Chloroquine resistance of Plasmodium falciparum in Brazil detected by a simple in vitro method*

Bull. Org. mond. Sant)1 1971, 45, 157-167Bull. WldHlth Org. J Chloroquine Resistance of Plasmodium falciparum in Brazil Detected by a Simple in vitro Method* KARL H. RIECKMANN1 & FRANCISCO J. LOPEZ ANTURiANO 2 The sensitivity ofPlasmodium falciparum to antimalarial drugs is primarily determined by observing changes in the level ofparasitaemia after administration ofdrugs to man. Per- formance of such a study and interpretation of its results may present certain difficulties, especially when conducted among persons living in malarious areas. The present study was conducted to evaluate the usefulness of a simple in vitro method for detecting chloro- quine-resistant strains among infected residents of Mato Grosso, Brazil. The findings indicated the participants were infected with chloroquine-resistant P. falciparum. Subse- quent administration of a standard regimen of chloroquine (25 mg/kg) to these subjects confirmed the findings observed in vitro by failing to cure 28 out of 30 infections. This in vitro technique seems to provide a useful method ofdetermining the presence orprevalence of chloroquine-resistant strains in a given area. Emergence of chloroquine-resistant strains of Plas- modiumfalciparum in South America and South-East Asia led to the adoption of a uniform in vivo proce- dure for determining the sensitivity of such strains to chIoroquine (WHO Scientific Group on Chemo- therapy of Malaria, 1967; WHO Expert Committee on Malaria, 1968). A standard regimen of chloro- quine is administered to infected subjects and the response of their asexual parasitaemia is determined by frequent examination of peripheral blood films. If parasites disappear within 1 week after treatment, observations are extended for an additional period of 3 weeks to exclude a recurrence of parasitaemia. While this procedure has been applied successfully, a simple in vitro method would facilitate appraisal of the problem of drug resistance for the following reasons: (1) Collection of a single specimen of blood for the in vitro test does not require repeated observations * This study was sponsored and supported by the Pan American Health Organization, Regional Office of the World Health Organization, Washington, D.C., USA, and was conducted while the first author served as a temporary adviser to this Organization. It was also supported, in part, by the Ministry of Public Health, Government of Brazil, and by the Medical Research and Development Command, Department of the Army, USA, under Contract DA-49- 193-MD-2413 with the University of Chicago. l Associate Professor of Medicine and Preventive Medi- cine, Rush Medical College, Chicago, Ill., USA. 2 Regional Parasitologist, Pan American Health Organi- zation, Rio de Janeiro, Brazil. as in the in vivo procedure and has the following advantages: (a) there is economy in funds and personne'l; (b) there is better co-operation from the communi- ty since the sampling causes only slight interference with normal activities; (c) the drug-sensitivity of parasites can be deter- mined in a higher proportion of participants; and (d) drug-resistant strains are rapidly detected. (2) Uncertainty regarding absorption of the drug in vivo (e.g., owing to vomiting or individual variation in the absorption or metabolism of a drug in vivo) need not be taken into consideration when the drug is added in vitro. (3) Recurrence of parasitaemia within 4 weeks after administration of chloroquine in vivo provides only presumptive evidence of a drug-resistant infection in areas where re-infection cannot be ruled out; determination of the sensitivity of parasites to chloroquine in vitro is a significant advantage of the method when the transmission of malaria cannot be excluded. Recent studies in the laboratory have shown that an in vitro method can be used to differentiate between a drug-sensitive and a drug-resistant strain of P. falci- parum (Rieckmann et al., 1968). The simplicity of the technique indicated that this method might be useful in determining the sensitivity of P. falciparumn 2714 157- 158 K. H. RIECKMANN & F. J. L. ANTUN4ANO to chloroquine under field conditions. The study described in this report was designed to test its po- tential value for this purpose. METHODS The study was conducted at Cuiaba, capital of Mato Grosso State, Brazil, between March and May 1969. Subjects participating in the study were usual- ly male adults (Table 2) who had recently arrived in Cuiaba' from one of the rural areas shown in Fig. 1. The locality from which the patient came and the number allocated to him are shown in the follow- ing tabulation: Locality Patient 1 29 2 30 3 1,9,20 4 27 5 2 6 3, 16 7 4 8 28 9 24 10 8 1 1 5,22,25 12 6, 15, 26 13 7, 10, 11, 12, 13, 14, 17, 18, 19, 21, 23 The patients presented themselves at the Sector Headquarters of the Malaria Eradication Campaign and reported that they had experienced episodes of fever during the preceding months. Although a few persons became