Abstract
Background
Trastuzumab deruxtecan (T-DXd) has shown efficacy in human epidermal growth factor receptor 2 (HER2)-positive and HER2-low metastatic breast cancer (MBC), but real-world prognostic data in heavily pre-treated patients are limited. This study evaluates T-DXd’s real-world efficacy and identifies predictive factors.
Methods
Our study included 317 patients (HER2-positive: n = 173; HER2-low: n = 144) treated with T-DXd between January 3, 2020, and September 9, 2024. Outcomes included real-world progression-free survival (rwPFS), overall survival (rwOS), objective response rate (ORR), and safety. A prognostic index was developed using clinical parameters.
Results
In the HER2-positive cohort, ORR was 44.5%, with a median rwPFS of 10.5 months and rwOS of 29.9 months. Early-line T-DXd use (first or second line) improved rwPFS and rwOS compared with later lines (P < .0001), while prior tubulin-inhibitor antibody–drug conjugates (ADCs) were associated with inferior outcomes. In the HER2-low cohort, ORR was 24.3%, with a median rwPFS of 5.6 months, and rwOS of 18.5 months. Prior exposure to topoisomerase-inhibitor-payload ADCs significantly reduced rwPFS (1.97 vs. 5.97 months, P < .0001) and rwOS (5.77 vs. 18.9 months, P < .0001). Primary resistance rates were higher in HER2-low disease (24.3% vs. 12.7%, P = .011). Prognostic index incorporating treatment lines, HER2 expression, and prior ADC exposure effectively stratified patients into risk groups with distinct survival outcomes.
Conclusions
T-DXd shows clinical benefit in HER2-expressing MBC, with efficacy influenced by treatment line, HER2 expression, and prior ADC payload type. The prognostic index could aid in personalizing therapy, optimizing patient selection for T-DXd in real-world practice.
Keywords: metastatic breast cancer, HER2-low, HER2-positive, trastuzumab deruxtecan, real-world study, prognostic index
Implications for Practice.
This real-world study demonstrates trastuzumab deruxtecan' s (T-DXd) efficacy in human epidermal growth factor receptor 2 (HER2)-expressing metastatic breast cancer, while identifying crucial clinical predictors. For HER2-positive disease, early-line use showed superior outcomes. In HER2-low disease, prior topoisomerase inhibitor-based antibody–drug conjugates (ADCs) were associated with significantly reduced efficacy (median rwPFS 1.97 vs. 5.97 months), suggesting alternative sequencing strategies. The prognostic models incorporating HER2 status, treatment lines, and prior ADC exposure provide practical decision-making tools. These findings optimize T-DXd use in clinical practice, particularly given its expanding indications and the need to balance efficacy with toxicity management.
Introduction
In recent years, antibody–drug conjugates (ADCs) have revolutionized the treatment of breast cancer, particularly human epidermal growth factor receptor 2 (HER2)-positive metastatic breast cancer (MBC).1,2 Trastuzumab deruxtecan (T-DXd) has shown significant benefits over traditional therapies like lapatinib and capecitabine (LX) or trastuzumab and capecitabine (HX) after the failure of trastuzumab emtansine (T-DM1).3 The DESTINY-Breast 02 and 03 studies have demonstrated T-DXd’s efficacy, with median progression-free survival (PFS) of 16.4-18.2 months in heavily pre-treated patients.3-5 These findings have led to T-DXd being recommended as a second-line therapy for HER2-positive MBC since 2022.4,5
Despite these advancements, real-world data on T-DXd's efficacy, especially in heavily pre-treated patients, remain limited. The Italian DE-REAL study reported a median real-world progression-free survival (rwPFS) of 16 months for HER2-positive MBC patients, with 86% receiving T-DXd as a third or subsequent line of therapy.6 However, the reasons for primary resistance to T-DXd are still unclear. Our previous research suggested that prior exposure to topoisomerase inhibitor payloads might be a risk factor for primary resistance.7
T-DXd has also shown promise in HER2-low MBC, defined as 1+ on immunohistochemical analysis or 2+ with negative in situ hybridization results.8 The DESTINY-Breast 04 and 06 studies indicated that T-DXd could benefit patients with HER2-low expression.8,9 However, these studies were limited by their restriction to earlier lines of treatment. Real-world data suggest that heavily pre-treated patients may have lower response rates, with primary resistance as high as 40.4% in HER2-low disease.10
Given the expanding use of T-DXd, it is crucial to identify the optimal patient population and understand the factors influencing treatment outcomes. This study aims to explore the real-world efficacy of T-DXd in both HER2-positive and HER2-low MBC, identify risk factors for primary resistance, and develop a prognostic index to predict treatment outcomes.
Materials and methods
Study design and patients
This retrospective, multicenter, real-world study included patients with inoperable, locally advanced, or MBC who received at least 1 dose of T-DXd at the Sun Yat-Sen University Cancer Centre (SYSUCC), Guangdong Nongken Central Hospital, Jiangmen Wuyi Hospital of Traditional Chinese Medicine, and Shandong Cancer Hospital and Institute between January 3, 2020, and September 9, 2024. Inclusion criteria were (1) histologically confirmed HER2-positive or HER2-low breast cancer, inoperable, locally advanced, or metastatic disease at enrollment; (2) received at least 1 dose of T-DXd; (3) had available response assessments; and (4) had complete clinical profiles. The study protocol was approved by the ethical committee of the SYSUCC. The requirement for individual informed consent was waived by the committee due to the retrospective nature of the study.
Data collection and definitions
Data were extracted from the medical records and included patient demographics, tumor characteristics, treatment, standard laboratory tests, and imaging scans. In general, the patients were treated with T-DXd until disease progression, intolerable toxicity, or death based on the experience of physicians. T-DXd was generally administered according to their instructions.
