640

TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1991) 85, 64&643

A placebo-controlled double-blind trial for the treatment of Bancroftian filariasis with ivermectin or diethylcarbamazine

M. Sabry’, H. Gamal*, N. El-Masry’ and M. E. Kilpatrick’ I United States Naval Medical Research Unit No. 3 (NAMRU-3), Cairo, Egypt; *Abbassia Fever Hospital, Ministry of Health, Cairo, Egypt

Abstract Theraoeutic efficacv and clinical side-effects of

ivermectm (single dose of 100 pg/kg) and diethylcar- bamazine (DEC) (3 mgikg for one day, then 6 mgikg dailv for 12 d) were evaluated for microfilaricidal effect in Bancroftian filariasis. Seventy-one microfilar- aemic consenting adult male patients (2100 micro- filariae (mf)/ml)~were randomiy assigned to receive ivermectin. DEC or nlacebo and kent in hosuital for 15 d. Those receiving placebo were treated with ivermectin on day 9. Ivermectin (19 ‘double-blinded’ and 22 ‘unblinded’ patients) caused an abrupt reduc- tion in mf count to 1.5% of the pre-treatment level 12 h after drug administration and to 0.06% on day 14, with recrudescence to 1.8% after one month and to 9.2% after 3 months. DEC (30 patients) caused a gradual drop in mf count to 1.1% of the pre-treatment level on day 14, which increased to 2.4% after one and 3 months. The total scores of side-effects were 77 (l%), 305 (2.1%) and 311.5 (3.0%) for placebo, ivermectin and DEC respectively; the differences between DEC or ivermectin and placebo were statisti- cally significant. Ivermectin produced lower side- reaction scores than DEC and the differences were highly significant at the 95% confidence level. Side- effects were mainly headache and body aches in the ivermectin patients, which appeared as early as 4 h after drug administration, resolved within 36 to 48 hours, and were significantly related to mf densities. Side-effects in DEC patients were mainly testicular and epididymal pain and swelling, unrelated to mf densities, which began at day 2 and continued to day 7. Single dose ivermectin was more effective in reducing microfilaraemia in Bancroftian filariasis than a 13 d course of DEC for a period of about 2 months after treatment, and side effects were of shorter duration.

Introduction Lvmuhatic filariasis is endemic in trouical and

subtroiical countries with approximately 965 million people at risk of infection, and 90.2 million people already infected; 81.6 million with Wuchereria ban- crofti and 8.6 million with Brugia malayi and B. timori (WHO. 1984).

Diethylcardamazine (DEC), a piperazine deriva- tive, has remained the drug of choice for the past 40 years, but because of its side-reactions, mainly testicular and epididymal pain and swelling, many patients refuse treatment (SASA et al., 1963; SUNDAR- AM et al., 1974; HAWKING, 1979). Ivermectin, a

Requests for offprints: Research Publications Branch, US NAMRU-3, FPO New York 09527-1600, USA. Attn. Code 101

semi-synthetic macrocyclic lactone derived from Streptomyces avermitilis, was extremely successful in the treatment of onchocerciasis at 150 pg/kg single dose.

The objective of the present study was to evaluate efficacv and side-effects of ivermectin and DEC in a ‘double-blinded’, placebo-controlled trial infected with W. bancrofti.

Patients and Methods Patients

Eighty male Egyptians (aged 15-55 years) volun- teered to participate in the study. They were initially identified by surveys conducted in the Qulyubia and Dakahlyia governorates, 2 areas endemic for Bancrof- tian filariasis. All had microfilariae (mf) counts 215 per 20 mm3 of finger prick blood (2100 mfiml of venous blood) and were admitted to hospital for 15 d in Abbassia fever hospital, Cairo. None of the individuals was symptomatic before admission and all denied any antifilarial treatment during the past 2 years. Informed consent was obtained from each individual after the study was explained to them.

Treatment Ivermectin (100 pglkg single dose on day 1, then

placebo for 12 d), DEC (3 mgikg on day 1, then 6 mgikg daily for 12 d), and placebo were each prepared as identical capsules in separate 13 d coded treatment kits in dosages appropriate for body weight ranges. Patients were randomly and ‘blindly’ assigned to one of the 3 treatment groups. Since all study individuals were infected, those patients assigned to placebo were identified by an independent monitor on day 8 of therapy and treated with a single 100 pg/kg dose of ivermectin on day 9, without disclosing their identity to the associate investigators, and they were observed for efficacv of treatment and side-effects for 6 d. All medication was given before breakfast on days 1-13 and patients were observed to swallow the capsules.