ill shortly after reaching Cuiaba, most individuals developed their first episode of fever before their arrival in the city. Many subjects had taken chloroquine or, less frequently, one or more of the following drugs: hydroxychloroquine, amodiaquine, mepacrine, pyrimethamine, sulphor- metoxine. Most subjects indicated that they had taken no antimalarial drugs for at least a week be- fore participating in the study. A finger-tip specimen of blood was obtained from each subject and a thick film was prepared. Blood films were stained with Giemsa stain using the rapid method described by Field, Sandosham & Fong (1963) and examined by conventional microscopy. Subjects whose blood contained large trophozoites of P. falciparum (96% of all subjects) were requested to participate in the study. Throughout the entire study, the clinical well-being of the subject always took precedence over the determination of the sensi- tivity of parasites to chloroquine. The study was designed to compare the effects of chloroquine upon P. falciparum in vitro with those observed in vivo. After withdrawal of blood for the in vitro test, subjects were given chloroquine (either 10 or 25 mg of base per kg of body weight) and follow-up observations were made over a period of 4 weeks to determine the response to treatment. Among 45 subjects who participated in the study, ob- servations were completed in 13 who received 10 mg of chloroquine per kg of body weight and in 30 who received 25 mg/kg. Determination of the sensitivity ofparasites to chloro- quine in vitro The procedure, previously described by Rieck- mann et al. (1968), involved the following steps: (1) A small sample of venous blood (about 8 ml) was collected into a plain silicone-coated vacuum tube,' transferred immediately to a sterile 25-ml Erlenmeyer flask containing glass beads, and defibri- nated by rotating the flask for 5 minutes. (2) Using a sterile 1-ml pipette, blood samples were then placed in screw-cap, flat-bottomed glass vials (6 cm by 1.5 cm internal diameter) that contained glucose (5 mg) and either no drug (control) or chloro- quine diphosphate. For convenience, glucose and chloroquine had been added to vials before the study began, according to procedures outlined previously (Rieckmann et al., 1968). The vials contained 0.5, 1.0, 1.5, 2.0, or 2.5 nanomoles (nmol) 2 of chloro- quine, and upon the addition of blood, these amounts yielded concentrations equivalent to 160, 320, 480, 640, or 800 ,ug of chloroquine base per litre of blood. (3) Blood in the vials was swirled gently to mix the contents well, and the vials were then placed in a water-bath at 38.0-40.00C for a period of 24 hours. (4) After incubation, the vials were shaken to resus- pend cells in plasma, and thick blood films were prepared and stained for 20 minutes with Giemsa stain (3 ml of Giemsa stain diluted with 60 ml of 0.9% saline plus 37 ml of phosphate-buffered water, pH 7.0). Maturation of parasites in vitro is inhibited by chloroquine. The extent of this inhibition can be assessed by comparing the degree of maturation in control samples of blood with that observed in 1 Vacutainer, supplied by Beckton Dickinson & Co., Rutherford, N.J., U.S.A. 2 1 nmol=0.320,ug of chloroquine base. DETECTION OF CHLOROQUINE RESISTANCE OF PLASMODIUM FALCIPARUM IN BRAZIL Fig. 1. Study area and probable origin of strains of P. falciparum. Inset: northern part of South America to show the location of Mato Grosso State. samples containing chloroquine. Previous studies had shown that the percentage of ring forms that mature to schizonts containing more than two nuclei provided a useful endpoint for the quantitative asess- ment of maturation and could be used to differentiate a chloroquine-sensitive strain of P. falciparum from a chloroquine-resistant strain (Rieckmann et al., 1968, 1969). The number of schizonts per 100 asexual parasites was determined in samples containing chlo- roquine and in control samples. Values noted for samples containing the drug were divided by values noted for corresponding control samples, and the results were expressed as percentages. Table 1 shows the method used to express the results. 159 K. H. RIECKMANN & F. J. L. ANTUfiANO Table 1. Example of method used to express findings in vitro Percentage No. of schizonts of parasites, relative Sample per 100 parasites to control samples after incubation (100 %), that matured to schizonts samples without chloroquine: 1 70 2 76 mean 73 sample with 48 chloroquine 48 4 x 100 = 6673 Determination of the sensitivity roquine in vivo. of parasites to chlo- Administration of 10 mg of chloroquine base per kg. After the collection of blood samples for study in vitro, 14 subjects received a single oral dose of 10 mg of chloroquine base per kg of body weight. Thick blood films were made daily for 1 week after treat- ment and parasites were counted according to the method of Walker (Manual of the microscopic diag- nosis of malaria, 1963). Samples of urine, plasma, and red cells for estimation of chloroquine levels were collected