HER2 expression was mainly detected by immunohistochemistry (IHC). The IHC scores were assessed according to the HER2 testing guidelines for breast cancer.11-13 The HER2 gene amplification could also be evaluated by fluorescence in situ hybridization (FISH), according to the HER2 testing guidelines for breast cancer.11-13 HER2-low status was defined as IHC 1+or IHC 2+and FISH negative. Hormone receptor-positive disease was defined as estrogen or progesterone receptor immunoreactive in ≥1% of tumor-cell nuclei according to local testing.
The primary outcome was rwPFS, defined as the time from the initiation of treatment with T-DXd to the date of the first documentation of disease progression or death from any cause, whichever occurred first.6 Disease progression was concluded by the treating clinicians on the basis of radiological, laboratory, pathological, or clinical assessment. Patients without an rwPFS event were censored at the last time they were known to be alive and free of progression. Duration of follow-up was defined as the number of months from the start of treatment with T-DXd to death from any cause or the data cutoff date. Objective response was defined as a response observed on at least 2 consecutive imaging assessments at least 4 weeks apart. Objective response was evaluated using RECIST version 1.1 for response assessment. CT was performed routinely at baseline and every 6 weeks. Follow-up CT scan data were collected for 2 years or until progressive disease (PD).
Survival was measured from the initiation of therapy to the event of interest (death or progression). Disease control rate (DCR), clinical benefit rate, objective response rate (ORR), PFS, and overall survival (OS) were analyzed. The DCR was calculated as the proportion of patients who achieved a complete response (CR), partial response (PR), or stable disease (SD). The ORR was calculated as the proportion of patients achieving a CR or a PR. The duration of response was defined as the time from the first evaluation of CR, PR, or SD to PD. Patients with PD as the best overall response were considered primary resistance to T-DXd. The adverse events (AEs) were graded according to the Common Terminology Criteria for Adverse Events version 5.0. The relation of each AE with T-DXd and treatments was considered possibly, probably, or likely related to treatment and estimated as the proportion of all toxicity-evaluable cycles in which toxicity was observed. Follow-up was censored on October 28, 2024.
Statistical analysis
Graph Prism 9.01 (GraphPad Software Inc.) and R 4.4.2 (The R Project for Statistical Computing, www.r-project.org) were used for statistical analysis. The study population for all analyses included the patients enrolled in the study who had received at least 1 dose of T-DXd. Descriptive statistics were used to summarize patient characteristics, treatment administration details, and antitumor activity. Hypothesis testing for categorical variables was performed using the chi-squared test or Fisher's exact test. A comparative analysis of primary resistance on different clinical parameters was conducted using the chi-square (χ2) test. In our study, rwOS and rwPFS were analyzed using the Kaplan–Meier method and Cox proportional hazard models. To identify independent prognostic factors, univariate Cox regression analyses and multivariate proportional hazards regression models were used. Discrimination was measured by Harrell's concordance index (C-index), which quantifies the likelihood of 2 random patients. The C-index was calculated by R version 4.4.2 via the survival and design packages. Median follow-up time was analyzed using reverse Kaplan–Meier method. Two-sided P < .05 were considered statistically significant.
Results
Baseline characters
A total of 354 MBC patients treated with at least 1 dose of T-DXd were screened, and 317 eligible patients with complete profiles were included in the study. Baseline characteristics of all patients were summarized in Table 1. The median age of the patients was 52 years (range: 28-87), and the median number of prior lines of therapy for metastatic disease was 4 (range: 0-19). Among the included patients, 108 (34.1%) had received prior ADC therapy, including 97 (30.6%) treated with tubulin-inhibitor-payload ADCs and 15 (4.7%) with topoisomerase-inhibitor-payload ADCs. Of the 58 patients with primary resistance to T-DXd, 30 (51.7%) had prior ADC exposure.
Table 1.
Baseline characters.
| HER2-low (N = 144) | HER2-positive (N = 173) | Overall (N = 317) | |
|---|---|---|---|
| Median age (range) | 50.5 [30, 87] | 52.0 [28, 82] | 52.0 [28, 87] |
| Menopausal | 99 (68.8%) | 110 (63.6%) | 209 (65.9%) |
| ECOG PSa | |||
| 0-1 | 137 (95.1%) | 161 (93.1%) | 298 (94.0%) |
| ≥2 | 7 (4.9%) | 12 (6.9%) | 19 (6.0%) |
| Disease status | |||
| De novo stage IV | 25 (17.4%) | 36 (20.8%) | 61 (19.2%) |
| Recurring disease | 119 (82.6%) | 137 (79.2%) | 256 (80.8%) |
| Hormone receptor status | |||
| Positive | 94 (65.3%) | 75 (43.4%) | 169 (53.3%) |
| Negative | 50 (34.7%) | 98 (56.6%) | 148 (46.7%) |
| Baseline metastasis | |||
| Bone | 84 (58.3%) | 84 (48.6%) | 168 (53.0%) |
| Non regional lymph nodes | 75 (52.1%) | 104 (60.1%) | 179 (56.5%) |
| Lung | 65 (45.1%) | 83 (48.0%) | 148 (46.7%) |
| Liver | 75 (52.1%) | 63 (36.4%) | 138 (43.5%) |
| Brain | 32 (22.2%) | 67 (38.7%) | 99 (31.2%) |
| HER2 expression statusb | |||
| IHC 0 | 11 (7.6%)c | 0 (0%) | 11 (3.5%) |
| IHC 1+ | 54 (37.5%) | 0 (0%) | 54 (17.0%) |
| IHC 2+ and ISH-negative | 79 (54.9%) | 1 (0.6%) d | 80 (25.2%) |
| IHC 2+ and ISH-positive | 0 (0%) | 39 (22.5%) | 39 (12.3%) |
| IHC 3+ | 0 (0%) | 133 (76.9%) | 133 (42.0%) |
| Lines of previous systemic therapy in the metastatic setting | |||
| 0-1 | 28 (19.4%) | 51 (29.5%) | 79 (24.9%) |
| 2 | 37 (25.7%) | 36 (20.8%) | 73 (23.0%) |
| 3 | 24 (16.7%) | 16 (9.2%) | 40 (12.6%) |
| 4 | 19 (13.2%) | 22 (12.7%) | 41 (12.9%) |
| ≥5 | 36 (25.0%) | 48 (27.7%) | 84 (26.5%) |
| Median number of lines (range) | 4[1, 19] | 3 [1, 19] | 4 [1, 19] |
| Previous cancer therapy | |||
| CDK 4/6 inhibitors e | 72 (50.0%) | 9 (5.2%) | 81 (25.6%) |
| Taxel-based chemotherapy | 133 (92.4%) | 159 (91.9%) | 292 (92.1%) |
| Antibody drug conjugate (ADCs) | 29 (20.1%) | 79 (45.7%) | 108 (34.1%) |
| T-DM1 | 4 (2.8%) | 62 (35.8%) | 66 (20.8%) |
| Tubulin-inhibitor-payload ADCs | 18 (12.5%) | 79 (45.7%) | 97 (30.6%) |
| Topoisomerase-inhibitor-payload ADCs | 13 (9.0%) | 2 (1.2%) | 15 (4.7%) |
aPerformance status scores on the Eastern Cooperative Oncology Group (ECOG) scale range from 0 (no disability) to 5 (death).
