Purpose:

MEDI3726 is an antibody–drug conjugate targeting the prostate-specific membrane antigen and carrying a pyrrolobenzodiazepine warhead. This phase I study evaluated MEDI3726 monotherapy in patients with metastatic castration-resistant prostate cancer after disease progression on abiraterone and/or enzalutamide and taxane-based chemotherapy.

Patients and Methods:

MEDI3726 was administered at 0.015–0.3 mg/kg intravenously every 3 weeks until disease progression/unacceptable toxicity. The primary objective was to assess safety, dose-limiting toxicities (DLT), and MTD/maximum administered dose (MAD). Secondary objectives included assessment of antitumor activity, pharmacokinetics, and immunogenicity. The main efficacy endpoint was composite response, defined as confirmed response by RECIST v1.1, and/or PSA decrease of ≥50% after ≥12 weeks, and/or decrease from ≥5 to <5 circulating tumor cells/7.5 mL blood.

Results:

Between February 1, 2017 and November 13, 2019, 33 patients received MEDI3726. By the data cutoff (January 17, 2020), treatment-related adverse events (TRAE) occurred in 30 patients (90.9%), primarily skin toxicities and effusions. Grade 3/4 TRAEs occurred in 15 patients (45.5%). Eleven patients (33.3%) discontinued because of TRAEs. There were no treatment-related deaths. One patient receiving 0.3 mg/kg had a DLT of grade 3 thrombocytopenia. The MTD was not identified; the MAD was 0.3 mg/kg. The composite response rate was 4/33 (12.1%). MEDI3726 had nonlinear pharmacokinetics with a short half-life (0.3–1.8 days). The prevalence of antidrug antibodies was 3/32 (9.4%), and the incidence was 13/32 (40.6%).

Conclusions:

Following dose escalation, no MTD was identified. Clinical responses occurred at higher doses, but were not durable as patients had to discontinue treatment due to TRAEs.

Translational Relevance

Prostate-specific membrane antigen (PSMA) overexpression is evident in >70% of patients with metastatic castration-resistant prostate cancer (mCRPC) and may have diagnostic and therapeutic utility. MEDI3726 is an antibody–drug conjugate targeting PSMA and carrying a pyrrolobenzodiazepine warhead. Preclinical in vivo studies indicated the potential for MEDI3726 to exhibit antitumor activity, with a high degree of specificity to PSMA-expressing tumors. This phase I study demonstrated limited clinical activity with MEDI3726 monotherapy in patients with mCRPC after progression on abiraterone and/or enzalutamide and taxane-based chemotherapy, a population that generally has poor outcomes.

Despite the increase in the range of treatment options for metastatic castration-resistant prostate cancer (mCRPC) over the past decade, including cytotoxic agents, anti-androgens, PARP inhibitors, targeted alpha therapy and autologous cellular immunotherapy, survival outcomes remain relatively poor (1). Clinical datasets suggest a 5-year overall survival (OS) rate of 25%–30% and a median OS of approximately 13 months (1–3). There is an urgent unmet need for novel therapeutic approaches.

Prostate-specific membrane antigen (PSMA) is a transmembrane protein that has limited normal tissue expression but is upregulated in nearly all prostate cancers (4, 5). PSMA overexpression is evident in >70% of patients with mCRPC and may have diagnostic and therapeutic potential (6, 7). MEDI3726 is a highly potent antibody–drug conjugate (ADC), comprising an engineered version of an anti-PSMA IgG1κ antibody (J591) site-specifically conjugated with pyrrolobenzodiazepine (PBD) dimers (SG3199) at a drug–antibody ratio of approximately 2 (8). MEDI3726 binds PSMA expressed on the tumor cell membrane and is internalized, whereupon intracellular cathepsin cleavage mediates release of the PBD dimer payload. The PBD dimer induces the formation of DNA cross-links, thereby triggering cell death via apoptosis. MEDI3726 antitumor activity has been demonstrated in a number of murine xenograft models of mCRPC (8).

Here we present safety and preliminary efficacy data from the dose-escalation portion of a first-in-human, open-label, phase I/Ib, multicenter study (NCT02991911).

Eligibility criteria

This study enrolled patients with a histologically confirmed diagnosis and documented evidence of mCRPC (defined as ≥1 metastatic lesion on either bone scan or CT/MRI) after prior receipt of abiraterone and/or enzalutamide (≥12 weeks) and taxane-based chemotherapy. Eligible patients had radiographic progression of soft-tissue disease by RECIST v1.1 and/or evidence of bone metastasis by Prostate Cancer Clinical Trials Working Group 3 (PCWG3) criteria and/or PSA progression confirmed by local laboratory testing (9). PSA progression was defined as an initial reference PSA level of ≥1 ng/mL followed by ≥2 consecutive recordings of rising PSA levels, taken ≥1 week apart; the final PSA value used to determine progression had to be within the 28-day screening period. Further eligibility criteria included serum testosterone <50 ng/dL (<2.0 nM) within 28 days of the first dose, an Eastern Cooperative Oncology Group performance status of 0–1, and adequate organ, bone marrow, and cardiac function.

Key exclusion criteria included previous exposure to a PSMA-directed therapy, neuroendocrine differentiation and/or small-cell prostate cancer and receipt of conventional or investigational anticancer treatment ≤21 days prior to the first dose of MEDI3726.

Study design

The dose-escalation portion of the trial comprised a series of cohorts (n = 1–12 patients each) to be recruited sequentially at planned doses of 0.015, 0.03, 0.06, 0.15, 0.3, and 0.6 mg/kg utilizing a modified toxicity probability interval algorithm, with a target dose-limiting toxicity (DLT) rate of 30% and an equivalence interval of 25%–35% (10). Patients received single-agent MEDI3726 by intravenous infusion over a 1-hour period every 3 weeks. Initially, 1 patient received MEDI3726 at the starting dose of 0.015 mg/kg. If this patient experienced an adverse event (AE) of grade ≥2, 2 additional patients would be enrolled at 0.015 mg/kg; otherwise, escalation to the next cohort could proceed. Further dose escalation was determined by emerging safety and pharmacokinetic data, with a minimum of 3 patients enrolled per cohort. The protocol was amended to alter the pattern of dose escalation from the approximately 100% proportional increase used initially to incorporate selected intermediate dose cohorts (0.2 and 0.25 mg/kg). All available safety, pharmacokinetic, and efficacy data were evaluated by a dose-escalation committee prior to each dose escalation.