just before treatment and 2 days after treatment and were kept frozen until they could be analysed at the Walter Reed Army Institute of Research, Washington, D.C., USA, by means of an automated method developed by Kazyak & Knob- lock (personal communication). A week after medi- cation with a single dose of chloroquine, 10 of these subjects received a standard dose of chloroquine (i.e., 25 mg/kg). Administration of 25 mg/kg chloroquine base. In accordance with the currently recommended proce- dures (WHO Scientific Group on Chemotherapy of Malaria, 1967; WHO Expert Committee on Malaria, 1968), 37 patients (7 of whom were later excluded) received 25 mg of chloroquine base per kg of body weight over a period of 3 days (10 mg/kg on day 0; 10 mg/kg on day 1; 5 mg/kg on day 2). The actual dosage of chloroquine varied from 24.3 to 25.9 mg/kg for persons weighing less than 60 kg. Heavier sub- jects (Table 2) were given 1500 mg ofchloroquine base and, with the exception ofpatient 5, the lowest dosage administered to these subjects was 22.1 mg/kg. The first daily dose was administered immediately after a sample of blood had been collected for study in vitro. Thick blood films were examined every day for a week following the start of treatment. When asexual parasitaemia was not evident by the end of 1 week, routine parasite counts were performed at weekly intervals for an additional period of 3 weeks. Specimens of blood were occasionally obtained on unscheduled days if a participant became sympto- matic or if he expected to leave the area before the study was completed. Observations for this study were usually discontinued upon the recurrence of asexual parasitaemia. Oral temperatures were recorded each time a finger-tip specimen of blood was obtained. Specimens for the estimation of chloroquine levels were collected just before treatment and 3 days after treatment started. Chloroquine levels after treatment were not determined in patients 1 and 12. Among 30 subjects who completed the study, 9 had received a single dose of chloroquine (10 mg/kg) 1 week before the standard regimen of chloroquine (25 mg/kg) was administered. The susceptibility of parasites to 25 mg/kg chloro- quine was graded according to the classification re- commended by the WHO Scientific Group on Chemo- therapy of Malaria (1967): (1) Sensitivity (S), clearance of asexual parasi- taemia within 7 days of the first day of treatment, without recrudescence. (2) Resistance (RI), clearance of asexual para- sitaemia as in sensitivity, followed by recrudescence of asexual parasites within 28 days of the first day of treatment. (3) Resistance (RII), marked reduction of asexual parasitaemia but no clearance. Patients with an RI level of resistance were classi- fied according to the week after drug administration during which they showed a recrudescence of patent parasitaemia. Transmission of malaria within Cuiaba' is considered minimal or non-existent and partici- pants did not leave the city during the period of study. Consequently, recurrence of parasitaemia probably originated from recrudescence of the treated infection and not from a new infection. RESULTS Sensitivity ofparasites to chloroquine in vitro Table 2 shows detailed results of studies per- formed in vitro with parasitized blood collected from 30 subjects. Data from 7 infected subjects whose response to a standard regimen of chloroquine could 160 DETECTION OF CHLOROQUINE RESISTANCE OF PLASMODIUM FALCIPARUM IN BRAZIL 161 Table 2. Effects of chloroquine on maturation of parasites in vitro, grouped according to the response of infections to chloroquine in vivo No. No. of schizonts Percentage of parasites, relative to control samples (100 %), that PatAge of asexual per 100 parasites matured to schizonts in samples containing the following amountsPatient and sex parasites after incubation of chloroquine (nmol/ml of blood): a (per MM3) in s~amples withoutchloroquine (controls) 0.5 1.0 1.5 2.0 2.5 RIl b 1 9M 30 800 15 87 93 100 60 20 2 34M 6 000 47 85 81 28 40 4 3 1oM 11 550 20 68 78 29 39 9 4 42M 5 700 32 96 81 31 18 0 5 61 M 7 700 13 118 59 29 7 0 6 1 1 M 13 800 45 84 67 1 8 2 0 7Cc 25Ml 1 8200 50od 92 44 1 0 2 - 7C 25M ~~~4000 236e 69 30 13 0 0 8 18M 3 000 73 107 66 16 0 0 9 19F 5 900 24 83 29 4 0 0 RI (1st week) b 10 33M 4 800 22 100 77 68 45 9 8 970 lod 80 65 20 15 - 11 e 40M 3 300 46 e 87 74 22 11 9 1 2 c 27M 4 300 734 95 53 37 20 - 13C 30M 6800 604 96 69 5 2 -1 100 556e 92 49 16 10 0 RI (2nd week) b 14 26M 6 600 20 105 75 70 35 20 15 29M 1 200 60 96 55 33 26 16 16 5F 12 000 21 95 71 47 14 0 17C 25M 17680 4 50 25 0 0 - 17500 206 78 43 34 9 0 1 8 c 15M 9 300 30 d 92 58 20 10 - 19 25M 3 850 29 101 52 7 0 0 20c 36M 5250 48d 79 35 12 0 -20Ce } 36M 5 250 45 e 64 0 0 0 0 RI (3rd week) b 21 c 19M 1 600 49d 94 63 61 18 - 300 SB6e 93 62 48 4 4 22 16M 1 350 90 104 51 33 16 2 23 41 M 2400 69 81 68 16 14 0 24 25M 1 900 35 77 42 14 1 1 0 25 39M 700 66 91 53 20 0 0 26 33M 800 79 82 22 5 0 0 RI (4th week) b 27 47M 500 42 116 10 10 0 0 28 48M 3500 35 68 2 0 0 0 S b 29 1 M 1 520 66 90 