TUXEDO-1 is an open-label, noncomparative, single-center, single-arm, phase 2 study evaluating the efficacy and safety of trastuzumab deruxtecan in HER2-positive breast cancer patients with newly diagnosed or progressing brain metastases who are deemed to be candidates for systemic therapy conducted at a tertiary care center. The present study was conducted in accordance with the Declarations of Helsinki and Good Clinical Practice and was approved by the local ethics committee of the Medical University of Vienna (EC no. 1359/2020). Written informed consent was obtained from each patient. None of the study participants received compensation for participation in the study. The trial is registered at ClinicalTrials.gov (NCT04752059) and the European Union Clinical Trials Register (EudraCT no. 2020-000981-41).

Patients

To be eligible for inclusion in the TUXEDO-1 trial, each patient had to fulfill all of the following criteria: histologically confirmed breast cancer; radiologically documented metastatic disease; HER2-positive as defined by IHC3+ and/or HER2/neu+ gene amplification; newly diagnosed brain metastases or brain metastases progressing after previous local therapy; measurable disease as defined by RANO-BM criteria35; no indication for immediate local treatment; no indication of leptomeningeal disease; Karnofsky’s index of performance status (KPS) > 70%/ECOG < 2; indication for systemic anti-HER2 treatment; previous exposure to trastuzumab and pertuzumab; previous exposure to T-DM1 allowed; life expectancy of at least 3 months; age ≥18 years; patient able to tolerate therapy and have adequate cardiac function (defined by baseline LVEF ≥50%); adequate bone marrow, liver and kidney function; adequate treatment washout period before enrollment, defined as: major surgery ≥4 weeks, radiation therapy ≥4 weeks, chemotherapy, small-molecule targeted agents, anticancer hormonal therapy ≥3 weeks, antibody-based treatment ≥4 weeks; and patient capable of understanding the purpose of the study and has given written informed consent. Patients who fulfilled any of the following criteria were excluded: metastatic breast cancer other than HER2-positve disease; use of any investigational agent within 28 d before initiation of treatment; history of malignancy other than squamous cell carcinoma, basal cell carcinoma of the skin or carcinoma in situ of the cervix within the last 3 years, including contralateral breast cancer; major surgery, other than diagnostic surgery, within the last 4 weeks; indication for immediate local therapy by local standard; leptomeningeal involvement; other anticancer therapy, including cytotoxic, targeted agents, immunotherapy, antibody, retinoid or anticancer hormonal treatment; concomitant radiotherapy; previous radiotherapy to the thorax other than breast irradiation or irradiation of bone metastases; a history of uncontrolled seizures, central nervous system disorders or psychiatric disability judged by the investigator to be clinically significant and adversely affecting compliance to study drugs; clinically significant cardiac disease including unstable angina, acute myocardial infarction within 6 months before randomization, congestive heart failure (New York Heart Association III–IV), LVEF <50%, arrhythmia unless controlled by therapy, with the exception of extrasystole or minor conduction abnormalities, and long QT syndrome (QTc interval >450 ms); subjects who have current active hepatic or biliary disease (with the exception of patients with Gilbert’s syndrome, asymptomatic gallstones, liver metastases or stable chronic liver disease according to investigator assessment), including acute and chronic infections with hepatitis B and C; inadequate hematological status at baseline before study entry: dependency on red blood cell and/or platelet transfusions, absolute neutrophil count (segmented + bands) <1.0 × 109 l−1; platelets <100 × 109 l−1; inadequate kidney function: serum creatinine >1.5× upper normal limit; hepatic dysfunction: total bilirubin >1.5× upper normal limit (>3 in patients with liver metastases or known history of Gilbert’s disease); alanine transaminase, aspartate aminotransferase >3× upper normal limit (>5 in patients with liver metastases); serum albumin <2.5 g dl−1; international normalized ratio ≥1.5; clinically severe pulmonary compromise resulting from intercurrent pulmonary illnesses, including, but not limited to, any underlying pulmonary disorder (that is, pulmonary emboli within 3 months of the study enrollment, severe asthma, severe chronic obstructive pulmonary disease, restrictive lung disease, pleural effusion, and so on), and any autoimmune, connective tissue or inflammatory disorders with pulmonary involvement (that is, rheumatoid arthritis, Sjögren’s syndrome, sarcoidosis and so on) or previous pneumonectomy (subjects with bronchopulmonary disorders who require intermittent use of bronchodilators (such as albuterol) not excluded from the present study); patients with active opportunistic infections; known human immunodeficiency virus infection; concomitant treatment with chloroquine or hydroxychloroquine; and pregnant or lactating women. Women with childbearing potential must have a negative pregnancy test at screening; also excluded are women with childbearing potential, including women whose last menstrual period was <1 year before screening, unable or unwilling to use adequate contraception from study start to 1 year after the last dose of protocol therapy. Acceptable contraception methods included the application of an intrauterine device, barrier method or total abstinence. Also included are patients with known hypersensitivity to trastuzumab, patients unable to provide written informed consent, patients with known substance abuse or any other medical conditions such as clinically significant cardiac or psychological conditions, which may, in the opinion of the investigator, interfere with the subject’s participation in the clinical study or evaluation of the clinical study results, and patients requiring concomitant use of chronic systemic (intravenous or oral) corticosteroids at doses >4 mg of dexamethasone per day or other immunosuppressive medications except for managing AEs (inhaled steroids or intra-articular steroid injections are permitted in the present study).

