An additional safety analysis of the global phase III DESTINY-Breast05 trial evaluating the HER2-directed antibody-drug conjugate fam-trastuzumab deruxtecan-nxki (T-DXd) has further established the pulmonary safety profile of T-DXd with radiotherapy, according to Michael Untch, MD, PhD, of Helios Hospital Berlin-Buch in Berlin, Germany, who presented the findings at the 2026 ASCO Annual Meeting.¹

[This further analysis] complements the superior efficacy of T-DXd and supports its adoption as a new postneoadjuvant standard of care.

— MICHAEL UNTCH, MD, PhD

Tweet this quote

The primary results of DESTINY-Breast05² led to the U.S. approval of T-DXd as adjuvant treatment of residual invasive HER2-positive breast cancer following neoadjuvant trastuzumab and taxane-based treatment. In that study, T-DXd demonstrated superior efficacy over ado-trastuzumab emtansine (T-DM1) in patients with residual invasive disease after neoadjuvant therapy and high risk of recurrence, yielding a hazard ratio for disease-free survival events of 0.47 (P < .001).

The current analysis focused on the incidence of adjudicated interstitial lung disease (ILD) and investigator-reported radiation pneumonitis, primarily as a first event. This further analysis “complements the superior efficacy of T-DXd and supports its adoption as a new postneoadjuvant standard of care,” Dr. Untch said.

Key Findings

The incidence of adjudicated drug-related ILD as a first event for patients receiving radiotherapy sequentially and for those receiving it concurrently was 10.7% and 9.6%, respectively, in the T-DXd arm and 2.6% and 1.0% in the T-DM1 arm.

“Importantly, adjudicated drug-related ILD and investigator-reported pneumonitis did not increase the risk or severity of subsequent adjudicated drug-related ILD or investigator-reported pneumonitis,” Dr. Untch said, noting all subsequent events were low-grade.

The incidence of ILD as a second event after a first ILD event was 0% and 0.5% with sequential and concurrent radiotherapy, respectively, in the T-DXd arm and 0.4% and 0% in the T-DM1 arm. Subsequent radiation pneumonitis events occurred in 0.9% and 0.2%, respectively, in the T-DXd arm and 0.4% and 0% in the T-DM1 arm.

Investigator-reported radiation pneumonitis as a first event occurred in the T-DXd arm in 34.5% of patients receiving sequential radiation and 29.2% receiving concurrent radiation; these rates in the T-DM1 arm were 37.4% and 26.7%, respectively.

As a second event after the first report of radiation pneumonitis, for sequential and concurrent radiation delivery, these rates were 2.2% and 0.9%, respectively, for ILD in the T-DXd arm and 1.1% and 0% in the T-DM1 arm. Subsequent pneumonitis occurred in 0% and 0.7%, respectively, with T-DXd and in 1.1% and 0% with T-DM1.

Other Key Findings

Adjudicated drug-related ILD cases were mostly low-grade and reversible with treatment according to guidelines; at the time of the analysis, these had resolved or were resolving in 78% of the T-DXd arm and 92% of the T-DM1 arm, independently of the timing of adjuvant radiotherapy.

Radiation pneumonitis events were grade 1 or 2 and, at the time of the analysis, had resolved or were resolving in 54% and 62%, respectively; mean duration of these events was similar between arms.

In the T-DXd arm, drug-related ILD events were fatal for one patient receiving sequential radiation and one receiving concurrent radiation.

Incidence Higher in Some Subgroups

In both treatment arms, the incidence of ILD and radiation pneumonitis was higher in patients from Japan and in patients with worse baseline renal function. For T-DXd, for Japan vs other Asian countries and vs the global population (including Japan), the incidence of ILD was 14.9%, 8.3%, and 8.9%, respectively; for investigator-reported radiation pneumonitis, rates were 47.1%, 38.5%, and 27.4%, respectively.