92 57 30 13 30c 65M 1 650 10d 80 50 10 10 - a 1 nmol -0.320 jAg of chloroquine base. b Response of infections to the administration of 25 mg/kg chloroquine base: RIl = marked reduction in asexual parasitaemia, but no clearance. RI = clearance of asexual parasitaemia as in sensitivity (S), followed by recrudescence of asexual parasitaemia within 28 days. The week after medication during which parasitaemia recurred is shown in parentheses. S = clearance of asexual parasitaemia within 7 days of the first day of treatment, without recrudescence. c Subjects received single dose of chloroquine (10 mg/kg) 7 days before receiving the standard regimen of chloroquine (25im7g/kg.d In vitro study performed before the single dose of chloroquine. e In vitro study performed before the standard regimen of chloroquine. K. H. RIECKMANN & F. J. L. ANTUNANO not be determined are not included; however, the in vitro results for these subjects were similar to those obtained with the other subjects. Asexual parasite counts prior to incubation varied from 250 to 30 800/mm3. Earlier studies had shown that, although total levels of parasitaemia do not change significantly during incubation, maturation of trophozoites in control (without drug) samples of blood varies considerably from one study to another (Rieckmann et al., 1968). In this study, the number of schizonts formed during incubation varied from 4 to 90 per 100 asexual parasites and, as was noted in earlier studies, the number seemed to be related to the proportion of older ring forms present at the outset of incubation. This variation did not inter- fere with the detection ofchloroquine-resistant strains of P. falciparum and emphasizes the need to compare maturation of parasites in samples containing the drug with that observed in corresponding control samples (Rieckmann et al., 1968). Samples of blood containing 0.5 nmol of chloro- quine showed little or no inhibition of the formation of schizonts. When samples contained 1.0 nmol, slight to marked inhibition of maturation was observ- ed but the formation of schizonts was not completely inhibited. Complete arrest of the formation of schizonts was observed in a few samples containing 1.5 nmol, in approximately one-third of the samples containing 2.0 nmol and in approximately two-thirds of the samples containing 2.5 nmol of chloroquine. Fig. 2 is a comparison of the effects of chloroquine in vitro upon the Uganda I and Malayan (Camp.) strains of P. falciparum with those observed in the present study. Studies with the Uganda I and Malay- an (Camp). strains were performed previously at the University of Chicago, Army Medical Research Pro- ject (Rieckmann et al., 1968). The Uganda I strain is sensitive to chloroquine (Martin & Arnold, 1967) whereas the Malayan (Camp.) strain is resistant to the drug (DeGowin & Powell, 1965). In this study of strains from Mato Grosso, results obtained in vitro with patients whose parasitaemia cleared temporarily (RI) after a standard regimen of chloroquine (25 mg/ kg) are depicted separately from those obtained with patients whose parasitaemia showed no clearance (RII). Maturation of parasites in the presence of chloroquine varied to a similar extent in samples of blood obtained from either group of subjects. In 2 cases with no recrudescence of parasitaemia (S), the findings in vitro were comparable to those ob- served in 28 cases with an RI or RII level of resis- tance (Table 2). These results were in marked contrast to previous findings with the chloroquine-sensitive Uganda I strain in which the formation of schizonts was com- pletely inhibited when 1-ml blood samples contained 0.5 nmol of chloroquine. Moreover, previous studies with the chloroquine-resistant Malayan (Camp.) strain showed that 1.5 nmol per ml of blood usually inhibited the formation of schizonts completely. Similar results were obtained when parasites of these two strains were incubated in unused vials returned to the University of Chicago after the field study. The findings indicated that strains of P. falciparum in Mato Grosso were considerably less sensitive to chloroquine added in vitro than the Ugan- da I strain and that their sensitivity to chloroquine was equal to or less than that of the Malayan (Camp.) strain. Results of studies with samples of blood obtained from subjects who received 2 courses of chloroquine are included in Table 2. Studies with patients 7, 11, 13, 17, 20, and 21 were performed before admini- stration of 10 mg/kg and, again, just before admini- stration of 25 mg/kg. Effects of chloroquine observed during the first and second studies seemed to differ significantly in only 2 cases. In the first study with patient 17, scanty formation of schizonts in the con- trol samples prevented adequate assessment of the maturation of parasites in samples containing chloro- quine. In the second study with patient 20, samples of blood obtained from this subject before the drug was added in vitro contained a substantial amount of chloroquine (see below) and it is likely that this accounted