Experimental chemotherapy in leprosy*

Experimental chemotherapy in leprosy * The Memorandum reviews the considerable progress that has been made in research on the chemotherapy of leprosy during the last 10-15 years, as a result of which it is now possible to study the same topics in leprosy as are studied in other bacterial diseases. Thus drugs have been screened in mice for their activity against Mycobacterium leprae. Those that have been found to have the greatest activity against M. leprae at acceptable dosages-dapsone, rifampicin, and clofazimine-have been characterized in terms of the minimal effective dosage and rate of bacterial kill. Similarly, their pharmacokinetics in man and in certain animals have been defined. The theoretical basis for drug trials in leprosy patients is discussed in terms of the number of viable and the number of dead M. leprae that remain at various stages of therapy. Much progress has been made in research on the chemotherapy of leprosy in the last 10-15 years. Previously, drugs against Mycobacterium leprae could be studied only in clinical trials and the concepts behind the trials were often ill defined or ambiguous. In spite of recent developments, research in this field is still hampered because M. leprae cannot be grown in vitro, the growth of the organism in animals is slow, and the course of the disease in man is protracted. Nevertheless, the recent progress has made it possible to study the same topics in leprosy as are studied in other bacterial diseases. These topics are: (a) laboratory screening of new drugs against M. leprae and determination of the minimal effective dosage; (b) laboratory characterization of a drug's anti- leprosy activity: bacteriostatic or " bactericidal- type" (defined under Methods of study, page 426); (c) measurement of drug concentrations in blood or tissue, pharmacokinetics (including repository effect), and determination of minimal inhibitory concentration; (d) toxicity of drug at the minimal inhibitory concentration and at therapeutically effective levels; (e) metabolism of drug: chief metabolites in blood and tissue; * This Memorandum was drafted by the signatories listed on page 432, who were the members of the Committee on Experimental Chemotherapy held at the Xth International Leprosy Congress, Bergen, Norway, 13-18 August 1973. Requests for reprints should be addressed to: Dr Charles C. Shepard, Center for Disease Control, Atlanta, GA 30333, USA. A French translation of the Memorandum will be published in a future issue of the Bulletin. (f) short-term clinical trials in lepromatous leprosy to determine whether the drug is also active in man and, if so, how rapidly M. leprae are killed during the early stages of treatment; (g) long-term clinical trials to determine whether the drug's activity continues until the patients are smear-negative; (h) very-long-term clinical trials to determine whether smear negativity is maintained or whether relapse occurs because of the emergence of drug- resistant M. leprae; and (i) trials in paucibacillary (nonlepromatous) lep- rosy. Assessment of drug activity and the required duration of treatment. Short cuts should not be taken and all the steps listed above should be followed in the development of antileprosy drugs. Patients should not be deprived of standard dapsone therapy in order to evaluate compounds that have not been tested against M. leprae in animals, that appear on the basis of results in animals to be clearly less efficacious than standard therapy is, or that, at their minimal inhibitory concentrations, cause toxicity in man. DRUG SCREENING AND CHARACTERIZATION OF ANTILEPROSY ACTION Experimental models In the absence of significant growth of M. leprae in vitro, all work must be done in animals. This is not as severe a handicap as might appear at first sight, because much of the information that must be obtained in animals (toxicity, dosage, absorption of the drug from the gut and blood, and 3455 - 425 - BULL. WORLD HEALTH ORGAN., Vol. 53, 1976 426 MEMORANDUM tissue levels of the drug) is relevant to man. The most research has been done in the mouse model. The infection is very consistent in these animals, and genetically uniform mice are readily available and easily maintained under standard conditions. Hence, the mouse continues to be the animal of choice for research in leprosy. Other animals are useful when particular findings must be checked in another species. For studies requiring larger numbers of M. leprae, the thymectomized and irradiated mouse (34), the neonatally thymectomized rat (9, 10, 11), and the armadillo (22) may provide suitable animal models. Methods of study The continuous method (administration of the drug from the day of infection to the end of the experiment) reveals whether a drug is active against M. leprae (47). The kinetic method (administration of the drug during a limited period, beginning late in the lag phase or early in the logarithmic phase of growth of bacilli) reveals whether a drug is active and, in addition, determines whether active drugs produce a bacteriostatic or a bactericidal- type effect a (45). Administration of drugs in a graduated dosage allows the minimal effective dosage and minimal inhibitory concentration to be determined. a In the kinetic test, a treatment may be described as bacteriostatic if multiplication of M. leprae is suspended only while the drug is present and begins again immediately after the drug has been eliminated from the mouse. Also, a treatment may be termed bactericidal if no subsequent bacterial growth is observed. When bacterial growth appears, but it begins significantly later than the time when the drug was eliminated from the mouse, the treatment may theoretically have been (a) bactericidal for a major fraction of the bacteria, or (b) bacteriopausal. The latter term refers to the phenomenon of temporary failure of bacteria to grow after the removal of the drug; it has been described for other organisms and termed " bacteriopause " by Videau (57), " slow recovery " by Eagle & Musselman (5), and " delay " or " lag period " by Dickinson & Mitchison (3). Moreover, with some drugs, available pharmacokinetic information may not allow one to exclude (c) a repository effect as the cause of the extended growth delay. In pre- liminary investigations, it has been convenient to term all instances of extended growth delay as " bactericidal-type ". In practice, differentiation of these three causes of extended growth delay has been accomplished in short-term trials in lepromatous patients, in which mouse inoculations were performed and morphological indices were measured. However, opinions differ with regard to interpretation. The results with dapsone and with clofazimine are sometimes interpreted as indicating a slow bactericidal effect; the results with rifampicin, as indicating a rapid bactericidal effect. According to another interpretation, the results with dapsone and with clofazimine indicate a bacteriopausal effect of the drug plus a slow bactericidal effect of host immune factors, the results with rifampicin being considered as bactericidal. Results of drug screening and characterization With these methods, more than 200 drugs have been screened. Only a few have exhibited bactericidal- type activity, and these appear to include most of the drugs of real promise in leprosy, which are: (a) dapsone and other sulfones yielding dapsone in the gut or in the tissues (as a result of hydrolysis or metabolism); (b) rifampicin (the related antibiotic strepto- varycin is distinctly less active); (c) clofazimine (B663) and another phenazine dye, B1912 (3-anilino-7-chloro-10-phenyl-2,10-di- hydro-2-cyclohexyliminophenazine); (d) long-acting sulfonamides (the minimal in- hibitory concentrations of these compounds appear to be close to toxic blood levels); and (e) ethionamide (in the dosage apparently required in man, gastrointestinal distress is frequent). Minimal effective dosage and minimal inhibitory concentration of dapsone. In the mouse, the minimal effective dosage of dapsone is 1 mg/kg in the diet- the dosage to which all strains of M. leprae from un- treated patients have been sensitive. Only a fraction of such strains are sensitive to a 0.1 mg/kg dose of dapsone (36, 48, 50). Hence the minimal inhibitory concentration of dapsone for M. leprae lies between 0.01 and 0.001 mg/l-the blood levels produced by these two dosages (7,26). Peters et al. (30) found that in rats the minimal inhibitory concentration of dapsone against one strain of M. leprae was 0.002- 0.004 mg/l-the plasma concentrations produced by the minimal effective dosage of the drug. Gordon et al. (16) have shown recently that constant plasma levels for 7-21 days were produced in both mice and rats by feeding them on diets containing dapsone. Furthermore, plasma levels of the drug were found to be directly related to dietary dapsone concentrations ranging from 0.1 mg/kg to 50 mg/kg. Tissue levels were determined in both species receiving relatively high levels of dietary dapsone. Tissue-to-plasma ratios of the drug in mice ranged from 0.6: 1 to 0.9: 1 for brain, muscle, and testes and from 1.3:1 to 1.6:1 for the ears, foot-pads, skin, and fat. The highest ratio found was 3.5:1 for the liver. Mouse tissues exhibited low levels of the metabolite, monoacetyl-dapsone, as did plasma. In rats, tissue-to-plasma ratios were 0.7: 1 to 1.0: 1 for muscle and testes, respectively, and ranged from 1.2: 1 to 2.2: 1 for the ears, foot- pads, nose, and skin. Tissue-to-plasma ratios of LEPROSY monoacetyl-dapsone in rats reflected the ratios of dapsone in the same tissues. These results in- dicated that monoacetyl-dapsone was a metabolite of dapsone in tissue and that tissue levels of dapsone were, in most cases, somewhat higher than con- currently determined plasma levels. The finding of this very low minimal inhibitory concentration in mice led to the introduction of treatment with acedapsone (48)-a drug that, following injections of the standard dosage of 225 mg every 75 days, releases dapsone at an