New MAC in a Patient on CFTR Modulators: A Guideline Written for Someone Else
A man with cystic fibrosis, newly diagnosed with MAC pulmonary disease, on elexacaftor/tezacaftor/ivacaftor that has transformed his lung function over three years. The disagreement is what to do about the rifamycin his NTM regimen would ordinarily include, since the guideline that specifies it was never written with his drug regimen — or his diagnosis — in mind.
Marcus O., a 34-year-old man, is getting married in ten weeks, and has spent the last several months more focused on cake tastings and a guest list than on the sputum sample he almost forgot to leave at his last CF clinic visit. He has lived with cystic fibrosis his whole life, and elexacaftor/tezacaftor/ivacaftor, started three years ago, has changed what that means for him more than any other single thing has — his FEV1 has climbed from 58% of predicted before starting the drug to 91% now, he has gained back weight he’d struggled with for years, and he has gone more than two years without a hospitalization for the first time in his adult life. That nearly-forgotten sputum sample grew Mycobacterium avium complex, and a follow-up chest CT shows new nodular infiltrates distinct from his baseline CF bronchiectasis pattern — together enough to diagnose MAC pulmonary disease layered onto his existing lung disease, a new problem arriving at what otherwise feels, to him, like the best period of his adult life.
The 2020 ATS/ERS/ESCMID/IDSA guideline that would ordinarily specify his regimen excludes cystic fibrosis from its scope, but it is not true that no guideline describes him: the CF Foundation and European CF Society consensus recommendations cover clarithromycin-sensitive MAC disease in CF directly, ask for a daily macrolide, rifampin and ethambutol, and specifically recommend against the thrice-weekly schedule used outside CF. So the document written for him names the very drug he cannot take. Rifampin reduces ivacaftor exposure by 89% in direct pharmacokinetic study — a drop that would functionally undo three years of a treatment that has reshaped his health. Rifabutin is the softer version of the same problem, and the evidence on it has recently firmed up: PBPK modeling had predicted reductions of roughly 35 to 69 percent across the three ETI components, and a prospective study in healthy volunteers has since measured them at 55%, 32% and 40% for elexacaftor, tezacaftor and ivacaftor respectively, with modeled concentrations still sitting above the half-maximal effective concentration. Marcus’s MAC was found on a surveillance culture he almost forgot to leave, with no new symptoms and no cavitation — which is precisely the low-burden picture in which the team now has to decide whether the third drug is worth any of this at all.
CF/ID joint clinic, planning NTM treatment
Substitute rifabutin for rifampin, keep ETI unchanged, and add monitoring — sweat chloride and, where available, ETI levels — to confirm exposure holds. Rifabutin induces CYP3A far less than rifampin — the measured ivacaftor AUC ratio is 0.60 with rifabutin against 0.11 with rifampin — and a published case series of three patients on concomitant rifabutin and ETI found no compromise in efficacy by lung function or sweat chloride.
Three years of ETI got him from 58% to 91% predicted. That’s not a number I want to gamble with on the strong-inducer option.
That case series is three patients — real, but not something to build a plan around as though it settles anything. I’ll grant you more than you asked for, though: there is now a prospective healthy-volunteer study measuring it, and the reductions came in at 55%, 32% and 40%, milder than the modeling predicted, with concentrations still above EC50. What that study doesn’t give us is Marcus. It was healthy volunteers, not a man whose entire clinical trajectory depends on those levels holding, and his induction remains unquantified without drawn levels.
I’d hold ETI during the treatment course, at least its early months, and restart once therapy is well underway. His infection may still be early — getting antibiotic exposure right matters most right now, and holding removes the whole layer of uncertainty rather than monitoring around it.
The interaction data genuinely doesn’t resolve to a clean answer either way — you’re both reading real uncertainty honestly.
But I want to challenge the premise you’re both arguing inside — that the rifamycin is a drug worth this much trouble. Ethambutol is the agent that actually prevents emergent macrolide resistance in MAC, and it stays in the regimen either way. Rifampicin’s in vitro activity against MAC is low, its PK-PD targets aren’t attainable at tolerable doses, and the published argument is now that it adds nothing to an azithromycin-ethambutol backbone and does not prevent macrolide resistance — which is the job it is usually said to be doing. For low-burden disease, non-cavitary and smear-negative, a two-drug regimen may be sufficient, and that describes him. Omit the rifamycin, dose azithromycin and ethambutol daily per the CF-specific recommendations, and the interaction question disappears rather than being managed.
Agreed: start a two-drug regimen — daily azithromycin and ethambutol, per the CF-specific recommendation against thrice-weekly dosing — without a rifamycin, and continue ETI unchanged. Ethambutol carries the resistance-prevention role; culture monitoring is for detecting failure, not for replacing a drug.
Not agreed: what to do if cultures haven’t converted by the next defined check. The CF specialist would try adding rifabutin with monitoring at that point; the infectious disease physician still leans toward holding ETI temporarily as the cleaner second-line move if the two-drug regimen isn’t working.