New insights with miltefosine unresponsiveness in Brazilian Leishmania infantum isolates

Visceral leishmaniasis (VL) is a neglected tropical disease caused by Leishmania donovani (Old world) and Leishmania infantum (New and Old world). It affects mostly the poorest of our society with the greatest burden in the Indian sub-continent, Eastern Africa and Brazil. The tens of thousands of people infected each year need to be treated but the choices are limited with 5 approved drugs (amphotericin B, antimonials, miltefosine, paromomycin, and pentamidine) each with limitations. Initially developed as an anti-tumor drug, Simon Croft first demonstrated the efficacy of miltefosine in experimental VL [1]. Miltefosine had the distinct advantage of being active orally while the 4 remaining drugs require parenteral administration. Further development led in 2002 to the registration of miltefosine for treating VL in India. Initial efficacy reached 95% against L. donovani, although for a number of reasons the efficacy is decreasing [2]. Nonetheless, the advantage of an oral route warranted clinical trials to test for miltefosine efficacy in other settings. Surprisingly, miltefosine was only 60% effective in a phase 2 dose-ranging trial in Brazil. In this issue of EBioMedicine, Carnielli and colleagues [3] investigated the molecular basis explaining the lack of miltefosine activity against L. infantum isolated from patients enrolled in this trial.

The authors had access to pretreatment isolates coming from VL patients either cured (14 strains) or relapsing (12 strains) after miltefosine treatment. They sequenced the genome of those 26 strains and performed Genome-Wide Association Studies. This led to identification of a region on chromosome 31 harboring four genes (coding for two putative 3′-nucleotidase/nucleases, a helicase-like protein and a 3,2-trans-enoyl-CoA isomerase) which was termed the Miltefosine Sensitivity Locus (MSL). This locus was present in most isolates from cured patients but deleted from the isolates derived from patients who relapsed. Statistical analyses supported the association between the presence of the MSL and cure with miltefosine treatment, and a 9.4-fold higher risk of relapse when patients were infected with a strain deleted for the MSL. The mechanism of MSL loss was studied and is likely mediated by homologous recombination between repetitive elements bordering the MSL locus. Finally, Carnielli et al. [3] investigated the prevalence of the MLS in additional Brazilian L. infantum isolates. They found that strains from the Centre-East of Brazil had in majority a deletion of the MSL but a majority of those from the North East retained that locus. Of note, they report [3] that the MSL is universally present in L. donovani genomes available in database (a species responding to miltefosine).

This manuscript clearly brings novel insights on miltefosine response in patients infected with Leishmania. Pending confirmation of the reported association with a greater number of isolates, this work may even have clinical implications where a simple PCR assay could indicate whether miltefosine treatment has a high likelihood of being effective. Our understanding of miltefosine resistance mechanisms in Leishmania is mostly derived from in vitro work where the main resistance mechanism is a loss of function mutation in the P-type ATPase miltefosine transporter (MT) [4]. Other mutations have been described but the loss of the MSL described by Carnielli et al. [3] is genuinely novel. This may highlight differences between in vitro selected Leishmania strains and clinical isolates although mutations in MT were described in clinical isolates of Old world Leishmania [5,6].

The discovery of the MSL is important but also brings several questions. How does its loss contribute to treatment failure? What is the selective advantage and driving forces for its deletion? There is no obvious link between the four genes coded by the MSL locus and known miltefosine mode of action. It remains to be determined whether the loss of the MSL has any impact on parasite susceptibility. In vitro susceptibility determinations (MIC, EC₅₀) are usually a reasonable measure of treatment outcome for infectious diseases. However, this is neither standardized nor validated for Leishmania but miltefosine susceptibility in MLS⁻ and MSL⁺ parasites need to be tested. It is also possible that the presence of the MSL, through the action of one of its 4 gene products, may lead to a more robust host response. The tools for manipulating the Leishmania genome are available for generating isogenic L. infantum strains −/+ MSL and test for the role of this locus in either parasite susceptibility or host response in animal models. If a functional link were established, this would justify clinical trials in the North East of Brazil where the majority of parasites are MSL⁺. If no functional link can be drawn, MSL could still serve as a biomarker for miltefosine response but this would require extensive validation.

In conclusion, Carnielli et al. [3] have found a new locus well correlated to miltefosine response in Brazilian VL patients and additional work will indicate whether this biomarker can be use for predicting patient response to miltefosine.

Disclosure

The authors do not have any conflict of interests to disclose.