for the marked inhibition of schizogony in vials containing chloroquine. Sensitivity ofparasites to chloroquine in vivo Administration of 10 mg/kg chloroquine base. Lev- els of asexual parasitaemia before treatment varied from 400 to 9 300/mm3 (mean: 3 600). In 9 cases, levels of parasitaemia showed virtually no change after treatment. The remaining 4 cases showed a reduction in levels of parasitaemia followed by a rise within 7 days of treatment. Chloroquine was not detected in any specimens of blood or urine collec- ted before treatment, but it was detected in all speci- mens collected 2 days after treatment. Administration of25 mg/kg chloroquine base. Table 3 shows the susceptibility of P.falciparum infections to chloroquine according to the currently recom- mended classification (WHO Scientific Group on Chemotherapy of Malaria, 1967). Among 30 patients who completed the study, 9 showed no clearance of 162 DETECTION OF CHLOROQUINE RESISTANCE OF PLASMODIUM FALCIPARUM IN BRAZIL 163 l 00 UGANDA I MALAYA (camp.) 80] 802 60- 60- 40 40 N 20 * 20C -= 0 002 05 LO 15 2.0 2.5 02 O.5 LO 15 2.0 2.5 BRAZIL(Mtos Grosso) BRAZIL (Mato Grosso) eo- ~~~~~~~~~~~~~~~~120- E R I *R II h)w long 0ea 04~~~~~ 0.) 80 .0So8- 0 00. 0.. 6 0 60 0 40 40-~~~ 200 0o 40 0. 4 0000 03 LO L5 2.0 2.5 0 0.5 LO L5 2.0 2.5 Amount of chloroquine ( nmol) added per ml of defibrinated blood WH103 Fig. 2. Effects of chloroquine added in vitro upon maturation of asexual erythrocytic forms of the Uganda I and the Malayan (Camp.) strains of P. falciparum and strains of P. falciparum from Mato Grosso, Brazil, with RI and RII levels of resistance. The ordinate represents the percentage of parasites, relative to the control sample (100%), that matured to schizonts. Points are individual values observed after incubation with the amounts of chloroquine shown along the abscissae. A line is drawn through the means of these values. Table 3. Results obtained in vivo after administration of 25 mg of chloroquine per kg of body weight Response No. No. of cases with fever b Concentration of chloroquine base of No. on following days: (jig/litre) on day 3 c parasitaemia a of cases Q ori1 2or3 I4or5 Red cells Plasma Rhl 9 6 5 2 184 (0-400) 279 (100-560) RI 19 8 3 3 281 (0-770) 435 (60-1 270) S 2 1 1 1 215 220 a Refer to Table 2. b Oral temperature above 37.2°C. c Average values; range of values shown in parenthesis. K. H. RIECKMANN & F. J. L. ANTUNANO parasitaemia (RII), and 2 showed a clearance of parasitaemia without subsequent recrudescence (S). Most patients showed a recrudescence of asexual parasites after the initial clearance of patent parasitaemia (RI). Among 19 cases in this cate- gory, 4 showed a recrudescence during the first week, 7 during the second week, 6 during the third week, and 2 during the fourth week after drug admini- stration (Table 2). The clinical response to chloroquine in sympto- matic infections was usually satisfactory, even in subjects whose parasitaemia showed no clearance after treatment. By the fourth or fifth day after medication, low-grade fever was recorded in only 6 cases (Table 3). Chloroquine was detected in all specimens of blood and urine collected after administration of the drug. Only specimens from patients 1 and 12 were unavail- able for examination. Levels ofchloroquine observed in different subjects varied considerably (Table 3) and relative concentrations of chloroquine in red cells and in plasma showed no consistent pattern. In comparing data from subjects with an RI and RII level of resistance, no significant differences in the average values of chloroquine were observed in red cells (t = 1.33; 0.1 <p<0.2) or plasma (t = 1.23; 0.2 <p <0.3). Chloroquine was detected in samples of blood collected from 7 subjects prior to the admi- nistration of 25 mg of chloroquine base per kg; 6 of these had received chloroquine (10 mg/kg) 7 days previously. Just before receiving 25 mg of chloro- quine per kg, patient 20 showed a level of 90 ,ug of chloroquine base per litre of red cells and 330 ,pg of chloroquine base per litre of plasma. The average concentration of chloroquine in the other 6 patients was 4 jug per litre of red cells (range: 0-20 ,ug/litre) and 100 jug per litre of plasma (range: 40-140 pg/litre). Symptoms suggestive of malaria among subjects with an RII level of resistance had usually started less than a month before treatment with chloroquine. On the other hand, most subjects with an RI level of resistance had noted their first symptoms more than a month before participating in the study (Table 4). The onset of symptoms reported by the two groups of subjects was significantly different (X2 = 4.670 with Yates' correction; P = 0.03 by Fisher's exact test). The findings suggested that subjects whose parasitaemia was less responsive to chloroquine (RII) had, in general, acquired their infections more recently than other participants in the study (RI and S). The virtual absence of game- tocytes in persons with an RII level of resistance provided further support for this hypothesis (Table 4). Low