Methods Blood samples (5 ml) were collected in ethylene-

diaminetetraacetic acid from all studv individuals between 2200 and 2400 h on days 0, 1; 2, 8 and 14; the placebo patients had samples taken on days 0, 1, 2, 8, 9, 10 and 14. After discharge similar samples were taken at 1 month and 3 months, and further follow-up samples will be taken at 69 and 12 months. Mf counts were made by filtering 1 ml aliquots through Nuclepore@ membrane filters, mounting the filter on a slide, staining with Giemsa stain solution diluted 1: 10 in phosphate buffer, and counting the mf at 100x magnification. Complete blood cell counts,

641

ivermectin on day 9). Side-effects of ivermectin treatment were also combined in these 2 groups. This was possible because days 0 and 8 were before treatment, and days 1, 2 and 9, 10 were after treatment. The results are shown in Table 1, which indicates that there was no significant difference in mf counts among the 3 treatment groups on day 0. On days A, 2 and 8, both the ivermectin and DEC groups had srgnificantly lower mf counts than the placebo group (P<O.OOl) and the ivermectin group mf counts were significantly lower than those in the DEC group (P<O.OOl). On day 14, the ivermectin group’s mf counts remained significantly lower than those in the DEC group (P<O.O5), but they were comparable at 1 month and DEC mf counts were significantly lower than those in the ivermectin group at 3 months (P<O.O05). Fig. 1 shows that the single dose of

determination of serum creatinine, alkaline phospha- tase, aspartate aminotransferase (SGOT), alanine aminotransferase (SGPT) and bilirubin, and urinaly- sis were performed on each individual on days 0, 8 and 14.

Side effects Patients were questioned and examined daily for

signs and symptoms of possible drug side effects. Any untoward effects were recorded on the individual’s clinical file and scored on a scale of 0 to 3 (nil, mild, moderate or severe). Blood pressure was recorded 3 times daily. Fever was scored on a scale of 1 to 3: 1, 37.6”C to 38.X; 2, 38.6 to 39.X; and 3, >39.6”C.

Statistical analysis Student’s t test was used to determine the signi-

ficance of differences between the means of 2 samples of unequal sizes (using their pooled variance) and the Wilcoxon test was used as a non-parametric alterna- tive (SNEDECOR & COCHRAN, 1967).

Results Seventy-five patients completed the 15 d in hospit-

al; 5 were discharged earlier, one for refusing to give a blood sample on day 1, one because he vomited on day 1; and 3 for family reasons. Another 4 patients were excluded from the analysis of drug efficacy and side-effects because their admission blood specimen (on day 0) contained no mf, although they had been positive in the survey before admission and all denied having received any antifilarial drug before admission to hospital. Of the 71 patients evaluated, 30 received DEC, 19 received ivermectin, and 22 received placebo until day 8 and ivermectin on day 9. At the one and 3 month follow-up examinations 69 and 68 individuals were evaluated respectively.

Mf counts from days 0, 1 and 2 for the ivermectin group were combined with the days 8,9 and 10 counts of the placebo group (who were then treated with

2 8 14Day 1 Month aMonth

TIME Fig. 1. Microfilariae counts per ml of venous blood of the three study groups (placebo: upper broken line; diethylcarbamazine, solid line; ivermectin, lower broken line), expressed as a percentage of the pre-treatment counts, for the period in hospital (1-14 days) and at the one month and three month follow-up examinations.

Table 1. Microfilariae counts of the three study groups during the period in hospital and at the one month and three month follow-up examinationsa’b

Placebo Ivermectin DEC n=22 ?I=41 n=30

CX (SEX)

CX CX (*we) Percentage (*awe) (S& Percentage (*we) (S& Percentage

Day 0’ 27228 1237.64 100 42527 1037.24 100 27187 906.23 100 (197-4044) (217.32) (153-4447) (153.37) (87-2667) (130.21)

Day 1 28635 1301.59 105.17 632 1542 1.49 4776 159.2 17.57 (411d154) (197.86) (G99) (3.27) (l-746 ) (33.34)

Day 2 24914 1132.46 91.50 1.66 0.16 1514 SO.47 5.57 (163-3402) (180.79) (02; (0.34) (6323) (12.09)

Day 8 27354 1243.36 100.46 1.00 0.10 387 12.9 1.42 (2274447) (249.52)

(CL!; (0.47) (G69) (3.49)

Day 14 NA - (CA) 0.58 0.06 311 10.37 1.14 (0.30) (O-91) (3.39)

1 month NA - - 755 18.88 1.82 635 21.90 2.42 (@54) (3.81) (O-154) (7.19)

3 month NA - 3797 94.93 9.15 605 21.61 2.38 (tK663) (17.27) (&163) (6.31)

“n=number of patients, mf=microfilariae, Zx=sum of mf counts, I=mean mf counts, SEH=standard error of I, NA=not applicable. %atistical significance: placebo vs ivermectin and vs DEC, P<O,OOl; ivermectin YS DEC at days 1,2 and 8, P<O,OOl; day 14, PtO-05; 1 month, P>O.O5; 3 months, P<O,OO5.