bHER2-low expression of human epidermal growth factor receptor 2 was defined as a score of 1+ on immunohistochemical analysis or as an IHC score of 2+ and negative results on in situ hybridization (ISH).
cPatients had previous HER2 1+ or 2+ expression, and presented HER2 0 in the latest tumor biopsy.
dOne patient showed a negative result of HER2 FISH but HER2 enrich of PAM50 sub-type.
eCDK4/6 denotes cyclin-dependent kinases 4 and 6.
Abbreviations: ADC, antibody–drug conjugate; T-DXd, trastuzumab deruxtecan; HER2, human epidermal growth factor receptor 2; IHC, immunohistochemistry; tubulin-inhibitor-payload ADCs including trastuzumab emtansine and disitamab vedotin; topoisomerase-inhibitor-payload ADCs included sacituzumab govitecan.
Patients were stratified into 2 cohorts based on HER2 expression: HER2-positive (n = 173, 54.6%) and HER2-low (n = 144, 45.4%). At the end of follow-up, 86 (27.1%) patients were still receiving T-DXd, while 6 discontinued due to intolerance and 11 due to financial constraints. Further details are provided in Figure S1.
HER2-positive cohort
In the HER2-positive cohort, the ORR was 44.5%, with a median rwPFS of 10.5 months (Table 2 , Figure 1A, Table S1, and Figure S2A). Tumor response details by treatment line and prior regimens are shown in Figure S3A and S3B and Tables S2 and S3.
Table 2.
Summary of tumor response and survival outcome.
| HER2-low (N = 144) | HER2-positive (N = 173) | Overall (N = 317) | |
|---|---|---|---|
| ORR | 35 (24.3%) | 77 (44.5%) | 112 (35.3%) |
| CBR | 63 (43.8%) | 115 (66.5%) | 178 (56.2%) |
| DCR | 108 (75.0%) | 150 (86.7%) | 258 (81.4%) |
| Primary resistance | 35 (24.3%) | 22 (12.7%) | 57 (18.0%) |
| rwPFS (months) | 5.6(4.2-7.0) | 10.5 (8.0-12.6) | 7.6(6.7-9.2) |
| rwOS (months) | 18.5(14.6-NA) | 29.9 (25.9-NA) | 28.4(21.1-NA) |
Abbreviations: CBR, clinical benefit rate; DCR, disease control rate; HER2, human epidermal growth factor receptor 2; ORR, objective response rate; rwOS, real-world overall survival; rwPFS, real-world progression-free survival
Figure 1.
Tumor response in patients with HER2-low and HER2-positive disease (A). HER2, human epidermal growth factor receptor 2.
Patients who received T-DXd as first- or second-line therapy had significantly longer rwPFS compared to those who received it as a later line (median rwPFS not reached vs. 7.3 months, hazard ratio [HR] = 0.27, P < .0001, Figure S4A). Prior exposure to tubulin-inhibitor-payload ADCs was associated with shorter rwPFS (median rwPFS 13.3 months vs. 7.53 months, HR = 0.44, P < .0001, Figure S5A).
The median rwOS was 29.9 months, with significantly longer survival observed in patients treated with T-DXd as first- or second-line therapy (Figures S2A and S6A). Prior exposure to tubulin-inhibitor-payload ADCs was associated with significantly inferior rwOS (Figure S5B). Patients with primary resistance to T-DXd had markedly shorter rwOS compared to those without primary resistance (median rwOS 5.8 months vs. 35.4 months, P < .0001, Figure S7A). Furthermore, HER2 expression significantly impacted survival outcomes, with HER2 IHC 3+ patients exhibiting longer rwPFS and rwOS compared to HER2 IHC 2+ and FISH-positive patients (Figure S8A and S8B).
Prognosis index for T-DXd efficacy in HER2-positive cohort
According to the univariate analysis, several clinical factors were found to significantly predict poor rwPFS in the HER2-positive cohort, including treatment lines ≥3, the prior treatment with tubulin-inhibitor-payload ADCs, received prior taxel-based chemotherapy, and an Eastern Cooperative Oncology Group (ECOG) performance status (PS) score ≥2. Conversely, having HER2 3+ was associated with superior rwPFS. After multivariate analysis, 3 variables were identified to maintain a negative prognostic impact on rwPFS: HER2 2+, treatment lines ≥3, and an ECOG PS score ≥2. Further details of these prognostic factors are presented in Table S4. The forest plot for multivariate analysis is shown in Figure S9.