The study, which was approved by relevant independent ethics committees and Institutional Review Boards, was conducted in accordance with the Declaration of Helsinki, International Council for Harmonisation, Good Clinical Practice, and applicable regulatory requirements. All patients provided written informed consent.

Objectives and endpoints

The primary objective was to evaluate the safety and tolerability of single-agent MEDI3726, describe DLTs and determine the MTD or the maximum administered dose (MAD) through assessment of AEs, serious AEs (SAE), and laboratory parameters/clinical tests.

Secondary objectives included the preliminary evaluation of antitumor activity through a composite response. Composite response was defined as a confirmed response by RECIST v1.1, and/or a ≥50% decrease in PSA after ≥12 weeks (PSA50 response), and/or a conversion in circulating tumor cell (CTC) count (defined as a decrease from ≥5 to <5 cells/7.5 mL blood), with a confirmatory assessment ≥4 weeks later (CTC response).

Additional secondary endpoints for the assessment of antitumor response included objective response per RECIST v1.1, time to response (TTR) and duration of response (DoR; by RECIST v1.1 and by the composite response), percentage change from baseline in PSA and percentage change from baseline in CTCs. Other secondary objectives included assessment of pharmacokinetic parameters including maximum observed concentration, area under the concentration–time curve, clearance, and terminal half-life; and evaluation of immunogenicity, defined as the percentage of patients who develop antidrug antibodies (ADA).

Exploratory objectives included identifying genomic alterations as potential non-PSMA biomarkers for response, assessed in baseline circulating tumor DNA (ctDNA) samples using a targeted sequencing panel.

Pharmacokinetics

Bioanalytical strategy for the characterization of pharmacokinetics and absorption, distribution, metabolism, and excretion (ADME) of ADC candidate therapeutics must be designed on the basis of structural characteristics and expectations regarding the potential catabolism of these complex biotherapeutics (11, 12). Because of the potential complexity of catabolites resulting from the structural design and conjugation strategy of MEDI3726, four different bioanalytical methods measuring six analyte classes were employed. Validated hybrid ligand-binding assay LC/MS-MS methods were employed for the quantification of total ADC (MEDI3726 and all forms of its catabolites that contain at least one heavy chain with one warhead), active ADC (MEDI3726 and all forms of its catabolites that contain at least one heavy chain with one warhead and one light chain), total IgG (MEDI3726 and all catabolites that contain at least one heavy chain), intact IgG (all forms of MEDI3726 and its catabolites that contain at least one heavy chain associated with one light chain) and conjugated SG3199 (all the warheads conjugated to the heavy chain among MEDI3726 and all forms of its catabolites; ref. 13). A validated LC/MS-MS method was used for the quantification of free (unconjugated) SG3199 (13). The concentrations were then analyzed using Phoenix WinNonlin software (Pharsight) to generate noncompartmental pharmacokinetic parameters. On day 1, plasma samples for pharmacokinetic analysis were collected pre-dose, immediately post-infusion, 2 hours post-infusion, and 6 hours post-infusion. During cycle 1, plasma samples were also collected on days 2, 3, 8, and 15. On designated days of MEDI3726 administration between day 22 (cycle 2) and day 106 (cycle 6), plasma samples were collected pre-dose and immediately post-infusion.

Immunogenicity

The potential of MEDI3726 to elicit an antibody response was evaluated using a tiered approach comprising screening, confirmation, and titer, via a validated bridging immunoassay. This method utilizes a mixture of the biotinylated MEDI3726 and ruthenylated MEDI3726, incubated with patient samples to allow the formation of immunocomplexes between drug and ADAs. This method has sufficient sensitivity (limit of detection = 88.54 ng/mL) and drug tolerance for adequate detection of ADAs (14). Serum samples for immunogenicity assessment were collected on day 1 (pre-dose, baseline), post-dose on day 15 and pre-dose on day 22, then every 3 weeks for cycles 3–6, then every 12 weeks thereafter (on designated days of MEDI3726 administration). ADA prevalence was calculated as the percentage of patients who had a baseline positive result only or both a baseline and a post-treatment positive result where the titer increase from baseline was <4-fold. ADA incidence was calculated as the percentage of patients with treatment-induced ADAs (baseline negative and post-treatment positive results) and those with treatment-boosted ADAs (both baseline and post-treatment positive results where the titer increase was >4-fold).

Statistical analyses

All analyses were based on the as-treated population (which included all patients who received any dose of investigational product). Evaluation of the MTD was based on the DLT-evaluable population (defined as patients who remained in the study for a minimum of 21 days from the first dose of MEDI3726). AEs were graded per NCI Common Terminology Criteria for Adverse Events v4.03. Efficacy analyses were based on PCWG3 criteria (9). Time-to-event data were summarized using Kaplan–Meier estimates (median time and corresponding 95% confidence intervals).

Data underlying the findings described in this article may be obtained in accordance with AstraZeneca's data sharing policy described at: https://astrazenecagrouptrials.pharmacm.com/ST/Submission/Disclosure.

Between February 1, 2017 and November 13, 2019, 33 patients were recruited and received ≥1 dose of MEDI3726. As of the data cutoff (January 17, 2020), median follow-up was 5.9 months (Table 1) and all patients had discontinued treatment. Patient demographics and baseline characteristics were similar across the dosage cohorts (Table 1). The majority of patients were white (29; 87.9%) and median age was 71 years. Patients had a median PSA of 182.0 ng/mL. The median number of prior therapy regimens was 4.

Table 1.

Patient demographics and disease characteristics at baseline.

