Antibody Drug Conjugates in Breast Cancer: Spotlight on HER2

Abstract

Antibody-drug conjugates (ADCs) are composed of monoclonal antibodies linked to a cytotoxic payload, enabling targeted delivery of more potent chemotherapy. In the past decade, there has been rapid development of ADCs aimed at different types of breast cancer. The success of the monoclonal antibody trastuzumab has led to the evolution of several antibody-drug conjugates targeting HER2-positive breast cancer. Trastuzumab-emtansine (T-DM1), the first approved ADC targeting HER2-positive breast cancer, has become standard of care for patients with high-risk early-stage HER2-positive breast cancer who have residual disease after neoadjuvant chemotherapy. More recently, the observation of the bystander effect, in which ADCs target both antigen-positive cells and adjacent antigen-negative cells, has led to the reclassification of “HER2-low” breast cancer and the development of trastuzumab-deruxtecan (T-DXd) to target this population. This article reviews the history of HER2-directed ADCs in breast cancer as well as ongoing ADCs in development.

Introduction

Antibody-drug conjugates (ADCs) are composed of monoclonal antibodies linked to cytotoxic agents with the goal of selective delivery of chemotherapy. Given the specificity of the antibody, these treatments seek to expand the therapeutic window beyond traditional chemotherapy by minimizing off-target toxicity. This mechanism can allow for target cells to see increased potency up to 100 to 1000 times more concentrated than traditional systemic chemotherapy (1) In addition, the improved specificity conferred by ADCs allows for potential use of more potent cytotoxic agents than would be possible in conventional chemotherapy (2). ADCs have three components: an antibody that targets a tumor-related antigen, a linker, and a cytotoxic payload. An ideal target is found in tumor cells but less frequently in non-malignant cells to limit off-target toxicity (2). Variations in linker biochemistry allow for linkers that can be cleavable or non-cleavable, with cleavable linkers releasing their payload based on tumor-specific factors such as pH or enzymatic changes. Cleavable linkers may deliver their cytotoxic payload more efficiently while non-cleavable linkers may be more specific but can require additional mechanisms of payload release such as lysosomal degradation which can affect the structure of the payload (3).

ADCs have been found to be effective in the setting of relapsed/refractory disease where tumor heterogeneity may contribute to resistance to previous treatments, where components of the tumor cell do not express the target of a prior therapy. One proposed explanation for the efficacy of ADCs in the setting of tumor heterogeneity is the “bystander effect,” where the cytotoxic payload crosses from within the targeted tumor cell and exerts effect on tissues adjacent to the target cells (2, 4, 5). One early observation of this phenomenon was in preclinical studies as far back as 2003, when an ADC known as BR96-DOX (SGN-15), with an antibody targeting the LewisY antigen was combined with a doxorubicin payload, was investigated in rat models and found to have toxicity against both antigen-positive cells and antigen-negative neighboring cells, although later phase studies were later discontinued to do lack of demonstrating of clinical benefit (6,7, 8). Similarly, trastuzumab-deruxtecan (T-DXd) is thought to act through the bystander effect, where HER2− cells near HER2+ cells were exposed the cytotoxic payload of T-DXd after it crossed the cell membrane of targeted HER2+ cells. This mechanism could potentially explain the efficacy of the treatment in patients previously treated with other anti-HER2 agents that target only HER2-expressing cells (1, 4). However, alongside potential for efficacy in refractory disease, agents that utilize the bystander effect may come with increased systemic toxicity due to lack of tumor specificity. In this review, we will discuss the history, clinical development, results from pivotal clinical trials, and future directions related to HER2 ADCs.

