A Phase 1b, Randomized, Double-Blind, Placebo-Controlled, Multiple-Ascending Dose Study to Investigate the Safety, Tolerability, and Pharmacokinetics of DSTA4637S in Patients With Complicated Staphylococcus aureus Bacteremia Receiving Standard-of-Care Antibiotics

Abstract

Background

Safe and effective therapies are needed for hospitalized patients with complicated Staphylococcus aureus infections, including approaches capable of eradicating intracellular bacterial reservoirs that contribute to treatment failure. This study evaluated the safety, tolerability, and pharmacokinetics of DSTA4637S, an antibody–antibiotic conjugate combining an anti-S. aureus antibody with a rifamycin-class antibiotic, in patients with complicated S. aureus bacteremia.

Methods

In this phase 1b, randomized, double-blind, placebo-controlled study, 25 patients with methicillin-sensitive (MSSA, 68%) or -resistant (MRSA, 32%) complicated S. aureus bacteremia received weekly intravenous doses of DSTA4637S at 15, 45, or 100 mg/kg in combination with at least 4 weeks of standard-of-care antistaphylococcal antibiotics.

Results

Most patients (92%) experienced ≥1 treatment-emergent adverse event (TEAE), of which 44% were considered treatment-related. Grade ≥3 TEAEs and serious adverse events occurred in 72% and 56% of patients, respectively. The most common grade ≥3 TEAEs were infusion-related reactions (IRRs), pneumonia, anemia, and respiratory failure. Four patients (21%) receiving DSTA4637S discontinued study treatment due to treatment-related IRRs. Systemic exposure of DSTA4637S in patients increased approximately dose-proportionally from 15 to 100 mg/kg but was 30%–60% lower than previously observed in healthy volunteers.

Conclusions

The safety profile of DSTA4637S, characterized by unpredictable IRRs, and its pharmacokinetic behavior in patients differed substantially from those observed in healthy volunteers, highlighting the critical importance of conducting early phase studies in the target patient population.

DSTA4637S, an antibody–antibiotic conjugate, was studied in 25 patients with complicated Staphylococcus aureus bacteremia. Infusion-related reactions were frequent and unpredictable, systemic exposure was 30%–60% lower than in healthy volunteers, and one patient developed rifampin-resistant S. aureus.

Graphical Abstract

Graphical Abstract.

This graphical abstract is also available at Tidbit: https://tidbitapp.io/institutional-portal/open-forum-infectious-diseases/tidbits/a-phase-1b-randomized-double-blind-placebo-controlled-multiple-ascending-dose-study-to-investigate-the-safety-tolerability-and-pharmacokinetics-of-dsta4637s-in-patients-with-complicated-staphylococcus-aureus-bacteremia-receiving-standard-of/update

Staphylococcus aureus (S. aureus) is a leading cause of invasive infections, including bacteremia, infective endocarditis, and osteomyelitis, and remains associated with substantial morbidity and mortality despite available therapies, contributing to an estimated 300 000 deaths annually worldwide [1–4]. Outcomes have not meaningfully improved over decades [5–7], in part due to an intracellular reservoir of S. aureus within host phagocytes that conventional antibiotics cannot reach [8, 9]. Within phagolysosomes, S. aureus becomes dormant, evades immune clearance, and, upon host cell lysis, contributes to persistent and relapsing infections.

DSTA4637S is a THIOMAB™ antibody–antibiotic conjugate (AAC) designed to target intracellular S. aureus. It comprises an engineered human IgG1 anti-S. aureus monoclonal antibody (MSTA3852A) directed against a β-O-linked N-acetylglucosamine (β-GlcNAc) modification of wall teichoic acid (β-GlcNAc WTA) linked via a protease-cleavable valine-citrulline linker to the rifamycin-class antibiotic dmDNA31 [10, 11]. Following binding to S. aureus, opsonization, and phagocytic uptake, intracellular proteolysis of the VC linker releases dmDNA31 within phagolysosomes, enabling killing of intracellular S. aureus.

In preclinical studies, DSTA4637S demonstrated activity against clinically relevant S. aureus strains in human phagocytes and reduced bacterial burden in a murine bacteremia model, alone and in combination with vancomycin [10]. In a phase 1 study in healthy volunteers, single intravenous (IV) doses of DSTA4637S up to 150 mg/kg were safe and well tolerated with acceptable pharmacokinetics (PK) [12].

Here, we report a phase 1b study evaluating the safety, tolerability, PK, and immunogenicity of DSTA4637S in patients with complicated MRSA or MSSA bacteremia receiving SOC antibiotics.

