Neoadjuvant treatment with the HER2-directed antibody-drug conjugate trastuzumab deruxtecan-nxki (T-DXd), given alone, with endocrine therapy, or followed by chemotherapy, produced pathologic complete response rates (pCR) in the single digits among patients with hormone receptor (HR)-positive, HER2-low early breast cancer in the phase II TALENT trial, results of which were reported at the European Society for Medical Oncology (ESMO) Congress 2026 (Abstract LBA23).
“Neoadjuvant T-DXd demonstrated modest clinical activity in HER2-low HR-positive localized breast cancer,” said Sara Hurvitz, MD, Senior Vice President and Director of the Clinical Research Division at Fred Hutchinson Cancer Center and Head and Professor of the Division of Hematology and Oncology at the University of Washington School of Medicine, who presented the study. “The addition of endocrine therapy to T-DXd did not improve [pathologic complete response], and sequencing T-DXd with multiagent chemotherapy only marginally improved it.”
Translational work from the study, however, offered insights into how T-DXd might best be used in this patient population. For instance, HER2 protein levels were shown to fall during treatment, hinting that baseline HER2 levels may help identify responders, she indicated.
T-DXd is established in metastatic HER2-low breast cancer, but its activity in localized disease had not been defined. The hope is that T-DXd would be a good alternative to standard neoadjuvant anthracycline- and taxane-based chemotherapy, which yield low pCR rates in HR-positive disease, possibly lower still in HER2-low tumors. Chemotherapy also results in modest radiologic response, frequent dose reductions, and toxicity that can be irreversible or life-threatening, Dr. Hurvitz said, as background to the study.
TALENT Design
TALENT enrolled men and pre- or postmenopausal women with HR-positive, HER2-low, stage II–III operable breast cancer. The first 58 patients were randomly assigned to one of two arms: Arm A: T-DXd alone; Arm B: T-DXd plus anastrozole, with a GnRH analog for men and premenopausal women.
Because of concern that a single antibody-drug conjugate might be insufficient for high-risk disease, a third, nonrandomized arm was added later: Arm C: four cycles of T-DXd followed by four cycles of docetaxel and cyclophosphamide.
Patients initially received six cycles; after 29 had enrolled, investigators increased this to eight cycles after near-complete responses were seen at six. Adjuvant therapy after surgery was at investigator discretion.
At least two pCRs were required for expanded enrollment into any arm. Arm A met that bar and enrolled 16 more patients; arm B did not. Arm C also met the bar, enrolling 58 patients in total.
The primary endpoint was pCR in the breast and lymph nodes, defined as residual cancer burden 0. No formal statistical comparisons were made between the arms. Tumor tissue, blood for circulating tumor DNA, and imaging were collected at several timepoints.
The final study population included 46 patients in arm A, 29 patients in arm B, and 58 patients in arm C. Baseline characteristics were broadly similar across arms, although arm C, which was enrolled later and not randomized, included fewer postmenopausal patients and more patients with node-positive disease. The population was deemed high-risk: most patients had a baseline Ki-67 ≥ 30%, and 44% had grade 3 tumors (poorly differentiated). Treatment was discontinued early by four patients in arm A, four in arm B and six in arm C.
Rates of pCR Low
Whereas objective response rates were relatively high, pCR rates were low in every arm (Table 1). Adding endocrine therapy did not improve pCR rates, and the addition of chemotherapy after T-DXd yielded only a marginal boost. When near-complete responses (residual cancer burden 0 or 1) were counted, the rates remained similar across the three arms.
TABLE 1: Response to T-DXd Neoadjuvant Therapy
|
Treatment Arm |
Objective Response |
Pathologic Complete Response |
|
A: T-DXd alone (%) |
62 |
7 |
|
B: T-DXd + anastrozole (%) |
66 |
3 |
|
C: T-DXd followed by chemotherapy * (%) |
72 |
9 |
*docetaxel/cyclophosphamide
Treatment duration appeared to matter. “No patients in arms A or B experienced a pathologic complete response with only six cycles of therapy,” Dr. Hurvitz said. The pCRs in those arms occurred only among patients who received eight cycles.
Side effects largely mirrored T-DXd’s known profile in metastatic disease. Nausea, fatigue, and loss of appetite were the most common treatment-related events in all three arms. Patients in arm C had more mucositis and peripheral neuropathy, as expected with chemotherapy, as well as more grade ≥ 3 treatment-related events and dose reductions. Three patients developed interstitial lung disease, all grade 2. One patient in arm A died of severe gastrointestinal toxicity and a myocardial infarction.