MEDI3726 dose, n (%)
0.015 mg/kg0.03 mg/kg0.06 mg/kg0.15 mg/kg0.2 mg/kg0.25 mg/kg0.3 mg/kgTotal
Characteristicn = 1n = 3n = 3n = 4n = 10n = 7n = 5N = 33
Median age (range), years 73.0 (73) 75.0 (71–81) 71.0 (69–72) 73.0 (66–75) 66.0 (55–71) 73.0 (59–79) 70.0 (54–78) 71.0 (54–81) 
Race, n (%) 
 Black or African American 1 (25.0) 1 (10.0) 2 (6.1) 
 White 1 (100) 3 (100) 2 (66.7) 3 (75.0) 8 (80.0) 7 (100) 5 (100) 29 (87.9) 
 Other 1 (33.3) 1 (10.0) 2 (6.1) 
Median number of prior regimens 4.0 4.0 4.0 6.0 4.0 4.0 5.0 4.0 
Prior regimensa, n (%) 
 Systemic therapy 1 (100) 3 (100) 3 (100) 4 (100) 10 (100) 7 (100) 5 (100) 33 (100) 
 Radiotherapy 1 (100) 2 (66.7) 3 (100) 3 (75.0) 7 (70.0) 5 (71.4) 3 (60.0) 24 (72.7) 
 Surgery 1 (100) 1 (33.3) 1 (33.3) 2 (50.0) 3 (30.0) 3 (42.9) 11 (33.3) 
 Abiraterone 1 (100) 3 (100) 2 (66.6) 2 (50.0) 4 (40.0) 4 (57.1) 3 (60.0) 19 (57.6) 
 Enzalutamide 1 (100) 1 (33.3) 2 (66.7) 4 (100) 8 (80.0) 7 (100) 4 (80.0) 27 (81.8) 
 Docetaxel 1 (100) 3 (100) 3 (100) 4 (100) 9 (90.0) 7 (100) 5 (100) 32 (97.0) 
 Cabazitaxel 1 (33.3) 3 (75.0) 5 (50.0) 5 (71.4) 4 (80.0) 18 (54.5) 
Median Gleason score 8.0 8.0 9.0 8.0 9.0 7.0 8.0 8.5 
Median time from initial diagnosis to study entry, months (range) 161 (161) 99.5 (50–197) 142 (53–181) 155 (19–202) 46 (35–173) 58 (31–170) 86 (46–117) 73 (19–202) 
Stage at initial diagnosisb, n 
 I 
 II 
 III 
 IV 17 
Median PSA, ng/mL 3,631.2 93.5 70.7 314.7 203.0 117.5 456.7 182.0 
Median duration of follow-up, months (range) 4.2 (4.2) 5.9 (4.1–22.8) 15.0 (11.7–23.5) 5.3 (1.8–10.5) 8.8 (1.7–25.3) 5.4 (3.5–8.7) 5.0 (2.6–19.5) 5.9 (1.7–25.3) 
MEDI3726 dose, n (%)
0.015 mg/kg0.03 mg/kg0.06 mg/kg0.15 mg/kg0.2 mg/kg0.25 mg/kg0.3 mg/kgTotal
Characteristicn = 1n = 3n = 3n = 4n = 10n = 7n = 5N = 33
Median age (range), years 73.0 (73) 75.0 (71–81) 71.0 (69–72) 73.0 (66–75) 66.0 (55–71) 73.0 (59–79) 70.0 (54–78) 71.0 (54–81) 
Race, n (%) 
 Black or African American 1 (25.0) 1 (10.0) 2 (6.1) 
 White 1 (100) 3 (100) 2 (66.7) 3 (75.0) 8 (80.0) 7 (100) 5 (100) 29 (87.9) 
 Other 1 (33.3) 1 (10.0) 2 (6.1) 
Median number of prior regimens 4.0 4.0 4.0 6.0 4.0 4.0 5.0 4.0 
Prior regimensa, n (%) 
 Systemic therapy 1 (100) 3 (100) 3 (100) 4 (100) 10 (100) 7 (100) 5 (100) 33 (100) 
 Radiotherapy 1 (100) 2 (66.7) 3 (100) 3 (75.0) 7 (70.0) 5 (71.4) 3 (60.0) 24 (72.7) 
 Surgery 1 (100) 1 (33.3) 1 (33.3) 2 (50.0) 3 (30.0) 3 (42.9) 11 (33.3) 
 Abiraterone 1 (100) 3 (100) 2 (66.6) 2 (50.0) 4 (40.0) 4 (57.1) 3 (60.0) 19 (57.6) 
 Enzalutamide 1 (100) 1 (33.3) 2 (66.7) 4 (100) 8 (80.0) 7 (100) 4 (80.0) 27 (81.8) 
 Docetaxel 1 (100) 3 (100) 3 (100) 4 (100) 9 (90.0) 7 (100) 5 (100) 32 (97.0) 
 Cabazitaxel 1 (33.3) 3 (75.0) 5 (50.0) 5 (71.4) 4 (80.0) 18 (54.5) 
Median Gleason score 8.0 8.0 9.0 8.0 9.0 7.0 8.0 8.5 
Median time from initial diagnosis to study entry, months (range) 161 (161) 99.5 (50–197) 142 (53–181) 155 (19–202) 46 (35–173) 58 (31–170) 86 (46–117) 73 (19–202) 
Stage at initial diagnosisb, n 
 I 
 II 
 III 
 IV 17 
Median PSA, ng/mL 3,631.2 93.5 70.7 314.7 203.0 117.5 456.7 182.0 
Median duration of follow-up, months (range) 4.2 (4.2) 5.9 (4.1–22.8) 15.0 (11.7–23.5) 5.3 (1.8–10.5) 8.8 (1.7–25.3) 5.4 (3.5–8.7) 5.0 (2.6–19.5) 5.9 (1.7–25.3) 

Abbreviation: PSA, prostate-specific antigen.

aOther therapies received by >10% of patients included bicalutamide (66.7%), leuprorelin acetate (36.4%), goserelin (24.2%), prednisone (24.2%), radium RA 223 dichloride (24.2%), sipuleucel-T (24.2%), denosumab (21.2%), and leuprorelin (12.1%).

bStage was unknown for 2 patients in the 0.03 mg/kg group and was not recorded for 1 patient in the 0.25 mg/kg group.

Safety and tolerability

Treatment-emergent AEs (TEAE) were observed in all 33 (100%) patients. Grade 3/4 TEAEs and treatment-emergent SAEs occurred in 23 (69.7%) and 18 (54.5%) patients, respectively.

