The Bispecific PSMAxCD3 Antibody CC-1 in Patients with Castration Resistant Prostate Carcinoma

Study rationale with regard to patient populations and test drug

CRCP is an aggressive malignant disease with poor prognosis after failure of established therapies. Any drug employed after third-line treatment is associated with only limited clinical benefit. Therefore, there is a high medical need for new therapeutic approaches in this patient population. While progress has been made in the field of immunotherapeutics for other tumor types, no relevant progress in immunotherapeutic options in CRPC has been achieved over the last years. The rationale for the therapeutic use of CC-1 is based on its proposed mode of action as a bsAb being specifically designed to direct T cells via its CD3 binding part towards tumor target cells via its PSMA binding part. Furthermore, CC-1 also reacts with tumor vessels of CRPC thereby allowing for a dual mode of anti-cancer action by also attacking tumor blood supply and allowing for improved influx of immune effector cells. Due to its unique ability to redirect T cells via CD3 for PSMA-expressing tumor cell lysis, CC-1 can elicit repeated target cell elimination by cytotoxic T cells and a polyclonal response of previously primed CD4+ and CD8+ T cells. Compared to other immunotherapeutics presently being approved or in development (bsAbs with alternative formats like the authorised bsAb blinatumomab or other antibodies or CART cells), CC-1 is expected to offer the following major advantages:

(i) Reduction of off-target T cell activation and thus reduction of side effects due to its optimized format (ii) The possibility to tightly steer anti-target activity via serum level-controlled antibody application which, in contrast to CART cells, allows for termination of activity if desired.

On the background of the poor prognosis of patients with CRPC after third-line therapy, the bsAb CC-1 holds promise as a new treatment option of immunotherapy for these patients.

Study rationale with regard to objectives and further development of CC-1 In nonclinical studies, in vitro and in vivo, proof of concept, preliminary PK and PK/PD effects as well as toxicology have been evaluated as described in detail in the IB, section 5. However, due to differences between animal models and the human situation, some aspects have to be assessed and further characterised in humans. For example, the target mediated drug disposition (TMDD), an effect that largely influences the serum half-life of antibody molecules particularly at low concentrations, cannot be properly addressed in mice. Furthermore, nonhuman primates (NHP) and rodents have several limitations as predictive models for toxicity and immunogenicity evaluation of CC-1. The CD3 binding part of CC-1 does not cross-react with CD3 of macaques and thus it is not possible to evaluate in these NHPs dose limiting side effects. Likewise, although CC-1 is cross-reactive with macaque-PSMA, PSMA distribution in macaques significantly differs from that in humans. The same holds true for rodents. Therefore, a First in Human study is planned for CC-1 to overcome these limitations and to characterise the effects of CC-1 in humans.

The planned phase I trial is designed to confirm and further explore the safety and tolerability of the PSMAxCD3 bsAb CC-1 in adult patients with CRPC after third-line therapy. The primary objective is incidence and severity of adverse events (AEs) under therapy with CC-1. Furthermore, the trial aims to expand experience on pharmacokinetics, pharmacodynamics and toxicology of CC-1 from nonclinical studies to the human situation in relation to the PK, expected efficacy and safety. A focus will be on the following specific aspects/parameters:

• Pharmacokinetics and pharmacodynamics of CC-1 in humans
• Immunogenicity of CC-1 in humans based on both absolute (number and percentage of subjects who develop human anti-human antibody (HAHA).
• Absolute changes from baseline in the tumor marker (PSA)
• Absolute changes from baseline in laboratory parameters
• Change in cytokines from baseline
• Assessment of response rate by ReCist on routine imaging
• Overall and progression free survival

Rationale for preemptive IL-6R blockade by Tocilizumab prevent development of CRS in the first place than to treat CRS once it has arisen. This strategy holds promise to increase the safety of study patients and timely study conduct.