TROP2 and HER2 Levels Informative
The translational program produced the trial’s most intriguing signals. Central pathology review at the University of Southern California found that estrogen receptor expression did not change with any of the treatments. Ki-67 fell significantly from baseline to cycle 2, and from baseline to surgery, in all arms, most notably in arm B, which included anastrozole, she reported.
Investigators at Yale University then measured TROP2 and HER2 protein in arms A and B using TROPLEX, a quantitative immunofluorescence assay standardized to cell lines with mass spectrometry–measured targets. The two proteins behaved differently; TROP2 levels were similar between arms at baseline and did not change after T-DXd; HER2 levels were similar at baseline but dropped significantly after T-DXd in both arms.
Baseline HER2 protein also tracked with pCR. “Interestingly, baseline HER2 protein level does appear to be associated with the achievement of pCR (P = .0329),” Dr. Hurvitz said, cautioning this observation was based on a small sample size. The finding raises the possibility that quantitative HER2 measurement could refine patient selection beyond conventional immunohistochemistry scoring.
Analyses of event-free survival, breast-conserving surgery rates, additional biomarkers, patient-reported outcomes, and ovarian function are ongoing and will be reported later.
Expert Point of View
“T-DXd has been evaluated in the TALENT trial as neoadjuvant therapy in patients with HER2-low HR-positive early breast cancer across three cohorts. Only arm A and arm C reached the prespecified criteria for stage two expansion, and overall the pCR rates remained low and comparable to those historically achieved with standard chemotherapy. However, TALENT offers some very interesting clinical and translational findings,” said invited discussant Barbara Pistilli, MD, Head of the Breast Cancer Unit at Gustave Roussy in Villejuif, France.
Dr. Pistilli summed up the key clinical takeaways:
- Pathologic complete response rates were higher with eight cycles of T-DXd than with six
- Four cycles of T-DXd followed by four of docetaxel/cyclophosphamide seems similar to eight of T-DXd monotherapy, but grade ≥3 toxicities and treatment discontinuations are more common
- The addition of the aromatase inhibitor does not seem to improve T-DXd activity.
She noted that TALENT’s results do not differ from the prior SOLTI-VALENTINE study that reported a low pCR rate when replacing the entire part of the neoadjuvant chemotherapy with six cycles of the antibody-drug conjugate patritumab deruxtecan (HER3-DXd) and, similar to TALENT, the addition of letrozole did not increase pCR. Low pCR rates were also seen with four cycles of Dato-DXd with or without durvalumab and with T-DXd with the novel bispecific anti–PD-1/anti-TIGIT antibody rilvagostomig in unselected luminal tumors in I-SPY trials.
The suggestion is that immune-enriched HR-positive breast cancers may derive greater benefit from the combination of an antibody-drug conjugate with immunotherapy than with endocrine therapy, “making this combination worthy of further evaluation in biomarker-selected luminal tumors,” she added.
“TALENT also offers some very interesting translational findings thanks to the longitudinal sample collection,” she said, focusing on the results of the quantitative assessment of TROP2 and HER2. It is important to quantitatively assess the target, she explained, because while target expression may correlate with antibody-drug conjugate activity, the associations are weak, reflecting the limited analytical precision of semi-quantitative immunohistochemistry (IHC) scoring. Quantitative high-sensitivity HER2 and TROP2 multiplex immunofluorescence assay assesses both the absolute target levels and relative expression within similar breast cancer populations.
The key findings from this assessment were these:
- HER2 protein level at baseline was associated with pCR
- Quantitative HER2 assessment may better identify patients most likely to benefit from T-DXd
- HER2 expression, not TROP2, decreased following T-DXd treatment.
“However, this assay and this study so far do not address the question if this depends on the target-specific selective pressure or a transient HER2 downregulation,” she added.
DISCLOSURE: Dr. Hurvitz had personal disclosures for Gilead, AstraZeneca, Breast Cancer Foundation New Zealand, Elsevier, McGraw, Sage, Springer, Wiley, Wolters Kluwer, and numerous Certified Medical Education entities. Dr. Pistilli had personal disclosures for Daiichi Sankyo, Pierre Fabre, AstraZeneca, Gilead, MSD, Pfizer, and Novartis.
