No easy choices in prophylaxis for MDR-TB
Walking tightrope between efficacy, toxicity
Oh, those quinolones. They’re like the bright kid in class with the smart mouth. Some people love them; others are wary but willing to give them a chance. Then there are those who’d prefer to pack them straight off to the principal’s office. They are quinolones.
Consider the case of adults likely to have been infected with multidrug-resistant tuberculosis (MDR-TB) and judged to be at high risk for developing disease. (For children, it should be noted, quinolones are almost never a good choice, since they may pose a threat to growing cartilage, says Jeffrey Starke, MD, assistant professor of pediatrics at Baylor College of Medicine in Houston).
Ranged against the two reliable old hands ethambutol and pyrazinamide, and considering recent reports of toxicity from quinolone-containing regimens, do the quinolones even have a place in a regimen for MDR prophylaxis?
"It’s a can of worms," says Anne Morrissey, MS, chief of clinical microbiology at University Hospitals at Cleveland.
The biggest problem is the dearth of real-life data on the subject, she says. What happens in people, Morrissey points out, doesn’t always correlate with the situation in test tubes, which lack the immune systems people come equipped with (unless they’re HIV-positive, a factor which adds several more layers of complexity).
Then there’s the puzzling finding that certain drugs are tolerated well by patients with active disease but poorly by those who take them for preventive therapy "a very strange thing," says Sonal Munsiff, MD, Director of Medical Affairs for the New York City department of TB control.
Finally, there’s the issue of patient reliability, or, as Morrissey puts it, "Are they having no side effects because the drugs aren’t giving side effects? Or because they’re not taking the drugs?"
Added together, it all makes for a confusing picture.
To see how various experts sort it out, TB Monitor asked some for their opinions. Here’s what they said:
• Expert opinion #1: Stick with what you know best, ethambutol (EMB) plus pyrazinamide (PZA).
Assuming, of course, that susceptibility testing of the source case shows these two drugs are available, that’s the first choice of Elsa Villarino, MD, MPH, section chief of the Research and Evaluation Branch of the Division for Tuberculosis Elimination at the Centers for Disease Control and Prevention.
Villarino’s preference for EMB and PZA stems from her belief that these two drugs have proved themselves trustworthy in the past. "These are the drugs for which we have the most information and the least evidence of toxicity," she says. (The Food and Drug Administration hasn’t even approved quinolones for use with TB, she notes.)
Reports of toxicity with OFL and PZA
In addition, Villarino cites data from New York and Chicago, where a second alternative regimen the quinolone ofloxacin (OFL) paired with PZA was linked to many reports of toxicity. In New York, where clinicians had no choice but to use a quinolone and PZA, the combination "was poorly tolerated," says Munsiff.
In Chicago, after that city’s recent bouts with MDR, candidates for prophylaxis placed on EMB and PZA experienced the least toxicity, with a 30% rate of adverse events by regimen, Villarino calculated. By comparison, the rate of adverse events among those who took PZA and a quinolone was 55%; and among those who took EMB and a quinolone was 38%.1
The results did not constitute a controlled trial in either setting, Villarino notes. "Nor will there ever be enough patients at one site" to do a trial, she adds. "So the most important question that of efficacy will remain unanswered."
Having authored the section on various alternative regimens for prophylaxis (in the CDC’s recent "National Action Plan to Combat Multidrug-Resistant Tuberculosis,"2 Villarino presently is at work compiling what’s happened to those who opted for one or another of the drug pairs listed. For now, she advocates a watch-and-wait strategy whenever feasible.
"If you are otherwise healthy and don’t meet criteria such as being HIV-infected, we’d just recommend supervision without drug therapy," she says. "Save your heavy artillery for active disease."
• Expert Opinion #2: Forget PZA and EMB; instead, bring on the quinolones, in combination with PZA.
If you’re talking to John Sbarbaro, MD, MPH, professor of medicine at the University of Colorado Health Sciences Center in Denver, there’s an interesting corollary to this principle: If you choose OFL as your quinolone, consider adjusting the dose of PZA downward (or decreasing its frequency to three times a week). That way, you’ll minimize the toxicity that troubled patients in New York and Chicago, Sbarbaro says.3
Or if you’d prefer to use PZA at the full dose, substitute ciprofloxacin for OFL, he adds.
The reason? While ciprofloxacin is eliminated through the liver, both OFL and PZA "come out through the kidney," Sbarbaro says. "If you give both [OFL and PZA] in normal doses, you’re going to drive PZA sky-high."
That sounds plausible to Charles A. Peloquin, PharmD, director of the Pharmacokinetics Laboratory at National Jewish Hospital in Denver. OFL is eliminated almost entirely through the kidneys, Peloquin says; so maybe it competes with PZA’s metabolites in the same way those metabolites can compete with uric acid. And then, maybe the PZA metabolites are causing toxicity. "It’s possible," he cautions, "but not proven."