History of HER2 ADCs

HER2 (also known as ERBB2) is a transmembrane glycoprotein that is part of the epidermal growth factor receptor (EGFR) family of receptors with tyrosine kinase activity (9, 10). When HER2 is dimerized, there is autophosphorylation of tyrosine residues in the receptor that activates many cell signaling pathways that leads to proliferation and tumorigenesis (10). Overexpression of HER2 is seen in a substantial minority of patients with breast cancer (15-20%) as well as gastric/gastroesophageal cancers (10-30%), and is found less frequently in a variety of other cancers (10, 11). In 1987, Slamon et al. described an association between HER2 amplification and greater risk of disease recurrence and death (10, 12). It was further confirmed that HER2 overexpression was associated with greater risk of recurrence in early stage disease (10, 13, 14). In breast cancers with HER2 amplification, HER2 is highly expressed with approximately 1-2 million copies per cell, making it an attractive therapeutic target. The monoclonal antibody trastuzumab binds to domain IV of the extracellular region of the HER2 receptor (10). This is thought to interfere with multiple mechanisms integral to HER2 function, including signal transduction through inhibiting interactions between HER2 and HER3, which is thought to disrupt the PI3K-AKT signaling pathway (15). In addition, trastuzumab is thought to induce cell-mediated toxicity by binding the FcyRIII (15, 16). Finally, additional proposed mechanisms involve trastuzumab inhibiting angiogenesis and inducing cell cycle arrest and antibody-mediated cellular cytotoxicity (9, 15).

With the success of targeted antibody treatments for HER2 positive breast cancer, focus shifted to the application of nascent ADC technology to this setting. A comprehensive list of different HER2 ADCs is provided in Table 1. A summary of clinical results from pivotal trials is outlined in Table 2 and discussed in detail below.

Table 1:

HER2-directed ADCs

Name	Antibody	Drug-Antibody ratio (DAR)	Linker	Payload
Trastuzumab-emtansine (T-DM1)	Trastuzumab	3.5:1	Non-cleavable	Maytansinoid derivative-disrupt microtubule function
Trastuzumab deruxtecan (T-DXd)	Trastuzumab	8:1	Cleavable	Topoisomerase-I inhibitor
Trastuzumab duocarmazine	Trastuzumab	2.8:1	Cleavable	DNA Alkylator
A166	Trastuzumab	n/a	Cleavable	Duostatin-5 (Anti-microtubule)
XMT-1522	HT-19- binds alternative HER2 binding site	12:1	n/a	AF-HPA-aurostatin derivative, inhibits tubulin polymerization
RC-48- Disitamab Vedotin	Hertuzumab- binds alternative HER2 binding site	4:1	Cleavable	MMAE (monomethyl auristatin E)
ALT-P7	Trastuzumab variant HM2	2:1	n/a	MMAE (monomethyl auristatin E)
ARX788	anti-HER2 antibody	1.9:1	Non-cleavable	AS269- tubulin inhibitor
PF-06804103	Trastuzumab derivative	4:1	Cleavable	AUR-06380101 (Auristatin derivative)

Table 2:

Summary of results from pivotal Phase II/III trials evaluating HER2-targeted ADCs

Name (Year)	Conclusion
EMILIA (2012)	Phase III trial investigating T-DM1 compared to lapatinib/capecitabine in advanced HER2+ breast cancer previously treated with trastuzumab and a taxane. Patients with T-DM1 demonstrated significantly improved PFS (9.6 months vs. 6.4 months) and at second interim analysis, median OS improvement in T-DM1 reached the stopping point for efficacy (30.9 months vs. 25.1 months).
TH3RESA (2017)	Phase III trial comparing T-DM1 against TPC in previously treated HER2+ advanced breast cancer. Significantly improved overall survival (22.7 months vs. 15.8 months) in T-DM1 group.
KATHERINE (2019)	Phase III trial for patients with early stage HER2+ breast cancer with residual disease after neoadjuvant chemotherapy with trastuzumab and a taxane. Patients were randomized to T-DM1 vs. T for 14 cycles after surgery. Patients who received T-DM1 had significantly improved invasive disease-free survival (88.3% vs. 77%).
KRISTINE (2019)	Phase III trial comparing T-DM1 and P to traditional neoadjuvant chemotherapy (TCHP) in patients with non-metastatic HER2+ breast cancer. Patients who received T-DM1 and P had significantly lower rates of pCR (44.4% vs. 55.7%) and lower rates of three-year event-free survival (85.3% vs 94.2%).
DESTINY-Breast 01 (2020)	Phase II study of T-DXd in patients with HER2+ mBC previously treated with T-DM1. 60.9% of patients were found to respond to treatment with median PFS of 16.4 months.
DESTINY-Breast 03 (2022)	Phase III study comparing T-DXd to T-DM1 in HER2+ mBC. Progression-free survival at 12 months was 75.8% in T-DXd vs. 34.1% in T-DM1 cohort.
DESTINY-Breast 04	Phase III trial for patients with HER2-low mBC comparing T-DXd to TPC. The median PFS in T-DXd group was 9.9 months vs 5.1 months in TPC, leading to OS of 23.4 months vs. 16.8 months in TPC.

