A Phase 0 Study to Assess the Biodistribution and Pharmacokinetics of a Radiolabeled Antibody Targeting Human Kallikrein 2 in Participants with Metastatic Castration-Resistant Prostate Cancer

Visual Abstract

Pat ients with metastatic castration-resistant prostate cancer (mCRPC) have limited treatment options as novel androgen receptor (AR) inhibitors and cytotoxic therapies move to the castrationsensitive setting (1,2).Recently, the mCRPC treatment landscape has expanded with the approval of the targeted radiopharmaceutical therapy 177 Lu-vipivotide tetraxetan (Pluvicto; Novartis), a systemic treatment that delivers radiation directly to the tumor cell, providing a novel mechanism for mCRPC treatment (3).Administration of such therapies requires a well-defined tumor-specific target to support effective payload delivery and minimal toxicity.In addition, other tumor-directed therapies, such as immunotherapies with chimeric antigen receptor T cells or T-cell-sequestering antibodies and antibody-drug conjugates, are promising avenues for drug development for mCRPC (4,5).Prostate-specific membrane antigen (PSMA) is the only cell surface target to be leveraged for clinical benefit in men with advanced prostate cancer; however, PSMA expression is heterogeneous among patients and even within patients on a lesion-to-lesion basis (6).PSMA is also putatively expressed in normal tissues such as the salivary and lacrimal glands, which can lead to off-tumor but on-target side effects in nonprostate tissue (6).Finally, patients who progress through PSMA-directed therapy may develop disease that does not express PSMA.New targets are therefore needed to address these shortcomings and to advance the treatment options for a disease as biologically dynamic and diverse as mCRPC.
Human kallikrein 2 (hK2; HUGO Gene Nomenclature Committee-approved gene symbol KLK2) is a trypsinlike antigen produced by columnar prostate epithelial cells with expression driven by AR signaling in a manner identical to that of the closely related gene encoding prostate-specific antigen (HUGO Gene Nomenclature Committee-approved gene symbol KLK3) (7-10).hK2 is both secreted and membrane-bound; however, unlike prostate-specific antigen, circulating hK2 is found at exceptionally low levels, where it can be bound by multiple protease inhibitor complexes (7,8,11).It has been shown that higher hK2 expression correlates with increased cell proliferation and lower apoptosis in castrationresistant prostate cancer specimens, thereby modulating the growth of castration-resistant disease (12).Additionally, a study reported 80% of mCRPC bone lesions to be AR-driven based on high expression of AR gene signatures, including both KLK2 and KLK3 (13).Increased AR signaling activity has demonstrated increased hK2 expression but decreased PSMA expression in multiple prostate cancer models (8,14,15).Specifically, treatment with the a-particle emitter 225 Ac generated a feed-forward mechanism with increased AR signaling and hK2 expression (8).This could impact therapeutic outcomes of targeting hK2 versus PSMA in mCRPC, because reactivation of AR signaling is commonly associated with mCRPC, despite castration serum levels of androgen.Indeed, preclinical observations in animal models showed improved survival after a single administration of hK2-targeting 225 Ac, suggesting a therapeutic benefit of targeting hK2 (8).Finally, hK2 expression shows prostate specificity and is relatively homogeneously expressed across disease stages, from localized to metastatic prostate cancer (16).Together, these data suggest hK2 to be a promising target for treatment.
In preclinical studies, reduction in tumor volume along with prolongation of survival were observed in mouse models of prostate cancer treated with the 225 Ac-radiolabeled humanized anti-hK2 monoclonal antibody, h11B6, where h11B6 bound only membraneassociated hK2 (8,17).This phase 0 first-in-human study evaluated the biodistribution, normal-tissue dosimetry, and tumor targeting of 111 In-radiolabeled h11B6 ([ 111 In]-DOTA-h11B6) to evaluate hK2 as a target in patients with mCRPC.