Treatment-related AEs (TRAE) occurred in 30 (90.9%) patients and were grade 3/4 in 15 (45.5%; Table 2). The most frequent grade 3/4 TRAEs were increased gamma-glutamyltransferase (GGT; 21.2%), thrombocytopenia, capillary leak syndrome (each 9.1%), increased alanine aminotransferse (ALT), and fatigue (each 6.1%; Table 3). One patient receiving 0.3 mg/kg MEDI3726 developed a DLT of grade 3 thrombocytopenia. Eleven (33.3%) patients had MEDI3726-related SAEs; the most frequent were pleural effusion (15.2%) and capillary leak syndrome (9.1%). All MEDI3726-related SAEs occurred at doses ≥0.15 mg/kg. Eleven (33.3%) patients discontinued study treatment due to TRAEs; 10 of them were receiving doses ≥0.2 mg/kg. Four patients died because of AEs (1 in the 0.15 mg/kg cohort and 3 in the 0.25 mg/kg cohort). These deaths did not occur on treatment and none were considered directly related to MEDI3726.

Table 2.

Safety summary.

MEDI3726 dose, n (%)
0.015 mg/kg0.03 mg/kg0.06 mg/kg0.15 mg/kg0.2 mg/kg0.25 mg/kg0.3 mg/kgTotal
Safety parametern = 1n = 3n = 3n = 4n = 10n = 7n = 5N = 33
All-cause AEs 
 All treatment-emergent AEs 1 (100) 3 (100) 3 (100) 4 (100) 10 (100) 7 (100) 5 (100) 33 (100) 
 Grade 3/4 treatment-emergent AEs 1 (100) 1 (33.3) 1 (33.3) 4 (100) 5 (50.0) 6 (85.7) 5 (100) 23 (69.7) 
 Treatment-emergent SAEs 1 (33.3) 3 (75.0) 5 (50.0) 5 (71.4) 4 (80.0) 18 (54.5) 
 AEs leading to discontinuation 1 (33.3) 5 (50.0) 3 (42.9) 2 (40.0) 11 (33.3) 
 SAEs and/or grade ≥3 AEs 1 (100) 1 (33.3) 1 (33.3) 4 (100) 8 (80.0) 7 (100) 5 (100) 27 (81.8) 
 Deaths (grade 5 AEs) 1 (25.0) 3 (42.9) 4 (12.1) 
MEDI3726-related AEs 
 Any AE 1 (100) 1 (33.3) 2 (66.7) 4 (100) 10 (100) 7 (100) 5 (100) 30 (90.9) 
 Grade 3/4 AEs 1 (100) 3 (75.0) 1 (10.0) 6 (85.7) 4 (80.0) 15 (45.5) 
 SAEs 1 (25.0) 4 (40.0) 3 (42.9) 3 (60.0) 11 (33.3) 
 AEs leading to discontinuation 1 (33.3) 5 (50.0) 3 (42.9) 2 (40.0) 11 (33.3) 
 AEs leading to interruption 2 (20.0) 1 (14.3) 3 (9.1) 
 AEs leading to dose reduction 2 (28.6) 2 (6.1) 
MEDI3726 dose, n (%)
0.015 mg/kg0.03 mg/kg0.06 mg/kg0.15 mg/kg0.2 mg/kg0.25 mg/kg0.3 mg/kgTotal
Safety parametern = 1n = 3n = 3n = 4n = 10n = 7n = 5N = 33
All-cause AEs 
 All treatment-emergent AEs 1 (100) 3 (100) 3 (100) 4 (100) 10 (100) 7 (100) 5 (100) 33 (100) 
 Grade 3/4 treatment-emergent AEs 1 (100) 1 (33.3) 1 (33.3) 4 (100) 5 (50.0) 6 (85.7) 5 (100) 23 (69.7) 
 Treatment-emergent SAEs 1 (33.3) 3 (75.0) 5 (50.0) 5 (71.4) 4 (80.0) 18 (54.5) 
 AEs leading to discontinuation 1 (33.3) 5 (50.0) 3 (42.9) 2 (40.0) 11 (33.3) 
 SAEs and/or grade ≥3 AEs 1 (100) 1 (33.3) 1 (33.3) 4 (100) 8 (80.0) 7 (100) 5 (100) 27 (81.8) 
 Deaths (grade 5 AEs) 1 (25.0) 3 (42.9) 4 (12.1) 
MEDI3726-related AEs 
 Any AE 1 (100) 1 (33.3) 2 (66.7) 4 (100) 10 (100) 7 (100) 5 (100) 30 (90.9) 
 Grade 3/4 AEs 1 (100) 3 (75.0) 1 (10.0) 6 (85.7) 4 (80.0) 15 (45.5) 
 SAEs 1 (25.0) 4 (40.0) 3 (42.9) 3 (60.0) 11 (33.3) 
 AEs leading to discontinuation 1 (33.3) 5 (50.0) 3 (42.9) 2 (40.0) 11 (33.3) 
 AEs leading to interruption 2 (20.0) 1 (14.3) 3 (9.1) 
 AEs leading to dose reduction 2 (28.6) 2 (6.1) 

Abbreviations: AE, adverse response; SAE, serious AE.

Table 3.

Most common MEDI3726-related AEs (overall incidence >10%).