METHODS

Study Design

This phase 1b, randomized, double-blind, placebo-controlled study (ClinicalTrials.gov: NCT03162250; EudraCT: 2016-001880-35) evaluated multiple ascending doses of DSTA4637S in combination with SOC antistaphylococcal antibiotics in patients with complicated MRSA and MSSA bacteremia requiring ≥4 weeks of SOC therapy. The study was conducted at 17 sites in the United States, Spain, and South Korea. The primary objective was to assess safety and tolerability; secondary objectives included PK and immunogenicity; the exploratory objective was to evaluate the emergence of resistance to DSTA4637S.

This study was conducted in full conformance with the ICH E6 guideline for GCP, the Declaration of Helsinki, and applicable local regulatory requirements. All patients provided written informed consent prior to initiation of any study procedures.

Protocol Availability Statement

The protocol used for this trial can be found in the Supplemental Materials.

Patients

Adults aged 18–80 years of age with BMI 18–40 kg/m² and complicated S. aureus bacteremia were eligible. At randomization, patients were required to have microbiologic or molecular evidence of S. aureus within 120 hours prior to enrollment and an anticipated need for ≥4 weeks of IV SOC therapy. Complicated S. aureus bacteremia was defined as positive blood cultures for S. aureus in patients who did not meet criteria for uncomplicated bacteremia, as specified in the protocol and consistent with prior definitions [13, 14].

Key exclusion criteria included the presence of an intravascular catheter; S. aureus bacteremia associated with an intracardiac device; infection of a prosthetic joint or vertebral hardware; known rifampicin-resistant S. aureus; polymicrobial bacteremia (≥2 bacterial species identified in blood culture); or significant immune suppression.

Interventions

DSTA4637S (Genentech, Inc., South San Francisco, CA) was administered as a 15, 45, or 100 mg/kg IV infusion once weekly for 4–6 doses in combination with investigator-selected SOC antibiotics based on the literature and local treatment guidelines [13, 15–17]. The total number of DSTA4637S doses (4–6) was determined by the planned duration of SOC therapy, with 1 dose of DSTA4637S administered for each week of SOC treatment. Rifamycin-class antibiotics were prohibited. Placebo-matched DSTA4637S in appearance and was administered on the same schedule (see Supplementary Materials for formulation and blinding details).

Randomization and Blinding

Patients were randomized in a 6:2 (active: placebo) ratio within each cohort using a voice or web-based response system (IxRS). Sentinel dosing (1:1) was implemented for the first 2 patients in each cohort. PK samples were collected from all patients to preserve blinding. PK sample handling was designed to maintain blinding (see Supplementary Materials for details).

Safety Assessments

Safety was assessed via adverse events (AEs), laboratory testing, vital signs, ECGs, and physical examinations. AEs were coded using MedDRA v23.0 and graded per NCI CTCAE v4.0. Dose-limiting AEs and AEs of special interest were defined per protocol (see Supplementary Materials for details).

Pharmacokinetics

PK parameters (C_max [maximum observed concentration], T_max [time to attain maximum observed concentration], AUC_tau [area under the concentration-time curve between time zero (predose) and time of next dose (7 days)], t_½ [terminal elimination half-life]) were estimated using noncompartmental analysis (Phoenix WinNonlin v8.2) and summarized descriptively. Three analytes were assessed: DSTA4637S conjugate, total antibody, and unconjugated dmDNA31, using validated bioanalytical methods (see Supplementary Materials for details).

Immunogenicity

Antidrug antibodies against DSTA4637S were measured using a validated bridging ELISA, as previously described [12]. Analyses included patients with at least 1 baseline and 1 postbaseline ADA assessment. Patients were classified as ADA-positive or ADA-negative based on baseline and postbaseline samples (see Supplementary Materials for details).

Resistance Assessments

S. aureus isolates from blood and sterile sites were evaluated for β-wall teichoic acid (β-WTA) epitope expression by flow cytometry to assess potential resistance to the antibody component of DSTA4637S and for dmDNA31 susceptibility by MIC testing using a CLSI broth microdilution assay.

Whole Genome Sequencing

Genomic DNA from selected isolates underwent Illumina sequencing and assembly with SPAdes, followed by annotation with Prokka (see Supplementary Materials for full pipeline and QC criteria).

Statistical Methods

The study was designed to enroll 24 patients, based on a planned treatment allocation of 6:2 (active: placebo) per cohort. No formal statistical hypothesis testing was performed; all statistical summaries were descriptive and performed in the safety-evaluable population, defined as all patients receiving ≥1 dose of study drug. Patients were grouped by treatment received. All patients assigned to placebo, regardless of dose cohort, were combined into a single group for analysis.