Promising new quinolone
The newest quinolone on the block, levofloxacin, has also caught Sbarbaro’s eye. Available in the United States now for only a couple of months, the drug has been touted by some as quite promising.4 Along with evidence of good activity, it lacks the annoying theophylline-like caffeine side-effects of other quinolones, Sbarbaro says.
"Since we now know quinolones kill deader than a doornail," Sbarbaro asks, "why don’t we just use levofloxacin?"
As for Villarino’s preferred regimen of PZA and EMB, Sbarbaro snorts disdainfully. "If you give water, you have no side effects," he says exactly what giving PZA and ethambutol amounts to.
He reasons thus: EMB, though bactericidal at 25 mg/kg, is usually lowered to 15 mg/kg after the first six to eight weeks to minimize the risk of ocular toxicity. For the long haul of a prophylactic regimen, that means the drug simply isn’t effective.
PZA more effective early on
Then there’s PZA. Acknowledged as an effective weapon, quite useful early on in therapy, the drug appears to operate poorly later on, and clearly isn’t much help in mopping up small quantities of remaining bugs in the continuation phase. The reason, Sbarbaro says, is because PZA works best in the high-acid environment found in the caseous matter of a lung cavity not exactly a situation common to patients undergoing chemoprophylaxis for infection.
Again, that’s plausible, Peloquin says. Once PZA is ingested, the liver goes to work and converts some of the drug to its soluble metabolites, one of which pyrazinoic acid (PoA) is thought to be toxic to the TB microbes, which evidently are crafty enough not to take up the drug in this toxic metabolite form. The bugs do, however, readily take up the parent compound, some of which manages to arrive still unmetabolized at the site of infection.
Finally, mycobacteria equipped with a specific enzyme (an amidase that’s apparently lacking in bugs that have become resistant to PZA) take up the PZA, convert it to PoA, and duly expire, all in accordance with the plan.5
But does the entire process really require the low pH of a lung cavity to work? "This part gets dicey," admits Peloquin. Perhaps in the lower pH of a cavity, confounded by the gradient, the microbes can’t pump out the excess acid, a fact which ultimately seals their fate, he speculates. Or . . . perhaps not.
Cavity or no cavity, some argue, the myco-bacteria have been engulfed by macrophages, where they’re swimming in the low-pH environment of the lysosome. Yet it has been proposed (and some recent evidence suggests) that "the bugs have a way of derailing the train, so to speak, so that the phagolysosomes don’t fuse," Peloquin says.
If so, Sbarbaro may be correct in saying that PZA doesn’t work well for prophylaxis. For that matter, Peloquin agrees that EMB at 15 mg/kg may well be too weak to have much effect, either.
• Expert Opinion #3: Forget the metaphysics, but use a quinolone and PZA anyway, if possible.
"This is all a bunch of theoretical stuff," says Henry F. Chambers, MD, chief of infectious diseases at San Francisco General Hospital and professor of medicine at the University of California in San Francisco. "People have their biases, but they don’t know the answers."
Better to stick to practical issues, Chambers says, and simply heed the following parameters:
• The more active the regimen, the better.
"I’d use ofloxacin and PZA if the patient could tolerate it," he says. "You need to be serious about treatment" especially when you’re dealing with HIV-positive patients."
• Since you’re using relatively weak drugs, use a combination.
Isoniazid "is by far the most potent drug we have; nothing else comes close," he says. Given that you can’t, by definition, include INH in your arsenal, don’t fiddle around with monotherapy.
• For the same reason, treat for longer, not shorter, durations of time. (Chambers suggests 12 months.)
• While you’re at it, watch out for toxicity.
All this said, Chambers isn’t immovable when it comes to his first choice. "I wouldn’t jump up and down if someone said use ethambutol and PZA" instead of OFL and PZA, he says. Since HIV-positive patients often have trouble tolerating stronger regimens, the regimen of EMB and PZA might offer advantages in some situations, he says.
References
1. Villarino M, Greene V, Paul W, et al. Preventive therapy for exposures to multidrug-resistant tuberculosis. Tubercle and Lung Disease, poster session abstracts, p. 60, Conference on Global Lung Health and the 1996 Annual Meeting of the International Union Against Tubercle and Lung Disease, Paris, October 2-5, 1996.
2. CDC. National Action Plan To Combat Multidrug-Resistant Tuberculosis. MMWR 1992; 41:1-71.
3. Sbarbo J, Iseman M, Crowle. Combined effect of pyrazinamide and ofloxacin with the human macrophage. Tubercle and Lung Disease 1996; 77:491-495.
4. Klemens SP, Sharpe CA, Rogge MC, et al. Activity of Levofloxacin in a Murine Model of Tuberculosis. Antimicrobial Agents and Chemotherapy. 1994; 38:1,476-1,479.
5. Sreevatsan S, Pan X, Zhang Y, et al. Mutations associated with pyrazinamide resistance in pnca of Mycobacterium tuberculosis complex organisms. Antimicrobial Agents and Chemotherapy 1997; 41:636-640.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content