Regarding the rwOS, univariate analysis revealed that treatment lines ≥3, an ECOG PS ≥2, the prior treatment with tubulin-inhibitor-payload ADC were associated with an inferior survival. According to the multivariate analysis, treatment lines ≥3 and an ECOG PS ≥2 were significantly associated with inferior rwOS. Detailed results are presented in Table S5 and Figure S10. Nomograms showing their predictive performance for rwPFS and rwOS in patients with HER2-positive breast cancer were shown in Figures S11 and S12.
Regarding the primary resistance of T-DXd, nomograms demonstrating its predictive performance were shown in Figure 2. It was shown that liver metastasis was significantly associated with primary resistance. Conversely, having HER2 3+ was significantly associated with response to T-DXd (Figure 2). Therefore, based on these 3 independent predictive factors (HER2 expression, treatment lines ≥3, ECOG PS ≥2) for rwPFS in the multivariate analysis, a new prognostic model was constructed for all 173 patients by combining factors as follows: Group 1 (38 cases, 22.0%), no adverse factor; Group 2 (98 cases, 56.6%), adverse factor; and Group 3 (37 cases, 21.4%), 2 or 3 adverse factors. The new predictive model for HER2-positive MBC effectively stratified patients according to prognosis. The median rwPFS for Groups 1, 2, and 3 was not reached, 8.1, and 6.0 months, respectively (P < .0001) (Figure 3A). In addition, the median rwOS of Groups 1, 2 and 3 was 47.9, 21.2, and 21.1 months, respectively (P < .0001) (Figure 3B). The C-index is 0.68 (95% CI, 0.63-0.73).
Figure 2.
The nomogram combining clinical feature to predict primary resistance in HER2 positive cohort. HER2, human epidermal growth factor receptor 2.
Figure 3.
Kaplan–Meier survival analysis for rwPFS (A) and rwOS (B) in Groups 1, 2, and 3 according to the predictive model in HER2 positive cohort. HER2, human epidermal growth factor receptor 2; rwOS, real-world overall survival; rwPFS, real-world progression-free survival.
HER2-low cohort
In the HER2-low cohort, the ORR was 24.3%, including 36 PD cases (Table 2 and Table S1). The primary resistance rate was significantly higher in HER2-low compared to HER2-positive disease (24.3% vs. 12.7%, P = .011, Figure 1A). Notably, patients with prior exposure to topoisomerase-inhibitor-payload ADCs had a significantly higher primary resistance rate compared to those without such exposure (84.6% vs. 19.1%, P < .0001, Figure 1B).
The median rwPFS was 5.6 months (Table S1 and Figure S2A). No significant difference in rwPFS was observed between patients receiving T-DXd as first-/second-line therapy versus later lines (median rwPFS 7.3 vs. 5.3 months, HR = 0.70, P = .14, Figure S4B). However, triple-negative breast cancer (TNBC) patients had significantly shorter rwPFS compared to HR-positive patients (median rwPFS 3.43 vs. 6.10 months, Figure S13A). Prior exposure to topoisomerase-inhibitor-payload ADCs was associated with markedly inferior rwPFS (median rwPFS 1.97 vs. 5.97 months, HR = 3.80, P < .0001, Figure S14A), while prior tubulin-inhibitor-payload ADC exposure did not significantly impact rwPFS (median rwPFS 5.5 vs. 5.6 months, HR = 1.03, P = .92, Figure S15A).
The median rwOS for all HER2-low disease was 18.5 months, with TNBC and HR-positive patients showing median rwOS of 11.3 and 34.5 months, respectively (Table S1, Figures S2B and S13B). Similar rwOS was observed across treatment lines (Figure S6B). Patients with prior topoisomerase-inhibitor-payload ADC exposure had significantly shorter rwOS compared to those without (5.77 vs. 18.9 months, P < .0001, Figure S14B), while prior tubulin-inhibitor-payload ADC exposure did not significantly affect rwOS (Figure S15B). Patients with primary resistance to T-DXd had significantly shorter rwOS compared to those without (median rwOS 5.2 vs. 21.3 months, P < .0001, Figure S7B). No significant differences in rwPFS or rwOS were observed based on HER2 IHC expression levels (0, 1+, or 2+) within the HER2-low cohort (Figure S16A and S16B).
Prognosis index for T-DXd efficacy in HER2-low disease
According to the univariate analysis, the exposure of topoisomerase-inhibitor-payload ADC was found to significantly predict poor rwPFS in the HER2-low cohort. Conversely, having HR-positive disease, menopausal status was associated with superior rwPFS. Subsequently, in multivariate analysis, exposure to topoisomerase-inhibitor-payload ADC and pre-menopausal status were identified to retain the same prognostic influence on rwPFS. Further details of these prognostic factors are provided in Table S6. The forest plot for multivariate analysis is shown in Figure S17.
Regarding the rwOS, univariate analysis revealed that more lines of treatment, an ECOG PS ≥2, and the prior treatment with topoisomerase-inhibitor-payload ADCs were associated with inferior survival, whereas HR-positive disease was associated with superior rwOS. According to the multivariate analysis, exposure of topoisomerase-inhibitor-payload ADCs and an ECOG PS ≥2 were significantly associated with inferior rwOS, while HR-positive disease was associated with superior rwOS. Detailed results are presented in Table S7 and Figure S18. Figures S19 and S20 present nomograms demonstrating their predictive performance for rwPFS and rwOS in HER2-low disease.