Study Design and Participants
This was a phase 0 imaging study conducted to evaluate the biodistribution, normal-tissue dosimetry, and tumor targeting of [ 111 In]-DOTA-h11B6 (ClinicalTrials.govidentifier NCT04116164).After written informed consent for study participation was provided by the patients, screening of eligible participants was conducted 14 d before administration of the study agent.Participants at least 18 y of age with progressive measurable/evaluable mCRPC as defined by Prostate Cancer Working Group 3 criteria were prospectively enrolled in the study.Eligible participants had an Eastern Cooperative Oncology Group performance status of 1 or less, a Karnofsky performance scale score of 70 or greater, castrate testosterone levels less than 50 ng/dL, and metastatic disease documented by CT, PET/CT, MRI, or a radionuclide ( 99m Tc-methylene diphosphonate or 18 F-NaF) bone scan.Participants with lesions identified by PSMA PET only and participants with pure small-cell or neuroendocrine prostate cancer were excluded.Participants could not have any condition that could impair the ability to comply with study procedures and could not have received radiotherapy or immunotherapy within 30 d, a therapeutic radioactive isotope within 28 d, or a single fraction of palliative radiotherapy within 14 d of administration of the study agent.Participants with a known allergy to antibodies were also excluded.The study was performed under a U.S. Food and Drug Administration-reviewed exploratory investigational new drug application and was approved by the Memorial Sloan Kettering Cancer Center Investigational Review Board, and all participants signed an informed consent form.Study data were collected and processed with adequate precautions to ensure confidentiality and compliance with data privacy protection laws and regulations.
The study enrolled participants in cohort 1, with the option to expand into an additional cohort at the discretion of the sponsor on the basis of findings from cohort 1.The focus of cohort 1 was to identify the optimal mass of [ 111 In]-DOTA-h11B6, which was defined as the mass at which there is minimal accumulation of radioactivity in parenchymal organs (liver, lung, spleen) and maximal accumulation in mCRPC lesions as assessed by imaging.Data collected from cohort 1 were used to estimate the serum clearance, the normal-tissue dosimetry, and the most favorable mass amount of [ 111 In]-DOTA-h11B6.Participants in cohort 1 received a single administration of [ 111 In]-DOTA-h11B6 (185 MBq) as a slow intravenous bolus over 1 min 6 30 s followed by a saline flush through an indwelling peripheral intravenous catheter.
The primary endpoints of the study were safety and [ 111 In]-DOTA-h11B6 dosimetry, including serum clearance, optimal mass, and [ 111 In]-DOTA-h11B6 biodistribution.Secondary and exploratory endpoints included pharmacokinetic parameters and serum hK2 levels, respectively.
The macrocyclic chelator DOTA was conjugated to h11B6 (Dia-Prost AB) through incubation in a p-SCN-Bn-DOTA solution (Macro-Cyclics Inc.) at 25 C for 20 h.[ 111 In]-DOTA-h11B6 was prepared by incubating 111 In in acetate buffer for 2 h at 37 C with DOTA-h11B6.The reaction mixture was purified and tested for sterility, endotoxins, and radiochemical purity.Immunoreactivity was measured through a validated immunoassay.Preclinical testing of [ 111 In]-DOTA-h11B6 demonstrated radiochemical purity of more than 95% and immunoreactivity consistently greater than 80%.In vitro studies confirmed antigen-mediated antibody internalization as previously reported (8).