A similar pattern was seen in the T-DM1 arm. ILD incidence for Japan vs other Asian countries and vs the global population was 6.7%, 1.3%, and 1.2%, respectively. For radiation pneumonitis, incidence rates were 45.0%, 38.7%, and 27.3%, respectively.

While the incidence of ILD with T-DXd was 9.7% in patients with normal baseline function and 8.5% in those with mild impairment, it rose to 14.3% in patients with moderate renal impairment. With T-DM1, it similarly increased from 1.0% in the normal group and 2.2% in the mildly impaired group to 5.0% in the setting of moderate impairment. Radiation pneumonitis was apparently not affected by renal function.

Monitoring Schedule

Dr. Untch described the protocol-specific guidance on CT requirements for identifying ILD and radiation pneumonitis in patients receiving adjuvant radiotherapy.All underwent low-dose noncontrast chest CT at baseline and prior to infusion at cycles 3, 7, and 11, then at follow-up around 40 days. Patients with any signs or symptoms underwent additional chest CTs; those with scans showing ILD or pneumonitis were referred for blinded central adjudication. Causal association between study drug and ILD/radiation pneumonitis events was based on the timing and location of radiographic abnormalities relative to the radiation treatment.

Dr. Untch described these guidelines for treatment:

• Grade 1: For drug-related ILD, interrupt T-DXd and consider steroids; restart T-DXd only after full resolution. For radiation-related pulmonary toxicity, maintain the dose.
• Grade 2: For drug-related ILD (symptomatic with radiographic abnormalities), permanently discontinue T-DXd and initiate steroids. For radiation-related pulmonary toxicity, interrupt treatment and manage per standard of care (eg, steroids) until recovery to grade 1 (asymptomatic).
• Grade 3 or 4: For either condition, discontinue treatment and initiate steroids.

Comments from Co-Investigator

DESTINY-Breast05 co-investigator Jame Abraham, MD, FACP, Enterprise Chair of the Department of Hematology and Medical Oncology at Cleveland Clinic and Professor of Medicine at the Cleveland Clinic Lerner College of Medicine, said the purpose of the current analysis was to help clinicians recognize and manage ILD and radiation pneumonitis optimally in the curative setting. “We need to make sure that in the adjuvant setting we are not causing any damage with T-DXd, and here our concern is mostly ILD and pulmonary toxicity,” he said. “We have seen grade 5 toxicity, and it’s important for clinicians to know whether there are risk factors. Also, how do we diagnose and manage ILD?”

Jame Abraham, MD, FACP

As previously documented, radiation is a compounding factor for ILD and pulmonary toxicity. In the analysis, he noted, one new risk factor did emerge: moderate renal impairment at baseline. Clinicians might think twice before using T-DXd in patients with creatinine clearance between 30 and 59 mL/min/1.73 m², Dr. Abraham advised.

Patients of Japanese ethnicity are another subset with a higher incidence of events; therefore, such patients deserve additional consideration when either T-DXd or T-DM1 are possible treatment choices, he said. Differences in pharmacogenetics among ethnic groups can influence toxicity risks and may well be responsible for this higher risk, but at this time, this discrepancy from the other patient subsets remains unexplained.

“Overall, for the North American patient population, there were no surprising findings in this analysis,” Dr. Abraham said. “We have the same consistent findings as previous studies.” 

DISCLOSURE: Dr. Untch has received honoraria from AstraZeneca, Daiichi Sankyo, Lilly, Eurbio Scientific, Myriad Genetics, Novartis, Pfizer, Roche, Sanofi-Aventis, Menarini Stemline, and MSD Merck and served on advisory boards for AstraZeneca, Daiichi Sankyo, Lilly, MSD Merck, Myriad Genetics, Novartis, Pierre Fabre, Pfizer, Gilead, Roche, Sanofi-Aventis, Seagen, Menarini Stemline, and BeOne Medicines Germany. Dr. Abraham has served on a scientific advisory board for ThinkBio.Ai; has served as a Medical Advisory Board member for OpenEvidence; has received institutional research support from Daiichi Sankyo; and serves as the National Study Chair and institutional Principal Investigator for NSABP B-60 (DESTINY-Breast05).