‘Pre-study values.

642

Table 2. Clinical reaction scoresa during the period in hospital for the three study groupsb

Total reaction score Score ratio (%)

Sign/ symptom

Placebo Ivermectin n=22 n=41

77 305 1 2.1

Population Daily Population Daily affected reaction affected reaction

(%I scores (%) (%) scores (%)

DEC fl=30 311.5

2.97 Population Daily

affected reaction W) scores (%)

Fever Hypotension Headache Nausea Vomiting Dry/bitter tongue Anorexia Diarrhoea Weakness Fa;y urination

Pain Chest

4.6 0.6 56.1 1.0 10

2?3 ST7 6i.z 23.6 0.6 30 3.3 ;:2” 6.9 4.6 1.1 12.2 1.7 10 1.7

;:; A:; 19.5 9.8 0.7 4.7 3.3 3.3 0.2 0.5 - 12.2 4.5 10

1;6 2:3 12 1.8 9.1 5.7 12.2 ;:;

1t: ;:; 2.6 6.7 1.7 3.3 1.4

0.6 1.0 Lower ribs pbigzgt;m

;:; Oy6 ;:;

3.5 2.0 g:; ;:;

;:; 4.9

Lower back 4.6 12.2 ;:; ;:; ;:; Body 22.7 4.0 34.2 14.1 Joints Testes ;6 lT7 2-4

:i ::;

Epididymes 2.4 ;:i 36.7 12.3 13.3 6.4

Swelling Inguinal nodes

266.: 4.1

Testes 10.8 Epididymes 20 9.3

“Clinical reactions are scored on a scale of 0 to 3 (nil, mild, moderate or severe), and fever on a scale of 1 to 3 (1,37.6”C to 38.5”C; 2, 38.6”C to 39.5”C; 3, >39.6”C).

bStatistical significance: placebo vs DEC on days 3, 5 and 7, PcO.05; placebo vs ivermectin on days 1 and 2, P<O.OOl; days 3 and 7, P<O.O5; ivermectin vs DEC on day 1, P<O.OOl; day 2, P<O.O05; day 4, P<O.O5; days 5 and 6, P<O.Ol.

H w ~ n

. I. B

: o w T 7 FWB

I

A 1 F

v, s

PLACEBO IVERMECTIN DEC

Fig. 2. Main reaction scores (percent per day) for each of the three study groups during the period in hospital. B, body aches; F, fever; H, headache; T, testicular and epididymal pain and swelling; W, weakness.

ivermectin caused an abrupt clearance of circulating mf while the 13 day regimen of DEC produced a slower drop in mf count. Recrudescence at 3 months was less in the DEC group.

The side-effects of treatment are presented in Table 2. The major side-effects in patients treated with DEC were pain and swelling of testes and epididymes, in approximately 50% of patients. Side-effects after DEC treatment occurred during the first week of

therapy. In patients treated with ivermectin, side- effects were comparable in the 19 patients treated in the ‘double-blind’ group and in the 22 in the ‘open’ group. The side-effects were headache, body aches, weakness and fever (37.8-38.7”C), occurring as early as 4 h after treatment and resolving in 36-48 h. Mild, self-resolving hypotension was noted in one DEC patient and 2 ivermectin patients; one DEC patient developed a prilritic rash on the penis, right side of the neck and the flexor areas of the right elbow and knee, lasting from day 2 until day 7. The statistical significance of the difference in reactions between the 3 study groups (Student’s t test) are presented in Table 2. The non-parametric Wilcoxon test also indicated highly significant (.z,=2.07) differ- ences between the DEC and ivermectin side-effects at the 95% confidence level. Fig. 2 shows the major side-effects expressed as percentage per day.

When reaction scores were analysed according to blood mf densities, the side-effects scores in the ivermectin-treated low-density group (100-1000 mfi ml) were significantly lower than those in the high- density group (1000-5000 mfiml) (ts9=6,3961, P<O.OOl). The differences in reaction scores between the low and high mf density groups receiving DEC were not significant (t*,=O.42298, E-0.6).

There were no significant changes in blood cell counts, blood chemistry or urine studies in the 3 groups during the period in hospital. Eosinophil counts did not change during or after treatment.

643

Discussion Ivermectin appears to be the only antifilarial drug

to consider for widespread general use. This study revealed 2 benefits of ivermectin treatment compared to DEC. First, it is given as a single tablet for a 60 kg person compared to 90 DEC tablets to be taken over 13 d. Second, ivermectin reactions were those of an influenza-like syndrome and lasted only 36-48 h, while DEC caused genital pain and swelling lasting for one or more weeks.

