Targeted therapy through the use of antibody drug conjugates holds great promise for the treatment of many malignancies, in particular, breast cancer.
Kelly M. Gaertner, PharmD
Department of Pharmacy
Roswell Park Cancer Institute
Kothai Divya Guruswamy Sangameswaran, MBBS
Department of Medicine
Roswell Park Cancer Institute
Thaer Khoury, MD, FCAP
Department of Pathology
Roswell Park Cancer Institute
Jessica S. Young, MD
Department of Surgery
Roswell Park Cancer Institute
Mateusz Opyrchal, MD, PhD
Roswell Park Cancer Institute
Targeted therapy through the use of antibody drug conjugates holds great promise for the treatment of many malignancies, in particular, breast cancer. The potential of this class of drugs for the treatment of trastuzumab-resistant, HER2-amplified metastatic breast cancer has been demonstrated in 2 pivotal phase III trials by trastuzumab emtansine (T-DM1). T-DM1 was recently evaluated in the neoadjuvant and adjuvant settings, as a firstline treatment for metastatic disease, and in combination with other agents. Although the study has failed to show improvement in progressionfree survival without the use of traditional chemotherapy, several additional agents that hold promise in the treatment of breast cancer are also reviewed in this article. CDX-011, an anti-GPNMB antibody conjugated to monomethyl auristatin E, has shown early efficacy in triple- negative breast cancer in an ongoing phase II clinical trial. IMMU-132, BAY94-9343, SGN-LIV1A, and BAY1129980 are compounds in the early stages of clinical development with potential in the treatment of breast cancer.
Breast cancer is one of the most common malignancies and the second-leading cause of cancer-related mortality in women in the United States. Despite many advances in the treatment of this disease, a significant portion of women have disease progression and recurrence. Metastatic disease continues to be incurable, leading to morbidity and increased mortality. Most women are diagnosed with metastatic disease due to failure of current standard treatments and inherent resistance of their breast cancer to systemic therapies. Therefore, there continues to be an urgent need for development of novel therapeutics with both increased efficacy and decreased toxicity.
Targeted therapy with antibodies (Abs) has demonstrated great success in the treatment of HER2- amplified breast cancers with the introduction of trastuzumab.1Antibodies and their derivatives exert their antitumor effect via several mechanisms, including signal blockade, antibody dependent cellular cytotoxicity (ADCC), phagocytosis, and activation of the complement system.2Unfortunately, despite the success of anti-HER2 Ab therapy, other Abs against the multitude of breast cancer-specific antigens have not shown the same efficacy.
Antibody drug conjugates (ADCs) are composed of a targeted antibody, linker, and potent cytotoxic agent (Figure). In order to deliver its toxic payload, the antibody must be targeted to a cancer antigen that is readily internalized. In contrast to Abs, ADCs are postulated to exert their antitumor effect solely through delivery of the toxic molecule directly to cancer cells. This novel mechanism of action has led to renewed interest in many breast cancer antigens as potential therapeutic targets.3,4
Antigen expression in normal tissues and nonspecific antibody binding are a major source of toxicities seen with ADCs. Therefore, selection of an antigen and Ab are critical to the development of an ADC. The linker binds the drug to the antibody. Linker instability and release of the drug systemically is another source of toxicity. Recent improvements in linker biochemistry have improved their systemic stability, resulting in decreased systemic release of the toxic drug and subsequently decreased systemic toxicity.5,6
There are 2 classes of linkers: cleavable, which become cleaved in the presence of low pH in the endosome, leading to release of the cytotoxic drug, and noncleavable, where the whole Ab is degraded before the drug is released and activated.7The process of attaching the drug to the Ab through a linker has undergone great improvements, with a more predictable number of drug molecules bound to the Ab moiety, resulting in enhanced efficacy and predictable toxicity.
