Intended for healthcare professionals


Evidence of Plasmodium falciparum malaria resistant to atovaquone and proguanil hydrochloride: case reports

BMJ 2003; 326 doi: (Published 22 March 2003) Cite this as: BMJ 2003;326:628
  1. Anna Färnert, medical doctor (anna.farnert{at},
  2. Johan Lindberg, associate professorb,
  3. Pedro Gil, post doctoral researcherd,
  4. Göte Swedberg, associate professorc,
  5. Yngve Berqvist, associate professore,
  6. Mita M Thapar, PhD studenta,
  7. Niklas Lindegårdh, PhD studente,
  8. Sándor Berezcky, PhD studenta,
  9. A Björkman, professora
  1. a Department of Medicine, Division of Infectious Diseases, Karolinska Institute, Karolinska Hospital, S-171 76 Stockholm, Sweden
  2. b Department of Infectious Diseases, Sahlgrenska University Hospital, Östra, Gothenburg, Sweden
  3. c Department of Pharmaceutical Biosciences, Division of Microbiology, Uppsala University, Uppsala, Sweden
  4. d Centro de Malária e outras Doencas Tropicais, Instituto de Higiene e Medicina Tropical, Universidade Nova de Lisboa, Lisboa, Portugal
  5. e Department of Clinical Chemistry, Falun Central Hospital, Falun and Dalarna University College, Borlänge, Sweden
  1. Correspondence to: A Färnert
  • Accepted 9 January 2003

The increased spread of drug resistant malaria highlights the need for alternatives for treatment and chemoprophylaxis. The combination of atovaquone and proguanil hydrochloride (Malarone, GlaxoSmithKline, NC) has shown high efficacy against Plasmodium falciparum with only mild side effects and has been registered for use in several countries, including Denmark, Germany, Sweden, the United Kingdom, and the United States.1 Treatment failures have been attributed to suboptimal dosage, reinfections, or to a point mutation in the cytochrome b gene. 1 2 Bioavailability of atovaquone depends on the concomitant intake of a fatty diet, yet drug concentrations were not analysed in these reports. We provide evidence of resistance in two patients treated with atovaquone and proguanil hydrochloride for P falciparum infection.

Details of three patients treated with atovaquone and proguanil hydrochloride (Malarone; GlaxoSmithKline) for Plasmodium falciparum malaria

View this table:

Methods and results

In September 2000, two boys and their mother were diagnosed as having P falciparum malaria at the university hospital in Gothenburg, after returning from an eight week visit to the Ivory Coast. They had taken chloroquine weekly and proguanil daily for chemoprophylaxis against malaria. Case 1, the youngest boy (18 months), had fever, convulsions, and 1% infected erythrocytes. He was treated with atovaquone and proguanil hydrochloride (table). His fever continued, and he had a few convulsions but no impaired consciousness or other signs of severe malaria. On day 2 the parasitaemia was 4%. Treatment was changed to mefloquine, which was successful. His 4 year old brother, case 2, was febrile with 0.5% infected erythrocytes. His symptoms resolved with atovaquone and proguanil hydrochloride, and the parasites were cleared after three days. On day 28 he was again febrile with P falciparum parasites and was successfully treated with mefloquine. The mother (case 3), although asymptomatic, had a few P falciparum rings. She was treated with atovaquone and proguanil hydrochloride, which cleared the parasites without recrudescence.

Drug concentrations were measured by high pressure liquid chromatography in repeated serum samples. The concentrations of atovaquone (7.6-13.9 μM), proguanil (300-1200 nM), and cycloguanil (125-400 nM) were all above the levels considered therapeutic in children and adults.3 Treatment failure could therefore not be attributed to poor bioavailability of the drug.

Cases 1 and 3 were infected with single clones whereas case 2 had five genetically diverse parasite populations, detected by analysis of merozoite surface proteins 1 and 2. Mutation A803G (changing tyrosine to serine in 268) in cytochrome b, related to resistance to atovaquone, was detected in cases 2 and 3 by polymerase chain reaction and restriction fragment length polymorphism of loci 133, 268, 272, 275, 280, and 284, and confirmed by sequencing (table).4 However, only wild types were found in case 1. Analyses of loci 51, 59, 108, and 164 in the dihydrofolate reductase gene, related to resistance to proguanil and cycloguanil, revealed wild types in all samples except those from case 2, in which triple mutation were found at recrudescence.5


Treatment of three patients with atovaquone and proguanil hydrochloride for P falciparum malaria was unsuccessful in two non-immune children but successful in an adult with probable partial protective immunity. The patients had adequate blood concentrations of the drugs, indicating resistance by P falciparum. Mutation in cytochrome b may have contributed to treatment failure but cannot be the only mechanism for resistance to the drug combination because it was also detected in the patient who responded well and was not detected in the patient with early treatment failure. Atovaquone and proguanil hydrochloride represents one of the main new developments in malaria chemotherapy, but because of the resistance shown at this early stage there is a need for careful surveillance of drug efficacy.


Contributors: JL treated the patients. AF coordinated the analyses and wrote the paper with the assistance of the other authors, specifically AB. NL, YB, and MT performed the pharmacological analyses. PG, GS, and SB performed the genetic analyses. AF will act as guarantor for the paper.


  • Funding None.

  • Competing interests JL has been reimbursed by GlaxoSmith Kline for speaking at a meeting.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract