-->> See the slides

 

Ongoing Trials with New Drugs/Regimens: The Fluoroquinolone Case

Dr Jacques GROSSET

Center for Tuberculosis Research Johns Hopkins University School of Medicine, USA

Although ethics is not in my field of expertise, let us consider the matter in relation with controlled clinical trials for tuberculosis. To make this presentation simple and fashionable, I will take the example of clinical trials with .fluoroquinolones.

Fluoroquinolone Overview

Fluoroquinolones are synthetic antimicrobial compoundss derived from nalidixic acid and characterized by a fluorine atom at position 6. They include ciprofloxacin, ofloxacin, levofloxacin, sparfloxacin, moxifloxacin, and gatifloxacin, and can be differentiated by the range of their minimum inhibitory concentrations (MICs) for M.tuberculosis, the lowest MIC90 being 0.5 µg/ml for the latter three compounds.

At standard human doses, moxifloxacin (MXF) achieves the highest maximal serum concentration (Cmax) and the highest area under the serum concentration time curve (AUC) that are the most relevant pharmacokinetic (Pk) parameters to evaluate the potential activity of an antibiotic. MXF has also the most favorable pharmacodynamic (Pd) parameters, the ratios Cmax/MIC and AUC/MIC, that are strongly correlated with bactericidal activity of the fluoroquinolones against a wide range of Gram positive and Gram negative pathogens. Considered on Pk and Pd parameters, MXF is therefore a very promising drug.

Experimental Anti-Tuberculosis Activity of Fluoroquinolones in Mice

The bactericidal activity of fluoroquinolones alone against M-tuberculosis in mice have been studied in several experiments. After four weeks of treatment in the mouse, only moxifloxacin exhibited a bactericidal activity similar to that of isoniazid.

The Anti-Tuberculosis Activity of Fluoroquinolones in Humans

In 1985, Tsukamura and his collaborators in Japan treated 19 patients with ofloxacin alone for six to nine months. None of the patients had drug side-effects and 5 of the 19 became culture negative. Many anecdotal reports confirmed these early findings. In 1993, the American Thoracic Society (ATS) and the CDC considered fluoroquinolones as useful for the treatment of MDR-TB. In 2003, the ATS and CDC guidelines went a little further in recommending fluoroquinolones for the treatment of MDR-TB but not for the first line treatment.. Why fluoroquinolones have not been officially recommended as first-line agents? The answer is simple. First, because all early bactericidal activities (EBA) studies failed to demonstrate powerful bactericidal activity of fluoroquinolones, other than moxifloxacin,. Second, because the addition of a fluoroquinolone to the standard regimen failed to demonstrate benefit in terms of time to culture conversion and relapse rate. However, in the TRC Chennai 2002 trial, the addition of ofloxacin to the standard regimen suggested that it might permit the shortening of treatment duration to four or five months. However, this trial raised a number of scientific issues.

Ethical Rules for Clinical Trials

Ethics in clinical trials cannot be reduced to the informed consent of patients. It involves also the rationale, the scientific basis, and the design of the trials. Controlled clinical trials should follow the four following rules:

  • The first rule is the autonomy or consent of the patient. In no case, can clinical trials be carried out without the genuine consent of the patient.
  • The second rule is beneficience. Patients should benefit or at least not suffer from being in a trial.
  • The third rule is equipoise. The investigators must be equally comfortable with the alternative treatment arms when randomizing patients. In other words, the investigators should be prepared, in the hypothetic case of having tuberculosis themselves, to be randomized in any particular arm of the trial.
  • The fourth rule is justice. The benefits and burdens of research should be shared fairly, as far as is possible.

Before translating these rules into practice, the following fact should be remembered: in the treatment of tuberculosis the current standard six-month drug regimen can cure 100% of newly diagnosed patients with drug susceptible organisms when regular drug intake is ensured. Consequently, in no case should patients be deprived of a 100% active treatment. Second, patients should not be exposed to undue risks of toxicity. Third, the trial should be scientifically founded. In other words, experimental and clinical evidence is required before any clinical trial is embarked upon. Fourth, the trial should be scientifically designed and performed.

No Risk of Depriving Patients of an Active Treatment

Any deviation in the drug content and the duration of the standard treatment is unacceptable without solid scientific experimental/clinical evidence. For example, there is no evidence (potential is not evidence) that the duration of treatment in humans might be reduced by the use of any of the available fluoroquinolones. Nor is there any evidence that the time to smear and culture conversion might be shortened by the use of fluoroquinolones.

No Exposition to Undue Risks of Toxicity

Except in a 6-month study recently conducted in Italy with no side effects (Valerio et al., 2003), moxifloxacin has never been administered for several weeks or months. To address the issue of tolerance/toxicity of log course administration of fluoroquinolones, two double-blind control clinical studies in which moxifloxacin is being substituted for ethambutol are currently in progress under the auspices of CDC and John Hopkins University. These studies have two primary end-points: culture conversion within two months and safety/tolerability.

The Trial Should be Scientifically Founded

Experimental and clinical data should provide rationale for undertaking the trial. Without such data, it is unethical to undertake a clinical trial. An example of experimental results that provide rationale for a clinical trial is given by the use of moxifloxacin in combination with the first-line drugs in the experimental chemotherapy of murine tuberculosis. The benefit of adding moxifloxacin to the standard six-month regimen (2RHZ/4RH) and of substituting moxifloxacin for each of the individual components of the standard regimen was evaluated. The addition of moxifloxacin to the standard regimen resulted in a modest though significant reduction of the viable counts (CFU) of M. tuberculosis in the lungs of mice at two, three, and four months. But culture conversion to negative did not occur earlier. Therefore the overall benefit of adding the most potent fluoroquinolone to the standard regimen was marginal. There is therefore no scientific (and ethical) experimental basis for undertaking controlled clinical trials to test the efficacy of regimens of less than 6-month duration, in which moxifloxacin or another fluoroquinolone would be added to the standard drugs. However, when moxifloxacin was substituted for isoniazid in the standard regimen, the time to culture conversion was dramatically reduced.. There is therefore experimental evidence that the substitution of moxifloxacin for isoniazid might shorten the time to culture conversion in humans.

Additional information supports the use of moxifloxacin in humans. Three studies of the early bactericidal activity (EBA) of antituberculosis drugs in pulmonary tuberculosis have been carried out with isoniazid, rifampicin and moxifloxacin. They evaluated the fall in log10 CFU counts during the first two days of treatment. With isoniazid and moxifloxacin used alone the fall was similarly between 0.4 logs to 0.7 logs. There is thus some evidence that moxifloxacin is as potent in humans as in mice. Such information provides additional support for undertaking a clinical trial.

As a consequences of this scientific evidence, CDC is currently working with the Russian Research Institute for Phthisio-Pulmonology to implement a Phase II trial to evaluate the potential of substituting moxifloxacin for isoniazid in the initial phase of treatment. The trial will consider the time to culture conversion and safety/tolerability. Investigators from John Hopkins University are also developing a similar approach.

The Trial Should be Scientifically Designed and Performed

The protocol should follow the rules of controlled clinical trials. That is, there should be a control group. An adequate number of patients are needed to obtain significant results. There must be an adequate organization of drug intake, and clinical and laboratory monitoring. Finally, the primary and secondary end points should be defined.

Conclusion

"At long last improving the treatment of tuberculosis is more than a distant dream (R. O'Brien, 2003)". To avoid such a dream to become a nightmare for some patients, controlled clinical trials should be rationally initiated and conducted, and comply with ethical rules.