Early ADCs concentrated on improved delivery of chemotherapy agents commonly used in the treatment of various malignancies. Frequently, these drugs did not achieve high enough concentrations in tumor tissues to allow for a meaningful cytotoxic effect.7ADCs are able to increase the therapeutic window for cytotoxic drugs. The drug moieties in use today are highly potent cytotoxic agents that could not be used alone due to their toxicities. The most common drugs in use with ADCs work through either inhibition of tubulin polymerization (auristatins and maytansinoids) or through DNA binding (calicheamicins).The optimal targets for ADCs are antigens expressed on the surface of cancer cells in high density, without expression on normal tissue.7,8The antigen also has to be readily internalized after forming a complex with the ADC to allow for release of the cytotoxic payload. There has been an enormous amount of research to identify possible targets on cancer cells. Unfortunately, even the most attractive cancer antigens have potential downsides: either they are not expressed at high enough density on the surface of cancer cells or only small percentages of tumor cells express it. Most of them are also expressed to some degree in normal tissue, which elicits concern for toxicity.9
The majority of nonconjugated Abs against breast cancer antigens have failed to show a clinical benefit. Their lack of efficacy could be due to incomplete signal blockade, insufficient ADCC, or complement response. The successful Abs became the most attractive for initial testing of ADCs. Because the ADCs’ mechanism of tumor killing is different from nonconjugated Abs, there is hope that they will expand the range of therapeutic cancer antigens.
Identification of HER2-amplified tumors and the development of trastuzumab has been one of the major success stories in the treatment of breast cancer. Trastuzumab is a very attractive Ab to use as a scaffold for the delivery of a cytotoxic payload. It is linked to the maytansinoid DM1 through a noncleavable linker.
The efficacy and safety of trastuzumab emtansine (T-DM1) in patients with metastatic breast cancer who were previously treated with trastuzumab has been evaluated in 2 phase III trials. The EMILIA study was a randomized, open-label, international trial including patients with HER2-amplified, unresectable, locally advanced or metastatic breast cancer who had previously received trastuzumab and a taxane. Patients were randomized 1:1 to receive T-DM1 or lapatinib (1250 mg daily) plus capecitabine (1000 mg per square meter of body-surface area every 12 hours). Primary endpoints included progression-free survival (PFS) by independent review, overall survival (OS), and safety. Secondary endpoints included PFS by investigator review, objective response rate, duration of response, and time to symptom progression.
A total of 495 patients were randomized to receive T-DM1, and 496 patients received lapatinib plus capecitabine. Baseline demographics were similar between groups. The median age was 53 years (range 24-84 years). The majority of patients had visceral involvement of disease (68%) and most (61%) had received 0 to 1 prior regimen for locally advanced or metastatic disease.
Progression-free survival by independent review was prolonged in the T-DM1 group compared with those receiving lapatinib plus capecitabine, 9.6 months versus 6.4 months, respectively (P = .001).10The benefit of T-DM1 was less clear among those aged 75 years or greater and with nonvisceral or nonmeasurable disease. Overall survival at the second interim analysis was a median of 30.9 months versus 25.1 months, favoring T-DM1; (hazard ratio [HR] 0.68; 95% CI, 0.55-0.85; P <.001). The estimated survival rates at 1 and 2 years were 85.2% and 64.7%, respectively, in the T-DM1 arm and 78.4% and 51.8%, respectively, in the lapatinib plus capecitabine arm.
For the secondary endpoints, investigator assessment revealed a similar benefit with T-DM1 as seen with the independent review, with a PFS of 9.4 months for T-DM1 and 5.8 months for lapatinib plus capecitabine (HR 0.66; 95% CI, 0.56-0.77; P <.001). The objective response rate was 43.6% for T-DM1 and 30.8% for lapatinib plus capecitabine, with a median duration of response of 12.6 months versus 6.5 months, respectively. Time to symptom progression, based on prespecified endpoints, was 7.1 months with T-DM1 compared with 4.6 months with lapatinib plus capecitabine.
In the T-DM1 group, 5.9% of patients discontinued T-DM1 due to adverse events (AEs). Serious AEs were reported less frequently in the T-DM1 group versus the lapatinib plus capecitabine group at 18% versus 15.5%, respectively. Grade 3 or greater AEs also occurred less frequently in the T-DM1 group, 40.8% versus 57%, respectively.