With regard to the primary resistance of T-DXd, Figure 4 presents nomograms demonstrating their predictive performance. It was shown that exposure of topoisomerase-inhibitor-payload ADCs was significantly associated with primary resistance. Therefore, based on these 3 independent predictive factors (HR expression, an ECOG PS ≥2, exposure to topoisomerase-inhibitor-payload ADCs) for rwPFS in the multivariate analysis, a new prognostic model was constructed for all 144 patients by combining the factors as follows: Group 1 (66 cases, 45.8%), no adverse factor; Group 2 (52 cases, 36.1%), 1 adverse factor; and Group 3 (26 cases, 18.1%), 2 or 3 adverse factors. The median rwPFS of Groups 1, 2, and 3 was 6.1, 6.7, and 2.1 months, respectively (P < .0001) (Figure 5A). In addition, the median rwOS of Groups 1, 2, and 3 was 18.9, 46.7, and 7.1 months, respectively (P = .00028) (Figure 5B). The C-index is 0.63 (95% CI, 0.57-0.68). Although Group 2 showed numerically longer median rwPFS and rwOS compared to Group 1, the difference was not statistically significant. Besides, both groups demonstrated superior outcomes relative to Group 3.
Figure 4.
The nomogram combining clinical features to predict primary resistance in HER2-low cohort. HER2, human epidermal growth factor receptor 2.
Figure 5.
Kaplan–Meier survival analysis for rwPFS (A) and rwOS (B) in Groups 1, 2, and 3 according to the predictive model in HER2-low cohort. HER2, human epidermal growth factor receptor 2; rwOS, real-world overall survival; rwPFS, real-world progression-free survival.
Safety
Treatment-related AEs were observed in 266 (83.9%) patients, with grade 3-4 AEs occurring in 67 (21.1%) patients (Table 3). The most common AEs included nausea, dyspepsia, vomiting, fatigue, leukopenia, neutropenia, thrombocytopenia, and anemia. Interstitial lung disease occurred in 29 (9.1%) patients, with 15 (4.7%) classified as severe AEs (SAEs). No treatment-related deaths were reported.
Table 3.
Treatment-related adverse events.
| HER2-low (N = 144) | HER2-positive (N = 173) | Overall (N = 317) | ||||
|---|---|---|---|---|---|---|
| Any grade | Grade 3/4 | Any grade | Grade 3/4 | Any grade | Grade 3/4 | |
| Any AE | 117 (81.3%) | 32 (22.2%) | 149 (86.1%) | 35 (20.2%) | 266 (83.9%) | 67 (21.1%) |
| SAE | 17 (11.8%) | 17 (11.8%) | 15 (8.7%) | 15 (8.7%) | 32 (10.1%) | 32 (10.1%) |
| Leukopenia | 57 (39.6%) | 14 (9.7%) | 51 (29.5%) | 6 (3.5%) | 108 (34.1%) | 20 (6.3%) |
| Neutropenia | 51 (35.4%) | 14 (9.7%) | 44 (25.4%) | 12 (6.9%) | 95 (30.0%) | 26 (8.2%) |
| Anemia | 51 (35.4%) | 6 (4.2%) | 60 (34.7%) | 4 (2.3%) | 111 (35.0%) | 10 (3.2%) |
| Thrombocytopenia | 19 (13.2%) | 5 (3.5%) | 27 (15.6%) | 4 (2.3%) | 46 (14.5%) | 9 (2.8%) |
| Nausea | 42 (29.2%) | 2 (1.4%) | 44 (25.4%) | 2 (1.2%) | 86 (27.1%) | 4 (1.3%) |
| Dyspepsia | 33 (22.9%) | 1 (0.7%) | 42 (24.3%) | 3 (1.7%) | 75 (23.7%) | 4 (1.3%) |
| Vomit | 23 (16.0%) | 3 (2.1%) | 31 (17.9%) | 2 (1.2%) | 54 (17.0%) | 5 (1.6%) |
| Fatigue | 18 (12.5%) | 2 (1.4%) | 19 (11.0%) | 4 (2.3%) | 37 (11.7%) | 6 (1.9%) |
| Pneumonia | 18 (12.5%) | 6 (4.2%) | 16 (9.2%) | 6 (3.5%) | 34 (10.7%) | 12 (3.8%) |
| Interstitial lung disease | 16 (11.1%) | 7 (4.9%) | 13 (7.5%) | 8 (4.6%) | 29 (9.1%) | 15 (4.7%) |
| Peripheral sensory neuropathy | 8 (5.6%) | 0 | 20 (11.6%) | 1 (0.6%) | 28 (8.8%) | 1 (0.3%) |
| Dizziness | 12 (8.3%) | 0 | 13 (7.5%) | 1 (0.6%) | 25 (7.9%) | 1 (0.3%) |
| Headache | 11 (7.6%) | 2 (1.4%) | 10 (5.8%) | 2 (1.2%) | 21 (6.6%) | 4 (1.3%) |
| Constipation | 7 (4.9%) | 0 | 9 (5.2%) | 0 | 16 (5.0%) | 0 |
| Diarrhea | 8 (5.6%) | 2 (1.4%) | 3 (1.7%) | 0 | 11 (3.5%) | 2 (0.6%) |
| Rash | 3 (2.1%) | 0 | 5 (2.9%) | 0 | 8 (2.5%) | 0 |
| Oral ulcer | 2 (1.4%) | 0 | 5 (2.9%) | 0 | 7 (2.2%) | 0 |
| Pruritus | 2 (1.4%) | 0 | 1 (0.6%) | 0 | 3 (0.9%) | 0 |
| AST elevation | 32 (22.2%) | 1 (0.7%) | 53 (30.6%) | 3 (1.7%) | 85 (26.8%) | 4 (1.3%) |
| ALT elevation | 21 (14.6%) | 1 (0.7%) | 42 (24.3%) | 0 | 63 (19.9%) | 1 (0.3%) |
| Hypoalbuminemia | 28 (19.4%) | 0 | 44 (25.4%) | 0 | 72 (22.7%) | 0 |
| Serum total bilirubin elevation | 12 (8.3%) | 1 (0.7%) | 13 (7.5%) | 1 (0.6%) | 25 (7.9%) | 2 (0.6%) |
| Serum Creatinine elevation | 8 (5.6%) | 0 | 13 (7.5%) | 0 | 21 (6.6%) | 0 |
Abbreviation: HER2, human epidermal growth factor receptor 2
Discussion
In our study, we comprehensively evaluated the real-world outcomes of T-DXd in both HER2-positive and HER2-low MBC cohorts. Additionally, we identified potential patient subgroups that may derive greater benefit from T-DXd therapy and elucidated key risk factors for primary resistance based on clinical parameters. Notably, in the HER2-low cohort, prior exposure to topoisomerase-inhibitor-payload ADCs was significantly associated with primary resistance. Conversely, in the HER2-positive cohort, liver metastasis and HER2 expression of IHC 2+ emerged as significant risk factors for primary resistance. Regarding the impact of prior therapies on T-DXd efficacy, our findings suggest that previous treatment with tubulin-payload ADCs did not significantly influence rwPFS, rwOS, or primary resistance in either HER2-low or HER2-positive disease. However, prior exposure to topoisomerase-inhibitor-payload ADCs was strongly associated with inferior rwPFS, rwOS, and increased rates of primary resistance across both cohorts. These findings underscore the importance of considering prior ADC payload classes when selecting patients for T-DXd therapy and highlight potential mechanisms of cross-resistance that warrant further investigation.
