Machines vs Malaria: A Flow-Based Preparation of the Drug Candidate OZ439.

Shing-Hing Lau,Steven V. Ley,Claudio Battilocchio,Alicia Galván,Rohan R. Merchant,Malcolm B. Berry,José A. Souto

doi:10.1021/acs.orglett.5b01307

Abstract

An efficient preparation of the antimalarial drug candidate OZ439, which was obtained by integrating a machine-assisted approach with batch processes, is reported. This approach allows a rapid and cost-effective production of the key intermediates that were readily elaborated into the target molecule.

Highlights

New antimalarial agents must be available at low cost in order to match the economies mostly afflicted by this disease
Recent investigations[7] have led to the identification of new types of antimalarial agents containing a trioxolane unit as main structural feature, with compound OZ277 (2) as one of the most promising hits arising from this research.2a,8 Further studies directed to improve the pharmacokinetic and pharmacodynamic properties resulted in the identification of the more potent analogue OZ439 (3).2a,9 The important property of 3 is that it provides a single oral dose treatment for the total eradication of the parasite in humans; it is currently undergoing phase IIa trials after having successfully completed phase I clinical trials.2a,10
The reported synthesis of this compound relies on the use of commercially available but relatively expensive 4-(4hydroxyphenyl)cyclohexan-1-one (4) and makes use of large quantities of pentane, which is an undesirable solvent for the preparation of the key trioxolane motif.[11]

Summary

Organic Letters

To achieve of our initial aim, we envisaged a new preparation of the intermediate trioxolane 67,9 which would take advantage of the use of enabling methods. This was avoided by decreasing the concentration of oxime 5 (from 0.4 to 0.2 M) and ketone 11 (from 0.2 to 0.1 M) while increasing both the liquid and gaseous flow rates (1 mL min−1 and 1 L min−1, respectively) in order to keep an identical throughput for the reaction system Under these diluted conditions we were able to continuously run the reaction without any clogging issue for over 3 h, isolating 78% of the model product with a calculated throughput of 1.6 g·h−1 (equating to 38.4 g·d−1).

■ ACKNOWLEDGMENTS

Findings

■ REFERENCES