Trastuzumab emtansine (T-DM1)

ADC Structure:

T-DM1, the first ADC directed against HER2, linked trastuzumab with the maytansinoid derivative DM1. Maytansinoids, similar to vinca alkaloids, disrupt microtubule function, but have not been used independently due to intolerable toxicity. DM1 is created by replacing the N-acetyl group in maytansine with a thiolpropanoyl group (15). Maytansinoids are thought to be 24 to 270 times more potent than paclitaxel and 2-3 times more potent than doxorubicin as was demonstrated in human breast cancer cell lines (15). T-DM1 has a drug-antibody ratio of approximately 3.5 (17). The linker is a nonreducible thioether bond (17). In human breast cancer cell links, the binding affinity of T-DM1 to HER2 was similar to trastuzumab, and T-DM1 was found to mediate antibody-dependent cellular cytotoxicity and also inhibit the AKT pathway (15).

Early phase trials:

Early Phase I studies of T-DM1 enrolled patients with locally advanced or metastatic HER2+ breast cancer who had progressed after treatment with trastuzumab. Krop et al. 2010 was a first in human study of T-DM1 that established 3.6 mg/kg as the maximum tolerated dose, enrolling 24 patients in the dose-escalation phase. Patients received T-DM1 infusions every 3 weeks. Almost all patients (22/24) were found to have at least one drug-related adverse event, most commonly thrombocytopenia (54.2%), transaminitis (41.7%), or fatigue (37.5%), although most AEs were grade 1 or grade 2, six of the 24 patients were found to have an objective partial response, with a median number of prior lines of treatment of 3.5. There was a clinical benefit rate for the 15 patients treated at the maximum tolerated dose of 73%, and for this cohort, the median duration of treatment was 238 days. This study demonstrated an impressive clinical benefit rate and also demonstrated a favorable safety profile, providing a pathway for further investigation (17).

Metastatic HER2 positive breast cancer:

The phase III TH3RESA study (N=602) compared trastuzumab-emtansine (T-DM1) with treatment of investigator’s choice in patients with HER2+ advanced breast cancer who had previously received at least two HER2-directed agents. Patients who received T-DM1 had significantly longer overall survival, with median 22.7 months (95% CI 19.4-27.5 months) compared to 15.8 months (95% CI 13.5-18.7 months) in the treatment of physician’s choice group (HR 0.68, 95% CI 0.54-0.85, p=0.0007). Patients treated with T-DM1 were more likely to have thrombocytopenia (6%) and hemorrhage of any type (4%), whereas patients in physician’s choice group had higher incidence of diarrhea (4%), neutropenia (16%) and febrile neutropenia (4%) (18). In the phase III EMILIA study (N=991), T-DM1 was compared to the small molecule tyrosine kinase inhibitor lapatinib in patients with advanced HER2+ breast cancer that had been already treated with trastuzumab and a taxane (19). Patients treated with T-DM1 had significantly improved PFS compared to patients treated with lapatinib and capecitabine (9.6 months versus 6.4 months, HR 0.65, 95% CI 0.55-0.77). Patients receiving lapatinib/capecitabine had higher incidence of grade 3 adverse events overall including diarrhea, nausea, and palmar-plantar erythrodysesthesia, whereas patients treated with T-DM1 had higher incidence of thrombocytopenia and elevated transaminase levels (19).

