Detection of Falsified Antimalarial Sulfadoxine-Pyrimethamine and Dihydroartemisinin-Piperaquine Drugs Using a Low-Cost Handheld Near-Infrared Spectrometer

Abstract

Falsified drugs are of serious concern to public health worldwide, particularly for developing countries where quality control of drugs is inefficient. In law enforcement against such fake medicines, there is a need to develop reliable, fast, and inexpensive screening methods. In this work, the ability of an innovative low-cost handheld near-infrared spectrometer to identify falsifications among two antimalarial fixed dose combination tablets, dihydroartemisinin/piperaquine and sulfadoxine/pyrimethamine, has been investigated. Analyzed samples were collected in Burkina Faso mainly in rural transborder areas that could be infiltrated by illicit drugs. A principal component analysis was applied on the acquired near-infrared spectra to identify trends, similarities, and differences between collected samples. This allowed to detect some samples of dihydroartemisinin/piperaquine and sulfadoxine/pyrimethamine which seemed to be falsified. These suspicious samples were semiquantitatively analyzed by thin-layer chromatography using Minalab® kits. Obtained results allowed to confirm the falsifications since the suspected samples did not contain any of the expected active pharmaceutical ingredients. The capacity of the low-cost near-infrared device to identify specifically a brand name of dihydroartemisinin/piperaquine or sulfadoxine/pyrimethamine has been also studied using soft independent modelling of class analogy (SIMCA) in the classical and data driven versions. The built models allowed a clear brand identification with 100% of both sensitivity and specificity in the studied cases. All these results demonstrate the potential of these low-cost near-infrared spectrometers to be used as first line screening tools, particularly in resource limited laboratories, for the detection of falsified antimalarial drugs.

Usefulness of medicine screening tools in the frame of pharmaceutical post-marketing surveillance

“…These results are similar to those of Ciza et al who found that the application of OCC models to NIR spectra allowed a good discrimination between samples containing different APIs [43]. Moreover, these findings are in accordance with those of Yabre et al who achieved 100% specificity with correct discrimination of falsified samples [5,44].…”