Pharmaceutical companies are increasingly leveraging the power of biocatalysis to enable the development of concise and robust commercial manufacturing processes for their therapeutics. Yet, implementation of biocatalysis via protein engineering often requires years to generate a production-ready enzyme. To enable rapid clinical supply, we engineered an enoate-reductase (ERED) to catalyze a key asymmetric reduction in a concise 4-step chemoenzymatic route toward the orally bioavailable STING agonist MK-2118. Within 2 months, we evolved a Z-selective ERED with high activity that delivered the desired product in high yield and exquisite enantioselectivity. The 4-step chemoenzymatic process was used to successfully deliver >10 kg of MK-2118 from readily available starting materials at 40% overall yield and >99% ee to accelerate clinical trials.

An industrial process for the conversion of the fungal cyclopeptide cyclosporine A (2) into voclosporin (1a,b) was optimized and scaled up to the kilogram scale. The combined optimization of reaction conditions and final purification made it possible to obtain voclosporin in excellent purity (HPLC % area >99.5%) and reasonable overall yield (40–45%), with an E/Z diastereomeric ratio complying with the pharmaceutical requirements (90–95% E-isomer dominance). A study of the stability of voclosporin (2) under basic conditions revised the nature of the process involved in its degradation, identifying two major reactions responsible for the formation of impurities in the final products, namely, dehydration at C-22/C-23 (residue 1) and L-to D-isomerization of the N-methyl leucines residues 4 and 9. These impurities were isolated, structurally characterized, and quantified in the final product.

A seven-step enantioselective route for the synthesis of benzyl (R)-10-oxo-2-oxa-7-azaspiro[4.5]decane-7-carboxylate (1) with >99% ee is reported on a multi-hundred-gram scale in 42% overall yield. The synthesis features an asymmetric allylic alkylation to establish the quaternary stereocenter on a piperidine ring, followed by an oxidation/reduction and Mitsunobu conditions sequence to construct the tetrahydrofuran ring within the chiral oxa-azaspirocycle.

The rigorous assessment of impurities in pharmaceutical drugs is vital to ensuring their safety and efficacy. Recent focus has centered on nitrosamines due to their known mutagenic and carcinogenic potential, initially highlighted by the presence of N-nitrosodimethylamine (NDMA) in a drug in 2018. Subsequently, similar nitrosamines have been found in other pharmaceutical drugs, raising concerns about their potential mutagenic impurities. This underscores the importance of evaluating both actual and potential impurities for bacterial mutagenicity. In alignment with ICH M7 (R2) guidelines, we have undertaken an in-silico study for the identification and control of potential mutagenic impurities in lansoprazole as a case study, which was one of the ranitidine-substituted drugs as recommended by the FDA. Through the synthetic scheme assessments, we identified the actual and potential impurities of lansoprazole. In silico bacterial mutagenicity predictions were carried out using a combination of expert rule-based and statistical rule-based methodologies according to the guidelines outlined in ICH M7 (R2). A total of 35 lansoprazole impurities were evaluated using Derek Nexus, Sarah Nexus, and Case Ultra software, wherein 9 impurities resulted in positive predictions and 26 impurities resulted in negative predictions. The impurities with positive predictions were subsequently subjected to further scrutiny using Mirabilis software to develop a control strategy. Based on the nature of the impurity and its stage of formation/introduction to the process, a hypothetical scenario using an initial concentration of each positively predicted impurity (intermediate/by-product/reagent) at its point of introduction was considered to understand its purge and fate at the final stage. This study provided critical insights into the possible presence of mutagenic impurities and facilitated the development of an appropriate control strategy to ensure the safety of the final drug substance.

Thermal runaway from chemical reactions is the result of competition between heat generation and heat dissipation in industrial chemical vessels. However, the worldwide used TD24 (the temperature at which time to maximum rate under adiabatic conditions equals 24 h) focuses solely on heat generation for abnormal conditions of cooling failure, which account for only an insignificant portion of thermal runaway incidents. In contrast, the temperature of no return (TNR), defined by the balance of heat generation and heat dissipation, provides a more relevant threshold for process upsets in industrial chemical vessels. As the operating temperature approaches the TNR, minor thermal disturbances can precipitate thermal runaway. To estimate this threshold, calorimetry is coupled with reactor-scale energy-balance modeling. Kinetic parameters are determined using accelerating rate calorimetry (ARC), with corroborating measurements from differential scanning calorimetry (DSC), vent sizing package (VSP) calorimetry, and micro reaction calorimetry (μRC). In this work, the downside of TD24 for runaway hazard assessment is first analyzed. Guided by the Semenov diagram, a heat-balance model for stirred-tank reactors (STRs) is established, and based on ARC tests on a 20 wt % solution of di-tert-butyl peroxide (DTBP) in toluene under adiabatic conditions, a heat-generation-rate expression and a coolant heat-dissipation-rate expression are derived. A computational procedure is then proposed to determine TNR. Relative to adiabatic TD24, TNR supports optimization of reactor-cooling design, enhancing alarms and interventions under abnormal conditions, and quantitative assessment of reaction hazards.

