Dihydroartemisinin–piperaquine treatment failure in uncomplicated Plasmodium falciparum malaria case imported from Ethiopia

Abstract

Dihydroartemisinin–piperaquine (DHA–PPQ) is the artemisinin combination therapy that was recently introduced for the treatment of Plasmodium falciparum uncomplicated malaria, but emerging resistance in South-East Asia is threatening its use. This report describes a case of DHA–PPQ treatment failure in uncomplicated malaria occurring in an immigrant living in Italy, after a travel to Ethiopia. Thirty days after malaria recovery following DHA–PPQ therapy, the patient had malaria recrudescence. According to the genotyping analysis, the same P. falciparum was responsible for both episodes. Thus, it seems important to consider possible malaria recrudescence occurring after DHA–PPQ therapy in patients from African countries

Background

Dihydroartemisinin–piperaquine (DHA–PPQ) is one of the five artemisinin combination therapy (ACTs) regimens recommended by World Health Organization (WHO) for uncomplicated malaria [1]. For decades, the South-East Asian region (mainly the Thailand–Cambodian border) has been the epicentre for the emergence of Plasmodium falciparum multidrug resistance, including that related to DHA–PPQ which was recognized in Cambodia and rapidly spread throughout the country (25% in 2010, 46% in 2014) [2].

According to studies performed between 2002 and 2010 (mainly in African countries) comparing clinical efficacy of DHA–PPQ with artemether–lumefantrine (AL), DHA–PPQ showed a lower treatment failure rate and a longer prophylactic effect on new infections [3] possibly because the half-life of PPQ (2–3 weeks) is longer than that of lumefantrine (LUM) (4.5 days). Furthermore, recent studies conducted in Burkina Faso [4], Kenya [5], Sudan [6] and Angola [7] confirmed the high efficacy of DHA–PPQ regimen, and showed very rare cases of recrudescence within 28 days. Currently, although some evidence of slow parasite clearance in patients treated with AL for uncomplicated malaria has been reported [6, 8], there is no confirmation of artemisinin resistance spread in Africa [9]. The present study describes a potential DHA–PPQ treatment failure in an Ethiopian immigrant living in Italy who developed uncomplicated malaria after visiting relatives and friends in her country of origin.

Case report

Clinical description

A 39-year-old Ethiopian woman (living in Italy) was admitted at the Infectious Disease Unit of Policlinico Umberto I in Rome (Italy) on 29th September 2016 because of a 4-day history of fever (up to 39 °C), headache, arthromyalgia, malaise, nausea, and diarrhoea. Symptoms appeared 10 days after returning from a trip to Ethiopia (Adama city, Centre region) during which no malaria prophylaxis was taken. Physical examination at admission was normal and the patient’s weight was 56.5 kg. Blood tests showed normal white blood cell (WBC) count (3840/μL), slight anaemia (haemoglobin 11.9 g/dL), thrombocytopenia (48,000/μL), increased aspartate aminotransferase (AST 183 IU/L), alanine aminotransferase (ALT 125 IU/L), lactate dehydrogenase (LDH 595 IU/L) and C-reactive protein (CRP 99,000 μg/L). Microscopic urine exam showed micro-hematuria (131/μL); creatinine was normal. Markers for hepatitis A, B, and C were negative. Peripheral blood films showed asexual forms of P. falciparum on microscopic exam with an estimated parasitaemia of 3%. Moreover, a routine X-ray showed a small pulmonary infiltrate. The patient was treated with DHA–PPQ 320–40 mg, three tablets/day for three consecutive days, plus levofloxacin (750 mg/day/7 days). Symptoms disappeared after the second dose of DHA–PPQ, and the parasitaemia dropped to < 1%. On the 4th day, the patient was discharged. A follow-up blood microscopic exam and chest X-ray were planned, however, the patient did not return for follow-up. Thirty days after treatment, a recurrence of symptoms (fever, malaise, nausea and vomit) was reported and, 2 days later the patient was readmitted to our hospital. No additional travel abroad or contact with persons with recent journey to malaria-endemic countries was reported. Physical examination was normal. The blood tests showed normal WBCs count (4900/μL), anaemia (haemoglobin 10 g/dL), thrombocytopenia (106,000/μL), increased LDH (453 IU/L) and CRP (70,000 μg/L); liver enzymes, creatinine, and microscopic urine exam were normal. Chest X-ray and abdomen ultrasound examination were normal. Microscopic diagnosis showed asexual forms of P. falciparum (parasitaemia of 1%). Treatment with i.v. quinine hydrochloride (QN) was started given that the patient was unable to take any medication orally. Two days after, the drug was shifted to 1-day oral mefloquine (MFQ) (total dose: 1500 mg). After 1 day, symptoms disappeared and the parasitaemia dropped to < 0.1%. After two negative blood microscopic exams, performed 24 and 48 h after stopping antimalarial therapy, the patient was discharged. At 1- and 2-month follow-up visit, all blood tests were normal, and blood microscopic exams remained negative.