The remarkable efficacy demonstrated in the DESTINY-Breast 01 and 02 trials has significantly influenced clinical practice, prompting physicians to utilize T-DXd for HER2-positive MBC patients, including those with extensive prior treatment histories.2,3 These pivotal trials reported a median PFS ranging from 16.4 to 18.2 months in heavily pre-treated patients who had received more than 3 prior lines of therapy.2,3 These findings align with the DE-REAL study conducted in Italy, which documented an rwPFS of 16 months for patients with HER2-positive breast cancer treated with T-DXd.6 Notably, in the DE-REAL study, 86% of patients received T-DXd as a third-line or subsequent therapy.6 In our study, the observed rwPFS for HER2-positive disease was 10.5 months, with a notable decline in efficacy among patients who had received more than 2 prior lines of therapy, resulting in an rwPFS of 7.3 months. The shorter PFS observed in our cohort compared to both clinical trials and the DE-REAL study may be attributed to our larger sample size and a more heavily pre-treated patient population. Despite these differences, our findings provide evidence on the efficacy of T-DXd in real-world clinical practice, particularly when used in the second-line setting. These results underscore the importance of considering treatment sequencing and prior therapy exposure when optimizing T-DXd therapy for HER2-positive MBC patients.
T-DXd has demonstrated clinical benefit across all HER2 expression levels, with particularly pronounced efficacy in HER2-positive MBC patients.14 In our study, HER2 expression exhibited differential impacts on treatment outcomes between HER2-positive and HER2-low cohorts. While previous studies of T-DXd have not fully elucidated the comparative efficacy between HER2 IHC 3+ and IHC 2+/ISH-positive subgroups,3,5 our findings corroborate that higher HER2 expression is strongly correlated with improved survival outcomes in HER2-positive disease. Conversely, the relationship between HER2 expression levels and treatment response in the HER2-low cohort remains less clear, consistent with observations from the DESTINY-Breast 04 and 06 trials.8,9
The therapeutic efficacy of ADCs is influenced by both the antibody and payload components, with the antibody's role potentially varying across different tumor types and molecular targets. For instance, HER-3-directed ADCs have shown clinical benefits in patients with varying levels of HER-3 expression and across different indications.15 Similarly, the relationship between TROP2 expression and response to TROP2-targeted ADCs remains incompletely understood.16 In the context of HER2-targeted ADCs, the antibody component appears to play a more significant role in HER2-positive disease, whereas in HER2-low disease, the payload's bystander killing effect and targeted delivery mechanism may contribute more substantially to therapeutic efficacy.14 To further elucidate these mechanisms, we plan to investigate the role of target expression in treatment efficacy, particularly focusing on how the antibody component may induce antitumor effects through distinct mechanisms, especially in HER2-positive disease. This research may provide valuable insights into optimizing ADC-based therapies for different HER2 expression subgroups.
T-DXd has been approved for the treatment of HER2-low MBC following progression on endocrine therapy or chemotherapy.8,9 However, prior clinical trials primarily included patients who had received no more than 2 lines of prior therapy,8 leaving a gap in understanding its efficacy in more heavily pre-treated populations. Our study aimed to address this gap by evaluating the real-world efficacy of T-DXd in a broader patient population with HER2-low MBC, including those who may not meet the stringent eligibility criteria typically applied in DESTINY-Breast clinical trials. In our HER2-low cohort, we identified that prior exposure to topoisomerase-inhibitor-payload ADCs significantly influenced both primary resistance and survival outcomes. This finding aligns with our previous research and a real-world study by Poumeaud et al. in France, which suggested a higher incidence of cross-resistance when ADCs with similar payloads are administered sequentially.7,10 These observations highlight the critical need for biomarkers to predict payload sensitivity, as the risk of cross-resistance appears to be elevated in patients previously treated with ADCs sharing similar payload mechanisms.