RESULTS

Patient Disposition and Characteristics

Between July 2017 and August 2019, 30 patients were screened, and 27 were randomized across 11 investigational sites in 3 countries (United States, n = 4; Spain, n = 5; and South Korea, n = 2). Twenty-five patients received study drug; 6 received placebo, 6 received 15 mg/kg DSTA4637S, 6 received 45 mg/kg DSTA4637S, and 7 received 100 mg/kg DSTA4637S (Figure 1).

Figure 1.

Patient Disposition. 2731 patients were prescreened; 30 patients were screened with 3 screen failures due to shortage of investigational product (n = 1), positive drug screen (n = 1), and presence of intravascular catheter that will not be removed (n = 1). Twenty-seven patients were randomized; 2 patients were removed from study due to physician decision related to eligibility criteria. Six patients were randomized to placebo and all 6 completed the study; 5 patients completed treatment; 1 patient discontinued treatment due to AE. Six patients were randomized to DSTA46375S 15 mg/kg cohort and 5 patients completed the study; 3 patients completed treatment; 3 patients discontinued treatment due to AE (n = 2) and patient withdrawal (n = 1). Six patients were randomized to DSTA46375S 45 mg/kg cohort and 5 patients completed the study; 3 patients completed treatment; 3 patients discontinued treatment due to AE (n = 2) and physician decision. Seven patients were randomized to DSTA46375S 100 mg/kg cohort and 5 patients completed the study; 4 patients completed treatment; 3 patients discontinued treatment due to AE (n = 2) and patient withdrawal (n = 1).

Baseline demographics were similar across treatment groups (Table 1). A greater proportion of patients had MSSA bacteremia (n = 17, 68%) compared with MRSA (n = 8, 32%). Fifteen patients (60%) had comorbidities associated with poorer outcomes in S. aureus bacteremia, including diabetes mellitus (n = 14, 56%), endocarditis (n = 5, 20%), and receipt of hemodialysis (n = 1, 4%).

Table 1.

Patient Demographics and Baseline Characteristics

	Placebo (n = 6)	Cohort A 15 mg/kg (n = 6)	Cohort B 45 mg/kg (n = 6)	Cohort C 100 mg/kg (n = 7)	All Patients (N = 25)
Age (y) mean, (range)	58.3 (45–69)	52.8 (25–66)	59.5 (44–70)	56.6 (44–68)	56.8 (25–70)
≥65 y n (%)	3 (50%)	1 (16.7%)	2 (33.3%)	2 (28.6%)	8 (32%)
Gender, n (%)
Male	5 (83.3%)	6 (100.0%)	4 (66.7%)	6 (85.7%)	21 (84.0%)
Race
American Indian or Alaska Native	0	0	0	1 (14.3%)	1 (4.0%)
Asian	0	0	1 (16.7%)	2 (28.6%)	3 (12.0%)
Black or African American	0	3 (50.0%)	0	0	3 (12.0%)
White	6 (100.0%)	2 (33.3%)	3 (50.0%)	2 (28.6%)	13 (52.0%)
Unknown	0	1 (16.7%)	2 (33.3%)	2 (28.6%)	5 (20.0%)
Ethnicity
Hispanic or Latino	4 (66.7%)	2 (33.3%)	2 (33.3%)	3 (42.9%)	11 (44.0%)
Weight (kg) at baseline mean, (range)	80.7 (73–89)	85.5 (58–103)	81.5 (53–114)	75.6 (59–97)	80.6 (53–114)
BMI (kg/m2) at baseline mean (range)a	27.3 (24.6–30)	27.4 (18.7–32)	29.9 (20.7–39.6)	25.9 (18.8–33.6)	27.6 (18.7–39.6)
S. aureus bacteremia
MRSA	1 (16.7%)	3 (50.0%)	1 (16.7%)	3 (42.9%)	8 (32.0%)
MSSA	5 (83.3%)	3 (50.0%)	5 (83.3%)	4b (57.1%)	17 (68.0%)
Diabetes mellitus	3 (50.0%)	4 (66.7%)	4 (66.7%)	3 (42.9%)	14 (56.0%)
Hemodialysisc	0	0	0	1 (14.3%)	1 (4.0%)
Indicationd
Endocarditis	2 (33.3%)	1 (16.7%)	0	2 (28.6%)	5 (20.0%)
Bone and joint	2 (33.3%)	5 (83.3%)	5(83.3%)	3 (43)	15 (60.0%)
Endovascular	2 (33.3%)	1 (16.7%)	0	1 (14.3%)	4 (16.0%)
Pneumonia	0	0	0	1 (14.3%)	1 (4.0%)

MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-susceptible Staphylococcus aureus.