Early HER2 positive breast cancer (adjuvant):

The phase III KATHERINE study (N=1486) established T-DM1 as treatment of choice for patients with residual HER2+ disease after neoadjuvant chemotherapy. Altogether, invasive disease-free survival was significantly higher in T-DM1 group compared to trastuzumab (HR 0.5, 95% CI 0.39-0.64, p<0.001) and the risk of distant recurrence was significantly lower in T-DM1 group (HR 0.60, 95% CI 0.45-0.79). In the T-DM1 group, most common AEs grade 3 or higher were decreased platelet count (5.7%) and hypertension (2%), compared to T-DM1 group with 1.2% hypertension, 1% radiation-related skin injury. 18% of T-DM1 group had adverse events leading to discontinuation of the drug, most commonly decreased platelet count (4.2%), elevated bilirubin level (2.6%), elevated AST (1.6%), elevated ALT (1.5%), peripheral neuropathy (1.5%), and decreased EF (1.2%). One patient in the T-DM1 group died of intracranial hemorrhage after a fall, although the rates of hemorrhage in both groups were similar (0.4% in T-DM1, 0.3% in trastuzumab). This data changed practice for patients with residual disease after NACT with trastuzumab, prompting a 1 year course of T-DM1. Notably, 5% of patients in both T-DM1 and trastuzumab group had central nervous system (CNS) as one of first sites of recurrence, raising the question of how active this agent was in the CNS (20).

Early HER2 positive breast cancer (neo-adjuvant):

The phase III KRISTINE study (N=444), also published in 2019, established some boundaries on the ability of ADCs to replace traditional chemotherapy. This trial enrolled patients with stage II-III HER2+ breast cancer. Altogether, pathologic complete response (pCR rate was lower in patients who received T-DM1 and pertuzumab (44.4%) compared to those who received chemotherapy with docetaxel, carboplatin, trastuzumab, and pertuzumab (TCHP) (55.7%), although the T-DM1 arm demonstrated fewer grade 3 events or higher (13% vs 64.4%) and fewer serious adverse events (AEs) (4.9% vs 28.8%). Overall, the three-year event-free survival rates were 85.3% in T-DM1 and P arm compared to 94.2% with TCHP. The rate of invasive disease-free survival IDFS was similar in both cohorts, 93.0% in T-DM1+P compared to 92% in the TCHP arm. Tumors with locoregional progression were found to have lower HER2 expression and higher HER2 heterogeneity, suggesting a resistance mechanism to HER-2 directed therapy that the investigators postulated might require chemotherapy (21).

Trastuzumab Deruxtecan (T-DXd):

ADC Structure:

Trastuzumab deruxtecan (T-DXd), the second ADC approved by the FDA to treat advanced HER2+ breast cancer, has shifted the treatment paradigm of advanced HER2+ disease. T-Dxd differs structurally from T-DM1 in several significant ways. The non-cleavable linker in T-DM1 is dependent on a complex process of lysosomal degradation and subsequent travel to the cytosol for the active cytotoxic agent to have its anti-tubulin effect. In cells with high HER2 expression, this is efficient, but may explain resistance to T-DM1 in cells with low/moderate HER2 expression. In contrast, T-Dxd has a cleavable linker with membrane-permeable payload which allows delivery to the extracellular environment via the “bystander effect”. (22, 23). In addition, there is a different payload in T-Dxd, an exatecan-derived topoisomerase-I inhibitor, rather than a microtubule inhibitor. Finally, T-DXd has a drug-antibody ratio of 8, compared to 3.5 in T-DM1. These differences likely explain the activity of T-DXd in patients who were refractory to T-DM1 as well as activity in patients with low or heterogeneous HER2 expression (23, 24).

Early Phase Trials:

The DESTINY-Breast trials, many of which are ongoing, have demonstrated the impact of T-DXd and are exploring the agent in a variety of settings. DESTINY-Breast 01 was a phase II study evaluating T-DXd in patients with metastatic HER2+ breast cancer who had been previously treated with T-DM1. 184 patients received T-Dxd, with 112 patients demonstrating a confirmed response (60.9%, 95% CI 53.4-68.0), with a median progression-free survival (PFS) of 16.4 months. Patients most frequently reported neutropenia (20.7% grade 3 or higher) along with anemia, but a smaller percentage were found to have interstitial lung disease (ILD), a less common and more worrisome toxicity associated with breast cancer treatment. 13.6% of patients were found to have ILD of any grade, mostly grade 1 or 2, but four deaths (2.2%) were attributed to ILD, a concerning finding particularly given the long median time to onset of this complication of 193 days (25).

