the sequential decision‐making process where clinicians adapt dose administration based on patient response and evolution monitoring. This paper aims to investigate the potentiality of coupling RL with population PK/PD models to develop precision dosing algorithms, reviewing the most relevant works in the field. Case studies in which PK/PD models were integrated within RL algorithms as simulation engine to predict consequences of any dosing action have been considered and discussed. They mainly concern propofol‐induced anesthesia, anticoagulant therapy with warfarin and a variety of anticancer treatments differing for administered agents and/or monitored biomarkers. The resulted picture highlights a certain heterogeneity in terms of precision dosing approaches, applied methodologies, and degree of adherence to the clinical domain. In addition, a tutorial on how a precision dosing problem should be formulated in terms of the key elements composing the RL framework (i.e., system state, agent actions and reward function), and on how PK/PD models could enhance RL approaches is proposed for readers interested in delving in this field. Overall, the integration of PK/PD models into a RL‐framework holds great promise for precision dosing, but further investigations and advancements are still needed to address current limitations and extend the applicability of this methodology to drugs requiring adaptive dosing strategies. [ABSTRACT FROM AUTHOR]

Poly-γ-glutamic acid (γ-PGA) is an attractive biopolymer for medical, agri-food, and environmental applications. Although microbial synthesis by Bacilli fed on waste streams has been widely adopted, the obtainment of efficient sustainable production processes is still under investigation by bioprocess and metabolic engineering approaches. The abundant glycerol-rich waste generated in the biodiesel industry can be used as a carbon source for γ-PGA production. Here, we studied fermentation performance in different engineered Bacillus subtilis strains in glycerol-based media, considering a swrA+ degU32Hy mutant as the initial producer strain and glucose-based media for comparison. Modifications included engineering the biosynthetic pgs operon regulation (replacing its native promoter with Physpank), precursor accumulation (sucCD or odhAB deletion), and enhanced glutamate racemization (racE overexpression), predicted as crucial reactions by genome-scale model simulations. All interventions increased productivity in glucose-based media, with Physpank-pgs ∆sucCD showing the highest γ-PGA titer (52 g/L). Weaker effects were observed in glycerol-based media: ∆sucCD and Physpank-pgs led to slight improvements under low- and high-glutamate conditions, respectively, reaching ~22 g/L γ-PGA (26% increase). No performance decrease was detected by replacing pure glycerol with crude glycerol waste from a biodiesel plant, and by a 30-fold scale-up. These results may be relevant for improving industrial γ-PGA production efficiency and process sustainability using waste feedstock. The performance differences observed between glucose and glycerol media also motivate additional computational and experimental studies to design metabolically optimized strains. [ABSTRACT FROM AUTHOR]

The integration of pharmacokinetic‐pharmacodynamic (PK‐PD) modeling and simulations with artificial intelligence/machine learning algorithms is one of the most attractive areas of the pharmacometric research. These hybrid techniques are currently under investigation to perform several tasks, among which precision dosing. In this scenario, this paper presents and evaluates a new framework embedding PK‐PD models into a reinforcement learning (RL) algorithm, Q‐learning (QL), to personalize pharmacological treatment. Each patient is represented with a set of PK‐PD parameters and has a personal QL agent which optimizes the individual treatment. In the training phase, leveraging PK‐PD simulations, the QL agent assesses different actions, defined consistently with the clinical knowledge to consider only plausible dose‐adjustments, in order to find the optimal rules. The proposed framework was evaluated to optimize the erdafitinib treatment in patients with metastatic urothelial carcinoma. This drug was approved by the US Food and Drug Administration (FDA) with a dose‐adaptive protocol based on monitoring the levels of serum phosphate, which represent a biomarker of both treatment efficacy and toxicity. To evaluate the flexibility of the methodology, a heterogeneous virtual population of 141 patients was generated using an erdafitinib population PK (PopPK)‐PD literature model. For each patient, treatment response was simulated by using both QL‐optimized protocol and the clinical one. QL agents outperform the approved dose‐adaptive rules, increasing more than 10% the efficacy and the safety of treatment at each end point. Results confirm the great potentialities of the integration of PopPK‐PD models and RL algorithms to optimize precision dosing tasks. [ABSTRACT FROM AUTHOR]

MEN1611 is a novel orally bioavailable PI3K inhibitor currently in clinical development for patients with HER2‐positive (HER2+) PI3KCA mutated advanced/metastatic breast cancer (BC) in combination with trastuzumab (TZB). In this work, a translational model‐based approach to determine the minimum target exposure of MEN1611 in combination with TZB was applied. First, pharmacokinetic (PK) models for MEN1611 and TZB in mice were developed. Then, in vivo tumor growth inhibition (TGI) data from seven combination studies in mice xenograft models representative of the human HER2+ BC non‐responsive to TZB (alterations of the PI3K/AkT/mTOR pathway) were analyzed using a PK‐pharmacodynamic (PD) TGI model for co‐administration of MEN1611 and TZB. The established PK‐PD relationship was used to quantify the minimum effective MEN1611 concentration, as a function of TZB concentration, needed for tumor eradication in xenograft mice. Finally, a range of minimum effective exposures for MEN1611 were extrapolated to patients with BC, considering the typical steady‐state TZB plasma levels in patients with BC following three alternative regimens (i.v. 4 mg/kg loading dose +2 mg/kg q1w, i.v. 8 mg/kg loading dose +6 mg/kg q3w or s.c. 600 mg q3w). A threshold of about 2000 ng·h/ml for MEN1611 exposure associated with a high likelihood of effective antitumor activity in a large majority of patients was identified for the 3‐weekly and the weekly i.v. schedule for TZB. A slightly lower exposure (i.e., 25% lower) was found for the 3‐weekly s.c. schedule. This important outcome confirmed the adequacy of the therapeutic dose administered in the ongoing phase 1b B‐PRECISE‐01 study in patients with HER2+ PI3KCA mutated advanced/metastatic BC. [ABSTRACT FROM AUTHOR]

Coastal lagoons and brackish ponds are extremely dynamic and temporary ecosystems that follow natural changes throughout their geological history. The correct management of the lagoons ensures their integrity and proper functioning. For this reason, their ecological status should be surveyed for assessing the most appropriate strategies of use. In the present study, historical datasets collected in 2003–2004 are used to investigate the spatiotemporal variation in the species composition and community structure of the macrobenthos of the Tortolì Lagoon (Sardinia, Italy) and to assess their relationship with key environmental variables. Owing to the presence of a riverine runoff at a site and confined areas at some distance from the sea inlet, we hypothesize the marked spatiotemporal changes of the macrobenthic community consistent with the high environmental variability typical of coastal lagoons. The results show a surprisingly high benthic biodiversity for a medium-sized lagoon (250 ha), with 101 species unevenly distributed across the lagoon. The environmental variables did not explain the zonation of the macrobenthic community as that typically found along a lagoonal gradient, due to a marked marine influence. The sampling sites were in fact discriminated by the species distribution according to their ecological affinity; in particular, the most distinctive characteristics of the Tortolì Lagoon emerged from the strictly marine species that represented the most abundant group, consistently with the high marinization of the lagoon. Our results show that the Tortolì Lagoon constitutes a peculiar ecosystem within Mediterranean lagoons, departing from the classic confinement theory. [ABSTRACT FROM AUTHOR]