^aEligibility criteria were amended prior to enrollment of Cohort C to permit inclusion of patients with BMI ≥40 kg/m².

^bOne patient with MSSA bacteremia in the 100 mg/kg DSTA4637S group was misclassified as having MRSA bacteremia at baseline.

^cEligibility criteria were amended prior to enrollment of Cohort C to permit inclusion of patients receiving hemodialysis.

^dIndication determined by the sponsor.

All patients received at least 4 weeks of antistaphylococcal SOC antibiotics. Among patients with MRSA bacteremia, definitive SOC therapy consisted of vancomycin and daptomycin, with ceftaroline and clindamycin each used in 1 patient. Among patients with MSSA bacteremia, definitive SOC therapy consisted of antistaphylococcal penicillins (cloxacillin, oxacillin, or nafcillin) and cefazolin, with ceftriaxone, ceftaroline, and daptomycin each used in 1 patient. Most patients received SOC antibiotic monotherapy, although a small number underwent sequential changes in their SOC regimen. Combination antibiotic therapy was used in only 1 patient with native valve endocarditis, who received adjunctive fosfomycin and daptomycin during treatment with ceftaroline.

Fifteen patients (60%) completed study treatment and received 4–6 doses of study drug, 5 patients (83%) in the placebo group, and 10 patients (53%) in the DSTA4637S group. Median treatment duration was 22 days (range 1–39), with median durations of 28.5 days in the placebo group and 18, 14, and 22 days in the 15, 45, and 100 mg/kg DSTA4637S groups, respectively. The median number of doses administered was 4 (range 1–6), including 5 in the placebo group, 3 in the 15 and 45 mg/kg groups, and 4 in the 100 mg/kg group.

Four of 25 patients (16%), all receiving DSTA4637S, discontinued the study: 1 each in the 15 and 45 mg/kg groups and 2 in the 100 mg/kg group. Reasons for discontinuations included withdrawal of consent (n = 3, 12%) and death (n = 1, 4%).

Safety

Twenty-three of 25 patients (92%) experienced ≥1 treatment-emergent adverse event (TEAE), with a higher incidence in the DSTA4637S treatment group (100%) than in the placebo group (83%) (Table 2). Grade ≥3 TEAEs occurred in 18 patients (72%), including 4 (67%) in the placebo group and 3 (50%), 5 (83%), and 6 (86%) in the 15, 45, and 100 mg/kg DSTA4637S groups, respectively. The most common grade ≥3 TEAEs were infusion-related reactions (IRRs), pneumonia, anemia, and respiratory failure (each in 2 patients). Eleven patients (44.0%) experienced ≥1 TEAE considered related to the study treatment, including 10 who received DSTA4637S.

Table 2.

Summary of AEs

	Placebo (n = 6)	Cohort A 15 mg/kg (n = 6)	Cohort B 45 mg/kg (n = 6)	Cohort C 100 mg/kg (n = 7)	All Patients (N = 25)
Total no. patients with ≥1 AE	5 (83.3%)	5 (83.3%)	6 (100.0%)	7 (100.0%)	23 (92.0%)
Total no. of AEs	24	27	36	67	154
Total no. of deaths	0	0	0	1 (14.3%)	1 (4.0%)
Grade ≥3 AE	4 (66.7%)	3 (50.0%)	5 (83.3%)	6 (85.7%)	18 (72%)
Related AE	1 (16.7%)	2 (33.3%)	3 (50.0%)	5 (71.4%)	11 (44.0%)
SAE	3 (50.0%)	3 (50.0%)	3 (50.0%)	5 (71.4%)	14 (56.0%)
Related SAE	0	1 (16.7%)	1 (16.7%)	0	2 (8.0%)
AE leading to interruption of study treatment	0	1 (16.7%)	1 (16.7%)	2 (28.6%)	4 (16.0%)
AE leading to withdrawal of study treatment	1 (16.7%)	2 (33.3%)	2 (33.3%)	2 (28.6%)	7 (28.0%)
IRRa	1 (16.7%)	1 (16.7%)	2 (33.3%)	2 (28.6%)	6 (24.0%)
IRRs leading to withdrawal from treatment	0	1 (16.7)	2 (33.3%)	1 (14.3%)	4 (16.0%)
AEs occurring in >10% of all patientsb
Serum color abnormal	0	1 (16.7%)	1 (16.7%)	3 (42.9%)	5 (20.0%)
Chest pain	1 (16.7%)	0	0	3 (42.9%)	4 (16.0%)
Dizziness	0	2 (33.3%)	1 (16.7%)	1 (14.3%)	4 (16.0%)
Pneumonia	1 (16.7%)	0	1 (16.7%)	2 (28.6%)	4 (16.0%)
Pyrexia	1 (16.7%)	0	0	3 (42.9%)	4 (16.0%)
Anemia	2 (33.3%)	0	0	1 (14.3%)	3 (12.0%)
Constipation	1 (16.7%)	1 (16.7%)	1 (16.7%)	0	3 (12.0%)
Pruritus	0	1 (16.7%)	1 (16.7%)	1 (14.3%)	3 (12.0%)
Skin discoloration	0	0	1 (16.7%)	2 (28.6%)	3 (12.0%)

AE, adverse event; IRR, infusion-related reaction; SAE, serious adverse event.

