Sara Nunnery, MD, MSCI, and Irene M. Kang, MD, discuss toxicity and monitoring considerations for the use of ADCs in metastatic TNBC.
Two phase 3 trials of TROP2-directed antibody-drug conjugates (ADCs)—ASCENT-03 and TROPION-Breast02—are reshaping the first-line treatment of patients with metastatic triple-negative breast cancer (mTNBC) for those who are not candidates for immunotherapy, according to Sara Nunnery, MD, MSCI, and Irene M. Kang, MD.
Speaking on
Nunnery is a breast medical oncologist and the director of Breast Cancer Research at Tennessee Oncology in Nashville, Tennessee, as well as the host of Breast Cancer Briefing.
Kang is the medical director of Women’s Health Medical Oncology at City of Hope Orange County and an assistant professor in the Department of Medical Oncology and Therapeutics Research at City of Hope in Irvine, California.
A: Subtyping every breast cancer is essential because it largely determines which treatments will be effective, Kang said. Triple-negative disease is defined by the absence of 3 targetable biomarkers: the ER, the progesterone receptor, and HER2. Since those targets are absent, chemotherapy, an untargeted therapy, was historically the only effective treatment option. TNBC accounts for
approximately 15% of breast cancers, carries a higher risk of early recurrence and metastasis, and is associated with worse survival outcomes than other subtypes, in part because of these limited options.
Nunnery framed the shift: “For so long, because we have no targets to go after, we always just had chemotherapy; now, for the first time, we’re actually getting to start to use different types of treatments.”
A: Immunotherapy harnesses the body’s own immune system rather than an external cytotoxic drug, and checkpoint inhibition has been paradigm-shifting in oncology, Kang explained. PD-L1 is a checkpoint protein that can be expressed on a cancer cell and acts as an immune brake.
“When an immune cell or a T cell binds that protein, it turns that T cell off,” she said. “If we can shut off that checkpoint or immune brake, we can keep the body’s immune system online, alert, and vigilant, and potentially help fight the cancer.”
Only patients whose tumors express PD-L1 are candidates for checkpoint inhibition. In practice, this is a minority of patients, so most with newly metastatic TNBC are not immunotherapy candidates, Nunnery noted.
A: Since immunotherapy revs up the immune system in a way that cannot be finely controlled, patients with autoimmune diseases or immune dysregulation are at increased risk of complications, Kang said. A patient with lupus or rheumatoid arthritis who already requires immunosuppression to control their disease, for example, could be harmed by checkpoint inhibition. She added that a prior severe immune-related toxicity, such as serious inflammation of a vital organ like the lungs or heart during earlier immunotherapy, would generally make repeating it not worth the risk. These patients, along with the PD-L1–negative majority, are the population addressed by the frontline ADC trials.
A: An ADC is “a smartly engineered delivery system of chemotherapy,” Kang said. It is an antibody directed at a protein preferentially expressed on cancer cells, conjugated to a cytotoxic payload that it delivers primarily to the tumor, allowing more effective and better-targeted therapy than chemotherapy given straight into a vein.
The rationale for using the most effective agent first is crucial in this disease. The median overall survival (OS) in mTNBC is less than 2 years, and many patients never reach a second line of therapy.
“We don’t want to lose our patients that early,” Nunnery emphasized. “If we can give our best drug first and give them more time, that’s always what we’re hoping.”
A: The phase 3 ASCENT-03 trial (NCT05382299) randomly assigned patients with previously untreated mTNBC who were ineligible for PD-(L)1 inhibitors 1:1 to receive sacituzumab govitecan or investigator’s choice of chemotherapy.1 Sacituzumab govitecan significantly improved the primary end point of progression-free survival (PFS; HR, 0.62; 95% CI, 0.50-0.77; P < .0001) and induced more durable responses than chemotherapy.
Two trial design features shaped the experts’ interpretation of these results. First, the trial required a disease-free interval of at least 6 months, which Kang viewed as clinically double-edged.2
“From a study-design perspective, you want to limit who gets on study to patients who are likely to benefit; but from a clinical-need perspective, it’s absolutely the other way around—you want to have these options available for the patients with the highest-risk disease,” she said.
Second, patients in the chemotherapy arm could cross over to receive an ADC upon progression, which Kang praised as ethical but noted would blunt any OS signal, since most patients in the control arm went on to receive subsequent ADC therapy.
Notably, the safety profile of sacituzumab govitecan in this trial was consistent with prior experiences with the ADC, with adverse effects [AEs] including neutropenia, nausea, diarrhea, and alopecia. Notably, alopecia was driven partly by the non–alopecia-inducing gemcitabine/carboplatin option in the control arm.
A: The phase 3 TROPION-Breast02 trial (NCT05374512) enrolled patients with previously untreated, locally recurrent inoperable or metastatic TNBC for whom immunotherapy was not an option, randomly assigning them 1:1 to receive Dato-DXd or investigator’s choice of chemotherapy.3 Dato-DXd improved both dual primary end points, including median PFS (HR, 0.57; 95% CI, 0.47-0.69; P < .0001) and median OS (HR, 0.79; 95% CI, 0.64-0.98; P = .0291).
Notably, TROPION-Breast02 imposed no minimum disease-free interval, which Kang called “a brave decision” that appropriately allowed the enrollment of the highest-need patients with rapidly recurrent disease. The trial also did not build in crossover, strengthening the OS analysis. The single-agent carboplatin option in the control arm drew some criticism as atypical of US practice, but Nunnery noted the global trial reflects the different practice standards around the world, and that the results remain valid.
A: Despite sharing a target and payload class, the 2 agents differ markedly in tolerability, Kang said. Sacituzumab govitecan’s dominant AEs are hematologic, along with nausea, diarrhea, and alopecia. Dato-DXd, by contrast, is rarely associated with clinically significant cytopenias but carries a learning curve around 2 distinctive toxicities: stomatitis and ocular surface effects, mostly dry eye but occasionally conjunctivitis or keratitis. The mouth sores can be debilitating if not prevented, according to Kang.
The key with Dato-DXd use is AE prophylaxis, Nunnery emphasized.
“For the stomatitis, the recommendation is that patients use a steroid mouthwash 4 times a day, and for the dry eye, it’s recommended that patients use a preservative-free lubricating eye drop 4 times a day,” she explained.
A baseline eye examination, which an optometrist can perform, is advised at or before Dato-DXd initiation, and the use of contact lenses should be avoided. Reassuringly, in TROPION-Breast02 ocular events were largely low grade and reversible with drug discontinuation without evidence of permanent eye damage.
A: Kang said she does not perceive a meaningful efficacy difference between the agents, noting that although the response rate was numerically slightly higher with Dato-DXd and the control arm performed worse in TROPION-Breast02, the durations of response between the ADCs were similar, and the trials differed in their disease-free-interval requirements.
“I don’t think there’s a true activity difference between the 2 agents, and I’ve mostly been selecting based on AE profile and what schedule makes the most sense for the patient,” she said.
Schedule and logistics are meaningful differentiators. sacituzumab govitecan is given for approximately 2 weeks on and 1 week off in 3-week cycles, with infusions of about 90 minutes and frequent need for interval growth factor support. Alternatively, Dato-DXd is given once every 3 weeks, with infusions as short as 30 minutes, but it requires the patient to maintain steroid mouthwash and eye drop use several times daily throughout treatment. Having more than one option is a welcome development, Nunnery said, though “it can make it a little more challenging to talk through everything with the patient.” Both agreed the agents can be sequenced across lines even though the trials studied them in the first-line setting.
A: On May 22, 2026,
On June 24, 2026,