average rate of 2.5 mg daily and produces plasma levels averaging 0.050 mg of dapsone per litre (26). Later work with a more specific and sensitive technique showed that the level varies from an average of 0.030 mg/l just before the next injection (42) to a level about twice as high, at the maximum about 4 weeks after an injection (J. H. Peters, unpublished observation, 1975). Acedapsone has been found therapeutically active in a trial that has lasted 6 years (43). Confirmation that the minimal inhibitory concentration is low in man also came from a short-term trial showing that dapsone in an oral daily dose of I mg was therapeutically active (7, 58). This dosage produces dapsone serum levels of 0.01-0.02 mg/l. The plasma half-life of the drug in man averages about 24 hours; hence, with constant daily dosage, plasma levels vary only about two-fold during that period. Bactericidal effect of rifampicin. The first work on the therapeutic activity of rifampicin against M. leprae found this drug to be more rapidly bactericidal in man than dapsone is, on the basis of observations of the morphological index and preliminary results of mouse inoculation (37). Later work with the kinetic method in mice and with clinical trials involving serial inoculations of mice showed the bactericidal action of rifampicin to be dramatically rapid (20, 52). The drug was found to produce as much bactericidal effect in the first few days as dapsone did in the first few months. The minimal inhibitory concentration appears to be about 0.3 mg/l in mice receiving rifampicin in the diet (18, 19). However, the appli- cability of this finding to human dosage is compli- cated because the concentration of rifampicin in the blood of patients on daily dosage varies over a range in excess of 50-fold in the 24-h period. The minimal effective dosage in mice on daily administration by stomach tube is as low as 1.4 mg per kg of body weight (53). Clofazimine. Determination of the minimal inhibi- tory concentration of clofazimine has not been possi- ble because the drug is very unevenly distributed in the tissues; thus, neither blood nor tissue levels necess- arily reflect the concentration of drug in the immedi- ate environment of the M. leprae. Likewise, because the drug is deposited in the tissues and is released only very slowly, kinetic studies in mice cannot be used to determine the type of drug action. Short- term trials in man, however, have shown that, during treatment with clofazimine, M. leprae are killed at a rate similar to that obtained when dapsone is administered, as determined by measurements of the morphological index (32), although somewhat slower, but equally extensive, killing is shown by mouse inoculation (23). Long-acting sulfonamides. Of these compounds, only sulfadimethoxine has been studied by the kinetic method, and it was found to have bacteri- cidal-type activity similar to that of dapsone (C. C. Shepard, unpublished data, 1972). Ellard et al. (6) found that the minimal inhibitory con- centrations of sulfadimethoxine and sulfadoxine were only one-fourth to one-fifteenth of the blood levels achieved in man with normal dosages, so the chances for emergence of drug-resistant forms would seem high. Cross-resistance with dapsone would be expected, since dapsone-resistant strains from patients treated with this drug are resistant to long-acting sulfonamides (1, 35). Ethionamide. By the kinetic method, this drug was found to have bactericidal-type activity at a dietary level of 1 g/kg, bacteriostatic activity at 100 mg/kg, and no activity at 10 mg/kg (46). In a short-term trial monitored by mouse inoculation, two patients were treated with a dosage of 250 mg three times a day; the treatment appeared to be as effective as dapsone was in a dosage of 50 mg daily (L. Levy et al., unpublished data, 1971). Demonstration of drug-resistant M. leprae. Mouse inoculation provides the only widely accepted method of proving drug resistance. Thus, from many patients who have relapsed on prolonged dapsone therapy, strains of M. leprae have been isolated that multiply in mice fed with dosages of dapsone in excess of the minimal effective dosage for strains from untreated patients (35, 50). By contrast, dapsone-sensitive organisms have been isolated from patients who relapsed because they had stopped taking the drug, either surreptitiously 427 MEMORANDUM or by direction. A study undertaken in Malaysia showed that the proportion of lepromatous patients who eventually relapsed owing to the presence of dapsone-resistant M. leprae was 7.8% of those beginning treatment with solasulfone and 2.5% of those beginning with dapsone (25). Similarly, a recent study in Costa Rica has shown that 3.2% of the lepromatous patients who had begun treatment 22 or more years earlier, and 5.3%o of those still living, had relapsed owing to M. leprae confirmed to be dapsone-resistant (C. C. Shepard et al., unpublished studies, 1973). The Costa Rican patients had begun treatment chiefly with glucosulfone, aldesulfone sodium, and sulfadiasulfone