^aIRR is defined as (1) all suspected cases of grade ≥2 IRRs or (2) related adverse events occurring during or within 24 h of infusion.

^bMedDRA preferred term.

Serious adverse events (SAEs) occurred in 14 patients (56%), including 5 patients (71%) in the 100 mg/kg group and 3 patients (50%) in each of the remaining groups (Table 2). One patient with pneumonia and a history of severe chronic obstructive pulmonary disease requiring supplemental oxygen died of respiratory failure, which was assessed as unrelated to study treatment.

IRRs were reported in 6 patients (24%), including 5 of 19 (26%) receiving DSTA4637S (1 patient [17%] in the 15 mg/kg DSTA4637S group, 2 patients [33%] in the 45 mg/kg group, and 2 patients [29%] in the 100 mg/kg group). These events occurred during or within 4 hours of infusion and were characterized by fever, chills, hypoxia, and hemodynamic changes, including tachycardia and hypertension. Three patients who received a second dose of DSTA4637S experienced recurrent IRRs, and no patient with an IRR received more than 2 doses. IRRs were the most common TEAE leading to treatment discontinuation, occurring in 4 patients receiving DSTA4637S. A fifth patient, who had experienced an IRR leading to study treatment interruption, discontinued treatment for congestive cardiac failure prior to receiving the second dose of study drug. All IRR events were resolved within 24 hours with supportive care, including antipyretics, antihistamines, and antihypertensive therapies.

Serum discoloration occurred in 5 patients (26%) receiving DSTA4637S, with associated skin discoloration in 3 patients. Events of artifactual hypoxia were of particular interest due to a theoretical risk for the blue serum color caused by dmDNA31 to interfere with pulse oximetry measurements, although this phenomenon was not observed in a prior phase 1 study in healthy volunteers (GV29259). Two patients (29%) in the 100 mg/kg DSTA4637S group experienced events of artifactual hypoxia without corresponding hypoxemia, assessed as related to study treatment. In 1 case, pulse oximetry readings indicated low oxygen saturation, but arterial blood gas measurements did not show hypoxemia. In the other case, the patient did not appear hypoxemic and was not responsive to supplemental oxygen. Both events resolved on the day of onset without changes to study treatment. Patient skin and serum discoloration and artifactual hypoxemia contributed to 4 instances of unintentional unblinding.

TEAEs led to treatment interruption in 4 patients. Artifactual hypoxia without corresponding hypoxemia and IRRs were the only events leading to treatment interruption considered related to study treatment (each in 1 patient). TEAEs led to study treatment discontinuation in 7 patients, including 1 in the placebo group and 2 each in the 15, 45, and 100 mg/kg DSTA4637S groups. IRRs were the most common TEAE leading to discontinuation among patients receiving DSTA4637S (21%). One patient discontinued due to elevated hepatic enzymes assessed as unrelated to study treatment, and 1 patient in the placebo group discontinued due to respiratory failure.

No clinically meaningful trends were observed in mean laboratory parameters, vital signs, or ECGs, and no dose-limiting AEs were identified in the study.

Pharmacokinetics

All 19 patients who received DSTA4637S were included in PK parameter calculations after the first dose. Patients who received 4–6 doses of DSTA4637S and reached steady state were included in PK parameter calculations after the last dose.

TAC Conjugate

Following the first IV administration of 15, 45, and 100 mg/kg DSTA4637S, median T_max ranged from 0.11 to 0.25 days (Table 3). Geometric mean C_max and systemic exposure (AUC_tau) generally increased in a dose-proportional manner (Figure 2). Half-life (t_1/2) was generally consistent across dose levels, although interpatient variability was high at the 100 mg/kg dose level; among 3 patients for whom t_1/2 could be estimated, one had a t_1/2 of 18.8 days, while the geometric mean t_1/2 for the other 2 patients was 5.79 days, similar to the t_1/2 values in the lower dose cohorts. Minimal accumulation was observed, consistent with a mean half-life of 8.58 days at the 100 mg/kg dose.

Table 3.