levels of asexual parasitaemia at the outset of the study were usually noted in persons whose parasites recurred during the third or fourth week after treatment or whose parasites appeared to be sensitive to chloroquine (Table 2). Patient 28 showed the highest level of asexual parasitaemia (3 500 para- sites/mm3) among this group of subjects; his first epi- sode of fever occurred 2 weeks before treatment and examination of his blood film revealed no gameto- cytes. DISCUSSION Resistance of P. falciparum to chloroquine pre- vails in various areas of Brazil (Da Silva & Lopes, 1966; Walker & Lopez Antufiano, 1968) and is known to occur at Tres Lagoas in the southern part of the State of Mato Grosso (Lopez Antuniano & Sao Thiago, personal observations). In the pre- Table 4. Duration of symptoms and gametocyte levels before treatment with 25 mg of chloroquine per kg of body weight No. of persons Response with following duration No. of persons with: Average no.No. of symptoms: of parasitaemia a of cases gametocytes >1 month <1 month Gametocytes gametocytes per mm RIl 9 2 7 1 8 20 RI 19 14 5 12 7 217 S 2 2 0 2 0 825 a Refer to Table 2. 164 DETECTION OF CHLOROQUINE RESISTANCE OF PLASMODIUM FALCIPARUM IN BRAZIL sent study, most infections were probably acquired in one of the areas shown in Fig. 1, many of which are frequently situated near rivers flowing either north into the Amazon River or south into the Parand River. One of these infections probably originated in a village close to the Bolivian border. The results of the study have established the widespread occur- rence of chloroquine-resistant strains of P. falci- parum in central and northern Mato Grosso, which is geographically the centre of the South American continent. The findings presented in this report show that an in vitro method can be used successfully to detect resistance of P. falciparum to chloroquine in resi- dents of an endemic area. Results obtained by this method indicated that each participant in the study was infected with a chloroquine-resistant strain of P. falciparum; the failure to cure 28 out of 30 cases by the administration of a standard regimen of chlo- roquine supported, in general, the in vitro findings. Both subjects whose infections were cured by chloro- quine reported episodes of fever for 2-3 months and showed high levels of gametocytaemia at the begin- ning of the study (Table 4). This suggests that they had been infected with malaria for some time and, consequently, had probably acquired a substantial degree of partial immunity before participating in the study. In studies with drug-resistant infections of P. falciparum in South America, Walker & L6pez Antuniano (1968) observed that clearance of para- sitaemia and lack of a subsequent recrudescence occurred more frequently when chloroquine was ad- ministered during the later stages of an infection than when it was given during the earlier stages. During a field study among indigenous populations in Col- ombia, chloroquine-resistant infections were ob- served among children but were not seen in adults (Comer et al., 1968). The authors of the report concluded that children in endemic areas provided a better model for determining the sensitivity of P. falciparurm to chloroquine than the more immune adults. It is well known that the response of P. falciparuin infections to chloroquine in partially immune subjects differs substantially from that ob- served in nonimmune subjects; a nonimmune per- son frequently shows a slower and less complete res- ponse of fever and parasitaemia to the same drug anddosage(WHO Scientific Group on Chemotherapy of Malaria, 1967). This is exemplified by the wide spectrum of response observed in volunteers in- fected with the Malayan (Camp.) strain of chloro- quine-resistant P. falciparum. The response of an infection to a standard regimen of chloroquine (25 mg/kg) varies from resistance at the RII level in some nonimmune volunteers to a radical cure (S response) in an occasional volunteer with a high degree of partial immunity and a low rate of para- sitaemia. In nonimmune volunteers infected with this strain, chloroquine may allow the infection to pursue an asymptomatic course despite its inability to achieve a radical cure (Rieckmann, 1970). In the present study, 2 out of 9 cases with an RII level of resistance to chloroquine became sympto- matic more than a month before the study and a much higher proportion of subjects with a lower level of resistance experienced symptoms for this period. Only 1 case with an RII level of resistance showed a gametocytaemia. This provides further support for the premise that persons in this category had, in general, acquired their infections more re- cently than other participants in the study. There- fore, subjects with an RII level of resistance to chloroquine, considered as a group, probably had de- veloped less partial immunity to the infections than their counterparts with an RI level of resistance and, accordingly, their infections were less responsive to treatment. In