sodium. Gelber et al. (15) have shown that the administration of solasulfone by injection resulted in plasma levels of dapsone between 40 and 90 ,ug/l, so that treatment with solasulfone was, in effect, treatment with dapsone in low dosage. Other sulfone regimens yielding equally low dapsone plasma levels, such as low-dose dapsone and acedapsone, are similarly liable to result in a high level of dapsone resistance. The earliest cases of relapse due to sulfone- resistant M. leprae have occurred after 5 years of therapy, but cases have continued to occur even after more than 20 years of sulfone treatment. Relapse due to thiambutosine-resistant M. leprae occurs commonly after 2 or 3 years of treatment with that drug, and strains of M. leprae resistant to thiambutosine (and to thioacetazone) have been isolated 2 or more years after the start of therapy [Rees (35, 36) and unpublished data, 1975]. Treat- ment of all lepromatous patients with a combination of effective antileprosy drugs appears to offer the most promise for the prevention of relapses. PHARMACOKINETICS AND METABOLISM OF DRUGS Comparative studies of the pharmacology of drugs in mice and in man are necessary for under- standing their antileprosy action and for determining the most favourable dosage and timing of medication. Dapsone In man, there are large individual differences in the rate of dapsone clearance from the body. The half-life of the drug in plasma varies from about 10 to about 50 hours and averages about 28 hours (13, 29, 31). Results of repeat tests (26 months apart) in the same subjects showed that this characteristic was stable. In patients of mixed race who harboured M. leprae that were confirmed to be dapsone-resistant, the drug was observed to exhibit, on the average, a significantly lower half-life than it did in all other single population groups (31), suggesting a relationship between the rapid clearance of dapsone and dapsone resistance. However, in subsequent studies carried out in more uniform populations-in Chinese by Gelber and Rees (13) and in Costa Ricans by J. H. Peters et al. (unpub- lished observations, 1975)-no significant differences in the mean half-life of dapsone were found between patients with dapsone-sensitive M. leprae and those with dapsone-resistant M. keprae. Blood levels of dapsone can be calculated on the basis of the individual's dapsone half-life, as shown in Table 1. For example, 50 mg of dapsone daily would ensure blood levels continuously well in excess of the minimal inhibitory concentration, even for a patient with a dapsone half-life of 12 h, whereas 350 mg once weekly (the same total dosage) would not. A dosage of 50 mg twice weekly would provide blood levels continuously in excess of the minimal inhibitory concentration for many patients but not for those with a dapsone half-life of 12 h or less. Since it is not practicable to determine the half-life of dapsone for all patients, regimens need Table 1. Estimated plasma concentrations of dapsone after ingestion of 100 mg of the drug by a 60-kg man, according to various dapsone half-lives Plasma concentration (mg/I) with a dapsone half-life of: 12 h 24 h 48 h Maximum a 1.7 1.7 1.7 1 day later b 0.42 0.85 1.2 2 days later 0.11 0.42 0.85 3 days later 0.027 0.212 0.60 4 days later 0.0066 0.106 0.42 7 days later 0.0001 0.012 0.15 a At 2-6 h, based on equilibration of a 1 00-mg dose in a 60-kg human being. Plasma concentrations after a 50-mg dose would be one-half of the tabulated values; after a 25-mg dose, one-fourth of the tabulated values, etc. Similarly, for a person weighing 70 kg, the plasma concentrations would be approximately 6/7 of the tabulated values, and for one weighing 50 kg, 6/5 of the tabulated values. Thus, J. H. Peters (unpublished data, 1973) found a group of adult Phi- lippine patients to have an average plasma concentration of 0.82 mg/l 6 hours after a dose of 50 mg. b The plasma concentrations at various times after the initia equilibration are calculated from the formula CT = CO/2n, where CT = the concentration at time T, CO = the original concentration at equilibration, and n = the number of half-lives that have elapsed. 428 LEPROSY to be designed for patients with a short dapsone half-life. The only metabolite of dapsone found in human blood is monacetyl-dapsone. Individuals have been found to be genetically polymorphic in their capacity to acetylate dapsone, and they may be divided into rapid and slow acetylators (14, 29). Rapid acetylators have higher monoacetyl-dapsone: dapsone ratios in their plasma, but do not eliminate dapsone more rapidly. Therefore, a priori, one would not expect the acetylator status to affect the response of leprosy patients to dapsone. In a study of persons of various races, Peters et al. (31) found an apparent excess of rapid acetylators among patients with dapsone-resistant infections-a finding that suggested a relationship between acetylator phenotype and dapsone resistance. However, when studies were carried out in more uniform populations -the studies