PK Profile of TAC Conjugate, Total Antibody, and Unconjugated dmDNA31

	Cohort A 15 mg/kg		Cohort B 45 mg/kg		Cohort C 100 mg/kg
	n	Geometric mean (CV%)	n	Geometric mean (CV%)	n	Geometric mean (CV%)
TAC Conjugate—after first dose
Cmax (ng/mL)	6	3310 (22.6)	6	7380 (35.7)	7	20 000 (37.1)
Tmax (d), median (range)	6	0.11 (0.08–0.27)	6	0.25 (0.10–1.09)	7	0.12 (0.10–0.24)
AUCtau (ng·day/mL)	6	7420 (29.5)	6	21 100 (29.1)	7	47 300 (32.5)
TAC Conjugate—after last dosea
Cmax (ng/mL)	3	4750 (13.6)	3	11 000 (25.9)	3	19 300 (17.8)
Tmax (d),b median (range)	3	0.11 (0.11–0.25)	3	0.12 (0.10–0.17)	3	0.25 (0.09–0.29)
AUCtau (ng·day/mL)	3	14 600 (8.8)	3	29 200 (23.6)	3	56 100 (16.7)
t1/2 (d)	3	4.81 (6.8)	3	4.30 (19.8)	3	8.58 (79.0)
Total antibody—after first dose
Cmax (ng/mL)	6	277 000 (23.5)	6	862 000 (35.9)	7	1 980 000 (34.9)
Tmax (d), median (range)	6	0.11 (0.08–0.26)	6	0.12 (0.10–0.24)	7	0.12 (0.10–0.29)
AUCtau (ng·day/mL)	6	860 000 (32.7)	6	2 830 000 (30.1)	7	6 750 000 (45.1)
Total antibody—after last dosea
Cmax (ng/mL)	3	557 000 (14.1)	3	1 320 000 (26.7)	3	2 940 000 (20.2)
Tmax (d),b median (range)	3	0.26 (0.11–0.90)	3	0.17 (0.12–0.25)	3	0.10 (0.09–0.29)
AUCtau (ng·day/mL)	3	2 710 000 (21.1)	3	4 800 000 (40.4)	3	11 100 000 (13.6)
t 1/2 (d)	3	18.4 (24.5)	3	13.7 (33.2)	3	22.4 (12.8)
dmDNA31—after first dose
Cmax (ng/mL)	6	0.40 (13.3)	6	1.27 (19.7)	7	3.49 (79.2)
Tmax (d), median (range)	6	1.04 (0.46–1.97)	6	1.58 (1.05–2.14)	7	1.09 (0.24–2.09)
AUCtau (ng·day/mL)	0	NE	3	5.88 (20.3)	5	11.9 (50.3)
dmDNA31—after last dosea
Cmax (ng/mL)	3	0.46 (14.7)	3	2.16 (46.5)	3	2.72 (34.7)
Tmax (d),c median (range)	3	0.97 (0.90–1.11)	3	1.10 (0.26–1.15)	3	0.98 (0.88–1.98)
AUCtau (ng·day/mL)	0	NE	3	9.62 (42.3)	2	12.9 (55.4)
t1/2 (d)	0	NE	3	4.60 (30.4)	2	5.46 (41.1)

AUC_tau, area under the concentration-time curve between time zero (predose) and time of next dose (7 d); C_max, maximum observed concentration; CV, coefficient of variation; NE, not estimated; PK, pharmacokinetic; t_1/2, terminal elimination half-life; T_max, time to attain maximum observed concentration.

^aIncludes patients who received 4–6 doses of DSTA4637S.

^b T _max converted to the time elapsed from the last dose given.

^c T _max converted to the time elapsed from the last dose given.

Figure 2.

Systemic exposure of DSTA4637S, measured as TAC conjugate, total antibody, and unconjugated dmDNA31 in patients with S. aureus bacteremia following 1–6 doses of TAC in addition to SOC antibiotics through treatment follow-up (day 155).

Total Antibody

Median T_max ranged from 0.11 to 0.12 days. Exposure (geometric mean C_max) increased approximately dose-proportionally (Table 3; Figure 2). Systemic exposure (AUC_tau) after 4–6 doses (n = 3 per dose cohort) increased in a dose-proportional manner following the first and last dose administrations, but steady state was not reached due to the limited dosing duration. The t_1/2 values were consistent across dose levels, with the geometric mean t_1/2 ranging from 13.7 to 22.4 days. Accumulation was observed in patients receiving multiple weekly doses, consistent with a mean half-life of 22.4 days at the 100 mg/kg dose.

