Two Antibody-Drug Conjugates, One Line of Therapy Left to Pick the Right One
Two antibody-drug conjugates, T-DXd and sacituzumab govitecan, both have real trial support in her HER2-low, endocrine-refractory metastatic disease. Her interstitial lung findings on a recent scan don't rule either one out, but they change how much margin she actually has for the wrong choice.
Denise F., a 55-year-old hospice volunteer who spends three shifts a week sitting with other people's dying, has a clear-eyed, unsentimental way of asking about her own prognosis that her oncology team has come to rely on rather than soften around. Her HER2-low, hormone-receptor-positive metastatic disease has now progressed through two lines of endocrine-based therapy, moving the conversation to antibody-drug conjugates — and a routine surveillance CT chest, done the same week, found scattered subpleural ground-glass opacities she has no symptoms from and no history to explain, findings that don't rule out either candidate drug but change how much margin she has if the wrong one turns out to be the harder one on her lungs.
DESTINY-Breast04 established trastuzumab deruxtecan as effective specifically in HER2-low disease — the trial that created the category by demonstrating a real progression-free and overall survival benefit in patients who would previously have been treated as HER2-negative — but that same trial carries a genuine, drug-specific interstitial lung disease signal: adjudicated drug-related ILD or pneumonitis in 12.1% of patients receiving trastuzumab deruxtecan, with 0.8% grade 5. The risk is not distributed evenly, which is the part that reaches Denise directly. A pooled analysis by Powell and colleagues of nine T-DXd monotherapy studies ran a stepwise regression against baseline characteristics and identified the presence of lung comorbidities as one of the factors associated with increased ILD risk, alongside oxygen saturation below 95% and moderate-to-severe renal impairment. Sacituzumab govitecan, evaluated in a broader hormone-receptor-positive metastatic population in TROPiCS-02 rather than a HER2-low-specific trial, targets a different antigen entirely — Trop-2, not HER2 — and its major toxicities run toward neutropenia and diarrhea rather than pulmonary disease, with no comparable ILD signal in its own trial data. Denise's ground-glass opacities are asymptomatic and her FEV1 is 88% of predicted, so neither drug is contraindicated by what the scan actually shows — but "not contraindicated" and "not a risk factor" are different claims, and only the first is true of her. Her abnormal baseline lung field costs her twice over: it puts her in the group that pooled analysis flagged, and it removes the clean comparison scan a future cough would otherwise be read against, since a new finding could then be attributed to the disease, the drug, or the change already there.
Medical oncology, third-line selection
T-DXd's efficacy data in HER2-low disease is the strongest thing available to her right now, and I don't want to talk myself out of the better drug over an asymptomatic finding on a surveillance scan. I'll concede the pooled data puts her in a higher-risk group rather than a neutral one — but higher risk within a 12% event rate that is mostly grade 1 to 2 is still a long way from a reason to withhold the drug that gave her population an extra six months of overall survival.
I'm not arguing the efficacy point, but I want to name plainly what those ground-glass opacities do to our ability to monitor her safely: if she develops a cough on T-DXd next month, we will not be able to tell, from imaging alone, whether that's early drug-induced ILD or simply progression of whatever's already there — the baseline scan we'd normally compare against isn't clean.
A monitoring plan doesn't move her out of the risk group, though, and that's where I part company with you. Better surveillance shortens the time to diagnosis; it doesn't lower the incidence. You're treating her opacities purely as an imaging inconvenience when the pooled analysis treats them as a baseline characteristic that changed the event rate — those are different objections and only one of them is answered by scanning her more often.
I don't think I can call this a monitoring-intensity decision after that exchange, and I was going to. If her lung comorbidity raises the incidence rather than just obscuring the detection, then a same-radiologist high-resolution baseline and a lower symptom-reporting threshold buy her earlier recognition and nothing else — worth having, and not the thing the pulmonologist is actually objecting to.
Which puts the choice back where it started, but honestly this time. Sacituzumab govitecan carries TROPiCS-02 behind it in a hormone-receptor-positive metastatic population, trades pulmonary risk for neutropenia and diarrhea, and is not a consolation prize. Denise sits with dying people three days a week and has never once asked us to round a number in her favor. She should be told she is in the flagged group, told the size of it, and allowed to pick — not handed the drug with the better median and a promise that we'll watch closely.
Agreed, after Denise was told plainly that her scan places her in the higher-risk group rather than merely complicating the picture, and weighed both options herself: trastuzumab deruxtecan, started with a formal baseline high-resolution CT chest and a standing instruction to call the clinic directly, rather than wait for a scheduled visit, at any new cough or breathlessness.
Not fully settled: the pulmonologist accepts the choice as hers but not the reasoning that got the room comfortable with it, and asked that the record show he considers an intensified surveillance plan to be a detection measure mistaken for a risk-reduction measure — a distinction he expects to matter if she does develop pneumonitis and the group looks back at what it thought it had covered.