Study procedures

In this trial, trastuzumab deruxtecan was administered at the standard dose of 5.4 mg per kg bodyweight on day 1 of each cycle once every 3 weeks until progression, unacceptable toxicity or withdrawal for any other reason (Fig. 1). Before the first administration of trastuzumab deruxtecan, cranial magnetic imaging resonance, a bone scan and a computed tomography scan of the chest and abdomen were conducted with further workup if indicated. Staging investigations were to be repeated before the third and fifth treatment cycles and every 9 weeks thereafter, or whenever symptoms of disease progression occurred. The trial protocol provides a detailed overview of study procedures at baseline and during the study.

If patients progressed on study therapy, they entered survival follow-up. If treatment was discontinued for any reason other than progression and no alternative anticancer therapy was initiated and consent not withdrawn, patients were eligible for analysis of PFS without censoring at the time of study drug discontinuation. If alternative anticancer therapy was initiated at the time of trastuzumab deruxtecan discontinuation, patients entered survival follow-up. If patients were withdrawn within the first 9 weeks of the study (that is, before the first response evaluation) for reasons other than progression or death, they were replaced.

Endpoints

The primary endpoint of the TUXEDO-1 study was the rate of best responses of brain metastases at any radiological assessment after the administration of at least one cycle of trastuzumab deruxtecan. Objective response was defined as complete remission, partial remission, stable disease and progressive disease according to the RANO-BM criteria, as determined by the central assessment of a single board-certified neuroradiologist35. Secondary endpoints consisted of: Clinical Benefit Rate in the central nervous system (CBR CNS as defined by RANO-BM; complete remission/partial remission/stable disease ≥6 months), extracranial response rate defined as complete remission, partial remission, stable disease, progressive disease according to RECIST 1.1 criteria36, PFS defined as the interval from study inclusion until progression or death, time to WBRT defined as the interval from study inclusion until WBRT, overall survival defined as the interval from study inclusion until death, safety and QoL as assessed with the EORTC QLQ-c30 questionnaire, the brain-specific tool (BN20), and the breast-specific tool (BR45). For ancillary biomarker studies, blood samples (3 ml of EDTA, 3 ml of serum) were drawn before first administration of trastuzumab deruxtecan at cycles 1 and 4 and at EOT. The biomarker substudy of TUXEDO-1 aimed to investigate changes in the extent of metastases-induced brain damage in patients with and without response to therapy by measuring the levels of sNSE and sS100, two proteins constitutively expressed in the human brain and established markers of brain damage in neuro-oncology and ischemic brain damage37,38,39. Both markers were quantified by means of electrochemiluminescence assays according to standard operating procedures on cobas e801 analyzers (Roche Diagnostics) by an International Organization for Standardization 15189:2012-accredited medical laboratory (Department of Laboratory Medicine, Medical University of Vienna).

Statistical analysis

TUXEDO-1 was designed as a phase 2 study evaluating the ability of trastuzumab deruxtecan to induce objective responses in patients with HER2-positive metastatic breast cancer and with newly diagnosed or progressive brain metastases based on a Simon’s two-stage design40. A response rate of ≤25% was considered to be of no clinical interest whereas a response rate of ≥61% was considered to be clinically relevant. The null hypothesis that the true response rate was 25% was tested against a one-sided alternative. Based on these assumptions, six patients were to be accrued in the first stage. If at least three responses were observed in the first stage, nine additional patients were to be accrued for a total number of fifteen patients. The null hypothesis could be rejected if seven or more responses were observed in these fifteen patients. This design yields a type I error rate of 5% and a power of 80% to reject the null hypothesis when the true response rate is 61%.

The response rate was analyzed on the ITT principle, wherein all patients who had received at least one dose of the study drug were included in the analysis. Responses were summarized using frequency counts and percentages with 95% CIs. Fisher’s exact test was used for the comparison of differences in response rates between patients with newly diagnosed, untreated brain metastases and brain metastases progressing after previous local therapy. PFS and overall survival were estimated using the Kaplan–Meier product limit method.

Safety and tolerability of treatment in terms of hematological and nonhematological side effects were assessed by the investigators at each visit and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) v.5.0. SAEs were defined according to the International Conference on Harmonization Good Clinical Practice guidelines. AEs were summarized using frequency counts and percentages. All patients who were eligible for the present study and received at least one dose of study drug were included in the safety analysis.

QoL was assessed at day 1 of cycles 1, 3 and 5 and every 9 weeks thereafter. A final QoL assessment was conducted at the first survival follow-up at 3 months after EOT. Changes from baseline were analyzed using a linear mixed-effect model and separately displayed for the overall patient population and for the respective responder and nonresponder groups. Data were expressed as the mean ± s.e.m.

For the biomarker substudy, sNSE and sS100 levels were compared between responders and nonresponders using Mann–Whitney U-tests. Two-sided P values <5% were considered statistically significant. No correction for multiple testing was performed due to the hypothesis-generating design of this biomarker substudy41.

Statistical analysis was conducted using R 4.1.3. and IBM SPSS Statistic v.28.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.