Unconjugated dmDNA31

Median T_max ranged from 1.04 to 1.58 days, with geometric mean C_max values of 0.40, 1.27, and 3.49 ng/mL at 15, 45, and 100 mg/kg, respectively (Table 3; Figure 2). PK parameter estimation was limited by missing or unquantifiable samples, and small sample sizes contributed to high interpatient variability. Geometric mean t_1/2 values were similar at 45 mg/kg (4.60 days) and 100 mg/kg (5.46 days) in patients who received 4–6 doses of DSTA4637S, reached steady state, and had sufficient PK samples for analysis.

Comparison With Healthy Volunteers

In the 100 mg/kg dose group, the mean C_max of the TAC conjugate after the first dose was 48% lower in patients with S. aureus bacteremia than in healthy volunteers (Supplementary Figure 1, left) [12]. Total antibody levels were also lower in patients (Supplementary Supplement Figure 1, right), whereas unconjugated dmDNA31 concentrations were similarly low in both populations (data not shown).

Immunogenicity

Baseline ADAs were detected in 1 of 24 patients (4%). For the patient who was ADA-positive at baseline, post-treatment ADA responses remained similar to or lower than baseline levels. Following DSTA4637S administration, 1 of 18 patients (6%) developed treatment-induced ADAs, which was transient; this patient was in the 45 mg/kg dose group. The ADA-positive patient did not experience any IRRs, and all 5 patients with IRRs were ADA-negative. After completing DSTA4637S treatment, PK parameters in the ADA-positive patient were within the range observed for ADA-negative patients.

S. aureus Resistance to DSTA4637S Antibody (MSTA3852A)

To assess for resistance to the antibody component of DSTA4637S, flow cytometry was used to characterize the expression of β-GlcNAc WTA, the antigen on the cell surface of S. aureus recognized by the antibody component of DSTA4637S, in S. aureus isolates obtained during screening from positive blood cultures of 24 patients. High levels of β-WTA expression were observed in isolates from 21 patients (88%), moderate expression in 1 patient (4%), and no detectable expression in 2 patients (8%). In 1 patient isolate lacking detectable β-WTA expression, a premature stop codon was identified in the TarS gene. No genomic explanation was found in the isolate from the other patient to explain the lack of β-WTA expression. Five patients had additional S. aureus isolates collected on study days 2 to 4 from positive blood cultures, and 1 patient had a second isolate collected on day 87 during a recurrent S. aureus infection. No differences in β-WTA expression were observed between screening and post-treatment isolates for these patients.

S. aureus Susceptibility to dmDNA31

S. aureus isolates from screening blood cultures were available for 24 patients; 5 patients had additional isolates collected on days 2–4, and 1 patient had a post-treatment isolate collected on day 87, during a recurrent S. aureus infection. All 30 isolates (24 baseline, 6 post-treatment) were susceptible to dmDNA31 (MIC ≤ 0.002 µg/mL; susceptible quality control strain MIC ≤0.015 µg/mL) and rifampin (MIC ≤0.008 µg/mL; susceptible breakpoint: ≤1.0 g/mL), with no differences observed between MRSA and MSSA isolates (Supplementary Table 1).

One rifampin-resistant S. aureus isolate (MIC ≥ 4 µg/mL) was isolated from a patient who received 15 mg/kg DSTA4637S and developed a recurrent S. aureus infection on day 165. The patient's baseline S. aureus isolate was rifampin-susceptible (MIC ≤1.0 µg/mL), and the patient had not received rifampin during treatment, suggesting potential cross-resistance due to prior exposure to dmDNA31, which shares RNA polymerase as a target.

DISCUSSION

AACs represent a novel strategy for targeted delivery of antibiotics to intracellular reservoirs. To our knowledge, this is the first clinical evaluation of an AAC in an infectious disease. In this study, DSTA4637S was administered weekly at 15, 45, and 100 mg/kg in combination with SOC antibiotics.

Twenty-five patients with complicated S. aureus bacteremia (68% MSSA, 32% MRSA) and substantial comorbidities were enrolled; most had bone and joint or endovascular infections. Interpretation of safety data in this population is challenging, as AEs may reflect both disease and treatment effects [18, 19]. Accordingly, TEAEs were frequent (92%), with Grade ≥3 TEAEs and SAEs reported in 72% and 56% of patients, respectively. The highest incidence of TEAEs, SAEs, and Grade ≥3 events occurred in the 100 mg/kg cohort, although dose–response interpretation is limited by protocol modifications that broadened eligibility in this group.

AEs consistent with dmDNA31 were observed, including transient blue discoloration of tissues and body fluids. Artifactual hypoxemia, characterized by low pulse oximetry readings not confirmed by arterial blood gases or clinical assessment, is plausibly attributable to optical interference from dmDNA31 absorption near 660 nm.