MEDI3726 dose, n (%)
0.015 mg/kg0.03 mg/kg0.06 mg/kg0.15 mg/kg0.2 mg/kg0.25 mg/kg0.3 mg/kgTotal
n = 1n = 3n = 3n = 4n = 10n = 7n = 5N = 33
AEOverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4
Any MEDI3726-related event 1 (100) 1 (100) 1 (33.3) 2 (66.7) 4 (100) 3 (75.0) 10 (100.0) 1 (10.0) 7 (100) 6 (85.7) 5 (100) 4 (80.0) 30 (90.9) 15 (45.5) 
Alanine aminotransferase increased 1 (100) 1 (100) 1 (33.3) 2 (50.0) 3 (30.0) 1 (10.0) 3 (42.9) 2 (40.0) 12 (36.4) 2 (6.1) 
Aspartate aminotransferase increased 1 (100) 1 (33.3) 1 (25.0) 3 (30.0) 2 (28.6) 3 (60.0) 11 (33.3) 
Fatigue 1 (25.0) 3 (30.0) 4 (57.1) 1 (14.3) 2 (40.0) 1 (20.0) 10 (30.3) 2 (6.1) 
Gamma-glutamyltransferase increased 1 (33.3) 3 (75.0) 2 (50.0) 1 (10.0) 1 (10.0) 2 (28.6) 2 (28.6) 3 (60.0) 2 (40.0) 10 (30.3) 7 (21.2) 
Pleural effusion 5 (50.0) 2 (28.6) 2 (40.0) 9 (27.3) 
Dyspnea 3 (30.0) 3 (42.9) 2 (40.0) 1 (20.0) 8 (24.2) 1 (3.0) 
Nausea 1 (25.0) 2 (20.0) 3 (42.9) 1 (20.0) 7 (21.2) 
Rash 1 (33.3) 1 (25.0) 2 (20.0) 3 (42.9) 7 (21.2) 
Decreased appetite 2 (50.0) 1 (10.0) 1 (14.3) 2 (40.0) 6 (18.2) 
Blister 4 (57.1) 1 (20.0) 5 (15.2) 
Edema peripheral 1 (33.3) 1 (33.3) 1 (10.0) 2 (40.0) 1 (20.0) 5 (15.2) 1 (3.0) 
Erythema 1 (33.3) 1 (10.0) 3 (42.9) 5 (15.2) 
Capillary leak syndrome 2 (28.6) 2 (28.6) 2 (40.0) 1 (20.0) 4 (12.1) 3 (9.1) 
Dermatitis exfoliative generalized 3 (30.0) 1 (14.3) 4 (12.1) 
Rash maculopapular 2 (20.0) 1 (14.3) 1 (20.0) 4 (12.1) 
Thrombocytopenia 3 (42.9) 2 (28.6) 1 (20.0) 1 (20.0)a 4 (12.1) 3 (9.1) 
MEDI3726 dose, n (%)
0.015 mg/kg0.03 mg/kg0.06 mg/kg0.15 mg/kg0.2 mg/kg0.25 mg/kg0.3 mg/kgTotal
n = 1n = 3n = 3n = 4n = 10n = 7n = 5N = 33
AEOverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4OverallGrade 3/4
Any MEDI3726-related event 1 (100) 1 (100) 1 (33.3) 2 (66.7) 4 (100) 3 (75.0) 10 (100.0) 1 (10.0) 7 (100) 6 (85.7) 5 (100) 4 (80.0) 30 (90.9) 15 (45.5) 
Alanine aminotransferase increased 1 (100) 1 (100) 1 (33.3) 2 (50.0) 3 (30.0) 1 (10.0) 3 (42.9) 2 (40.0) 12 (36.4) 2 (6.1) 
Aspartate aminotransferase increased 1 (100) 1 (33.3) 1 (25.0) 3 (30.0) 2 (28.6) 3 (60.0) 11 (33.3) 
Fatigue 1 (25.0) 3 (30.0) 4 (57.1) 1 (14.3) 2 (40.0) 1 (20.0) 10 (30.3) 2 (6.1) 
Gamma-glutamyltransferase increased 1 (33.3) 3 (75.0) 2 (50.0) 1 (10.0) 1 (10.0) 2 (28.6) 2 (28.6) 3 (60.0) 2 (40.0) 10 (30.3) 7 (21.2) 
Pleural effusion 5 (50.0) 2 (28.6) 2 (40.0) 9 (27.3) 
Dyspnea 3 (30.0) 3 (42.9) 2 (40.0) 1 (20.0) 8 (24.2) 1 (3.0) 
Nausea 1 (25.0) 2 (20.0) 3 (42.9) 1 (20.0) 7 (21.2) 
Rash 1 (33.3) 1 (25.0) 2 (20.0) 3 (42.9) 7 (21.2) 
Decreased appetite 2 (50.0) 1 (10.0) 1 (14.3) 2 (40.0) 6 (18.2) 
Blister 4 (57.1) 1 (20.0) 5 (15.2) 
Edema peripheral 1 (33.3) 1 (33.3) 1 (10.0) 2 (40.0) 1 (20.0) 5 (15.2) 1 (3.0) 
Erythema 1 (33.3) 1 (10.0) 3 (42.9) 5 (15.2) 
Capillary leak syndrome 2 (28.6) 2 (28.6) 2 (40.0) 1 (20.0) 4 (12.1) 3 (9.1) 
Dermatitis exfoliative generalized 3 (30.0) 1 (14.3) 4 (12.1) 
Rash maculopapular 2 (20.0) 1 (14.3) 1 (20.0) 4 (12.1) 
Thrombocytopenia 3 (42.9) 2 (28.6) 1 (20.0) 1 (20.0)a 4 (12.1) 3 (9.1) 

Abbreviation: AE, adverse event.

aThis grade 3 event was a dose-limiting toxicity.

No MTD was identified during the study; the MAD was 0.3 mg/kg.

Clinical activity

The composite response rate was 12.1%, with two responders in the 0.2 mg/kg cohort and one responder in each of the 0.25 and 0.3 mg/kg cohorts (Supplementary Table S1). These composite responses were seen early, with a TTR of 0.3–2.1 months. The median composite DoR was 3.8 months. There were no RECIST responders; however, 12 (36.4%) patients had stable disease and 1 (3%) patient had an unconfirmed partial response after receiving two doses of study drug. The majority (12/13) of patients with stable disease or better were in cohorts receiving doses ≥0.2 mg/kg. One patient from the 0.2 mg/kg cohort had a PSA50 response (Fig. 1). The median PFS was 3.6 months and median OS was 8.9 months. There were no CTC responses among patients with at least one CTC measurement of >50 CTC/7.5 mL blood (Supplementary Fig. S1). Reductions in on-treatment CTCs were seen in patients with low CTCs (≤50 CTC/7.5 mL blood), with 4 patients achieving a confirmed CTC response (2 in the 0.2 mg/kg cohort and 1 each in the 0.25 and 0.3 mg/kg cohorts; Fig. 2).

Figure 1.

PSA best percentage change from baseline after ≥12 weeks. Figure includes all patients with available data ≥12 weeks postbaseline. PSA, prostate-specific antigen.

Figure 1.