By starting the study treatment with CC-1 directly with prophylactic tocilizumab, all study patients will benefit from the expected advantage of this combination with regard to safety and can be treated with sufficiently high doses of CC-1 to achieve dose levels high enough to hopefully result in efficacy effects.The rationale for preemptive IL-6R blockade by tocilizumab treatment is based on i. The firmly established efficiency and safety of tocilizumab for the treatment of CRS ii. Lack of clear evidence for increased tumor growth as potential drawback of IL-6R blockage iii. Observations that IL-6 activity, while being responsible for the undesirable sequelae of CRS, appears not to be required for the therapeutic activity of CC-1 CRS that was induced by therapy with the approved bsAb blinatumomab was reported to be successfully treated by tocilizumab. Most importantly, despite rapid disappearance of clinical CRS symptoms, the therapeutic activity of the bsAb blinatumomab was maintained. Furthermore, tocilizumab was also used in the very recent FIH study with the REGN1917 (CD20xCD3) antibody (Bannerji et al., 2018). Our own nonclinical studies demonstrate that tocilizumab does not impair the therapeutic activity of CC-1, neither in vitro nor in vivo. This is in contrast to steroids which are currently recommended and used as pre- and concomitant treatment to prevent CRS upon blinatumomab therapy. Due to the mechanism of action of tocilizumab, there is a theoretical risk of tumor development or tumor progression due to immune modulation. On the basis of the current literature derived from large studies conducted in Japan, the USA and Europe, however, there is no evidence for an increased tumor risk upon application of tocilizumab. Only one Japanese study described a minimally increased risk of de novo lymphoma development. However, this could not be confirmed in any other study. Especially for prostate cancer, there is no evidence for an increased incidence rate. Furthermore, tocilizumab is currently being investigated in several Phase I/II studies for the treatment of solid and hematological neoplasias without available evidence for an influence on tumor pathophysiology. Based on these findings, no relevant negative effects of tocilizumab on the efficacy and safety of CC-1are expected.

Trial Duration and Schedule A treatment cycle consists of 21 days, comprising 7 days (d1-7) of CC-1 application as continuous infusion, followed by a treatment free period with further close monitoring for safety. The duration of the trial for each patient is expected to be at least 3 months including the first treatment cycle of 21 days and a safety follow-up on day 90 (end of study, EOSf). Thereafter, a continuous survival follow-up every three months is conducted for one year after EOSf. In case of clinical benefit, treatment period can be longer.

Screening period, enrollment and baseline The screening period is part of the study and begins 14 days prior to first treatment. The patient is informed about the study by the investigator and has to sign informed consent. In the following, further measures and analyses are performed to check whether the patient meets all inclusion/exclusion criteria. If a patient meets all criteria, he is enrolled in the study. If he does not meet all criteria, he is not enrolled but documented as a screening failure.

Baseline values are the following:

• Routine imagingimaging and analysis of HAHA are both assessed in the screening period (up to day -14).
• Baseline values for Quality of Life are assessed in the screening period (up to day -14)
• Baseline values for PSA, cytokines, pharmacokinetics and liquid biopsy are taken directly before start of CC-1 infusion (on day 1)
• Baseline values for hematology and blood chemistry as well as coagulation parameters are taken directly before start of CC-1 infusion (on day 1)
• Baseline for signs and symptoms of disease as well as vital signs (for evaluation of CRS) are documented directly before start of CC-1 infusion (on day 1)

Treatment Cycle Each treatment cycle of an individual patient is 21 (±2) days. Every cycle comprises a treatment period (d1-7) with a continuous infusion (daily) of CC-1 after preemptive tocilizumab application and an infusion free period (d8-21). While for d1-9 the patients are hospitalized and closely monitored, day 10-21 is planned as outpatient period with outpatient self-monitoring and visits at the hospital at d15 and d21 (EOT). Discharge of the patient is planned at day 9 but depending on the patient´s condition. At the discretion of the physician, the patient can be also discharged at day 10 (or later). Justification of continuous infusion Continuous infusion has been utilized in several phase I/II studies with the bsAb blinatumomab: In a phase I/II study in children with refractory ALL, blinatumomab was applied as 24 continuous infusion over 4 weeks. A 24h continuous infusion of blinatumomab was also applied in first investigations in adults (NCT00274742) as well as in a phase II study in adults with refractory B-ALL. Overall, continuous infusion is expected to provide reasonable safety due to avoidance of peak serum levels with subsequent rapid decline.