cited on page 428 and that by Ellard et al. (8)-no significant difference in the distribution of phenotypes was found between patients with dapsone-resistant M. keprae and those with dapsone- sensitive M. leprae. Rifampicin Pharmacokinetic studies of rifampicin are compli- cated because the rate of elimination increases with decreasing drug concentration in the blood and because blood levels tend to be lower after the patient has been receiving the drug for several weeks (12). These observations and the rapid bactericidal effect of rifampicin suggested that large doses given intermittently might be highly effective. Studies with M. leprae in mice, however, failed to reveal that intermittent dosage was any more effective than daily dosage with the same total amount of drug (53). Intermittent schedules have given favourable results in tuberculosis (2, 17, 56). Unfortunately, potentially serious side- effects caused by the formation of antibodies to rifampicin were encountered, especially with higher doses and more widely spaced administration (2, 33). To minimize these side-effects, a trial in leprosy is being undertaken with intermediate doses (900 mg) once a week for 3 months (27); in other trials, a stan- dard daily dosage (600 mg) is being given in short initial courses combined with longer courses of dap- sone or acedapsone. Results from these trials are not yet available. To be practical in leprosy therapy, chemo- therapeutic regimens should not require prolonged institutionalization or expensive drugs. Clofazimine Pharmacokinetic studies are hampered by the tendency of this drug to accumulate in the tissues. When administered orally, the drug has repository activity against M. leprae in mice (51). In a short- term trial in man, the same total amount of drug was administered every 4 weeks as one dose or divided into smaller doses on more frequent sched- ules (54). The latter were somewhat more effective. TRIALS IN MAN Background Number of bacilli in the body. The application of experimental chemotherapeutic findings to man has been inadequately understood. For better understanding, the bacterial populations have to be considered (Table 2). In the table, an average lepromatous patient is considered at various times during treatment with dapsone, and various typical laboratory findings are interpreted in terms of bacterial population. In line 1, a lepromatous patient with a bacterial index (BI) of 4+ (38) and a morphological index (MI) of 1000 is seen to have an estimated 1011 M. leprae in his body, of which 1010 are viable (44). After he has had dapsone treatment for 1-3 months (line 2), the bacterial index has not changed appreciably, but the mor- phological index has decreased to 100; consequently, he has the same total number of M. leprae, but the number of viable organisms has decreased to 109. Mouse inoculations were not performed in the example on line 2, but if they had been done they would have been positive. Line 3 shows a situation in which the morphological index has decreased to less than 100; the number of viable M. keprae may then be put at less than 109. In line 4, the mouse inoculations are considered to have been performed with a weakly positive result, indicating that the number of viable M. leprae has decreased to bare detectability-that is, about 100 of the original number; the resultant interpretation is that 108 viable M. leprae remain. In line 5, the mouse inoculation result is considered to be negative, so the number of viable M. leprae is less than 108. In successive lines, situations are considered in which treatment has been continued long enough to result in a decrease in the bacterial index; with each decrease by one unit, the corresponding total number of bacilli falls to one-tenth of the preceding value (decrease by one exponent). When the bacterial index is less than 2+, measurement of the mor- 5 429 MEMORANDUM Table 2. Estimated number of M. Ieprae in a typical lepromatous patient at various times during response to regular dapsone therapy Findings Interpretation Period of treatment total viabletreamentB I M IaMouse M. M.inoc.b Ieprae leprae 1. untreated 4+ 10% pos. 1011 10 10 2. 1-3 months 4+ 1% n.d. c 1011 10 9 3. 1-3months 4+ <1% n.d. 1lo1 <1O9d 4. 1-3 months 4+ <1% (pos.) e 1011 10 8 5. 