contrast, maturation of parasites in vitro during incubation with chloroquine does not seem to be markedly affectedby partial immunity. Earlier studies to assess the effects of chloroquine upon the Uganda I or Malayan (Camp.) strains revealed that the extent of formation of schizonts during the early course of an infection was similar to that observed during its later stages (Rieckmann et al., 1968). Recent studies performed in vitro with P. knowlesi have shown that immune serum, although preventing re-invasion of erythrocytes, has little effect on the intracellular ma- turation of trophozoites to schizonts (Cohen, But- cher & Crandall, 1969). These observations suggest that immune factors do not interfere with the res- ponse of parasites to chloroquine in vitro to the extent observed in vivo. Consequently, the drug sensitivity of a strain of P. falciparum may be more accurately determined in vitro than is possible in vivo, and this factor may represent an additional advantage of the in vitro test. Maturation of parasites in the presence of chloro- quine was more variable in this study than had been noted previously (Rieckmann et al., 1968) with the Malayan (Camp.) strain (Fig. 2). Infections were acquired in various localities of Mato Grosso (Fig. 1) and the differences noted in vitro may reflect 165 K. H. RIECKMANN & F. J. L. ANTUF4ANO dissimilarities in the sensitivity of these strains to chloroquine. On the other hand, the variability ob- served in the present study may be due to other factors. During studies with human volunteers in- fected with the Uganda I strain, Rieckmann et al. (1968) found that the formation of schizonts varied widely in samples of blood containing 0.2 nmol of chloroquine (Fig. 2). In view of this observation, it seems reasonable to expect even greater variability in a field study where a large number of infections are studied, where levels of parasitaemia vary more widely, and where experimental conditions are gene- rally not as stable as those encountered at a malaria centre. Variability observed in vitro was similar among each group of subjects with an RII, RI, or S response to chloroquine in vivo (Table 2, Fig. 2). Lack of correlation between findings observed in vitro and those observed in vivo is not surprising because participants had presumably acquired vary- ing degrees of immunity before entering the study. Furthermore, individual variation in response to treatment with chloroquine may be unrelated to an acquired immunity to malaria. This has been shown to occur in subjects infected with the same strain of chloroquine-resistant P. falciparum (DeGowin & Powell, 1965; Powell et al., 1964a; Walker & Lopez Antufiano, 1968). In persons whose infections recurred more than 2 weeks after treatment or appeared to be sensitive to chloroquine, levels of parasitaemia were usually below 2 000/mm3 and frequently below 1 000/mm3 (Table 2). Such low-grade levels of parasitaemia may or may not have been related to a degree of partial immunity. Our study tends to support the statement that " tests carried out on semi-immune asymptomatic carriers with less than 1 000 tropho- zoites/mm3 of blood probably do not provide a sound basis for the full assessment of the action of the drug on nonimmune subjects" (WHO Scientific Group on Chemotherapy of Malaria, 1967), and indicates the value of the in vitro method in detecting chloroquine-resistant strains under such circumstances. In vitro findings in patient 28 (Table 2) suggest that he was probably infected with a strain of P. falciparum that was more susceptible to chloro- quine than strains that infected other subjects in the study. By contrast with the second study of patient 20 (see above), no chloroquine was detected in samples of blood collected before the addition of chloroquine in vitro. The level of asexual parasi- taemia (Table 2) with no concurrent gametocy- taemia observed before treatment and the recent onset of symptoms tend to indicate that this subject had probably acquired less immunity to the infec- tion than other persons with a late recrudescence. Consequently, the delayed recrudescence of para- sitaemia may reflect the response of a strain of P. falciparum that is less resistant to chloroquine, both in vivo and in vitro, than most strains of P. falciparum encountered in the present study. A few years ago one of the authors investigated the response of P. falciparum to chloroquine among 4 infected subjects in Guatemala (Rieckmann, personal observation). Results obtained in vitro showed that the sensitivity of parasites to chloro- quine was similar to that observed with the Uganda I strain. It is conceivable that other chloroquine-sen- sitive strains may be less susceptible to chloroquine and studies should be conducted in various areas to establish baseline data on the sensitivity of such strains to chloroquine and other drugs. Other antimalarials have also shown a marked difference in their effects upon parasites of the Malayan (Camp.) strain and upon those of the Uganda I strain of P. falciparum (Rieckmann