The most common grade 3 or 4 AEs with T-DM1 were thrombocytopenia (12.9%), elevated aspartate aminotransferase (4.3%), and alanine aminotransferase (2.9%). The first occurrence of grade 3 or 4 thrombocytopenia was most commonly reported during the first 2 cycles of T-DM1. Subsequently, a higher incidence of bleeding events was also reported with T-DM1 (29.8% vs 15.8%).10
TH3RESA was a phase III, randomized, multicenter, open-label trial in patients with HER2-amplified, unresectable locally advanced or recurrent breast cancer or metastatic breast cancer who had previously received both trastuzumab and lapatinib in the advanced setting and a taxane in any setting with progression after treatment, with a minimum of 2 HER2-directed regimens for advanced breast cancer.11Patients were randomized 2:1 to receive T-DM1 or treatment of physician’s choice (TPC). TPC was limited to chemotherapy (any single agent), hormonal therapy for HR-positive disease (single-agent or dual therapy), or HER2-directed therapy (single-agent, dual HER2-targeted therapy, or combination with either single-agent chemotherapy or single-agent hormonal therapy). Best supportive care alone was not allowed.
Primary endpoints included investigator-assessed PFS and OS. Secondary endpoints included investigator- assessed objective response, duration of objective response, 6-month and 1-year survival, safety, general health status or quality of life (QOL) and health-related QOL, symptom severity and interference, and pain ratings. A total of 404 patients were randomized to receive T-DM1 and 198 patients to receive TPC. The majority of patients were less than 65 years of age, with a median age of about 53 years, with metastatic disease and visceral disease involvement.
The median number of previous regimens for advanced breast cancer was 4 (range 1-19), excluding single-agent hormonal therapy. Among the group who received TPC, the majority (83%) received combination therapy with a HER2-directed agent (83%), which most commonly included trastuzumab plus chemotherapy (68%) or trastuzumab plus lapatinib (10%). The remaining 17% received single-agent chemotherapy, which most commonly included vinorelbine (32%), gemcitabine (16%), or other (17%).11
The primary endpoint of PFS was 6.2 months with T-DM1 compared with 3.3 months with trastuzumab plus lapatinib. Progression-free survival continued to be significantly improved in the T-DM1 group when compared with the subgroup of patients who received a trastuzumab-containing regimen as the TPC; however, the benefit with T-DM1 was consistent across all subgroups.
Overall survival at the first interim analysis was not statistically significant. Of those in the T-DM1 group, 15% had died versus 22% in the TPC group (HR 0.552; 95% CI, 0.369-0.826; P = .0034). However, estimated 6-month and 1-year survival rates were higher in the T-DM1 group. Regarding the secondary endpoints, the objective response was 31% with T-DM1 versus 9% with TPC (P <.0001). The median duration of response was 9.7 months with T-DM1, but had not yet been reached in the TPC group at the time of data cutoff.
Adverse events of any grade occurred in 94% of patients receiving T-DM1 versus 89% of those receiving trastuzumab plus lapatinib. Similar to the EMILIA trial, a higher incidence of grade 3 or greater thrombocytopenia was seen with T-DM1. The most common AEs with T-DM1, occurring in 10% or more of patients, included fatigue, asthenia, thrombocytopenia, dyspnea, and diarrhea. Regarding patient-reported outcomes, 57.8% of patients in the T-DM1 arm compared with 47.1% in the TPC arm experienced a clinically meaningful improvement in global health status based on prespecified criteria. The most bothersome symptoms in the T-DM1 group were fatigue and pain, while diarrhea and nausea/vomiting were reportedly more tolerable. Time to pain progression was not significantly different between groups.12
These 2 pivotal phase III trials have demonstrated that T-DM1 is active in HER2-amplified tumors that have progressed on trastuzumab. T-DM1 is generally well tolerated, with the most common AEs including thrombocytopenia, elevated alanine and aspartate aminotransferases, fatigue, diarrhea, and nausea/ vomiting.