Although IRRs were anticipated based on the known safety profile of antibody-based therapeutics [20–23], they occurred more frequently and with greater clinical impact than in healthy volunteers, where a single moderate IRR was observed at 150 mg/kg, beginning 26 hours postinfusion and characterized by chills, fever, and hypoxia [12]. In this study, IRRs (fever, chills, hypoxia, and/or hypertension) occurred in 26% of patients receiving DSTA4637S and led to treatment discontinuation. No dose–response relationship or predictive factors were identified, including baseline characteristics, PK exposure, ADA status, duration of bacteremia, or predose cytokines. Although reversible with supportive care, the frequency, unpredictability, and impact on continuity of therapy of IRRs would be a limitation for clinical use.

The PK profile of DSTA4637S reflects the complexity of AACs, with 3 analytes: TAC conjugate (1–2 dmDNA31 molecules), total antibody (fully conjugated, partially deconjugated, or fully deconjugated), and unconjugated dmDNA31. Exposure increased approximately dose-proportionally, and drug–antibody ratio and deconjugation were comparable; however, AUCs for TAC conjugate and total antibody were 30%–60% lower in patients than in healthy volunteers, consistent with other biologics [24]. The mechanism of reduced exposure remains incompletely understood but may reflect disease-related factors, with no clear associations observed with patient characteristics, disease state, inflammatory biomarkers, or methodological differences between studies. Nonclinical data similarly did not suggest intrinsic PK differences between healthy and infected states. Further studies are needed to clarify mechanisms, including potential contributions of tissue distribution and target-mediated clearance.

Interpretation of exploratory observations, measured by time to bacteremia clearance, was limited by the small number of subjects who remained bacteremic at dosing (n = 4), antecedent antibiotic therapy, and confounding effects of source control (see Supplementary Materials for details).

Consistent with findings in healthy volunteers, low systemic levels of unconjugated dmDNA31 were observed in patients with S. aureus bacteremia [12], suggesting that DSTA4637S has a relatively stable linker, with cleavage occurring predominantly within phagolysosomes, including during infection. The short mean half-life of unconjugated dmDNA31 further supports formation-rate–limited kinetics, in which deconjugation is slower than elimination of the released drug.

Evidence of possible pre-existing resistance was suggested by 2 baseline S. aureus isolates lacking β-WTA expression. In 1 case, this was associated with a premature stop codon in tarS; in the other, no alterations in tarS, tarP, or tarM were identified, suggesting potential environmental modulation of β-WTA expression as previously described [25] and warranting further investigation.

As with other rifamycin-class antibiotics, resistance to dmDNA31 is a key consideration for DSTA4637S. The in vitro frequency of spontaneous resistance was comparable to rifampin (3.9 × 10⁻⁷ vs 7.7 × 10⁻⁷). No baseline dmDNA31 resistance was detected, and concomitant SOC therapy was expected to limit the emergence of resistance during treatment. However, prolonged low-level exposure after treatment, due to slow deconjugation, could theoretically exert selection pressure. One patient treated with 15 mg/kg developed rifampin-resistant S. aureus (MIC ≥4 µg/mL) from a recurrent spinal infection 18 weeks post-treatment; although dmDNA31 susceptibility could not be directly assessed, this finding raises the possibility of treatment-associated cross-resistance in the setting of incomplete source control.

Conducting interventional clinical trials in S. aureus bacteremia is inherently challenging due to patient complexity, including substantial comorbidities, limited eligibility, need for rapid treatment initiation, and requirement for specialized expertise in managing severe infections [26]. In this study, 2731 patients with positive S. aureus cultures were prescreened across 17 global sites, resulting in the enrollment of 25 patients over 2.5 years. Expansion of eligibility criteria to include older, immunocompromised patients, those with end-stage renal disease or on hemodialysis, and individuals with recent substance use did not meaningfully improve enrollment and further complicated safety assessment. Additional operational challenges, included multidisciplinary care, heterogeneous management practices, early transition to outpatient care, complicated monitoring, and AE attribution. These findings align with prior studies and highlight the difficulty of conducting trials in this population [5, 27–29].

AACs represent a novel approach to targeting intracellular S. aureus reservoirs that may contribute to treatment failure in complicated infections. In this study, DSTA4637S was associated with an unexpectedly high incidence of IRRs and 30%–60% lower systemic exposure than in healthy volunteers, limiting its therapeutic potential. Based on these safety and PK findings, further clinical development of DSTA4637S for S. aureus bacteremia is not planned. Nonetheless, these results emphasize the importance of early phase evaluation in the target patient population to fully characterize clinical risk and PK.

Supplementary Data

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.