Justification of length of treatment (infusion) with CC-1: A treatment duration of 7 days was chosen for the following reasons:

It is expected that no maximum tolerated dose (MTD) will be identified in the dose expansion phase and that patients will receive the maximum tested dose. At higher dose levels, for reasons of safety we implemented step dosing on day 1 and day 2 of each cycle followed by application of the target dose (to be determined, possible maximum tested dose) over 5 days. This is based on our published observations that CD3-mediated activation of T cells within PBMC populations in vitro requires 3 days until full activation is achieved. Notably, in these in vitro analyses high antibody concentrations (3μg/ml and thus several orders of magnitude more than those that will be achieved in our study patients) were utilized. It was also observed in a previous study that the presence of human serum reduces CD3-mediated T cell activation. For these reasons, we conclude that exposure to active drug levels for additional 2 days (and thus 5 days in total after day 2) may ascertain optimal T cell activation.

Justification of length of cycle:

A cycle length of 21 days was chosen for the following reasons:

The calculated half-life in mice of about 48h is based on the simplified assumption of an exponential decline within the first 24h. Thereafter serum concentration declined at a slower rate indicating a biphasic elimination. For calculation of the cycle interval we therefore assumed a half-life of 48h. Considering that CC-1 application is completed on day 8, a margin of 5 half-lifes (equaling to 10 days) results in an overall interval of 18 days. Additionally, based on our reported findings that after achieving maximal T cell activation a decline of activity over 3 further days is observed, for reasons of safety we plan to allow for 3 additional days until the end of treatment (EOT) visit is conducted (18 plus 3 days) and treatment can potentially be repeated. Notably, in the dose escalation cohort, approval of the DSMB is required before a new patient can be treated with the next dose level, for which we calculate an additional 7 days. Thus, a second cycle of treatment can be applied at day 21 to the same patient, but treatment of new patients, for reasons of uniformity, is generally conducted at day 28 of the antecedent patient at the earliest.

Additional Cycles It is assumed that additional treatment cycles do not increase the risk but would rather benefit patients. Therefore, in case of positive response, the patient is allowed to continue treatment with up to 5 further treatment cycles. This is intended to allow patients to benefit better from therapy in case it is effective. Positive response to CC-1 treatment in previous cycle comprises (but is not limited to) a substantial PSA drop (>20% reduction compared to day 1 pretreatment as determined on day 15). To obtain additional cycles, patients must still meet the initial key eligibility criteria assessed between 6-8 days prior to the next cycle (excluding cardiac ultrasound, lung function, virology, tuberculosis testing, CT/MRI imaging). Patients are required to sign the informed consent form for additional CC-1 cycles. Study visits and investigations are performed as described for the first cycle of CC-1 treatment. This includes all obligations for safety-reporting and documentation. The dose of CC-1 in additional cycles is the same as in the patient´s first cycle. In case of additional cycles, the end of safety follow-up (EOSf) visit is prolonged to 90 days after start of the last CC-1 cycle. Routine imaging is planned every 90 days (+/- 7 days).

Post-Treatment Follow-Up until End of safety follow-up (EOSf)

After the treatment cycle (s), follow-up of patients is planned as follows:

After the last visit of the last treatment cycle at d21 (EOT), a follow-up visit is planned at d 90 (± 7) (69 days after EOT). In case of only one treatment cycle, this visit is the EOSf for this patient. This time point was chosen based on clinical practice and assessment according to RECIST criteria (see secondary endpoints). A comprehensive workup is planned on this visit. In case of additional cycles, the EOSf visit is prolonged to 90 days after start of the last CC-1 cycle.

Follow-up after EOSf until EOS After EOSf, each patient is followed every 90 days (± 7) for up to one 1 year. EOS will be the last visit one year after EOSf, unless study treatment is terminated earlier.

In case the patient is reported to have progressive disease on routine imaging, the investigator may decide to perform an additional visit 28-42 days later.

Patient care after last follow-up After finishing all study-relevant procedures, therapy and follow-up period as described above, the patient are followed in terms of routine care and treated if necessary, by the primary responsible oncology center. Any attempt should be made by the investigator to follow the patient in particular with regard to any safety issues even after the last assessment at EOS.

Estimated timelines The recruitment period may vary substantially based on observed toxicities, but is expected to be 2 years, and an overall study duration of 3.25 years is anticipated. Recruitment of patients starts in the fourth quarter of 2019