PSA best percentage change from baseline after ≥12 weeks. Figure includes all patients with available data ≥12 weeks postbaseline. PSA, prostate-specific antigen.

Close modal
Figure 2.

CTC while on treatment (patients with baseline CTC measurement ≤50). CTC, circulating tumor cell.

Figure 2.

CTC while on treatment (patients with baseline CTC measurement ≤50). CTC, circulating tumor cell.

Close modal

Pharmacokinetics and immunogenicity

A total of six analyte classes (total ADC, active ADC, conjugated SG3199, free SG3199, total IgG, and intact IgG) were measured in the MEDI3726 pharmacokinetic samples. A comprehensive assessment of MEDI3726 catabolism was reported previously (13). Further analyses focused on total and active ADC measurements. Following a single intravenous infusion, maximum observed concentration (Cmax) was attained at the end of infusion. Median time to maximum observed concentration (Tmax) for all dose levels occurred 0.1 days post-infusion (Supplementary Table S2). Plasma concentrations of MEDI3726 declined in a biphasic manner characterized by a fast initial distribution phase, followed by a slow elimination phase (Fig. 3). On the basis of mean plasma concentration–time profiles and dose-normalized area under the concentration–time curve (AUC) calculations, pharmacokinetic profiles were determined to be nonlinear. This was also true after a single intravenous infusion. Dose-normalized AUC from the start of dosing to infinity values generally increased with increasing dose, indicating that MEDI3726 pharmacokinetics were not dose proportional. This was also reflected in the mean plasma concentration–time profiles. Systemic clearance was determined to be dose dependent. Mean systemic clearance decreased when the dose of MEDI3726 was increased from 0.015 to 0.3 mg/kg. Terminal half-life was overall short (0.3–1.8 days) and increased with dose.

Figure 3.

Mean (SD) concentration–time profiles of MEDI3726 (active ADC) following the first intravenous dose of 0.015–0.3 mg/kg. Data below LLOQ are presented for illustration purposes only. ADC, antibody–drug conjugate; LLOQ, lower limit of quantitation; SD, standard deviation.

Figure 3.

Mean (SD) concentration–time profiles of MEDI3726 (active ADC) following the first intravenous dose of 0.015–0.3 mg/kg. Data below LLOQ are presented for illustration purposes only. ADC, antibody–drug conjugate; LLOQ, lower limit of quantitation; SD, standard deviation.

Close modal

Among 32 ADA-evaluable patients (those who had both baseline and post-treatment data), the ADA prevalence was 3 (9.4%) and the incidence was 13 (40.6%). Sixteen patients (50%) were ADA-negative (Supplementary Tables S3 and S4). Furthermore, among the 13 patients with treatment-induced or treatment-boosted ADAs, 12 had persistent positive ADA responses (i.e., positive at ≥2 post-baseline assessments, with ≥16 weeks between the first and last positive results, or positive results at the last post-baseline assessment), with titers ranging from 1 to 16. Because of the limited available data, no definitive conclusions could be reached regarding the impact of ADAs on pharmacokinetics. There was no correlation between ADA incidence and safety or antitumor activity.

Mutation profiles in ctDNA

Baseline ctDNA samples were available for 22 patients who received doses ≥0.2 mg/kg. Genomic analysis indicated that 2 patients had insertion/deletion (indel) mutations in BRCA2, a DNA damage response (DDR) gene that has been shown to sensitize to DNA-damaging agents, including the PBD warhead (15, 16). Overall, 9 of 22 (41%) patients had mutations in DDR or DDR-related genes. Mutations in DDR and DDR-related genes were observed in 3 of 4 patients with composite response as well as in 6 of 18 nonresponders (Supplementary Fig. S2). The limited number of patients with evaluable ctDNA samples precludes firm conclusions regarding the association between observed mutations and clinical activity.

In the dose-escalation portion of this trial, an MTD was not identified and TRAEs (particularly effusions and skin toxicities) prevented further escalation of the dose over 0.3 mg/kg. Evidence of clinical activity was observed with MEDI3726, mostly at higher doses, which also had the worst toxicity profiles (including treatment-related grade 3/4 AEs and SAEs), limiting the number of cycles of study drug patients could tolerate. As a consequence, the duration of any observed clinical activity was limited. The incidence of ADAs postbaseline was 13 (40.6%), suggesting that adverse immune responses to MEDI3726 did not contribute significantly to the toxicity profile.

The majority of the AEs observed in this trial were similar, qualitatively and quantitatively, to those reported in other published studies of agents featuring PBDs, either alone (e.g., SJG-136) or conjugated to an antibody as a warhead (forming an ADC), e.g., rovalpituzumab tesirine (Rova-T; refs. 17–19). They can be categorized into four groups: myelosuppression, liver enzyme abnormalities, skin toxicities, and effusions. Typically, myelosuppression was observed first, followed by increases in ALT and aspartate aminotransferase. Asymptomatic GGT elevation has also been observed in this and other trials of PBD-based therapies. Skin toxicities and effusions generally manifested after several cycles of study drug; they tended to be cumulative toxicities that consequently took longer to resolve.

Extracellular warhead release from an ADC may also lead to “bystander killing,” in which cytotoxic activity is observed in surrounding cells which may or may not express the target antigen (20). This effect may be favorable to antitumor activity; for example, through the killing of tumor cells that do not express the target epitope. The antitumor effect of the bystander killing activity of MEDI3726 was demonstrated to be robust in both in vitro and in vivo preclinical studies (8). However, due to the loss of target specificity, it is also conceivable that nonmalignant cells may be impacted (21). Importantly, the low levels of free PBD (SG3199, data not shown) observed in this study, as well as previously observed similarities between concentration–time profiles for total/intact IgG compared with total/active ADC, do not support extensive release of free warhead (13). Therefore, any bystander effect on observed safety or efficacy outcomes was likely minimal.

Genomic analysis was performed on patient samples from cohorts dosed at ≥0.2 mg/kg MEDI3726; indel mutations in the DDR gene BRCA2 were present in 1 of the 4 patients who had a clinical response and provided evaluable plasma samples. Interestingly, previous preclinical studies have suggested an association between BRCA2 mutation positivity and therapeutic activity with PBD-containing ADCs (15). In addition, previous data suggest that DDR aberrations may be associated with higher levels of PSMA expression (6). Elevated PSMA expression in tumor tissue may also contribute to antitumor activity with MEDI3726; complete and durable tumor regressions have been observed in highly PSMA-positive patient-derived xenograft models of prostate cancer (8). However, the limited sample size and availability of patient biopsy tissue in the present study precludes any firm conclusions.