Assessments
Anterior and posterior planar 111 In whole-body images were obtained immediately after study agent administration (0-4 h) and on day 2 (24 6 6 h), days 3-4 (42-78 h), and days 6-9 (114-198 h).A SPECT/CT scan from the inferior portion of the lungs to the kidneys was performed at one or more of these time points.For all time points except 0-4 h, participants were instructed to void urine before imaging.Imaging was performed on a dual-head hybrid g-camera (Siemens Intevo SPECT/CT; Siemens Healthcare).
Blood samples were collected before treatment; 15, 30, 60, 120, and 240 min (65 min) after study agent administration; at each imaging session; and at the 2-wk follow-up visit to measure the radioactivity and/or serum concentrations of h11B6.Vital signs (blood pressure, pulse, respiratory rate, temperature, and weight) were obtained before and up to 2 h after study agent administration as well as at imaging.Complete blood count and comprehensive metabolic panels were obtained at baseline and at follow-up.
Serum samples were measured in duplicate using an NaI (Tl) g-well-type detector (Wallac Wizard 1480 automatic g-counter; PerkinElmer) together with appropriate standards to assess radioactivity.The measured radioactivity concentrations were converted to a percentage of injected dose per liter, and serum volume of distribution was estimated as 100% per time-zero intercept of the percentage of injected dose per liter.Serum concentrations of h11B6 were determined using a validated immunoassay.
Absorbed radiation doses to normal tissues from [ 111 In]-DOTA-h11B6 were estimated on the basis of serum clearance of radioactivity and by image analysis of tissue uptake and retention of radioactivity over time.Data of the time-integrated activity coefficient derived from this analysis were used with the U.S. Food and Drug Administrationapproved software package application OLINDA/EXM version 2.0 (Hermes Medical Solutions) to estimate absorbed radiation doses to a panel of normal tissues using the MIRD method.
Accumulation of radioactivity was defined as an increasing ratio of tumor-to-nontumor signal over time in known sites of disease.The number of lesions that demonstrated accumulation of radioactivity was also measured.
Adverse events were assessed using the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.0, and participants were monitored for side effects during infusion and at every follow-up visit for up to 2 wk after agent administration.Participants were contacted 1-2 d after study agent administration to ascertain occurrence of any treatment-emergent adverse events when symptoms were not assessed during an on-site visit.

Statistical Analysis
Descriptive statistics were calculated for key parameters, reported as the mean value and SD or SE.

Safety
All participants who received the study agent were included in the safety and tolerability analysis.[ 111 In]-DOTA-h11B6 was well tolerated.Of the 22 participants enrolled, 2 participants (9.1%) in Serial whole-body images showed gradual clearance of radioactivity the vascular compartment and gradual though normal accumulation in the liver at later time points.There was no evidence of increased physiologic [ 111 In]-DOTA-h11B6 accumulation in bones, salivary glands, or the thyroid at either antibody mass level.Time-integrated activity analyses confirmed predominant localization of [ 111 In]-DOTA-h11B6 in the vascular compartment rather than in critical normal organs (Table 2), and absorbed radiation doses of critical normal organs were low for both cohorts (Table 3).A complete list of absorbed radiation doses of normal organs is included in Supplemental Table 2.A relationship between administered antibody mass and biodistribution was not observed (Fig. 1A).Serum clearance kinetics showed a gradual decline in radioactivity over time (Fig. 1B), with no apparent difference between the 2-mg antibody mass (subcohort 1A) and 10-mg antibody mass (subcohort 1B), suggesting independence of antibody mass.The mean volume of [ 111 In]-DOTA-h11B6 distribution 6 SE was 2.82 6 0.13 L and 2.87 6 0.26 L for subcohort 1A (n 5 15) and subcohort 1B (n 5 6), respectively.
Targeting of [ 111 In]-DOTA-h11B6 to known lesions was observed in 18 participants via SPECT/CT (Fig. 2) and was confirmed via 68 Ga-PSMA-11 PET/CT (data not presented).In the 4 participants with no evidence of lesion targeting, 1 participant had no evidence of disease by PSMA PET and 1 had incomplete imaging.