The mechanisms underlying resistance to ADCs are multifaceted, involving factors such as antigen expression, ADC processing, and payload activity.17,18 For instance, HER2 heterogeneity has been implicated in resistance to T-DM1.19 To address resistance to T-DXd, several strategies have been explored, including combinations with immunotherapy, endocrine therapy, and chemotherapy, although their efficacy remains uncertain.20 Our findings align with previous reports suggesting that alternating payload classes (eg, topoisomerase inhibitor- and tubulin-payload ADCs) may help overcome cross-resistance, though optimal sequencing requires further investigation. Moreover, our study proposes a novel approach to optimizing T-DXd therapy by evaluating distinct clinical parameters in HER2-low and HER2-positive disease. Specifically, we identified that patients with 2 or 3 adverse factors—such as prior exposure to topoisomerase-inhibitor-payload ADCs, higher treatment lines, or poor PS—may be refractory and showed limited benefit from T-DXd. Our findings suggest the importance of treating T-DXd earlier, before the accumulation of adverse prognostic factors, to optimize clinical outcomes.
T-DXd has an incremental cost-effectiveness ratio of up to $82 112 per quality-adjusted life-year,21 which, coupled with limited reimbursement in some regions, can restrict its widespread use in clinical practice. Our study suggests that T-DXd may not be equally effective for all patients, especially those facing significant financial toxicity, potentially exacerbating access disparities. The financial burden of T-DXd underscores the need for broader reimbursement policies and cost-effective strategies to ensure equitable access to this promising therapy. While the safety profile of T-DXd observed in our real-world cohort was consistent with findings from clinical trials and the DE-REAL study,3,4,6 the 21.1% incidence of grade 3-4 AEs warrants careful consideration. We observe a lower incidence of nausea and vomiting in our cohort compared to the reports in the DESTINY-Breast trials, which is likely due to the optimized use of antiemetic regimens in real-world settings. However, for patients with multiple adverse factors, the potential benefits of T-DXd should be carefully weighed against its toxicity profile and quality-of-life considerations.
This study has certain limitations, primarily stemming from its retrospective design and the inherent heterogeneity in baseline characteristics and treatment factors, which may introduce potential biases. Despite these limitations, the study's key strength lies in its comprehensive analysis of T-DXd efficacy in advanced breast cancer, particularly in heavily pre-treated patients and those with HER2-low disease—a population often underrepresented in clinical trials. Our findings provide valuable real-world insights into the clinical utility of T-DXd and support the importance of identifying predictive factors for treatment response. In our study, the prognostic model could help identify high-risk patients from lower-risk patients in HER2-low disease. While Group 2 trended toward better outcomes than Group 1, the lack of statistical significance suggests that a single adverse factor may not uniformly diminish T-DXd efficacy. Future studies should explore whether specific factors differentially impact outcomes within these subgroups. Besides, to further validate the prognostic model developed in this study and enhance its generalizability, prospective clinical trials and larger validation cohorts are essential. Such efforts will help evaluate the predictive value of our model and refine its application in guiding personalized treatment strategies for patients with advanced breast cancer.
Conclusion
This real-world study highlights the efficacy of T-DXd in both HER2-positive and HER2-low MBC, with significant differences in outcomes based on prior therapy and clinical characteristics. The developed prognostic index effectively stratified patients into risk groups, providing valuable insights for identifying patients most likely to benefit from T-DXd therapy.
Supplementary Material
Acknowledgments
The authors thank all the patients, their families, and the institutions for supporting this study. They acknowledge all medical staff, staff nurses, and research nurses who strongly contributed to the study’s success.
Contributor Information
Cong Xue, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Qianyi Liao, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Riqing Huang, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Yunjie Huang, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Rishang Chen, Medical Oncology Department III, Central Hospital of Guangdong Nongken, Zhanjiang 524002, China.
Zhenhua Yang, Department of Medical Oncology, Jiangmen Wuyi Hospital of Traditional Chinese Medicine, Jiangmen 529000, China; Department of Medical Oncology, Guangdong Provincial Hospital of Chinese Medicine, Zhuhai 519015, China.
Xiujiao Shen, Department of Pathology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Haifeng Li, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Qixiang Rong, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Ditian Shu, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Fei Pan, Department of Breast Medical Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, China.
Yanxia Shi, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Meiting Chen, Department of Medical Oncology, State Key Laboratory of Oncology in South China, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou 510060, China.
Author contribution
Yanxia Shi, Cong Xue, Meiting Chen (Conceptualization); Meiting Chen, Riqing Huang, Qianyi Liao, Yunjie Huang, Rishang Chen, Fei Pan, Zhenhua Yang, Haifeng Li, Ditian Shu, Qixiang Rong, Xiujiao Shen (Data curation); Meiting Chen, Riqing Huang (Formal analysis); Haifeng Li (Software); Meiting Chen (Visualization); Cong Xue (Methodology; Supervision); Meiting Chen, Cong Xue (Validation); Yanxia Shi, Meiting Chen (Funding acquisition); Qianyi Liao (Project administration); Yanxia Shi (Investigation); Cong Xue (Resources); Meiting Chen, Riqing Huang (Writing—original draft); Meiting Chen, Yanxia Shi, Riqing Huang, Cong Xue (Writing—review & editing)
Funding
This work was supported by the grants from the Natural Science Foundation of China (No. 82403969, No. 82473400), the National Key Research and Development Program of China (No. 2021YFE0206300), Sun Yat-sen University Clinical Research 5010 Program (No. 2024002), and Cancer Innovative Research Program of Sun Yat-sen University Cancer Center (CIRP-SYSUCC-023).
Conflicts of interest
The authors have no relevant financial or non-financial conflicts of interest to disclose.
Data Availability
The datasets generated during the current study are available from the corresponding author upon reasonable request.
