A Clinical Trial to Evaluate the Safety and Efficacy of Ciprofloxacin in the Treatment of Plague in Humans

Plague is a life-threatening disease that requires immediate treatment with effective antimicrobials. Drugs with demonstrated efficacy include streptomycin, doxycycline, and to a lesser extent, gentamicin. Animal and in vitro studies suggest that fluoroquinolones, including ciprofloxacin, may be effective agents for treatment of plague. However, with the exception of a single case report, there are no published instances in which fluoroquinolones have been used successfully to treat plague in humans. There is a critical need to evaluate newer antimicrobials that have been shown to be efficacious in vitro and in animal studies, used extensively in the treatment of other illnesses in humans, work by a separate mechanism than antimicrobials currently used for treatment, and are widely available and affordable.

The objective of this study is to conduct a randomized, open-labeled, non-inferiority study comparing the safety and efficacy of ciprofloxacin to doxycycline, the national treatment standard in Uganda. This study is designed to test the following hypothesis: the safety and efficacy of ciprofloxacin for the treatment of plague in humans is not inferior to doxycycline. This study will result in a better understanding of the safety and efficacy of ciprofloxacin for the treatment of plague in humans. Information from this study will be used to expand the experience and knowledge base for the treatment of plague in humans and guide national and international treatment guidelines. Information from this study also will possibly add to the antiquated list of effective antimicrobials used to treat plague and possibly lead to a reduction in the overall morbidity and mortality caused by plague and its treatment.

The randomized clinical trial provides the best design to distinguish treatment effects from other effects such as bias in allocation, observation, and measurement; spontaneous changes in disease course or healing; and improvement from participating in a study (i.e., placebo effect). An open-labeled design is being used because the opportunities for investigators or patients to bias the outcome are limited (i.e., patient randomized after enrollment and mortality is objective outcome) and a double blind design would greatly increase the complexity and cost of the trial.

The primary outcome for this trial will be patient outcome 14 days from enrollment and initiation of treatment, which includes outcomes of patient recovery without complications, recovery with complications, withdrawal, or death. Patient outcome will be evaluated only for those patients with laboratory-confirmed plague illness. Time to defervesence and occurrence of antimicrobial-associated adverse events will be evaluated as secondary outcomes in patients with laboratory-confirmed plague illness and all enrolled patients, respectively.

When conducting a non-inferiority study, it is important to consider the appropriate selection of a comparator treatment, non-inferiority margin (M), and analysis strategy. For non-inferiority trials, it is critical to select a standard treatment that has proven to be the best historically or been used in prior randomized clinical trials as the comparator treatment. Of the possible options, doxycycline is most optimal choice for this trial because it is the current treatment standard in Uganda (the country location of the proposed study) and its treatment regimen and side effect profile are the similar to that of ciprofloxacin. For this trial, the non-inferiority margin (M) is the acceptable difference between the probability that ciprofloxacin cures plague as compared to the probability that doxycycline cures plague. Because of the severity of illness and high case fatality rate for plague in general, and the potential for pneumonic transmission with pneumonic plague, the choice of an acceptable M should be no more than 10% with a most conservative choice of 2%. If we assume an overall Type I error rate = 0.05, a power (power = 1 - Type II error rate) of 80%, πs = 0.90 and we choose M = 0.10, then the number of patients required for each study arm is well over 1,000 (calculations not shown). From our previous experience with a similar trial (i.e., The safety and efficacy of gentamicin vs. streptomycin in the treatment of plague in humans, CDC IRB #3700) and our current experience with surveillance data in Uganda, we anticipate being able to enroll approximately 100-200 total patients per year. Therefore, we computed the power using different values of M (0.02, 0.05, 0.10; where 0.02 is the most conservative and 0.10 is the most forgiving but still within the accepted non-inferiority margin), to represent a range of reported efficacy of the standards (table not shown). Using the table, we would have approximately 61% power to detect non-inferiority of ciprofloxacin treatment with a non-inferiority margin, M, = 0.10 if the true efficacy of the standard is = 90% and if we were to enroll 148 patients total, for example.