The TH3RESA and EMILIA studies also included a biomarker analysis. Across all biomarker subgroups, median PFS was improved with T-DM1 versus TPC. A greater relative risk reduction for PFS was seen in those patients expressing greater than median HER2 mRNA levels.
Abnormalities in the PI3K-Akt pathway have been associated with resistance to trastuzumab therapy. PIK3CA mutations are the most common alteration in this pathway and have been found to confer resistance to anti-HER2 therapy. PIK3CA mutations were not associated with a decreased PFS in the T-DM1- treated arm. In the EMILIA analysis, the PIK3CA mutations were associated with a shorter PFS in the control-treated arm, although this was not seen in the TH3RESA analysis.
While the biomarkers need to be confirmed in a prospective study, these data present early evidence that T-DM1 can overcome resistance to anti-HER2 therapy due to PIK3CA mutations. This is most likely due to the difference in mechanism of action of T-DM1.13,14
TDM4874g (NCT01196052) was a phase II trial that investigated the safety and feasibility of T-DM1 following anthracycline-based chemotherapy in the adjuvant or neoadjuvant setting for patients with HER2-amplified early-stage breast cancer (EBC). Patients received 4 cycles of adjuvant or neoadjuvant T-DM1 after doxorubicin plus cyclophosphamide (every 2 or 3 weeks for 4 cycles) or 5-fluorouracil plus epirubicin plus cyclophosphamide (every 3 weeks for 3 to 4 cycles). Patients could continue T-DM1 for up to 17 cycles of HER2-directed therapy. After the first 4 cycles of T-DM1, radiotherapy and 3 to 4 cycles of docetaxel with or without trastuzumab were optional before continuing T-DM1.
The primary endpoints were safety and rate of prespecified cardiac events occurring within the first 12 weeks of treatment with T-DM1. Of the 153 patients included, 99 received treatment in the adjuvant setting. The majority of patients (73.9%) were between 41 and 64 years of age. Almost all patients (96.7%) received T-DM1, for a median of 14 cycles, and 82% completed the planned approximate year of HER2-directed therapy.15
Of the 50 patients treated with neoadjuvant therapy and undergoing surgery, the pathologic complete response (pCR) rate was 56%. The pCR rate was 51.7% for patients with hormone receptor-positive disease and 61.9% for patients with hormone receptor-negative disease. No protocol prespecified cardiac events occurred and the mean LVEF remained stable throughout therapy. The most common AEs included nausea, headache, epistaxis, asthenia, pyrexia, fatigue, arthralgia, thrombocytopenia, and myalgia.
This study concluded that T-DM1 appears feasible and safe with concurrent radiotherapy or hormonal therapy and should be further investigated in EBC. These results compared favorably with the pCR rates achieved in the TRYPHENA trial (approximately 50%), which evaluated double anti-Her2 therapy.16Neoadjuvant T-DM1 is also being evaluated as part of the ongoing I-SPY protocol (NCT01042379).
The MARIANNE trial is the first phase III study evaluating the combination of a targeted antibody and an ADC for first-line metastatic breast cancer. Patients are randomized 1:1:1 to one of three arms: trastuzumab plus docetaxel, T-DM1 plus pertuzumab, or T-DM1 plus placebo. The primary endpoint is PFS by independent review and secondary endpoints include safety, overall response rate, OS, duration of response, and QOL. The company sponsoring the trial has announced in news release that the results did not reach the primary end point for either of the T-DM1 containing arm although it was noted that the noninferiority goals were met. The full results will have to be evaluated before the full impact of these results can be assessed. The company is planning to release the full results in the near future.17
chemo indicates chemotherapy; DFS, disease-free survival; DLT, dose-limiting toxicity; LA, locally advanced; MBC, metastatic breast cancer; MTD, maximum tolerated dose; ORR, overall response rate; pCR,pathologic complete response; PFS, progression-free survival; PK, pharmacokinetics; QTc, heart-rate corrected QT, TDMI, trastuzumab emtansine; TRR, tumor response rate.
