Malarone: Atovaquone and Proguanil for Treatment and Prevention of Malaria


By Maria D. Mileno, MD, and Frank J. Bia, MD, MPH

The global malaria threat continues with rapid spread of multidrug-resistant P. falciparum, which is now present in more than 90 countries, in addition to the recent emergence of chloroquine-resistant P. vivax. Falciparum malaria has caused deaths of individuals who clearly used currently recommended malaria prophylaxis. The problem of multidrug resistance, along with the obvious lack of effective malaria vaccines for the traveling public, has created a continuing and urgent need for new antimalarial agents, for both prophylaxis and treatment. An informative evening symposium was presented at the 45th Annual Meeting of the American Society of Tropical Medicine and Hygiene in Baltimore, MD, on December 2, 1996, reviewing a recent advance in combination antimalarial drug research, Malarone (atovaquone and proguanil hydrochloride), for both prevention and treatment of malaria.1 This combination of agents represents an important new advance in antimalarial drug development, and this review provides a summary of the information presented at that session.

In an initial review of preclinical evaluations of atovaquone/proguanil, W.E. Gutteridge of WHO discussed pharmacology, efficacy, absorption, metabolism, distribution, excretion, and toxicology for this combination of agents.2 Briefly, in vitro, atovaquone is a novel hydroxynaphthoquinone with a fairly broad antiparasitic spectrum, including potent blood schizonticide activity against both sensitive and resistant strains of P. falciparum (IC50s for atovaquone range from 0.7 to 4.3 nM).

In the early 1940s, interest in the hydroxynaphthoquinones as potential antimalarial agents quickly evolved as part of the war effort in the United States.3 Hydrolapachol was shown to be active against avian malaria, and this initial observation led to synthesis of several other analogues during World War II as part of the United States Antimalarial Drug Development Program. So, it became known in the 1940s that at least one other hydroxynaphthoquinone, lapinone, was effective against P. vivax, but because parenteral administration had been necessary for clinical efficacy, further drug development was abandoned. Chloroquine resistance emerged in the 1960s and led to evaluation of another new agent in this class, menoctone, which, unfortunately, appeared to be ineffective against P. falciparum infections; interest again waned. By the 1970s, the Wellcome Research Laboratories had solved the problem of rapid metabolism of one parent agent selected for investigation and ultimately produced atovaquone, which had metabolic stability and high activity against P. falciparum infections.

The hydroxynaphthoquinones possess a rather unique mechanism of action, acting as inhibitors of parasite electron transport systems, thus blocking parasite respiration and inhibiting de novo pyrimidine biosynthesis. Proguanil is the prodrug of cycloguanil, a dihydrofolate reductase inhibitor. In malaria parasites, it blocks the conversion of dUMP to dTMP in the thymidylate synthase reaction. Acting together with atovaquone, the resulting reduced pool of pyrimidine nucleotides causes a concomitant reduction of dUMP levels, which then potentiates the cycloguanil blockade of dTMP synthesis. While this biochemical synergy may be significant, it does not entirely explain the extraordinary antimalarial potentiation occurring with the combination of agents. Proguanil is an important antimalarial in its own right. Also, the proguanil molecule can form a Schiff base, which may enhance the immune response by activation of T-helper cells, and further enhance proguanil’s potential mechanisms of antimalarial action. Another proguanil metabolite, p-chlorophenylbiguanide, has also been shown to be markedly synergistic with atovaquone in vitro.

Treatment efficacy of atovaquone was tested in vitro against a variety of P. falciparum strains and clones.4 Assays in mice revealed that it was efficacious against P. yoelii using several dosing schedules. A single oral dose given 24 hours prior to murine infection resulted in significant antimalarial activity. Drug-resistant strains of P. yoelii or P. berghei (i.e., chloroquine-resistant, pyrimeth-amine-resistant, and mefloquine-resistant) were fully susceptible to this quinone. In P. falciparum infected monkeys, 1 mg/kg of oral atovaquone each day for 3-7 days cleared all infected animals of circulating parasites 2-4 days after beginning treatment. Animals receiving seven days of treatment were completely cured, while those treated with three or five doses recrudesced (except for 1 monkey), only to later self-cure.