Metastatic HER2-positive breast cancer:

DESTINY-Breast 03 (N=524), a phase III study, directly compared T-DXd to T-DM1 in metastatic HER2+ breast cancer. An overall response occurred in 79.7% of patients who received T-DXd (95% CI = 74.3-84.4) and 34.2% of patients who received TDM1 (95% CI = 28.5-40.3). The median progression-free survival in the T-DXd arm could not be calculated at time of publication (95% CI 18.5 months-could not be estimated) whereas the T-DM1 arm had a median PFS of 6.8 months (95% CI 5.6-8.2). The most common drug-related adverse events in the T-DXd arm were nausea (72.8%), fatigue (44.7%), and vomiting (44.0%) (26). Most recently, early results from the DESTINY-Breast 02 trial were released demonstrating improved PFS in patients with unresectable or metastatic HER2+ breast cancer previously treated with T-DM1 who received T-DXd compared to treatment of physician’s choice (27).

Metastatic HER2 low breast cancer:

Recent data has demonstrated that HER2-targeted agents have a reach beyond traditional HER2-positive disease. While the pathologic HER2+ classification remains under active debate, it is now thought that about 40-50% of patients with breast cancer would meet current classification for HER2-low status (IHC 1+ or IHC 2+ without HER2 FISH amplification) (28). The DESTINY-Breast 04 trial demonstrated that patients with HER2-low breast cancer, previously considered ineligible for HER2-directed therapies, can clinically benefit from T-DXd. This phase III trial included 557 HER2-low patients with metastatic disease who had received 1-2 prior lines of chemotherapy, the large majority (88.7%) of which were hormone receptor-positive. Patients were randomized 2:1 to receive T-DXd compared to treatment of physician’s choice. The median PFS among all patients was 9.9 months in the T-DXd group compared to 5.1 months in the treatment of physician choice (TPC), leading to an overall survival difference of 23.4 months for patients receiving T-DXd versus 16.8 months for patients receiving treatment of physician’s choice (HR 0.64, p=0.001). As in prior trials, there was an increased incidence of ILD (12.1%) in T-DXd compared to chemotherapy, but overall patients receiving T-DXd had fewer grade 3 events (52.6% compared to 67.4%) (29).

Trastuzumab duocarmazine

ADC Structure:

Trastuzumab duocarmazine (SYD983) is an ADC under investigation in traditional HER2+ and HER2-low breast cancer.. This ADC contains an anti-HER2 antibody covalently bound to a payload of duocarmycin, a DNA alkylator, whichinduces cell death in both dividing and non-dividing cells. In preclinical studies, the drug demonstrated high activity against tumors and improved safety profile. In vitro, compared to trastuzumab, SYD983 induced cytotoxicity whereas trastuzumab appeared to partially inhibit cell proliferation. In mouse xenograft models, SYD983 used at a dose five times lower than standard-dose trastuzumab demonstrated equivalent anti-tumor activity. SYD983 was well tolerated in monkeys, with the highest non-severely toxic dose of >30 mg/kg. Hepatotoxicity, thrombocytopenia, and neuropathy were not seen, whereas these toxicities are found in T-DM1 (30). This ADC is also membrane permeable and would be expected to induce the bystander effect (24). The drug-antibody ratio of trastuzumab duocarmazine is relatively lower at 2.8:1 compared to T-DXd, which has a ratio of 8 (28).

Early Phase trials:

In a phase I first in human study, trastuzumab duocarmazine was tested in disease-agnostic solid tumors with HER2 IHC 1+ or greater. One death from pneumonitis occurred at the highest administered dose (2.4 mg/kg), leading to a phase 2 dose set at 1.2 mg/kg. Common toxicities that were seen included fatigue, conjunctivitis, and dry eye. 71% of patients had at least one ocular adverse event, a relatively unusual toxicity in this class of agents. In the breast cancer dose expansion group, ⅓ of patients with HER2+ disease achieved objective response by RECIST. Nine (28%) of 32 patients with HER2-low/HR+ disease and six (40%) of 15 patients with HER2-low/HR− negative disease achieved an objective response, notable given that this was in a heavily pretreated population including patients who had already received T-DM1 (31).