The cause of the side-effects could be pharmacolo- gical reactions of the drugs or the host reacting to antigens from killed mf. Prior studies have suggested that DEC may cause histamine release from lung mast cells (DELINE et al., 1973), initiate the release of the slow-reacting substance of anaphylaxis (SRS-A) from mast cells (ORANGE et al., 1971), neutralize the effects of histamine, once the latter has been released (HARADA er al., 1971); or initiate the development of eosinophilia (THEVATHASAN & LITT, 1971). This study, comparing ivermectin, DEC and placebo, suggested that the host response to mf destruction, rather than drug or metabolite reactions, was the major cause of side-effects because ivermectin, which kilis mf quickly, produced more side-effects at higher mf densitv. while DEC. which kills mf more slowlv, had delayed side-effects not related to mf density:

The recrudescence between one and 3 months after ivermectin treatment suggests a novel endemic-com- munity treatment approach with this single-dose drug. Culex pipiens molestus is the main vector of filariasis in Egypt and human carriers with low mf densities are not epidemiologically important (SABRY, 1987). Two ivermectin doses one month apart to microfilaraemic individuals would most probably result in a sustained low level of tnicrofilaraemia, thus removing these infected individuals from the trans- mission cycle.

Acknowledgements We arc indebted to the Ministry of Health officials, Dr A.

Khazindar (Under-Secretary of Health for endemic dis- eases), Dr M. Harb (Director-General, filariasis and leish- maniasis), Dr A. Khater (Under-Secretary of Health, Qalyubia governorate), Dr M. Safwat (Head, Banha Health Directorate) and Dr E. Bishav (Head. Abbassia Fever Hospital), for their invaluable” administrative assistance. Thanks are also due to all patients who participated in the study. Funding and drugs were generously supplied by Merck Share and Dohme Research Laboratories.

and Development Command, Bureau of Medicine, Bethes- da, Maryland, USA, Work Unit No. 3M162770A870.AR. 322. The opinions and assertions contained herein are the private ones of the authors and are not to be construed as official or as reflecting the views of the Navy Department, Department of Defense, or the US Government.

References Deline, T. R., Eyre, I’. & Wells, I’. W. (1973). Histamine

releasing properties of the anthelmintic l-diethylcar- bamyl-4-methyl-piperazine citrate (diethylcarbamazine). International Archives of Pharmacodynamic Therapeutics, 205, 192-198.

Harada, M., Takeuchi, M. & Katagiri, K. (1971). Charac- terization of various types of rat cutaneous anaphylaxis from their drug-susceptibilities. Dermacologica, 142, 193-201.

Hawking, F. (1979). Diethylcarbamazine and new com- pounds for the treatment of filariasis. Advances in Pharmacology and Chemotherapy, 16, 129-194.

Orange, I’. R., Austen, W. G. & Austen, K. F. (1971). Immunological release of histamine and slow-reacting substance of anaphylaxis from human lung. 1. Modula- tion by agents influencing cellular levels of cyclic 3’,5’-adenosine monophosphate. Journal of Experimental Medicine, 134, supplement, 136-148.

Sabry, M. (1987). A study of the quantitative relationship between the blood densiry of Wuchereria bancrofti micro- filariae and development of infective larvae in Culex pipiens molestus in Egypt. PhD Thesis, Faculty of Medicine, University of London.

Sasa, M., Oshima, T., Sato, K., Mitsui, G., Sugata, F., Nishi, S., Yamamotq, H., Tada, I. & Motoi, E. (1963). Studies on epidemtology and control of filariasis: observations on the carriers of Wuchereria bancrofti in the Amami Islands with soecial reference to the effects and side reactions of diethylcarbamazine. JapaneseJournal of Experimental Medicine, 33, 213-243.

Snedecor, G. W. & Cochran, W. G. (1967). Statistical Methods. 6th edition. Ames, Iowa: The Iowa State University Press.

Sundaram, R. M., Koteswara, Rao, N. & Krishna Rao, C. (1974). Studies on bancroftian filariasis control with diethylcarbamazine: I. Frequency and nature of drug reactions. Journal of Communicable Diseases, 6, 290-300.

Thevathasan, 0. I. & Litt, M. (1971). Inhibition by diethylcarbamazine (Hetrazan) of the eosinophil re- sponse which trichina extract elicits in guinea-pig lymph nodes. Clinical and Experimental Immunology, 9, 657- 665.

WHO (1984). Lymphaticfilariasis. Fourth Report of the WHO Expert Committee on Filariasis. Geneva: World Health Organization, Technical Report Series, no. 702.

Received 1 November 1990; revised 19 March 1991; accepted for publication 21 March 1991 This work’ was supported by the Naval Medical Research