Most hydroxynaphthoquinone compounds are extensively metabolized to derivatives that have less antimalarial activity than the parent compound; however, atovaquone is exceptionally resistant to metabolism. Pharmacokinetic studies in human volunteers show that atovaquone plasma levels for escalating doses were above the lowest detectable levels for all volunteers from 2-96 hours after the first 25 mg dose. No red-colored urine, which would imply metabolism of atovaquone, was detected. The mean elimination half-life is approximately 70 hours. In fasted subjects, a single oral atovaquone dose of 450 mg rapidly produced blood levels of 0.1-0.3 g/mL within one hour, which remain elevated for at least seven days. Food increases blood levels even further, and maximum plasma concentrations following 500 mg after food were above 4 g/mL.4 Regarding toxicology, the lethal oral dose of atovaquone is greater than 1825 mg/kg. No untoward effects, either reproductive or teratogenic, had been seen in chimpanzees. Some toxicity data for proguanil are lacking, yet it does have a long-established history of safety.

Overall, atovaquone is more active against P. falciparum than any of the currently established antimalarial drugs. However, in Thailand, clinical studies showed that recrudescence occurred in 25-30% of treated patients by day 28. Sensitivity analysis of recrudescent parasites shows that atovaquone resistance predictably develops when it is used alone, with IC50s markedly increased from pretreatment values by sensitivity analysis of the recrudescent parasites.3

Drug interactions were studied in detail by Canfield et al.5 Atovaquone was shown to be antagonistic if combined with chloroquine, quinine, pyrimethamine, mefloquine, and artemisinin compounds, and it was weakly antagonistic with primaquine. However, synergism is achieved with both the tetracyclines and proguanil. Proguanil was selected as a preferred antimalarial partner for atovaquone due to its impressive safety record and because it shows marked synergy with all tested clones and isolates, including at least one hydroxynaphthoquinone resistant isolate.

Anderson et al conducted a randomized, placebo-controlled, double-blinded study of atovaquone plus proguanil in more than 200 recipients as a suppressive prophylactic combination agent against P. falciparum in western Kenya.6 Enrolled subjects first received a course of radical treatment with 1000 mg atovaquone/400 mg proguanil once daily for three days. Subjects then received either a placebo, low-dose (250 mg atovaquone/100 mg proguanil daily) suppression, or high-dose (500 mg atovaquone/200 mg proguanil daily) suppression for 10 weeks, under supervision, with a four-week post-treatment follow-up. Weekly malaria smears were used to detect the end point event of confirmed parasitemia occurring after the initial radical course of therapy. Both combinations were 100% efficacious over the 10-week suppressive prophylactic period, during which 28 of 56 placebo-treated controls developed malaria. Preliminary drug sensitivity testing of subsequent P. falciparum isolates did not demonstrate generation of atovaquone resistant parasites. Headache appeared to be the most common side effect, and the combination was otherwise well-tolerated.

Thai patients showed excellent clinical responses following earlier studies in the United Kingdom which had demonstrated atovaquone’s consistent ability to clear falciparum parasitemias rapidly. However, a majority of the U.K. patients had experienced recrudescence of infection following monotherapy.3 Thai patients did the same and showed relapse rates of 33%. In vitro studies supported the hypothesis that atovaquone resistance was the cause since the IC50s for recrudescent parasites were found to be markedly elevated. At this point, drug combinations were evaluated, and despite a low level of proguanil sensitivity in Thailand, cure rates exceeded 95% in patients treated with atovaquone/proguanil. The few recrudescent parasites that could be tested after combination therapy showed minimal increases in their IC50s for atovaquone. Vivax malaria was also found to respond to this combination, but only the erythrocytic stages of vivax were eliminated, leaving the potential for relapse from exoerythrocytic hypnozoites.

Looareesuwan et al evaluated atovaquone alone and various combination regimens in 317 patients treated at the Bangkok Hospital for Tropical Diseases between 1990 and 1993.3 When used alone, atovaquone produced mean parasite clearance and fever times of 62 and 53 hours, respectively, with overall cure rates of approximately 67%, even with large doses (750 mg every 8 hours for 7 days). When proguanil was administered alone, in doses as high as 1 g/d, cure rates were less than 10%. Yet, when either tetracycline or proguanil was added to atovaquone, in either three- or seven-day regimens, cure rates exceeded 90% consistently. One-day therapy also produced cure rates of 83%, with only three total doses of atovaquone (500 mg) and proguanil (200 mg).

