Clinical Investigation of a Humanized Anti-CD47 Antibody in Targeting Cancer Stem Cells in Acute Myelodi Leukaemia (and Solid Tumours).

Most normal tissues are maintained by a small number of stem cells that can both self-renew to maintain stem cell numbers, and also give rise to progenitors that make mature cells. We have shown that normal stem cells can accumulate mutations that cause progenitors to self-renew out of control, forming cancer stem cells (CSC). CSC make tumors composed of cancer cells, which are more sensitive to cancer drugs and radiation than the CSC. As a result, some CSC survive therapy, and grow and spread.

We sought to find therapies that include all CSC as targets. We found that all cancers and their CSC protect themselves by expressing a 'don't eat me' signal, called CD47, that prevents the innate immune system macrophages from eating and killing them. We have developed a novel therapy (anti-CD47 blocking antibody) that enables macrophages to eliminate both the CSC and the tumors they produce. This CIRM Grants System - Administration https://grants.cirm.ca.gov/applicant/grant_app/print/3385 anti-CD47 antibody eliminates human cancer stem cells when patient cancers are grown in mice. At the time of funding of this proposal, we will have fulfilled FDA/MHRA requirements to take this antibody into clinical trials, showing in animal models that the antibody is safe and well-tolerated, and that we can manufacture it to MHRA/FDA specifications for administration to humans.

Here, we propose the initial clinical investigation of the anti-CD47 antibody with parallel first-in-human Phase 1 clinical trials in patients with either Acute Myelogenous Leukemia (AML) or separately a diversity of solid tumors, who are no longer candidates for conventional therapies or for whom there are no further standard therapies. The primary objectives of our Phase I clinical trials are to assess the safety and tolerability of anti-CD47 antibody. The trials are designed to determine the maximum tolerated dose and optimal dosing regimen of anti-CD47 antibody given to up to 42 patients with AML and up to 70 patients with solid tumors. While patients will be clinically evaluated for halting of disease progression, such clinical responses are rare in Phase I trials due to the advanced illness and small numbers of patients, and because it is not known how to optimally administer the antibody. Subsequent progression to Phase II clinical trials will involve administration of an optimal dosing regimen to larger numbers of patients. These Phase II trials will be critical for evaluating the ability of anti-CD47 antibody to either delay disease progression or cause clinical responses, including complete remission. In addition to its use as a stand-alone therapy, anti-CD47 antibody has shown promise in preclinical cancer models in combination with approved anti-cancer therapeutics to dramatically eradicate disease.

Thus, our future clinical plans include testing anti-CD47 antibody in Phase IB/II studies with currently approved cancer therapeutics that produce partial responses. Ultimately, we hope anti-CD47 antibody therapy will provide durable clinical responses in the absence of significant toxicity.

Many cancers are organized as a cellular hierarchy initiated and maintained by cancer stem cells (CSC). CSC comprise a subset of tumor cells that are relatively resistant to current therapy, which often shrink tumors, but do not lead to long-term cures. Such cures require strategies to target all the CSC. CD47 is a immunomodulatory protein expressed on the surface of CSC and cancer cells from nearly all human tumors. CD47 binds SIRP-alpha on phagocytes, which delivers an inhibitory signal to block phagocytosis. CSC and cancer cells upregulate CD47 expression, presumably to suppress phagocytic attack. This proposal outlines a clinical development plan to evaluate the safety, tolerability, and efficacy of our therapeutic, the humanized anti-CD47 antibody Hu5F9-G4. Hu5F9-G4 is a blocking anti-human CD47 antibody that prevents the binding of CD47 to SIRP-alpha, disrupting its inhibitory signal, and enabling the phagocytic removal of CSC and cancer cells. In xenograft models, we have demonstrated that Hu5F9-G4 has potent efficacy against AML. In non-human primate toxicokinetic studies, Hu5F9-G4 was well-tolerated at doses able to achieve sustained serum levels in a potentially therapeutic range. Simultaneously, we generated a cGMP-qualified master cell bank, and have initiated GLP/GMP manufacturing of Hu5F9-G4. We have had positive initial regulatory interactions with the FDAand MHRA in the UK, and and are on track to file two separate IND/IMPDs, one for AML, and one for solid tumors by early 2014. Here, we propose parallel first-in-human Phase 1 clinical trials in the US and UK of Hu-5F9-G4 in patients with either AML or separately a diversity of solid tumors for whom there are no further standard therapies. The primary objective will be to assess the safety and tolerability of Hu5F9-G4, to identify the maximum tolerated dose (MTD) or optimal biologically effective dose (OBED), and to get a preliminary assessment of tumor reduction for further efficacy studies.