Molecular analysis

Total genomic DNA was extracted using PureLink Genomic DNA Kits-Invitrogen, from 200 μL of whole blood samples collected from the patient at both hospital admittances (pretreatment and 30 days post-treatment). To determine if the two malaria infections were caused by the same parasite isolates (recrudescence), genotyping of the pre-treatment and 30 days post-treatment P. falciparum isolate(s) was performed by amplification of three polymorphic markers, allowing the detection of multiclonal P. falciparum infections: the merozoite surface protein 1 (Pfmsp1) and merozoite surface protein 2 (Pfmsp2), as previously described [10], and glutamate-rich protein (Pfglurp) genes as described by Viriyakosol et al. [11]. In addition, seven neutral microsatellite markers were also analysed. Genetic characterization confirmed the presence of the same parasite isolates in both the pre-treatment and 30 days post-treatment samples implying that the second infection was a recrudescence (Table 1).

Table 1.

Neutral microsatellites analysis for the P. falciparum isolates from patient samples (first and second malaria episodes)

Microsatellite markers	TA1	PolyA	PFPK2	TA109	2490	C2M34	C3069
September 29, 2016 sample	169	146	183	175	85	236	124
November 3, 2016 sample	169	146	183	175	85	236	124

Each number represents the size of the microsatellite repeat for each of the 7 markers

Sanger sequencing of known genetic point mutations associated with various antimalarial drug resistance (artemisinin: P. falciparum Kelch 13 gene, PfK13; chloroquine: P. falciparum chloroquine resistance transporter gene, Pfcrt; amodiaquine: P. falciparum multidrug resistance 1 gene, Pfmdr1; sulphadoxine–pyremithamine: P. falciparum dihydropteroate synthase and dihydrofolate reductase genes, Pfdhps and Pfdhfr, respectively; and atovaquone: P. falciparum cytochrome B, PfCytB) was performed as previously described [12]. In addition, the copy number of P. falciparum plasmepsin 2 (PfPM2) gene, a known marker of piperaquine resistance was investigated using a PET-PCR based assay as described by Souza et al. [13]. The two sequential P. falciparum isolates had a single copy of PfPM2 gene. No mutations associated with artemisinin resistance (PfK13 gene) and atovaquone resistance (PfCytB) were detected; however, previously described mutations in the Pfcrt, Pfdhps, Pfmdr1 and Pfdhfr genes were observed (Table 2).

Table 2.

Analysis of the molecular markers of P. falciparum associated with drug resistance

Marker		Pfcrt (a)				Pfmdr1 (a)					PfK13 (b)	Pfdhfr (c)			Pfdhps (c)					PfCytB (d)
codons		74	75	76	97	86	184	1034	1042	1246	580	51	59	108	436	437	540	581	613	268
Wild type		M	N	K	H	N	Y	S	N	D	C	N	C	S	S	A	K	A	A	Y
		atg	aat	aaa	cac	aat	tat	agt	aat	gat	tgt	aat	tgt	agc	tct	gct	aaa	gcg	gcc	tat
Mutant type		I	E	T	Y	Y	F	C	D	Y	Y	I	R	N	A/F	G	E	G	S/T	S
		att	gaa	aca	tac	tat	ttt	tgt	gat	tat	tat	att	cgt	aac	gct/ttt	ggt	gaa	ggg	tcc/acc	tct
Present work	Sample September 29, 2016	I	E	T	H	N	F	S	N	D	C	N	R	N	S	A	E	A	A	Y
		att	gaa	aca	cac	aat	ttt	agt	aat	gat	tgt	aat	cgt	aac	tct	gct	gaa	gcg	gcc	tat
	Sample November 3, 2016	I	E	T	H	N	F	S	N	D	C	N	R	N	S	A	E	A	A	Y
		att	gaa	aca	cac	aat	ttt	agt	aat	gat	tgt	aat	cgt	aac	tct	gct	gaa	gcg	gcc	tat
Gobbi, 2016 [12]	Sample November 27, 2014	I	E	T	H	N	Y	S	N	Y	C	N	R	N	S	G	E	A	A	Y
		att	gaa	aca	cac	aat	tat	agt	aat	tat	tgt	aat	cgt	aac	tct	ggt	gaa	gcg	gcc	tat
	Sample January 7, 2015	I	E	T	H	N	Y	S	N	Y	C	N	R	N	S	G	E	A	A	Y
		att	gaa	aca	cac	aat	tat	agt	aat	tat	tgt	aat	cgt	aac	tct	ggt	gaa	gcg	gcc	tat

^aMutations in Pfcrt and Pfmdr1 gene are associated to some quinoline-based antimalarial resistance (e.g., chloroquine)