A marked decrease in susceptibility to atovaquone was evident in recrudescent parasites.3 This study clearly demonstrated the danger of attempting to use atovaquone alone for either the treatment of falciparum malaria or for retreatment of recrudescent infections. Fifteen paired isolates, obtained on admission and at recrudescence, were tested for susceptibility to atovaquone. The mean IC50 for admission isolates was 3.3 ng/mL. Those parasites obtained as recrudescent isolates from patients treated with atovaquone alone had a mean IC50 of 4947 ng/mL. However, for three important recrudescent isolates obtained from patients treated with atovaquone plus proguanil, the mean IC50 was 3.8 ng/mL. Proguanil clearly seems to prevent the emergence of resistance to atovaquone.

Radoff et al recently conducted a comparison study of atovaquone/proguanil combination (1000 mg/400 mg, respectively, daily for 3 days) vs. amodiaquine (600 mg on 3 consecutive days, total 1800 mg over 48 hours) in Lambare´ne´, Gabon for the treatment of uncomplicated P. falciparum malaria.7 Amodiaquine, despite its known adverse side effects on both liver and bone marrow function, is still a cheap and often-used alternative to other antimalarials throughout Africa. These investigators already knew that in Gabon the expected cure rate with chloroquine was only 36%, and the expected therapeutic response to amodiaquine was 65%.

In their unblinded study, with random allocation of patients to either of two treatment groups, 142 patients satisfied their inclusion criteria, which were based on documented, unmixed, and uncomplicated P. falciparum infection, quantified levels of parasitemia, and negative urine testing for chloroquine or sulfonamides. Also, women who were either pregnant or breast-feeding were excluded from the study. A rigorous intention-to-treat analysis showed that 62 (87%) of 71 patients treated with the atovaquone/proguanil combination were cured, and only one demonstrated recrudescent malaria on day 28 following treatment showing 25 parasites/L. In contrast, 51 (72%) of the 71 patients treated with amodiaquine were cured, and 12 recrudescences were documented over the 28-day follow-up period—on days 14 (3 patients), 21(2 patients), or 28 (7 patients). Sixteen (11%) patients, (8 in each group) were lost to follow-up, and the intention-to-treat analysis takes this into account, thereby possibly diminishing the measured differences in treatment outcomes, although this difference in cure rates is still significant. Fever clearance times and symptom durations were not different. Parasite clearance times were slightly shorter in the amodiaquine-treated group of patients. (See Table.) At this dosing regimen, adverse side effects were noted in both groups of treated patients. Nausea and abdominal pain were more common in the atovaquone/proguanil group, whereas amodiaquine was associated with more complaints of pruritus, weakness, insomnia, and dizziness.


Efficacy of treatment regimens for P. falciparum malaria

Atovaquone plus Amodiaquine

proguanil (n = 71) (n = 71) P

Cure (95% CI) 52 (87%; 80-95) 51 (72%; 61-82) 0.022§

Recrudescence 1 12

Mean (SD) time to 72 (23) 67 (16) 0.114‡

parasite clearance (h)

P values calculated with §x2 test and ‡Mann-Whitney U test.

Adapted from: Radloff PD, et al. Lancet 1996;347:1511-1514.

This symposium enabled participants to hear from the pharmaceutical industry (Glaxo Wellcome) regarding what appears to be an important and responsible marketing plan for Malarone, the fixed combination of atovaquone/ proguanil (2.5:1.0). While providing access to the drug, there should be no attempts to replace existing antimalarial therapy with Malarone. It is viewed as reserved for second- or third-line therapy of resistant malaria. Atovaquone itself is an agent about which more information is necessary for use during pregnancy. It is known to cause some maternal toxicity in rabbits, although it is not teratogenic, nor did it cause reproductive toxicity in rats. Atovaquone is currently classified in FDA pregnancy category C.

Glaxo Wellcome is establishing a controlled access system, and Malarone is to be donated free of charge to those programs in which both patient education and use of bednets are part of control programs. It plans to donate 1 million treatment doses each year to ensure that it is used in a responsible and controlled manner.


1. Symposium: Am J Trop Med Hyg 1996;56:35.

2. Gutteridge WE. Preclinical evaluation of atovaquone and proguanil for combination antimalarial therapy. Abstract presented at Malarone symposium (ref 1).

3. Loorareesuwan S, et al. Am J Trop Med Hyg 1996;54:62-66.

4. Hudson AT, et al. Drugs Exptl Clin Res 1991;17: 427-435.

5. Canfield CJ, et al. Experimental Parasitol 1995;80: 373-381.

6. Anderson S, et al. Efficacy and safety of atovaquone and proguanil for prophylaxis of malaria. Presented at Malarone symposium (ref. 1).

7. Radloff PD, et al. Lancet 1996;347:1511-1514.