1-3months 4+ <1% neg. 1011 <108 6. ca.l1year 3+ <1% n.d. 1010 <108 7. ca.1 year 3+ <1% neg. 1010<107 8. ca.2 years 2+ <1% neg. 10i <107 9. ca. 2 years 2+ <1% neg. 109 <108 10. ca. 3 years 1+ n.p.f n.p. 2 108 <107 11. ca. 5 years 0 n.p. n.p. <107 <106h 12. required for cure 0? a When carried out according to the specifications described for 'solid ratios - so that the proportion of - solid - bacilli closely approaches the proportion of bacilli infective for mice. b When 1 x 103 to 1 x 104 bacilli are inoculated. c Not done. It For example, <10 9 means that the number may lie between 0 and 10 a (inclusive). e Weakly positive (long incubation period and irregular results in mice), indicating that the number of viable M. lepree is near the limit of detectability. f Not possible. g Not possible to recover enough M. leprae to inoculate the numbers indicated in footnote b. Of course, mice may be inoculated with numbers smaller than 1 x 10 3. With such low inocula, negative results in mice do not necessarily signify that the proportion of viable bacilli is smaller than it was before the start of treatment; positive results, on the other hand, would reveal the presence of viable M. lepree. h Since the Ml cannot be determined with a BI of less than 2 + the estimate of viable M. leprae is based on the supposition that not more than 10 % of the total are viable. This does not imply an increased number of viable M. lepree. phological index is not practicable and mice cannot be inoculated with the indicated number of bacilli. Consequently, it is not technically possible, with present procedures, to estimate the number of viable M. leprae present in the body at any number less than 107. These considerations allow one to understand how there can be many viable bacilli present in the body if treatment is stopped after the mor- phological index has reached baseline values and infectivity for mice can no longer be demonstrated. To explain relapse in such a patient, it is clearly not necessary to assume that nonsolid bacilli have become viable. Similarly, in a patient with negative smears, it is not necessary to assume that non-acid- fast viable forms of M. leprae exist, since there could be as many as 105 typical viable, but un- detected, M. leprae in the body. Mechanisms of bacillary survival during treatment. The survival of M. keprae during treatment appears to occur by two mechanisms. One, which is not unusual with certain other drug-bacteria combina- tions (4), is the survival of a fraction of drug- sensitive bacilli despite the continuing presence of the drug in concentrations well in excess of the minimal inhibitory concentration. Such bacilli do not multiply and, presumably because they are dormant or metabolically inactive, they remain insensitive to the drug until they resume normal metabolism. Moreover, the location of the bacilli may be important, and some believe that the location of M. leprae in nerve or muscle favours their survival (28, 59). Because of the large numbers of bacilli originally present in lepromatous disease, factors affecting even a very small fraction of the M. keprae become important. If adequate treatment is continued without interruption, the number of surviving bacilli presumably decreases slowly. In general, the clinical picture associated with this type of bacillary survival is (a) relapse when treatment is stopped, or (b) cessation of improvement in spite of continued treatment when the number of viable bacilli reaches a sufficient proportion of the total number of bacilli-i.e., one sufficient to maintain the total number of bacilli at a constant level. Evidence for the survival of dapsone-susceptible M. leprae has recently been obtained in two studies, one of which concerned patients treated for 3-5 years with acedapsone (24, 43) and the other, patients treated for at least 10 years with dapsone (59). The second, apparently unrelated, mechanism of survival of M. leprae is drug resistance. A small proportion of bacilli, genetically insensitive to the drug, multiply in its presence. Again, because large numbers of bacilli are present in lepromatous leprosy, a small proportion of resistant bacilli may constitute a large absolute number. The clinical picture associated with this type of bacillary survival is the worsening of existing lesions or appearance of new lesions in spite of treatment. Multiple drug therapy in the initial stages, which is essential in the treatment of tuberculosis if 430 LEPROSY relapse due to drug-resistant organisms is to be avoided, is probably important in lepromatous leprosy also. Clinical trials In the treatment of leprosy, distinct differences exist between (a) the rate of loss of viability (mea- surements of the morphological index and mouse inoculations) and (b) the rate of disappearance of acid-fast bacilli (measurements of the bacterial index). Nearly all drugs that have been tried in leprosy have been selected on the basis of their ability to carry out process (a), and such drugs do not affect process (b). During early treatment, (a) is much faster than (b). Except for special studies in relapsed patients, all trials should be carried out in previously un- treated patients. Otherwise, it is difficult to disen- tangle the effect of the previous treatment from that of the test treatment. Furthermore, trials are best limited to lepromatous patients (with the exception noted below under Trials in paucibacillary leprosy). Some trials have been limited to lepromatous (LL) patients (41). Others have involved all patients with plentiful bacilli in the skin, which, according to the Ridley-Jopling scale, has meant the inclusion ofLL, BL/LL (borderline-lepromatous/lepromatous), and most BL patients. Whichever practice is followed, it is essential that the patients be