Mark Pegram, MD, and Sara Tolaney, MD, review the DESTINY-Breast03 study design, clinical efficacy, and safety data.
Mark Pegram, MD: DESTINY-Breast03 is a phase III study investigating T-DXd [trastuzumab deruxtecan] versus T-DM1 [ado-trastuzumab emtansine] in patients previously treated with trastuzumab and a taxane. The key eligibility criteria were: HER2 [human epidermal growth factor receptor 2]-positive metastatic or unresectable locally advanced [disease], previously treated with trastuzumab and a taxane in the advanced or metastatic setting, or progression during or within 6 months after completed adjuvant therapy including trastuzumab with a taxane. There were 500 patients, randomized 1:1 to treatment with T-DXd at 5.4 milligrams per kilogram on IV every 3 weeks versus T-DM1 at the FDA approved dosing schedule of 3.6 milligrams per kilogram IV every 3 weeks. The key stratification variables included hormone receptor status, prior treatment with pertuzumab, and history of visceral metastasis. These patients were well balanced for baseline clinical pathological and demographic factors including patient age; median about 54. World region; about 20% European, 60% Asian, 6.5% North American, and 15% rest of the world. Hormone receptor status; approximately 50/50. HER2 status; almost 90% had IHC3 plus, and about 10% had IHC2 plus with in situ hybridization amplification. ECOG [Eastern Cooperative Oncology Group scale of performance] status; all the patients had very good to excellent ECOG status. About 20% of the patients had stable or treated brain metastasis and visceral disease was common in this cohort with about 70% in both arms. The primary end point of this study is a progression free survival [PFS] by a blinded independent central review. The key secondary end point is overall survival. Secondary end points also included objective response rate, PFS as recorded by the investigators, and response duration. It’s important to point out that this was not a true second line only population. Only about half the patients in the study were second line. It is notable, however that about 100% had prior trastuzumab. That was an eligibility criteria. Sixty percent had prior pertuzumab and 16% had received a HER2 tyrosine kinase inhibitor.
Sara Tolaney, MD: Destiny-Breast03, which compared trastuzumab deruxtecan to T-DM1 really demonstrated very amazing results. We saw a dramatic improvement in progression-free survival that was about 6.8 months in the T-DM1 arm and was not reached in the T-DXd arm. This was consistent with a hazard ratio of 0.28. A stark difference in progression-free survival favoring T-DXd over T-DM1. What’s interesting is at this early time point in patients who’d only had a median follow-up of about 15 months we’re also seeing a very strong trend towards overall survival benefit. We saw that there was an improvement in overall survival with a hazard ratio of about 0.56. This is pretty unique to see such a strong trend towards survival at such an early time point. Making one believe that it will reach statistical significance with just a little bit of longer follow-up. What’s interesting is that if you look at the subgroups from this trial, you can see that all subgroups of patients really benefit, whether they had hormone receptor-positive disease [or] hormone receptor-negative disease. Interestingly, even in the patients who had a history of stable treated brain metastases, you are also seeing benefit favoring trastuzumab deruxtecan over T-DM1, with the hazard ratio in that subgroup of about 0.38. No subgroup did better with T-DM1. Everyone did dramatically better with T-DXd in this trial.
One other interesting thing about trastuzumab deruxtecan that’s really unique is that it has had activity not just in metastatic HER2-positive breast cancer, but in fact, also in metastatic HER2-low-positive breast cancer. This is a new term to many of us because we usually think of breast cancer as being HER2-positive or HER2-negative. HER2-low-positive has meant that there’s a little bit of HER2 expressions, a one-plus or two-plus staining, but that the tumor is not FISH [fluorescence in situ hybridization] amplified. Even in these patients who are not HER2-positive but have some HER2 expression, we are seeing robust response rates. Data from a basket study, looking at T-DXd in these HER2-low-positive breast cancer patients found a response rate that was about 40% and a progression-free survival of around 11 months. Given this very promising early data, there is a randomized registration study, DESTINY-Breast04. We’re hoping to see results from next year, which compared trastuzumab deruxtecan to chemotherapy—a physician’s choice—in patients who have metastatic HER2-low-positive breast cancer. These were patients who had received either 1 or 2 prior lines of chemotherapy in the metastatic setting. I think T-DXd could also become an option in the not too distant future, not just for HER2-positive disease but now actually maybe even for HER2-low-positive breast cancer. That would really change the space dramatically because it could then become an option for our hormone receptor-positive patients as well as our metastatic triple-negative patients. It is really nice to see that we can be expanding options for all of our patients.
Mark Pegram, MD: When you are considering a new efficacy analysis with a new phase II trial, it’s also important to consider safety signals since they’re paramount in as much as that metastatic disease treatment intent is indeed a palliative intent. Drug discontinuation was higher in the trastuzumab deruxtecan arm, about 13% compared to 5% for T-DM1. There were more chemotherapy like toxicities with trastuzumab deruxtecan, such as nausea, vomiting, cytopenia, fatigue, alopecia, etc. However, it’s important to remember that patients in the experimental arm of DESTINY-Breast03 were on treatment for a much longer time period than the T-DM1 control arm. Therefore, they had more time to accumulate cumulative toxicities. Whereas patients on the T-DM1 arm were more likely to discontinue the drug due to disease progression. The most common treatment emergent adverse events associated with treatment discontinuation for T-DXd was ILD [interstitial lung disease] or pneumonitis, about 8.2%, and for T-DM1, thrombocytopenia at 2.7%. The most common treatment emergent adverse events associated with dose reduction for T-DXd were nausea at 6.2%, neutropenia 3.5%, and for T-DM1, thrombocytopenia [at] 4.2% and transaminase elevation [at] 2.7% each. Finally, I’ll mention that the T-DXd alopecia rate was 25% grade 1 and 10% grade 2, which is lower that what we had seen in the DESTINY-Breast01 trial which was about 50% overall alopecia incidents. So maybe in it’s less heavily pretreated population alopecia is less as a result of a prior treatment.
Fortunately, in terms of adverse events with special interest, the most important adverse event for T-DXd has been drug-induced pneumonitis. In this phase III trial there was no grade 4 or 5 drug-induced pneumonitis as compared to the 2.7% grade 5 events that were seen in the DESTINY-Breast01 nonrandomized phase II trial. This is really important. It’s interesting to speculate why this may be. We know that prior treatment with cytotoxic agents is a risk factor in general for all forms of drug-induced pneumonitis. This is a less heavily treated patient population compared to DESTINY-Breast01, which had [a] median recall of 5 to 6 prior lines of therapy for metastatic disease. That may be one factor. Also, per protocol, the investigators were much more engaged to be on the lookout for drug-induced pneumonitis and they were required to have early intervention including interruption of drug treatment and consideration of steroid treatment depending on the severity of the interstitial lung changes. Consequently, it’s probably a combination of these factors that led to better outcomes and lower incidence of drug-related ILD in this phase III study as compared to DESTINY-Breast01. Now it’s important also to point out that the drug-induced pneumonitis was not 0 even in the T-DM1 arm. There was still about 2% all grade ILD incidence compared to 10.5% in the T-DXd arm.
Transcript edited for clarity.