This study will attempt to include all suspect cases of plague aged 8 years or older identified through clinic-based surveillance. Most patients, including plague cases, are treated at the "level 3" health centers, mainly due to a lack of motorized transport. Between 3 and 10 "level 3" clinics in Arua and Nebbi Districts, Uganda will be chosen to participate in this study based on the reported number of cases of plague in the last 5 years and logistic considerations. Only UMOH clinics will be considered for inclusion.

The "level 3" clinics are staffed by 1-2 clinical officers and are very busy, seeing 50-150 patients each day. Because of the excess patient volume, the study collaborators will hire and train one additional clinical officer for each collaborating clinic to work alongside the UMOH clinical officer(s) stationed at the collaborating clinic so as not to impair normal clinic function. The hired clinical officer will be trained on the clinical aspects of plague and the study protocol, and will be responsible for evaluating suspect plague patients for study enrollment, consenting patients, and providing patient treatment and monitoring per the study protocol.

The diagnosis of suspected plague patients who consent and enroll in the study will follow the national guidelines of the UMOH. The only adaptations are that, after receiving informed consent, UMOH physicians will collect biologic samples and patients will be randomized and receive ciprofloxacin or doxycycline. Of note, a urine pregnancy test will be conducted on every female of child-bearing age at time of enrollment and randomization and antimicrobial treatment will not be initiated unless the results are negative. Based on our experience conducting a similar trial in Madagascar, results from the urine pregnancy test are available in less than 5 minutes so this should not cause any substantial time delays. Antimicrobial treatment per UMOH national guidelines will be started immediately should there be instances where there are delays in obtaining consent or collecting biologic samples, or if the treating physician deems it necessary.

Suspect cases of plague will be eligible and will be asked to give consent for study enrollment using the following criteria:

1. any person, including women and persons who are minorities, who;
2. must be aged 8 years or older, and;
3. must have had potential exposure to rodents and/or fleas or contact with a confirmed plague case, and;
4. must have a fever of at least 38ºC that developed rapidly, and have at least one of the following:
5. One or more buboes, defined as a tender lymph node swelling > 1cm in diameter, or;
6. Clinical suspicion of pneumonic plague (e.g. prostration, cough, increased respiratory rate, hemoptysis and/or purulent sputum), or
7. Clinical suspicion of cutaneous plague (lesion)
8. Clinical suspicion of plague and epidemiologic link with other cases.

Patients with suspected plague illness will be considered ineligible for the study and will be excluded from study enrollment using the following criteria:

1. Any women who is pregnant, or;
2. Any woman who is breast-feeding, or;
3. Any person aged < 8 years of age, or;
4. Any patient with:

• signs of plague meningitis
• hypotension unresponsive to fluid therapy
• an illness severity score of > 16 at time of enrollment (see below)
• known allergy to ciprofloxacin or doxycycline
• taken tetracyclines, quinolones, gentamicin, streptomycin, trimethoprim-sulfamethoxazole, or chloramphenicol in the 24 hours preceding study enrollment.

We will enroll suspected plague cases during the high transmission season between September and April, 2010 through 2012. Based on enrollment for a similar study performed previously by these investigators in Madagascar and Uganda (unpublished data, CDC) and given the existing epidemiological data on plague in Uganda, we anticipate recruitment of between 125 and 200 suspected cases of plague per year, where approximately 40% to 50% of cases will have a confirmed plague illness, with a 20-30% case-fatality rate.

Patients will be monitored daily. Participants may voluntarily withdraw from this study for any reason at any time. The protocol and consent forms must be approved by all three supervising IRBs (i.e., US Centers for Disease Control and Prevention, Uganda Virus Research Institute Scientific Ethics Committee, and Uganda National Council for Science and Technology) prior to starting the trial. A Data Safety and Monitoring Board will review all serious adverse events and interim and final trial data to ensure patient safety and good trial conduct.