The Mass That Wasn’t Cancer: Choosing a Nine-Month Regimen
A mass suspected to be cancer turns out to be a treatable infection. What remains genuinely contested is how aggressively to start, given an immune system that has been suppressed for fifteen years.
Beatrice N., a 68-year-old retired librarian who has managed rheumatoid arthritis for over fifteen years on chronic low-dose prednisone, was referred for what her primary team initially suspected was lung cancer — a spiculated 3cm mass on chest CT, ordered after months of a mild cough she’d attributed to seasonal allergies. A CT-guided biopsy, done specifically to stage the presumed malignancy before any oncology referral, instead grew Nocardia on culture, an unexpected and, for her pulmonology team, genuinely reassuring result once the shock of a wrong initial impression wore off. Staging workup since the biopsy, including brain MRI and abdominal imaging, has shown no evidence of central nervous system or other disseminated involvement — the infection appears confined to the single pulmonary site.
Her chronic corticosteroid use for rheumatoid arthritis is the clear predisposing factor, years of immunosuppression having created the vulnerability nocardiosis typically exploits, and it remains a live consideration even with a clean staging workup, since occult microscopic dissemination isn’t always visible on imaging at presentation. She lives independently, still walks to the library twice a week to volunteer shelving books, and is relieved beyond words that this isn’t the cancer diagnosis she’d spent the past two weeks preparing herself for — though she is now facing a treatment course, nine to twelve months by current practice standards, considerably longer than she’d expected for an infection rather than a malignancy.
Nocardia’s own biology is part of why the duration runs this long regardless of which induction strategy is chosen: the organism grows slowly, forms filamentous branching colonies that can wall themselves off within tissue, and has a well-documented tendency to relapse if treatment stops too early, even after apparent clinical resolution — the reason current practice treats nine months, not the six weeks a more conventional bacterial pneumonia would require, as the floor rather than an unusually cautious upper bound. What none of that supplies is a way to know she is actually free of disseminated disease rather than merely free of visible disease: the brain MRI that reassured everyone is a snapshot of what is large enough to see, and nocardial seeding is characteristically neither large nor early, which is why the duration is set by the organism’s reputation rather than by anything her own imaging can confirm.
Infectious disease and pulmonology, joint consult
I’d start her on treatment-dose TMP-SMX monotherapy. It’s the agent with the longest track record for nocardiosis, and her staging workup is clean — no CNS involvement, no evidence of dissemination — which is exactly the presentation where combination therapy’s added benefit is least supported. Monotherapy also lets her manage this as an outpatient rather than requiring IV access for an induction phase she may not actually need.
I want to weigh her actual immunosuppression more heavily before committing to monotherapy from day one. Fifteen years of chronic steroid exposure is a real, not theoretical, risk factor for a more severe course or occult dissemination than a single point-in-time scan can rule out completely. I’d start her on combination therapy — a carbapenem alongside TMP-SMX — for the first several weeks, and de-escalate to TMP-SMX alone once she’s clearly responding, which hedges against a worse course without locking her into combination therapy for the full nine to twelve months.
Agreed: begin with combination therapy (meropenem plus TMP-SMX) for the first three to four weeks, with a clinical and repeat-imaging reassessment at that point to confirm the expected response before de-escalating to TMP-SMX monotherapy for the remainder of a nine-to-twelve-month course.
She tolerated the induction phase well, with imaging at week four showing clear improvement in the pulmonary lesion; meropenem was discontinued as planned and she was transitioned to outpatient TMP-SMX to complete treatment, with rheumatology continuing to co-manage her steroid dosing throughout.