T-DM1 is being evaluated alone and in combination with chemotherapy and other targeted agents in multiple trials (Table 1). Over the next several years, we will be learning more about its full scope of activity, modes of resistance, and possibly better methods of choosing patients for therapy. We will continue to learn more about this class of drugs and the best possible partners for combination therapies.
Additional ADC therapies are currently in early clinical trials (Table 2). One of these, glembatumumab vedotin (CDX-011), is an antibody-drug conjugate composed of CR011, a fully human IgG2 monoclonal antibody againstglycoprotein nonmetastatic gene B(GPNMB), conjugated via a valine-citrulline link to the potent microtubule inhibitor monomethyl auristatin E (MMAE). GPNMB has been found to be expressed in a multitude of cancers18-20and overexpressed in aggressive breast cancer cell lines.21GPNMBexpression has also been associated with high endothelial cell density and was shown to induce endothelial cell migration in vitro.21It has also been found to upregulate metalloproteases.22,23Its exact function in breast cancer is not known, although it may play a role in invasion and metastasis.GPNMBis also expressed in normal tissue, with its mRNA detected in bones, adipose, thymus, skin, placenta, heart, kidneys, pancreas, lungs, liver and skeletal muscle,24-26which increases the potential for toxicity with any anti-GPNMBtherapy.
It appears that normally functioningGPNMBis localized in the intracellular component, while its expression in tumor cells is mostly on the cytoplasmic membrane.27-29This characteristic makes it a very attractive target for ADCs and minimizes the potential for toxicities from this therapy. In preclinical studies, a single dose of CDX-011 led to the regression of theGPNMB-expressing breast cancer cell line MDAMB- 468 in an animal model.29CDX-011 is currently being investigated in patients with melanoma and breast cancer.
A phase I/II study assessed the safety and activity of glembatumumab in patients with locally advanced and metastatic breast cancer and examined the relationship betweenGPNMBexpression and response. Patients included those with unresectable or metastatic breast cancer who had undergone at least 2 prior chemotherapy regimens. Use of a taxane, anthracycline, and capecitabine (as well as trastuzumab if HER2 amplified) were required. A 3+3 dose escalation design was used. A total of 42 heavily pretreated patients were enrolled, with a median of 7 prior regimens. Three patients were treated at low dose and 10 at the maximally tolerated dose (MTD) had triplenegative breast cancer. Eighty-four percent of tumors tested were positive forGPNMB. Worsening neuropathy was identified as the dose-limiting toxicity.
The primary endpoint was reached in the phase II study, with 33% of the 27 evaluable patients achieving PFS at 12 weeks. At the phase II dose, median PFS was 9.1 weeks for all patients, 17.9 weeks for those with triple-negative disease, and 18 weeks forGPNMB-expressing tumors. Nondose-limiting hematologic toxicity and mild to moderate rash were observed. The investigators concluded that glembatumumab vedotin has an acceptable safety profile as well as antitumor activity in heavily pretreated patients with metastatic breast cancer.30Although the response rates in nonselected patient populations were low, the results from theGPNMB-expressing triple-negative breast cancer cohort were encouraging. These results speak to the targeted nature of this therapy and bring hope for new treatment options in this difficult-to-treat subset. Further studies evaluating glembatumumab are ongoing inGPNMB-expressing tumors (NCT01997333).
Trop-2 is another potential cancer antigen that is a subject of interest in breast cancer. Trop-2 is a calcium signal transducer that has been associated with stimulation of growth of malignant cells.31Its full biological function has not yet been elucidated. Trop-2 over-expression has been found in a variety of cancers, including breast,32ovarian,33prostate,34endometrial,35lung,36colon,37and thyroid,38but it has limited cytoplasmic membrane presence in normal tissue. Its expression on the cytoplasmic membrane, as opposed to expression in the intracellular component, is associated with worse outcomes in patients with breast cancer.32Therefore, Trop-2 is an attractive target, especially in more aggressive breast cancer phenotypes, such as the triple-negative subtype.