REFERENCES

1. Untch M, Shao Z, Huang C-S, et al: Secondary safety analysis of trastuzumab deruxtecan (T-DXd) vs trastuzumab emtansine (T-DM1) in DESTINY-Breast05: Clinical and demographic risk factors of interstitial lung disease (ILD) and radiation pneumonitis (RP). 2026 ASCO Annual Meeting. Abstract 516. Presented June 1, 2026.

2. Loibl S, Park YH, Shao Z, et al: Trastuzumab deruxtecan in residual HER2-positive early breast cancer. N Engl J Med 394:845-857, 2026.

EXPERT POINT OF VIEW

The ASCO Post interviewed two breast cancer specialists who still had some concerns about fam-trastuzumab deruxtecan-nxki (T-DXd)–associated interstitial lung disease (ILD). William Sikov, MD,FACP, FNCBC, Director of Breast Medical Oncology at Women & Infants Hospital, Providence, Rhode Island, and Professor of Medicine at the Warren Alpert Medical School of Brown University, noted that while DESTINY-Breast05 is a “strongly positive study,” and indicates the ability of this agent to reduce the risk of cancer recurrence in high-risk patients, the incidence of ILD and the two drug-related deaths remain concerning to him.¹ He believes the risk for ILD would almost certainly be reduced with a shorter duration of treatment.

William Sikov, MD, FACP, FNCBC

“The question is, how many cycles do you need to obtain the desired benefit? Because we know that while ILD can occur at any point during treatment with T-DXd, the risk continues to increase with time,” Dr. Sikov said. “While we don’t know the optimal number of cycles of T-DXd in this setting, 14 is probably not necessary.” He pointed out that DESTINY-Breast11 showed a substantial benefit from the administration of T-DXd in the neoadjuvant setting in patients with bulky disease, despite limiting treatment to 4 cycles.

A study that stratifies T-DXd treatment duration according to the extent of residual disease could be informative, he suggested. “The question is whether we can give less and still see this sizable benefit, and I would be surprised if the answer is not ‘yes,’” he said. “I believe that giving less drug would probably result in improved safety without sacrificing efficacy.”

Echoing this sentiment was John Cole, MD, FACP, Director of Clinical Cancer Research at Ochsner Health in New Orleans, who stated, “The risk of fatal complications is important in any setting, but especially when we are treating patients in the adjuvant or neoadjuvant space. When there is a movement to de-escalate treatments, especially in the HER2-positive space, it would make sense to drop the T-DXd after maybe six cycles or so, if possible, because clearly the risk of side effects will increase with more treatment.”

John Cole, MD, FACP

Dr. Cole also noted that the monitoring schedule in DESTINY-Breast05 was likely more aggressive than that undertaken in typical clinical practice (see “Monitoring Schedule,” on page 4). “I think most clinicians will perform a CT scan at the first sign of a potential problem, such as cough or shortness of breath,” he said.

While the subsequent analysis of DESTINY-Breast05 did not reveal additional safety concerns and provided reassurance that a first event did not increase the risk of a second or more severe event, these specialists advised others to remain vigilant when administering T-DXd as it is currently prescribed. To trial investigators, their suggestion was to conduct studies using T-DXd for shorter durations. 

DISCLOSURE: Dr. Sikov reported no conflicts of interest. Dr. Cole was a site principal investigator for DESTINY-Breast05.

REFERENCE

1. Untch M, Shao Z, Huang C-S, et al: Secondary safety analysis of trastuzumab deruxtecan (T-DXd) vs trastuzumab emtansine (T-DM1) in DESTINY-Breast05: Clinical and demographic risk factors of interstitial lung disease (ILD) and radiation pneumonitis (RP). 2026 ASCO Annual Meeting. Abstract 516. Presented June 1, 2026.