accurately placed in the Ridley-Jopling classification since, on effective treatment, BL patients show an earlier and faster decrease in the bacterial index than LL patients do (39, 55). Short-term trials. These are carried out to confirm whether a drug's activity against M. leprae in the mouse is also observed in man. Two criteria may be applied: (a) Measurements of the morphological index. These provide immediate results but are difficult to standardize between laboratories and are tech- nically demanding. (b) Mouse inoculations. These provide firm evidence of bacterial viability and are 10-100 times more sensitive than measurements of the morpho- logical index, but they require greater investment of personnel and facilities and the results are available much later. Rapid bactericidal effects can often be demonstrated only by mouse inocula- tions because observable changes in bacterial morphology may lag several weeks behind loss of infectivity-a difference particularly evident with rifampicin. Short-term trials may be limited to 6 months, or even much less, depending on the regimen and the method of measurement. The bacterial index changes little in this period and is, therefore, of no value in such trials. Long-term trials ("five-year trials"). These are carried out to determine whether a drug's activity continues until smear negativity and clinical and histological quiescence are reached. Not many lepromatous patients reach this stage within 5 years. Mouse inoculations are particularly helpful when treatment failure is suspected, in which case tests of drug sensitivity provide crucial information. The first indication of treatment failure may be provided by measurements of the morphological index, if they can be performed reliably, or by histological observation of solid-staining bacilli, especially in dermal nerve bundles (40). Very-long-term trials. Because drug-resistant M. leprae may take many years to appear, a complete picture of the therapeutic efficacy of a drug cannot be obtained unless patients are followed for long periods-perhaps for the rest of their lives. There- fore, leprosy services that successfully practice very-long-term follow-up of lepromatous patients can provide invaluable information on the final efficacy of a regimen. As pointed out above, smear negativity does not signify that the patient is free of bacilli, but rather that the number of bacilli is 107 or less. The minimum number of viable M. leprae needed to cause a relapse in a lepromatous patient may be very small, since such a patient would not be expected to possess effective immunity against M. leprae. In these studies, it will be essential to determine whether relapse is caused by drug- sensitive or drug-resistant M. leprae. Experience with sulfone therapy has shown that relapses due to M. leprae that have been proved to be drug- resistant may occur 5 to more than 20 years after the commencement of treatment and after many of the patients have become smear-negative for various periods (21, 25). Trials in paucibacillary (nonlepromatous) leprosy. The term " paucibacillary " refers to leprosy patients whose bacterial index, before treatment, is low. Most of the leprosy patients in the world have pauci- bacillary disease, and information is needed on the time required in the various clinical varieties before 431 432 MEMORANDUM drug therapy may be safely stopped and the patient released from control. Few trials have been carried out in paucibacillary leprosy, and the most efficient form of the trial has yet to be determined. In one trial in progress, patients are being treated by one of two regimens for each of the following categories of leprosy: BL + BB, BT, TT, indeterminate-Mitsuda- positive, and indeterminate-Mitsuda-negative (the letter designations refer to the Ridley-Jopling scale). The end-point in the treatment of each patient will be 3 years of quiescence, as determined by the absence of clinical and histological signs of disease activity and by negative skin smears. CONCLUSION The application of the mouse model has at last placed the chemotherapy of leprosy on an objective bacteriological and pharmacological basis. It has provided sensitive procedures for the assessment of new drugs, the response to treatment, and the detection of drug resistance. It has also led to clarification of the theoretical basis of long-term and very-long-term clinical trials. These are difficult and expensive to carry out, but without them the final value of a regimen cannot be established. * * * Charles C. Shepard, Center for Disease Control, Atlanta, GA, USA (Chairman) Gordon A. Ellard, Medical Research Council Unit for Laboratory Studies of Tuberculosis, Hammer- smith Hospital, London, England Louis Levy, United States Public Health Service Hospital, San Francisco, CA, USA V. de Araujo Opromolla, Sanitario Aimores, Bauru, Brazil Stefaan R. Pattyn, University and Institute for Tropical Medicine, Antwerp, Belgium John H. Peters, Stanford Research Institute, Menlo Park, CA, USA R. J. W. Rees, National Institute for Medical Research, London, England M. F. R. Waters, Leprosy Research Unit, Sungei Buloh Leprosarium, Selangor, Malaysia REFERENCES 1. ADAMS, A. R. D. & WATERS, M. F. R. British medical journal, 2: 872 (1966). 