Although limited clinical activity was demonstrated, particularly at the higher doses tested, in a patient population that generally has poor outcomes, emerging treatment-related toxicities limited the duration of treatment and prevented further planned dose escalation.

J.S. de Bono reports personal fees and other from Amgen, Bioxcel Therapeutics, Boehringer Ingelheim, Eisai, Menarini Silicon Biosystems, Qiagen, and Terumo and grants, personal fees, and other from Astellas, AstraZeneca, Bayer, Cellcentric, Daiichi, Genentech Roche, Genmab, GlaxoSmithKline, Harpoon, Janssen, Merck Serono, Merck Sharp & Dohme, Orion Pharma, Pfizer, Sanofi Aventis, Sierra Oncology, Taiho, and Vertex Pharmaceuticals outside the submitted work; in addition, J.S. de Bono has a patent for DNA damage repair inhibitors for treatment of cancer (patent no. WO 2005 053662 - no personal income issued and licensed to AstraZeneca) and a patent for 17-substituted steroids useful in cancer treatment (patent no. US5604213 - no personal income issued and licensed to Janssen). M.T. Fleming reports personal fees from Janssen outside the submitted work. J.S. Wang reports personal fees and other from AstraZeneca; personal fees from Eisai; and other from Acerta, ADC Therapeutics, Agenus, Aileron, Bicycle, BioNTech, Boehringer Ingelheim, Celgene, CicloMed, Clovis, Cyteir, Daiichi Sankyo, Eli Lilly, EMD Serono, Evelo, Forma, Erasca, Genentech/Roche, GlaxoSmithKline, H3Bio, Hengrui, Hutchinson MediPharma, Ignyta, Incyte, Jacobio, Janssen, Jounce, Klus Pharma, Kymab, LOXO, LSK BioPharma, Macrogenics, Merck, Mirati, Moderna, Pfizer, Phoenix Molecular Designs, Prelude, Puget Sound Biologics, Revolution, Ribon, 7,8 Pharma, Syndax, Stemline, Taiho, Takeda, Tesaro, TopAlliance, Vedanta, Vigeo, Xencor, and Artios outside the submitted work. R. Cathomas reports personal fees from Astellas, AstraZeneca, Janssen, Bayer, Roche, MSD, and BMS outside the submitted work. M. Selvi Miralles reports other from AstraZeneca outside the submitted work. J. Bothos reports other from AstraZeneca outside the submitted work and is an employee of AstraZeneca. M.J. Hinrichs reports other from AstraZeneca and Ipsen Biopharmaceutical outside the submitted work. P. He reports other from AstraZeneca during the conduct of the study and other from AstraZeneca outside the submitted work. A.I. Rosenbaum reports employment with and has stock ownership/options with AstraZeneca. M. Liang reports other from AstraZeneca outside the submitted work. K. Vashisht is an employee of AstraZeneca and has stock ownership in the company. D.P. Petrylak reports personal fees from Ada Cap (Advanced Accelerator Applications), Amgen, Astellas, AstraZeneca, Bayer, Bicycle Therapeutics, Boehringer Ingelheim, Bristol Myers Squibb, Clovis Oncology, Eli Lilly, Exelixis, Incyte, Janssen, Mirati, Monopteros, Pfizer, Pharmacyclics, Roche, Seattle Genetics, and Urogen; grants from Ada Cap (Advanced Accelerator Applications), Agensys Inc, Astellas, AstraZeneca, Bayer, BioXcel Therapeutics, Bristol Myers Squibb, Clovis Oncology, Eisai, Eli Lilly, Endocyte, Genentech, Innocrin, MedImmune, Medivation, Merck, Mirati, Novartis, Pfizer, Progenics, Replimune, Roche, Sanofi Aventis, and Seattle Genetics; and other from Bellicum (sold 7/2020) and Tyme (sold 10/2019) during the conduct of the study, as well as personal fees from Ada Cap (Advanced Accelerator Applications), Amgen, Astellas, AstraZeneca, Bayer, Bicycle Therapeutics, Boehringer Ingelheim, Bristol Myers Squibb, Clovis Oncology, Eli Lilly, Exelixis, Incyte, Janssen, Mirati, Monopteros, Pfizer, Pharmacyclics, Roche, Seattle Genetics, and Urogen, and grants from Ada Cap (Advanced Accelerator Applications), Agensys Inc, Astellas, AstraZeneca, Bayer, BioXcel Therapeutics, Bristol Myers Squibb, Clovis Oncology, Eisai, Eli Lilly, Endocyte, Genentech, Innocrin, MedImmune, Medivation, Merck, Mirati, Novartis, Pfizer, Progenics, Replimune, Roche, Sanofi Aventis, Seattle Genetics, Bellicum (sold 7/2020) and Tyme (sold 10/2019) outside the submitted work. No disclosures were reported by the other authors.

J.S. de Bono: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. M.T. Fleming: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. J.S. Wang: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. R. Cathomas: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. M.S. Miralles: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. J. Bothos: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. M.J. Hinrichs: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. Q. Zhang: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. P. He: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. M. Williams: Conceptualization, formal analysis, investigation, visualization, writing–original draft, writing–review and editing. A.I. Rosenbaum: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. M. Liang: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. K. Vashisht: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. S. Cho: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. P. Martinez: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. D.P. Petrylak: Conceptualization, formal analysis, investigation, methodology, writing–original draft, writing–review and editing.

We would like to thank Luz Vanegas and Brandon Lam for their role in the development and validation of ADA assays, and Jessica Chen for ADA sample analysis. Medical writing support for the development of this article, under the direction of the authors, was provided by James Holland of Ashfield MedComms, an Ashfield Health company, and funded by AstraZeneca.