1. Health care and economic benefits to UK and world wide: There are 2200 new AML cases/year in the UK. 80% of patients are >60 years of age; in an aging population, the disease is becoming more common. Current curative therapy is restricted to medically fit patients, usually <70 years old (~30% of patients). It consists of 2-4 cycles of toxic chemotherapy. A patient spends 3-4 weeks/cycle receiving expensive in-patient care (cost ~£20000/cycle/patient). In ~30% of patients this is followed by an allogeneic blood stem cell transplant (cost £45-60 000/patient). Even this programme only cures 40-50% patients <60 years old. Direct treatment cost to NHS hospitals ~£60-120 000/patient. Annual cost for 660 patients (30% of 2200 who receive curative approach) ~£40M-£80M. This does not include costs: (i) in primary care (ii) loss of earning (iii) tax revenue (iii) care at home (iv) care of problems secondary to initial treatment - e.g. chemotherapy-induced second malignancies (5-10% of patients). ~70% of AML do not receive curative therapy and 98% of these patients die within 4 months. They receive palliative supportive care. Some are entered into Phase I/II/III trials of newer therapies; data so far show none are curative and may benefit 10-40% of patients.

We plan to proceed to 2 Phase II trials (i) elderly untreated AML population (ii) younger patients with minimal residual disease. If Phase II trials show an efficacy signal we will move rapidly to a randomised UK AML Working Party driven Phase II/III looking for an impact on (i) overall survival (currently only 2% in elderly unfit AML patients and 30% in MRD+ve young patients), (ii) improved quality of life, and reduced use of health care resources. We will build in QoL assessment and health care utilisation into the randomised Phase II/III trials.

2. Establishing new class of therapy for AML (and cancers more broadly): If safe, and if the AML and solid trials shows an efficacy signal, it establishes CD47 as a novel therapeutic target and targeting programmed cell removal as a novel target pathway. This will impact the whole cancer community.

3. Ability to shape the Anti-CD47 Trial Programme: If Phase I/II programmes are successful it will allow UK investigators to lead, design and deliver pivotal licencing Phase III studies. This will impact on and build (i) knowledge, experience and training of UK teams (ii) a trial programme addressing UK health care needs (iv) world-wide reputation of UK AML clinical and translational programmes (iv) potentially allow to access anti-CD47 therapeutics at reduced cost.

4. Intellectual Property (IP) of clinical and laboratory dataset: The clinical and laboratory dataset from the Phase I/II trials will be important for design, submission and delivery of licencing Phase III studies. This IP will be jointly owned by Universities of Oxford and Stanford. With Stanford we will seek further funding for Phase III trials. The Stanford and Oxford groups want to lead a joint long-term programme for taking anti-CD47 to market.

5. Intellectual Property (IP) of LSC monitoring flow cytometry panel: University of Oxford have IP on aspects of the immunophenotyping panel and joint IP with Becton Dickinson and Stanford on other aspects of the flow cytometry panel. This panel has potential commercial value to identify patients who will respond to (i) anti-CD47 therapy (ii) new therapeutics directed against other stem cell antigens and (iii) identify antibody combination directed against different leukaemic stem cell antigens. Please see letter from Dr Richard Reschen from Issis Innovation.

6. Training of all staff directly involved in the programme and boosting the knowledge economy

7. Provide a paradigm (and thus incentive) for academics take therapeutic targets into clinical practice (particularly through the Phase I study setting) without commercial support thereby increasing value of the therapeutic product.