Ethics approval and consent to participate
The clinical data were acquired with the approval and permission of the Institutional Review Board of the Sun Yat-Sen University Cancer Center. Informed consent was not required because this study was a retrospective report of cases, a retrospective analysis of clinical data with no relevance to human biological ethic problems. The study protocol was approved by the Institutional Review Board of the Sun Yat-Sen University Cancer Centre, and the study was performed in accordance with the principles of the Declaration of Helsinki. All methods were performed in accordance with the relevant guidelines and regulations.
References
- 1. Tarantino P, Carmagnani Pestana R, Corti C, et al. Antibody-drug conjugates: smart chemotherapy delivery across tumor histologies. CA Cancer J Clin. 2022;72:165-182. [DOI] [PubMed] [Google Scholar]
- 2. Shanu M, Cristina S, Toshinari Y, et al. Trastuzumab deruxtecan in previously treated HER2-positive breast cancer. N Engl J Med. 2020;382:610-621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. André F, Hee Park Y, Kim S, et al. Trastuzumab deruxtecan versus treatment of physician’s choice in patients with HER2-positive metastatic breast cancer (DESTINY-Breast02): a randomised, open-label, multicentre, phase 3 trial. Lancet. 2023;401:1773-1785. [DOI] [PubMed] [Google Scholar]
- 4. Hurvitz S, Hegg R, Chung W, et al. Trastuzumab deruxtecan versus trastuzumab emtansine in patients with HER2-positive metastatic breast cancer: updated results from DESTINY-Breast03, a randomised, open-label, phase 3 trial. Lancet. 2023;401:105-117. [DOI] [PubMed] [Google Scholar]
- 5. Javier C, Sung-Bae K, Wei-Pang C, et al. Trastuzumab deruxtecan versus trastuzumab emtansine for breast cancer. N Engl J Med. 2022;386:1143-1154. [DOI] [PubMed] [Google Scholar]
- 6. Andrea B, Roberta C, Simone S, et al. Real-world outcomes of trastuzumab deruxtecan in patients with HER2+ metastatic breast cancer: the DE-REAL Study. Oncologist. 2023;29:303-310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Meiting C, Riqing H, Rishang C, et al. Optimal sequential strategies for antibody-drug conjugate in metastatic breast cancer: evaluating efficacy and cross-resistance. Oncologist. 2024;29:e957-e966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Modi S, Jacot W, Yamashita T, et al. Trastuzumab deruxtecan in previously treated HER2-low advanced breast cancer. N Engl J Med. 2022;387:9-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Aditya B, Xichun H, Rebecca D, et al. Trastuzumab deruxtecan after endocrine therapy in metastatic breast cancer. N Engl J Med. 2024;391:2110-2122. [DOI] [PubMed] [Google Scholar]
- 10. Poumeaud F, Morisseau M, Cabel L, Gonçalves A.. Efficacy of administration sequence: sacituzumab govitecan and trastuzumab deruxtecan in HER2-low metastatic breast cancer. Br J Cancer. 2024;131:702-708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Zhang H, Moisini I, Ajabnoor RM, Turner BM, Hicks DG.. Applying the new guidelines of HER2 testing in breast cancer. Curr Oncol Rep. 2020;22:51. https://doi.org/ 10.1007/s11912-020-0901-4 [DOI] [PubMed] [Google Scholar]
- 12. Wolff A, Hammond M, Allison K, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update. J Clin Oncol. 2018;36:2105-2122. [DOI] [PubMed] [Google Scholar]
- 13. Wolff A, Hammond M, Hicks D, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol. 2013;31:3997-4013. [DOI] [PubMed] [Google Scholar]
- 14. Colombo R, Tarantino P, Rich J, LoRusso P, de Vries EGE.. The journey of antibody-drug conjugates: lessons learned from 40 years of development. Cancer Discov. 2024;14:2089-2108. [DOI] [PubMed] [Google Scholar]
- 15. Krop I, Masuda N, Mukohara T, et al. Patritumab deruxtecan (HER3-DXd), a human epidermal growth factor receptor 3-directed antibody-drug conjugate, in patients with previously treated human epidermal growth factor receptor 3-expressing metastatic breast cancer: a multicenter, phase I/II trial. J Clin Oncol. 2023;41:5550-5560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bardia A, Hurvitz S, Tolaney S, et al. Sacituzumab govitecan in metastatic triple-negative breast cancer. N Engl J Med. 2021;384:1529-1541. [DOI] [PubMed] [Google Scholar]
- 17. Abelman R, Wu B, Spring L, Ellisen L, Bardia A.. Mechanisms of resistance to antibody-drug conjugates. Cancers (Basel). 2023;15:1278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Chang H, Schwettmann B, McArthur HL, Chan IS.. Antibody-drug conjugates in breast cancer: overcoming resistance and boosting immune response. J Clin Invest. 2023;133:e172156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Filho O, Viale G, Stein S, et al. Impact of HER2 heterogeneity on treatment response of early-stage HER2-positive breast cancer: phase II neoadjuvant clinical trial of T-DM1 combined with pertuzumab. Cancer Discov. 2021;11:2474-2487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lorenzo G, Laura B, Dario T, Giuseppe G.. Antibody-drug conjugates in metastatic breast cancer: sequencing, combinations and resistances. Curr Opin Oncol. 2024;36:487-494. [DOI] [PubMed] [Google Scholar]
- 21. Yang J, Han J, Zhang Y, Muhetaer M, Chen N, Yan X.. Cost-effectiveness analysis of trastuzumab deruxtecan versus trastuzumab emtansine for HER2-positive breast cancer. Front Pharmacol. 2022;13:924126. https://doi.org/ 10.3389/fphar.2022.924126 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during the current study are available from the corresponding author upon reasonable request.