DISCUSSION
This first-in-human study demonstrated the favorable distribution of [ 111 In]-DOTA-h11B6 in participants with mCRPC and highlighted the feasibility of hK2-directed imaging.Tumor uptake was noted in 18 of 22 participants.Of the 4 participants with no evidence of tumor targeting, 1 had no evidence of disease by PSMA PET and 1 participant had incomplete imaging.Tumor accumulation of [ 111 In]-DOTA-h11B6 was visually observed approximately 4 d after agent administration and was more evident via imaging with increasing time after injection.
As with other monoclonal antibodies, serum clearance kinetics and time-activity analyses demonstrated antibody localization predominantly to the vascular compartment, with clearance characteristics comparable to those of other intact humanized IgGs (18,19).Imaging data further supported vascular compartment localization, with no evidence of significant parenchymal targeting or any extratumoral antigen sink.Importantly, the antibody mass amounts studied (2 and 10 mg) had no effect on serum clearance, volume of distribution, or tumor targeting.
[ 111 In]-DOTA-h11B6 was tolerated in participants with mCRPC.Two treatment-emergent adverse events were observed; however, these events were not related to the study treatment.The patient who experienced tumor-related bleeding discontinued the study early.Normal-organ and whole-body absorbed radiation doses were well within the limits of use for imaging radiopharmaceuticals. Unlike PSMA-targeted agents that are accompanied by an uptake in the lacrimal and salivary glands and in the renal tubules, which can lead to dose-limiting toxicity (20), targeting of hK2 with [ 111 In]-DOTA-h11B6 showed no significant uptake in the salivary glands and a modest renal uptake (0.22-0.23 mGy/MBq); therefore, low radiation doses to these organs is expected.Phase 1 studies assessing hK2-targeting therapies, including 225 Ac radioimmunotherapy, are currently ongoing and may further support hK2 as a therapeutic target in advanced prostate cancer (21).As off-target localization to parenchymal organs and bone marrow is a potential area of concern, especially for radioactive payloads, hK2-targeting radiolabeled agents will need careful monitoring.To this end, the risks of myelosuppression, renal and liver toxicity, and gastrointestinal effects have been identified with appropriate mitigation strategies in the ongoing phase 1 study assessing JNJ-6240 ([ 225 Ac]-DOTA-h11B6) (21).
The tumor-targeting specificity of [ 111 In]-DOTA-h11B6, its mass-independent biodistribution, and its tolerable safety profile together suggest a role for h11B6 as a carrier for targeted delivery of a conjugated payload for the treatment of mCRPC.Through  characterization of h11B6 biodistribution in human participants, this study further adds to previous preclinical prostate cancer studies (8,22) by validating the expression of hK2 in mCRPC, providing data to support the use of h11B6 in a clinical setting.

CONCLUSION
This phase 0 trial demonstrated that [ 111 In]-DOTA-h11B6 was well tolerated with mass-independent biodistribution and tumor targeting in These data not only confirm the validity of hK2 as both an imaging and a potential therapeutic target for mCRPC but also support the continued investigation of h11B6 for directed delivery of payloads such as therapeutic radionuclides to address an unmet need for additional mCRPC treatment options.Novartis (institutional), and Astellas Pharma (institutional); and travel, accommodations, and expenses from AstraZeneca, APCCC, and Memorial Sloan Kettering Cancer Center.No other potential conflict of interest relevant to this article was reported.

FIGURE 1 .
FIGURE 1. (A) Representative anterior and posterior whole-body images over time showing biodistribution after administration of 2 mg (top) or 10 mg (bottom) of [ 111 In]-DOTA-h11B6.Bars on right represent percentage of maximum pixel value over all images.(B) Average serum clearance kinetics are based on decay-corrected percentage of injected 111 In activity (percentage of injected dose per liter 6 SE) across subcohort 1A (2 mg, n 5 15) and subcohort 1B (10 mg, n 5 6).%ID/L 5 percentage of injected dose per liter.

TABLE 1
Participant Characteristics Across Subcohorts Participant was unable to complete remaining imaging per protocol because of noncompliance and discontinued study early.No results on tumor location or targeting are available.
† ‡ Participant had no obvious disease on imaging.ECOG PS 5 Eastern Cooperative Oncology Group performance status; PSA 5 prostate-specific antigen; N/A 5 not available.Subcohort 1A received 2 mg of protein, and subcohort 1B received 10 mg of protein.

TABLE 3
Absorbed Radiation Doses for Key Organs by Subcohort 1A (n 5 15*) and Subcohort 1B (n 5 6) *One participant was unable to complete remaining imaging per protocol because of noncompliance and discontinued study early.Data are mean with SE in parentheses.Subcohort 1A received 2 mg of protein, and subcohort 1B received 10 mg of protein.