^bPfK13 is the molecular markers for artemisinin resistance

^cMutations in Pfdhfr/Pfdhps genes are responsible for P. falciparum resistance to antifolate-cycloguanil

^dPolymorphisms in PfCytB gene were associated to atovaquone resistance

Discussion

Resistance to artemisinin and its partner drugs is a serious threat to malaria control programs [8, 14]. Reliable molecular markers for artemisinin resistance (mutations in the propeller domain of PfK13 gene) and piperaquine resistance (multiplication of the plasmepsin 2–3 gene) have been observed in P. falciparum circulating in South-East Asia [15, 16]. Although some mutations in the PfK13 gene and slow clearing parasites in patients treated with AL for uncomplicated malaria have been reported in several African countries [6, 8, 17], there is no confirmation of artemisinin resistance in Africa [9]. Resistance to PPQ has also not been confirmed in any African parasites. In this study, we report on P. falciparum recrudescence, 30 days post-treatment with DHA–PPQ, of a traveller returning from Ethiopia after visiting friends or relatives (VFR). No known mutations in the PfK13 gene were observed nor did we observe an increase in the P. falciparum plasmepsin-2 gene, associated with PPQ resistance, implying that resistant parasites harbouring these known polymorphisms were not responsible for the recrudescence. There may be other explanations for this DHA–PPQ treatment failure such as sub-therapeutic drug exposure due to lack of efficient drug absorption by the patient or suboptimal DHA–PPQ formulation, or other unknown parasites or host genetic factors. Malabsorption due to the underlined medical reason was ruled out according to the anamnestic and clinical presentation of the patient. Moreover, DHA–PPQ was administered directly by the health personnel who ensured that the fasting condition necessary for the intake of the drug was properly carried out. Pharmacokinetic data was not collected for this patient and therefore the drug-related factors cannot be ruled out. However, the DHA–PPQ formulation used to treat the patient, Eurartesim^®, is a drug regularly distributed in Italy, and currently, AIFA (the Italian competent authority for drugs) has not reported any concerns about the drug’s quality. Furthermore, considering that levofloxacin is primarily excreted unchanged in the urine and its metabolism is minimal, the concurrent administration with DHA–PPQ will not result in an under-exposure of the antimalarial drug. To date, there is no evidence of different pharmacogenetic/pharmacokinetic profiles of DHA–PPQ in different ethnic populations [18], although individual absorption rates would differ.

Interestingly, a similar case of DHA–PPQ treatment failure was reported in Verona, Italy, in 2014 [12], in which an Italian woman also returning from a travel to Ethiopia (Omo River Valley, Southwest region) had a malaria recrudescence 35 days post DHA–PPQ therapy. In this case, pharmacokinetic data excluded under-exposure to DHA–PPQ as the cause of treatment failure, whereas genetic analysis of these P. falciparum isolates showed mutations in the Pfmdr1, Pfcrt, Pfdhfr and Pfdhps genes [12] (Table 1). Analysis of these genes in the P. falciparum isolates from our patient revealed similar point mutations (Table 1). This is consistent with what is commonly observed in Africa where previously used antimalarials such as chloroquine, sulphadoxine–pyrimethimine, and amodiaquine led to selection of these known polymorphisms, which are now common in circulation. Pfmdr1 mutations and/or amplification has been proposed to confer resistance to artemisinin partner drugs such as mefloquine (MFQ) and lumefantrine (LUM) [14]. In addition, Pfmdr1 86N/184F mutations together with Pfcrt CVIET (codons 72–76) and Pfcrt101F haplotype are associated with a slight decrease in PPQ susceptibility in field isolates [14, 19].

The main limitation of this study is the lack of pharmacokinetic data related to DHA–PPQ. Notwithstanding that we cannot rule out the possibility of pharmacological factors in this case report, this study, together with the previously published case report of recrudescence post DHA–PPQ treatment [12] raises the possibility of a potential DHA–PPQ failure in an Ethiopian isolate. We cannot rule out the possibility that other unknown factors, e.g., mutations in other known or unknown genes are involved. Multicopy occurrence of the PfPM2 was shown to be a molecular marker associated with PPQ resistance in Cambodia [16]; however, the Ethiopian P. falciparum isolates in our study had only one copy of the PfPM2.

Our study highlights the need to consider possible malaria recrudescence after DHA–PPQ therapy for patients coming from African countries like Ethiopia, for which a longer clinical follow-up seems necessary. It is important to determine the cause of these recrudescences associated with DHA–PPQ treatment in travellers returning from Ethiopia. To this end, it is worthwhile for clinicians presented with similar cases to preserve such parasite isolates to further investigate if the failure is associated with reduction in in vitro drug sensitivity. This will further help to investigate parasite genetic factors if there is any in vitro evidence to suggest potential resistance to one of the drugs.