Malaria is a life-threatening disease that caused an estimated 241 million cases and 627,000 related deaths in 2020 among 85 countries, Africa the most affected continent[1]. Consequently, the World Health Organization (WHO) has prioritized the reduction of malaria cases to lower morbidity and mortality associated with this disease by 90 % by 2030[2]. A such goal requires an approach aimed to greatly reduce its prevalence and transmission in endemic populations primarily by using drugs for both prophylaxis and treatment along with vector control and early diagnosis procedures. Current antimalarial drugs can be categorized into multiple structural families such as aryl-amino alcohols, artemisinins, antifolates, antibiotics, 4-aminoquinolines (4AQ), and 8-aminoquinolines (8AQ). In the last few decades, six drugs have been approved by the FDA and are currently available by prescription for malaria prophylaxis. Several are in widespread use today. They are, in terms of the year of approval: chloroquine, primaquine, mefloquine, doxycycline, the combination of atovaquone and proguanil, and tafenoquine[3]. Among all antimalarial drugs, primaquine, and tafenoquine are the only 8AQ drugs currently available to treat relapsing forms of P. vivax malaria, and chloroquine, which is the 4AQ is the most common drug to treat uncomplicated P. vivax malaria[4].

Specifically, Tafenoquine is an 8AQ derivative that was approved by the FDA in 2018 as a single-dose treatment and was termed the “radical cure for P. vivax malaria”. It was also approved for chemoprevention but several doses are needed for this latter indication. This 8AQ antimalarial drug was first synthesized and named WR238605 by scientists at the Walter Reed Army Institute of Research in 1978 as a part of the US antimalarial drug program started in 1963[5], [6], [7]. It has been proven to be the most potent and less toxic 8AQ analog against P. vivax dormant liver stages. Although this compound shows inhibitory activity for both blood and liver stages of P. falciparum and P. vivax, its main interest relies on relapse prevention for the latter[8], [9]. Indeed, the challenges in controlling and eliminating P. vivax malaria are normally related to its ability to relapse from long-lasting dormant liver stages (hypnozoites) and its high transmission capability caused by the continuous production of gametocytes along with the shorter growth cycle in the vector host compared to other Plasmodium species[10].

Nevertheless, as a primaquine analog, it has inherited one of the main drawbacks of its predecessor: the potential to cause severe hemolytic anemia (breakdown of the red blood cells) in people with glucose-6-phosphate dehydrogenase (G6PD) deficiency[11], [12], [13] which may result in arrhythmias, cardiopathies or even heart failure. Consequently, as no robust test has been developed to quantitatively measure the G6PD levels in endemic areas, tafenoquine administration has not been widely adopted yet. Suitably, a safer alternative to this original hit would be necessary to lower this adverse effect and optimally find a more active drug candidate.

Since the 8AQ family has been the subject of extensive research efforts worldwide over the years, it is reasonable to assume that Tafenoquine has a undergone thorough investigation. Its discovery was presumably the result of meticulous and exhaustive exploration of the 8AQ structural framework. However, this study aims to critically assess the chemical space really studied and dive into the unexplored region. In pursuit of this objective, we examined the disparity in coverage between the compounds chosen for analysis based on a rational selection process and the comprehensive set of tafenoquine derivatives that have publicly been reported to date.

When looking into the patents containing Tafenoquine’s structure, such as the already expired original patent from 1986 (US4617394[14]) or a latter one from 2002 (US6376511[15]), the common Markush formula can easily be found in their concluding claims section. The combination of both patents would lead to a combinatorial library described by the general Markush structure stated in Fig. 1.

It is worth noting that while patents typically employ Markush structures in their claims to describe broad chemical spaces, the process of optimizing a novel main active ingredient often relies on a relatively simplistic Free-Wilson approach. Our research group conducted an in-depth examination of seven patents, including the one related to Tafenoquine[16]. Through this investigation, we have realized that this extended procedure tends to concentrate primarily on the chemical space proximate to a specific hit compound. Consequently, numerous molecular modifications are frequently disregarded and potential reservoirs of highly biologically active compounds typically remain unexplored. This limited exploration may inadvertently restrict the identification of novel and potent derivatives within the 8AQ class, highlighting the need for a more comprehensive and systematic approach to compound discovery and development.

The selection of this case study regarding Tafenoquine has been deliberate, primarily due to its focus on a rare disease (malaria). Such diseases have typically yielded a wealth of publicly available information, thus providing researchers with ample data to conduct thorough investigations. This accessibility to information has ensured transparency and fostered collaboration within the scientific community (unifying forces with not-for-profit drug research partnerships such as Medicines for Malaria Venture[17]), allowing for a comprehensive exploration of the subject matter without the constraints typically associated with more common diseases or commercially-driven research endeavors. As a result, this case study has offered a unique opportunity to delve into the intricacies of drug discovery and development in a manner that has been both informative and conducive to advancing scientific knowledge.

In this study, we contrast the application of rational analysis to conventional drug discovery processes, to improve the representativeness of the selected compounds. This approach has been applied to Tafenoquine’s chemical space to assess what would have happened if it had been implemented from the beginning. The most representative seven compounds were selected, synthesized and biologically tested. Results evince that the approach based on the clustering analysis and rational selection of candidates from large combinatorial databases effectively allows for a significant improvement in the early stages of drug discovery.