IMMU-132 is an ADC consisting of hRS7, a readily internalized Ab against Trop-2, which is conjugated through a cleavable linker to SN-38, an active metabolite of CPT-11. Early data from an ongoing phase I/II clinical trial in epithelial cancers reported at the 2014 American Society of Clinical Oncology annual meeting were encouraging. In a 3+3 design, neutropenia was found to be the dose-limiting toxicity, and the phase II component proceeded at doses of 8 mg/kg and 10 mg/kg in patients with colorectal, small-cell lung, and triple-negative breast cancers. Of the 36 patients in the phase II component, 14 had an assessment of response and 8 were found to have at least stable disease. Interestingly, 5 patients were found to be homozygous UGT1A1 *28/*28, with 2 of them experiencing more severe hematologic and gastrointestinal toxicities. This study is ongoing. A new phase II study evaluating IMMU-132 in combination with carboplatin in patients with triple-negative breast cancer is opening soon.39
Three other ADCs with potential for the treatment of breast cancer are in very early clinical development. BAY94-9343 is an ADC consisting of IgG1 directed against the cell surface glycoprotein mesothelin, which in turn is conjugated through a cleavable disulfide-containing linker to the maytansinoid DM4. Mesothelin has been found to be expressed in over 60% of triple-negative breast cancers.40,41BAY94- 9343 is currently undergoing phase I clinical testing (NCT01439152).
SGN-LIV1A is an ADC with a backbone of Ab against the antisolute family 39 zinc transporter member 6 (LIV-1) conjugated with a cleavable linker to MMAE. It is currently being tested in a phase I clinical trial (NCT01969643). LIV-1 expression is induced by estrogen stimulation and has been shown to lead to epithelial-to-mesenchymal transition (EMT), progression of the disease, and metastasis.42,43Expression of LIV-1 has been found in postendocrine treated tumors and, to a lesser degree, in triple-negative breast cancer cells. Preclinical results have shown very robust activity against cells expressing LIV-1.44
DLTs indicates dose-limiting toxicities; MTD, maximum tolerated dose; TNBC, triple-negative breast cancer.
Lastly, BAY1129980 is an anti-C4.4a IgG1 antibody conjugated to auristatin. C4.4a has been found to be strongly expressed in lung and breast cancers.45In breast cancer, C4.4a expression has been associated with better survival and has been found mostly in the HER2-amplified subset.46The full biologic function of C4.4a is unknown at this time, although it has been linked to invasion and metastasis. BAY1129980 is currently in phase I testing (NCT02134197). Results of these early clinical trials are eagerly awaited.ADCs offer enormous potential for developing targeted therapies against breast cancer, as well as other malignancies. The targeted nature of this therapy results in an improved therapeutic window for the cytotoxic payload, which allows for the use of highly toxic molecules that have a prohibitive toxicity profile if used alone. Currently, for the ADC to be functional it needs to be internalized after binding to its intended antigen. Developing new methods of release of a toxic payload may lead to identification of new targets, as well as increased bystander effect with resulting enhanced antitumor effect.
Although Abs have the advantage of long in vivo half-lives, their large size may inhibit penetration into solid tumors. ADCs encounter the same limitation. Further work on improving tumor penetration may lead to novel combinations along with the development of conjugates to antibody fragments or other targeted delivery systems with much smaller size.
Currently, there are over 50 ADC compounds being investigated at various stages of development. It is quite conceivable that more compounds will show efficacy in the treatment of breast cancer as new antigens are discovered. Theoretically, as more of these compounds reach the clinic, they should be active against any cancer cell expressing high enough density of the given antigen, irrespective of tissue of origin. Molecular pathology will continue to increase in importance in identifying treatment options for our patients. Scientists and clinicians will have to decide which tumors have sufficient probability of expressing a given antigen to be tested. As the era of personalized medicine arrives, tumor antigens should be a part of the gamut of tests for choosing the best treatment options for patients.