This study was funded by AstraZeneca and ADC Therapeutics.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Francini
E
,
Gray
KP
,
Shaw
GK
,
Evan
CP
,
Hamid
AA
,
Perry
CE
, et al
Impact of new systemic therapies on overall survival of patients with metastatic castration-resistant prostate cancer in a hospital-based registry
.
Prostate Cancer Prostatic Dis
2019
;
22
:
420
7
.
2.
Moreira
DM
,
Howard
LE
,
Sourbeer
KN
,
Amarasekara
HS
,
Chow
LC
,
Cockrell
DC
, et al
Predicting time from metastasis to overall survival in castration-resistant prostate cancer: results from SEARCH
.
Clin Genitourin Cancer
2017
;
15
:
60
66
.
3.
National Cancer Institute
. 
SEER cancer stat facts: prostate cancer
.
Available from
: https://seer.cancer.gov/statfacts/html/prost.html.
4.
Bostwick
DG
,
Pacelli
A
,
Blute
M
,
Roche
P
,
Murphy
GP
. 
Prostate specific membrane antigen expression in prostatic intraepithelial neoplasia and adenocarcinoma: a study of 184 cases
.
Cancer
1998
;
82
:
2256
61
.
5.
Silver
DA
,
Pellicer
I
,
Fair
WR
,
Heston
WD
,
Cordon-Cardo
C
. 
Prostate-specific membrane antigen expression in normal and malignant human tissues
.
Clin Cancer Res
1997
;
3
:
81
5
.
6.
Paschalis
A
,
Sheehan
B
,
Riisnaes
R
,
Rodrigues
DN
,
Gurel
B
,
Bertan
C
, et al
Prostate-specific membrane antigen heterogeneity and DNA repair defects in prostate cancer
.
Eur Urol
2019
;
76
:
469
78
.
7.
Donin
NM
,
Reiter
RE
. 
Why targeting PSMA is a game changer in the management of prostate cancer
.
J Nucl Med
2018
;
59
:
177
82
.
8.
Cho
S
,
Zammarchi
F
,
Williams
DG
,
Havenith
CEG
,
Monks
NR
,
Tyrer
P
, et al
Antitumor activity of MEDI3726 (ADCT-401), a pyrrolobenzodiazepine antibody–drug conjugate targeting PSMA, in preclinical models of prostate cancer
.
Mol Cancer Ther
2018
;
17
:
2176
86
.
9.
Scher
HI
,
Morris
MJ
,
Stadler
WM
,
Higano
C
,
Basch
E
,
Fizazi
K
, et al
Trial design and objectives for castration-resistant prostate cancer: updated recommendations from the prostate cancer clinical trials working group 3
.
J Clin Oncol
2016
;
34
:
1402
18
.
10.
Ji
Y
,
Liu
P
,
Li
Y
,
Bekele
BN
. 
A modified toxicity probability interval method for dose-finding trials
.
Clin Trials
2010
;
7
:
653
63
.
11.
Mou
S
,
Huang
Y
,
Rosenbaum
AI
. 
ADME considerations and bioanalytical strategies for pharmacokinetic assessments of antibody-drug conjugates
.
Antibodies
2018
;
7
:
41
.
12.
Faria
M
,
Peay
M
,
Lam
B
,
Ma
E
,
Yuan
M
,
Waldron
M
, et al
Multiplex LC-MS/MS assays for clinical bioanalysis of MEDI4276, an antibody-drug conjugate of tubulysin analogue attached via cleavable linker to a biparatopic humanized antibody against HER-2
.
Antibodies
2019
;
8
:
11
.
13.
Huang
Y
,
Del Nagro
CJ
,
Balic
K
,
Mylott
WR
,
Ismaiel
OA
,
Ma
E
, et al
Multifaceted bioanalytical methods for the comprehensive pharmacokinetic and catabolic assessment of MEDI3726, an anti-prostate-specific membrane antigen pyrrolobenzodiazepine antibody–drug conjugate
.
Anal Chem
2020
;
92
:
11135
44
.
14.
U.S. Food and Drug Administration
. 
Assay development and validation for immunogenicity testing of therapeutic protein products - guidance for industry
; 
2016
.
Available from
: https://www.fda.gov/media/77796/download.
15.
Zhong
H
,
Chen
C
,
Tammali
R
,
Breen
S
,
Zhang
J
,
Fazenbaker
C
, et al
Improved therapeutic window in BRCA-mutant tumors with antibody-linked pyrrolobenzodiazepine dimers with and without PARP inhibition
.
Mol Cancer Ther
2019
;
18
:
89
99
.
16.
Barrett
A
,
Chen
H
,
Howard
P
,
Koopman
J
,
Kagiampakis
I
,
Sridhar
S
. 
Analysis of large-scale, high-throughput screens to identify genes which confer specific sensitivity to pyrrolobenzodiazepine (PBD) warheads for ADCs
.
3rd Exploring DNA Repair Pathways as Targets for Cancer Therapy Conference
.
Feb 23–29
; 
2019
;
Nassau, Bahamas
.
17.
Hochhauser
D
,
Meyer
T
,
Spanswick
VJ
,
Wu
J
,
Clingen
PH
,
Loadman
P
, et al
Phase I study of sequence-selective minor groove DNA binding agent SJG-136 in patients with advanced solid tumors
.
Clin Cancer Res
2009
;
15
:
2140
7
.
18.
Morgensztern
D
,
Besse
B
,
Greillier
L
,
Santana-Davila
R
,
Ready
N
,
Hann
CL
, et al
Efficacy and safety of rovalpituzumab tesirine in third-line and beyond patients with DLL3-expressing, relapsed/refractory small-cell lung cancer: results from the phase II TRINITY study
.
Clin Cancer Res
2019
;
25
:
6958
66
.
19.
Mantaj
J
,
Jackson
PJM
,
Rahman
KM
,
Thurston
DE
. 
From anthramycin to pyrrolobenzodiazepine (PBD)-containing antibody-drug conjugates (ADCs)
.
Angew Chem Int Ed Engl
2017
;
56
:
462
88
.
20.
Staudacher
AH
,
Brown
MP
. 
Antibody drug conjugates and bystander killing: is antigen-dependent internalisation required?
Br J Cancer
2017
;
117
:
1736
42
.
21.
Tolcher
AW
. 
Antibody drug conjugates: lessons from 20 years of clinical experience
.
Ann Oncol
2016
;
27
:
2168
72
.