N-(2,6-Dimethylphenyl)-6-hydroxypicolinamide (1) has been identified as a highly effective ligand for copper-catalyzed C–N and C–O bond-forming reactions, with prevailing application in the industrial synthesis of pharmaceutical intermediates and fine chemicals. We report an innovative synthetic route to 1 via three sequential steps: (i) formation of an acyl chloride intermediate, (ii) amidation by the Schotten–Baumann reaction, and (iii) copper-catalyzed hydroxylation, executed in a telescoped one-pot process. The optimal reaction conditions for the hydroxylation of step 3 were obtained with the aid of design of experiments (DoE) study. Three generations of the process were developed and are reported. This telescoped process was used for multiple kilo-batch productions, and a 67% reduction in material costs compared to the previous reports was realized. The improvement mainly stems from superior atom economy, minimized solvent consumption, and the avoidance of intermediary purification operations.

The presence of effervescent gas bubbles in liquid–gas biphasic streams adversely affects liquid chromatography-based PAT (process analytical technologies) in all critical steps of the analysis, from injection to measurement, presenting significant obstacles for CPV (continuous process validation). This article describes a unique, multiconfiguration rotary valve capable of adopting configurations essential for the removal of the gas bubbles from the biphasic stream using an automated trap-purge technique. The multiple, function-specific configurations of the valve prevent the gas bubbles entering the chromatography stream and minimizes system dead-volume in the analytical workflow enabling precise execution of the trap-purge method for inline analysis. The currently disclosed PAT reliably reported purity of the desired product in the output stream of a continuous transfer-hydrogenation process. This work paves the way for high-frequency continuous process validation of multiphase flow reactions in line with process validation guidance of regulatory agencies that oversee fine-chemical manufacturing.

In this study, we performed a hazard assessment of actual and potential impurities involved in the synthesis of a newly developed LpxC inhibitor T-1228 in accordance with the ICH M7(R2) guideline. We identified eight mutagenic/carcinogenic impurities in need of control: benzyl chloride, 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride, O-(tetrahydro-2H-pyran-2-yl)hydroxylamine, 2-(4-iodophenyl)-2-oxoethyl acetate, (S)-1-(4-ethynylphenyl)ethane-1,2-diol, isopropyl methanesulfonate, hydroxylamine, and (S)-4-((4-(1,2-dihydroxyethyl)phenyl)ethynyl)-N-hydroxybenzamide. We then conducted risk assessments of the first six impurities using purge factors, and determined that Option 4 control of ICH M7(R2) guideline was the optimal approach for controlling these impurities. Furthermore, as (S)-1-(4-ethynylphenyl)ethane-1,2-diol and isopropyl methanesulfonate are impurities introduced during the later stages of active pharmaceutical ingredient (API) synthesis, we obtained additional experimental data (verification of reactivity, solubility, spike experiments, and analytical experiments) to verify the validity of the purge-based risk assessment. Regarding the final two impurities, hydroxylamine may be produced as a byproduct in the final intermediate and final processes; therefore, we determined that Option 1 control of ICH M7(R2) guideline was the optimal control strategy and set appropriate API criteria. Although (S)-4-((4-(1,2-dihydroxyethyl)phenyl)ethynyl)-N-hydroxybenzamide is a degradation product of API, in vivo mutagenicity assay conducted according to the ICH M7(R2) guideline showed negative results. Therefore, we determined that (S)-4-((4-(1,2-dihydroxyethyl)phenyl)ethynyl)-N-hydroxybenzamide can be managed as a nonmutagenic impurity according to the ICH Q3A(R2) guideline. The risk assessment of mutagenic impurities related to the T-1228 API manufacturing method, performed in accordance with ICH M7(R2) guideline, can guarantee the quality of this manufacturing method for initial clinical trials.

The novel LpxC inhibitor T-1228 is a candidate drug molecule for multidrug-resistant Gram-negative bacterial infection. This report describes the synthesis of the diol unit of T-1228, (S)-1-(4-ethynylphenyl)ethane-1,2-diol, from commercially available 2-bromo-1-(4-iodophenyl)ethan-1-one, achieved in 5 steps and 4 isolations. The stereogenic center of the diol unit was created by means of a highly scalable form of the Corey–Bakshi–Shibata reduction; process parameters were optimized using the Design of Experiments methodology. Using this newly developed process chemistry route, we successfully synthesized 12.5 kg of (S)-1-(4-ethynylphenyl)ethane-1,2-diol in 60% total yield and optical purity >99% ee.