Despite a recent publication describing boronic acids as a "novel class of bacterial mutagen," process and analytical chemists may not be aware of the potential worker exposure hazards, or of the need to assess boron compounds as potential genotoxic impurities (GTIs) per ICH M7. This publication provides new Ames data for 44 commercially available boronic acids, boronic acid derivatives, and boron containing reagents. Trends in the Ames data are discussed from a structure-activity perspective. Common reagents such as bis(pinacolato)diboron and bisboronic acid were shown to be mutagenic in the Ames assay. Currently available in silico computational models were found to provide little value in predicting the outcome of the Ames assay for boronic acids and derivatives. We propose oxygen mediated oxidation of boron compounds to generate organic radicals as a potential mechanism for mutagenicity. It is hoped that this paper will result in increased awareness of this class of GTIs, and prompt publication of additional Ames data for boronic acids and their derivatives that will lead to improved (Q)SAR models, and an understanding of the mechanism of mutagenicity.KEY WORDS GTI, mutagen, Ames, boron, boronic acid, (Q)SAR

The design, development, and implementation of a pilot-scale continuous hydrogenolysis in a catalytic packed bed to generate a starting material is described. Control of a critical defluorination impurity under the reaction conditions has been achieved by reducing residence time inside the catalyst bed to 15−30 min. A reactor volume throughput of 206 kg/h·m 3 was attained in a 3 L reactor (1.5 kg of 5% Pd/C catalyst) over a 9 h demonstration period, superior to the 1.3 kg/h·m 3 volume throughput obtained in batch. The reaction was successfully scaled up from 9 g/h to 550 g/h in packed beds ranging from 20 to 1500 g catalyst, demonstrating heat/mass transfer sufficiency at all examined scales. The process was monitored by online HPLC, providing real-time reaction information, using an internally developed automation cart coupled to a standard HPLC. Significant technical and business drivers for running the process in continuous flow mode were proposed and examined during development, demonstrating superior control of critical impurities and catalyst utilization with minimized risk to product and increased safety due to reduced handling of hydrogen and of palladium catalyst relative to equivalent substrate throughputs in a typical batch process.

An efficient synthesis of LY2886721 (1) in five steps and 46% overall yield from the chiral nitrone cycloadduct 2 is presented. Minimizing formation of a des-fluoro impurity during hydrogenolysis to cleave the isoxazolidine ring and remove the benzyl chiral auxiliary was a key challenge. Installation of the aminothiazine moiety required careful stoichiometry control of the reagents BzNCS and CDI, including in situ conversion monitoring, to minimize byproduct formation. A remarkably regioselective peptide coupling afforded 1 without competing acylation at the aminothiazine nitrogen or bis-acylation. Consideration of the combined chemistry and crystallization process identified an optimal solvent system for the peptide coupling and a reactive crystallization that afforded 1 in high purity and with physical property control. A slurry milling operation near the end of the crystallization, followed by "pH cycles" to digest fines formed during milling, significantly reduced the crystal aspect ratio and provided desirable API bulk density and powder flow properties.

A pilot-plant scale desymmetrization of the cyclic meso-epoxide 4b, using a chiral lithium amide prepared from symmetrical diamine 17, was designed and implemented to provide allylic alcohol 3b in high yield and greater than 99% ee. This chiral alcohol was converted to ketone 2b, a key intermediate in a new asymmetric synthesis of LY459477. Chiral diamine 17 was prepared from a readily available chiral precursor, (R)-rmethylbenzylamine, and could be recovered from the reaction mixture and reused. Studies performed to probe the mechanism of the rearrangement reaction of epoxide 4b showed that diamine 17 provided an optimal combination of selectivity and scaleability for this process.

The synthesis of a MET kinase inhibitor in an overall yield of 22% was achieved over eight steps starting with 3-hydroxybenzaldehyde, an improvement from the initial 12-step process with a 5.4% yield. Highlights of the process chemistry design and development are a Cu-catalyzed cyclization to form an important N1-methylindazole ring, a selective nitro reduction in the presence of an aryl bromide, a late-stage Suzuki cross-coupling, and a base-promoted Boc deprotection to form the desired drug candidate.