et al., 1968). In malarious areas where dihydrofolate reductase inhibitors such as pyrimethamine, pro- guanil, or cycloguanil are being used extensively, the in vitro method may be useful in detecting early emergence of drug-resistant strains. Determination of the sensitivity of parasites to these " antifolates " may also be of value in areas where chloroquine- resistant strains are prevalent because some of them could be sensitive to such drugs (Powell et al., 1964b; Walker & Lopez Antufiano, 1968; Young, 1962). The findings presented in this report emphasize the need for further in vitro studies in malarious areas. The simplicity of the technique represents a potential major advantage in determining the presence or prevalence of drug-resistant parasites in a particular geographical area. During further assessment of the value of this method, studies should always be performed in conjunction with the standard procedure (WHO Scientific Group on Chemotherapy of Malaria, 1967) for determining the sensitivity of parasites to chloroquine. 166 DETECTION OF CHLOROQUINE RESISTANCE OF PLASMODIUM FALCIPARUM IN BRAZIL 167 ACKNOWLEDGEMENTS We are grateful to Dr Mario de Oliveira Ferreira, Superintendent, and to Dr Djalma Almeida, Chief, Tech- nical Division, Malaria Eradication Campaign, Brazil, for their active support of the study. Our study at Cuiaba was facilitated by the efforts of Dr Joao de Almeida, Chief of the Sector of Mato Grosso, and members of his staff. Mr Bento Madalena provided valuable technical help. We thank Colonel Edward Knoblock, Director, Division of Biochemistry, Walter Reed Army Institute of Research and members of his staff for conducting the chloroquine determinations. Finally, we are grateful to Dr P. Carson and Dr G. Garcia Martin for reviewing this manuscript and to our other colleagues for their helpful suggestions. RESUME RtSISTANCE A LA CHLOROQUINE DE PLASMODIUM FALCIPARUM AU BRESIL DtTECTEE IN VITRO PAR UNE METHODE SIMPLE En 1969, une 6tude sur le terrain a e faite au Br6sil en vue de determiner la valeur d'une m6thode simple pour la detection in vitro des souches de Plasmodium falciparum resistantes A la chloroquine. Les sujets de l'etude etaient pour la plupart des adultes de sexe masculin ayant contract6 l'infection dans les regions du centre ou du nord de l'Etat de Mato Grosso. La methode in vitro consiste essentiellement a faire incuber I ml d'echantillon de sang parasite avec du glucose pendant 24 heures, et A constater dans quelle mesure la maturation des trophozoites en schizontes est inhibee par l'addition de diverses quantit6s de chloroquine. Le fait que des schizontes ont pu se former en presence de certaines concentrations de chloroquine atteste que chacun des sujets etait infecte par une souche de P. falciparum resistante A la chloroquine. Apres la collecte des echantillons hematologiques pour 1'6preuve in vitro, les sujets infect6s ont recu un traitement standard de 25 mg/kg de chloroquine-base pendant trois jours (soit 1500 mg en trois jours). Ce traitement a echou6 dans 28 cas sur 30. Les resultats obtenus in vivo ont donc confirme les observations in vitro. Les deux sujets gueris par la chloroquine avaient probablement acquis un degr6 eleve d'immunite partielle avant 1'epreuve. Les resultats de cette experience font ressortir la n6cessite de mener d'autres etudes in vitro dans les zones impalud&es. La simplicite de la technique represente un -avantage de premier plan pour la d6termination de la presence ou de la prevalence de souches de P. falciparum resistantes a la chloroquine. REFERENCES Cohen, S., Butcher, G. A. & Crandall, R. B. (1969) Nature (Lond.), 223, 368 Comer, R. D., Young, M. D., Porter, J. A., Jr, Gauld, J. R. & Merritt, W. (1968) Amer. J. trop. Med. Hyg., 17, 795 Da Silva, J. R. & Lopes, P. F. A. (1966) Hospital (Rio de J.), 69, 967 DeGowin, R. L. & Powell, R. D. (1965) Amer. J. trop. Med. Hyg., 14, 519 Field, J. W., Sandosham, A. A. & Fong, Y. L. (1963) The microscopical diagnosis of human malaria, Kuala Lumpur, Economy Printers, p. 64 Manual of the microscopic diagnosis of malaria, 1963, Washington, D.C., Pan American Health Organization (Scientific Publication No. 87) Martin, D. C. & Arnold, J. D. (1967) Trans. roy. Soc. trop. Med. Hyg., 61, 331 Powell, R. D., Brewer, G. J., Alving, A. S. & Millar, J. W. (1964a) Bull. Wld Hlth Org., 30, 29 Powell, R. D., Brewer, G. J., DeGowin, R. L. & Alving, A. S. (1964b) Bull. Wld Hlth Org., 31, 379 Rieckmann, K. H., McNamara, J. V., Frischer, H., Stockart, T. A., Carson, P. E. & Powell, R. D. (1968) Amer. J. trop. Med. Hyg., 17, 661 Rieckmann, K. H., McNamara, J. V. & Powell, R. D. (1969) Milit. Med., 134 (special issue), p. 795 Rieckmann, K. H. (1970) Lancet, 1, 82 Walker, A. J. & L6pez Antufnano, F. J. (1968) Trans. roy. Soc. trop. Med. Hyg., 62, 654 WHO Expert Committee on Malaria (1968) Wld Hlth Org. techn. Rep. Ser. No. 382 WHO Scientific Group on Chemotherapy of Malaria (1967) WId Hith Org. techn. Rep. Ser. No. 375 Young, M. D. (1962) Trans. roy. Soc. trop. Med. Hyg., 56, 252