2. AQUINAS, M. ET AL. British medical jour-nal, 1: 765 (1972). 3. DICKINSON, J. M. & MITCHISON, D. A. Tubercle, 47: 370 (1966). 4. EAGLE, H. American journal of medicine, 13: 389 (1952). 5. EAGLE, H. & MUSSELMAN, A. D. Journal of bacteri- ology, 58: 475 (1949). 6. ELLARD, G. A. ET AL. Leprosy review, 41: 223 (1970). 7. ELLARD, G. A. ET AL. Leprosy review, 42: 101 (1971). 8. ELLARD, G. A. ET AL. Nature, 239: 159 (1972). 9. FIELDSTEEL, A. H. International journal of leprosy, 41: 509 (1973). 10. FIELDSTEEL, A. H. & GARTNER, S. International journal of leprosy, 42: 121 (1974). 11. FIELDSTEEL, A. H. & MCINTOSH, A. H. Proceedings ofthe Society for Experimental Biology and Medicine, 138: 408 (1971). 12. FURESZ, S. ET AL. Arzneimittelforschung, 17: 534 (1967). 13. GELBER, R. H. & REES, R. J. W. American journal of tropical medicine and hygiene, 24: 963 (1975). 14. GELBER, R. H. ET AL. Clinical pharmacology and therapeutics, 12: 225 (1971). 15. GELBER, R. H. ET AL. Leprosy review, 45: 308 (1974). 16. GORDON, G. R. ET AL. International journlal of leprosy, 42: 375 (1974). 17. GYSELEN, A. & VERBIST, L. Praxis der Pnieumologie, 26: 269 (1972). 18. HOLMES, I. B. International journal of leprosy, 42: 289 (1974). 19. HOLMES, I. B. & HILSON, G. R. F. Journal of medical microbiology, 5: 251 (1972). 20. HOLMES, I. B. & HILSON, G. R. F. International journal of leprosy, 41: 508 (1973). 21. JACOBSON, R. R. International journal of leprosy, 41: 684 (1973). 22. KIRCHHEIMER, W. F. & STORRS, E. H. Inter-national journal of leprosy, 39: 693 (1971). 23. LEVY, L. ET AL. American journal of tropical medicine and hygiene, 21: 315 (1972). 24. MCRAE, D. H. ET AL. Internationaljournal of leprosy, 41: 487 (1973). 25. MEADE, T. W. ET AL. International journal of leprosy, 41: 684 (1973). 26. OZAWA, T. ET AL. American journal of tropical medicine and hygiene, 20: 274 (1971). 27. PATTYN, S. R. Annales de la Societe belge de medecine tropicale, 54: 43 (1974). 28. PEARSON, J. M. H. Leprosy review, 41: 155 (1970). 29. PETERS, J. H. ET AL. American journal of tropical medicine and hygiene, 21: 450 (1972). LEPROSY 433 30. PETERS, J. H. ET AL. International journal of leprosy, 40: 467 (1972). 31. PETERS, J. H. ET AL. American journal of tropical medicine and hygiene, 23: 222 (1974). 32. PETTIT, J. H. S. ET AL. International journal of leprosy, 35: 28 (1967). 33. POOLE, G. ET AL. British mnedicaljournal, 3: 343 (1971). 34. REES, R. J. W. Nature, 211: 657 (1966). 35. REES, R. J. W. International journal of leprosy, 35: 625 (1967). 36. REES, R. J. W. Transactions of the Royal Society of Tropical Medicine and Hygiene, 61: 581 (1967). 37. REES, R. J. W. ET AL. British medical journal, 1: 89 (1970). 38. RIDLEY, D. S. In: Cochrane, R. W. & Davey, T. F., ed. Leprosy in theory and practice, 2nd ed. Baltimore, Williams & Wilkins, 1964, p. 620. 39. RIDLEY, D. S. International journal of leprosy, 35: 187 (1967). 40. RIDLEY, D. S. International journal of leprosy, 41: 641 (1973). 41. RIDLEY, D. S. & JOPLING, W. H. International journal of leprosy, 34: 255 (1966). 42. RUSSELL, D. A. ET AL. International journal of leprosy, 41: 486 (1973). 43. RUSSELL, D. A. ET AL. Anmerican journal of tropical mnedicine and hygiene, 24: 485 (1975). 44. SHEPARD, C. C. In: Transactions of the Leonard Wood Memorial-Johns Hopkins University Sym- posium on Research in Leprosy, Baltimore, 8-10 May 1961. Washington, DC, Leonard Wood Memo- rial, 1961, p. 230. 45. SHEPARD, C. C. International journal of leprosy, 37: 389 (1969). 46. SHEPARD, C. C. Proceedings of the Society for Experimental Biology and Medicine, 132: 120 (1969). 47. SHEPARD, C. C. & CHANG, Y. T. Proceedings of the Society for Experimental Biology and Medicine, 109: 636 (1962). 48. SHEPARD, C. C. ET AL. Proceedings of the Society for Experimental Biology and Medicine, 122: 893 (1966). 49. SHEPARD, C. C. ET AL. American journal of tropical medicine and hygiene, 17: 192 (1968). 50. SHEPARD, C. C. ET AL. American journal of tropical medicine and hygiene, 18: 258 (1969). 51. SHEPARD, C. C. ET AL. Proceedings of the Society for Experimental Biology and Medicine, 137: 725 (1971). 52. SHEPARD, C. C. ET AL. American journal of tr-opical medicine and hygiene, 21: 446 (1972). 53. SHEPARD, C. C. ET AL. International journal of leprosy, 40: 459 (1972). 54. U.S. LEPROSY PANEL AND THE LEONARD WOOD MEMORIAL, COLLABORATIVE EFFORT OF THE. Inter- national journal of leprosy, 40: 233 (1972). 55. U.S. LEPROSY PANEL AND THE LEONARD WOOD MEMORIAL, COLLABORATIVE EFFORT OF THE. American journal of tropical medicine and hygiene, 24: 475 (1975). 56. VERBIST, L. ET AL. Chest, 61: 555 (1972). 57. VIDEAU, D. Annales de l'Institut Pasteur, 108: 602 (1965). 58. WATERS, M. F. R. ET AL. Interniational journal of leprosy, 36: 651 (1968). 59. WATERS, M. F. R. ET AL. Leprosy review, 45: 288 (1974).