Metastatic HER2 positive breast cancer:

The Phase III TULIP trial is assessing trastuzumab-duocarmazine compared to TPC (with HER2-directed therapy) in patients with HER2+ metastatic breast cancer who progressed on either 2 lines of HER2-directed therapy or during/after T-DM1 (28). One arm of the I-SPY trial is testing trastuzumab duocarmazine with adriamycin and cyclophosphamide (NCT01042379) as neoadjuvant treatment looking at rates of pCR. Other trials are evaluating trastuzumab duocarmazine with paclitaxel (NCT04602117) and niraparib (NCT04235101) (24).

Ongoing Clinical Trials and Future Directions:

There are several other ADCs in clinical development, as outlined in Table 3. Ongoing trials are evaluating T-DXd in early stage breast cancer in the neoadjuvant and adjuvant settings. DESTINY-Breast 05 is comparing T-DXd against T-DM1 in patients with high-risk early stage HER2+ breast cancer who have residual disease following neoadjuvant chemotherapy with trastuzumab and a taxane (Clinicaltrials.gov). The DAISY trial is investigating biomarkers in patients with T-DXd associated with response to treatment and progression, and has found that efficacy of the agent is associated with degree of HER2 expression (32).

Table 3:

Select ongoing clinical trials of HER2-directed ADCs

Name	Description	ClinicalTrials.gov Identifier
TALENT	Phase II trial investigating use of T-DXd with and without anastrozole in early stage HR+ HER2 low breast cancer.	NCT04553770
DESTINY-Breast 02	Phase III trial looking at use of T-DXd in HER2+ mBC previously treated with T-DM1.	NCT03523585
DESTINY-Breast 06	Phase III trial comparing T-DXd with TPC in HR+ HER2 low mBC progressing on endocrine therapy	NCT04494425
DESTINY-Breast 09	Phase III trial for T-DXd in HER2+ mBC in patients with no prior chemotherapy or HER-2 directed therapy.	NCT04784715
TULIP	Phase III trial comparing trastuzumab-duocarmazine to TPC (with HER2-directed therapy) in patients with HER2+ mBC previously treated with 2 lines of HER2-directed therapy or T-DM1.	NCT03262935
I-SPY 2	Phase II trial, with one arm investigating trastuzumab-duocarmazine as part of a neoadjuvant treatment regimen.	NCT01042379

Given the relatively high molecular weight, it was initially considered uncertain whether ADCs would have activity across the blood brain barrier, but recent data has demonstrated potential for CNS effect (24). Brain metastases develop in up to 50% of patients with metastatic HER2+ breast cancer, and the presence of brain mets is associated with worse prognosis (33, 34). In the KAMILLA trial, T-DM1 was studied for CNS metastases in patients without prior cranial irradiation, demonstrating an ORR of 49.3% with a median duration of treatment 9.5 months (35). In DESTINY-Breast 01, the progression-free survival of patients treated with T-DXd with asymptomatic brain metastases was 18.1 months compared to 16.4 months for all patients (24, 25). The TUXEDO trial is a phase II study specifically investigating the effect of T-DXd on brain metastases in patients with HER-positive breast cancer (NCT04752059).

Summary and conclusions

In the past 10 years, rapid development of antibody-drug conjugates in the breast cancer space have offered promising novel treatment options for patients. HER2-positive breast cancer, historically associated with a worse prognosis in the setting of fewer treatment options, has proven to be a particularly active space for development in ADCs. In addition, approximately 40-50% of patients with breast cancer have been recently re-classified as HER2-low, a heretofore undefined category of patients who are now eligible for some HER2-directed therapies. Active areas of investigation include the role of these agents in brain metastases as well as sequencing ADCs in patients who are candidates for multiple therapies.

Figure 1:

Mechanism of action of three HER2-directed antibody-drug conjugates including differences in linker structure, payload, and drug-antibody ratio.

Data availability:

Data sharing is not applicable to this article as no datasets were generated or analyzed during the writing of this article.