Using Preclinical Information for Clinical Development of Immunotherapy

Taking anti-CD137 agonistic monoclonal antibodies to the clinic it is a difficult but not an impossible task. As represented in Figure 8.2, it first involves the production of a panel anti-human CD137 mAbs through classical production of mouse hybridomas or in transgenic mice that produce fully human antibodies. If the classical technology is chosen, genetic engineering will be required to humanize the sequence. At this time the antibodies should be selected for being able to costimulate the in vitro proliferation of human T cells (). However this property on its own will not ensure their therapeutic behaviour in vivo. To more accurately foresee this feature, two types of experiments would be feasible: (i) testing the effect of the antibody on CTL responses of SCID mice reconstituted with human lymphocytes, and (ii) generating Knock-in mice in which the human CD137 sequence would be regulated by the same factors that induce mouse CD137. In the latter case the effects of the selected humanized antibodies should match the ones already achieved against established transplanted tumors. Humanized monoclonal

ROAD MAP TO CLINICAL TRANSLATION OF ANTI CD137 (4-1BB) MONOCLONAL ANTIBODIES

ROAD MAP TO CLINICAL TRANSLATION OF ANTI CD137 (4-1BB) MONOCLONAL ANTIBODIES

GMP-BIOPROCESING -j-► TROUBLE SHOOTING

RESULT EVALUATION:

1. PHARMACOKYNETICS

2. SAFETY

3. OPTIMAL DOSING

4. ASSESSMENT OF IMMUNITY

MORE ADVANCED CLINICAL TRIALS (PHASE II)

RESULT EVALUATION:

1. PHARMACOKYNETICS

2. SAFETY

3. OPTIMAL DOSING

4. ASSESSMENT OF IMMUNITY

MORE ADVANCED CLINICAL TRIALS (PHASE II)

Í IMMUNIZATION PROCEDURES

RADIOTHERAPY/CHEMOTHERAPY

OTHERS

POTENCIAL INDICATIONS (CANCER, CHRONIC VIRAL INFECTION)

Figure 8.2. Stepwise development and decision making for development of anti-CD137 agonistic mAb for clinical application.

antibodies are mandatory towards translation, but we should not forget that reported preclinical efficacy involves only rat immunoglobulins injected into mice. However to the best of our knowledge there is no reason to forecast any advantage for heterologous antibodies.

Once a particular antibody is selected, bioprocessing experts would have to find a way to produce enough quantities for toxicology, quality control, and pilot clinical trials. Classical protein production in cell cultures is the easiest possibility, while other approaches to be considered could use more creative biotechnology in transgenic plants or mammals. A continuous improvement of these techniques is to be implemented to satisfy regulations and the potential demand.

The clinicians designing and leading clinical development will have to perform the dose finding trials in patients with advance melanoma and renal cell carcinoma. Another possibility would be to include in these pilot trials patients with a broad spectrum of malignancies. Each alternative has potential advantages and drawbacks. Melanoma and renal cell carcinoma are probably the patients with the highest likelihood to detect some encouraging clinical efficacy at an early stage. Good early data would foster investment and will make the assessment of antitumor immunity an easier task. Combination with immunization strategies is a temptation and probably a wise alternative, since autoimmunity is not expected. Introducing model antigens for CTL and antibody responses would provide proof of the biological effects and give an idea of its intensity. However, artificial antigens might unbalance immunodominance (Yewdell and Del Val, 2004) and endanger the response to tumor antigens. Therefore the introduction of these model antigens in early trials is a debatable choice.

We should bear in mind that autoimmunity with anti-CTLA-4 monoclonal antibodies was not anticipated before reaching the clinical arena already in combination with cancer vaccines. However in that case, the severe autoimmunity pheno-type ofCTLA-4-/- mice (Tivol etal., 1995; Waterhouse etal., 1995) should have advised a more prudent course for development. Nonetheless, careful assessment of autoimmunity is a must in any case in CD137 development.

Dose finding is going to be very interesting. The antibodies would inhibit the response against themselves prolonging their half life even if not completely humanized. A humanized monoclonal antibody has already shown these properties in monkeys (Hong et al., 2000). A range from 0.1 mg/Kg to 40 mg/Kg of weight is to be tested with close follow up of the pharmacokinetics that is anticipated to be excellent. Readministration of doses should be decided from non-human primate pharmacokinetics but will probably consist in weekly or two-weekly repetitions. Studies in monkeys should have thoroughly pre-tested an overdose of these antibodies.

Unfortunately, there are not reliable dosing studies in the mouse models, so we do not know the optimal preclinical regime of dosing for cancer treatments. If an IgG4 tail is chosen for the humanized antibody, no adverse effects are anticipated beyond the biological effects on the immune response. We should be confident that the prevention of autoimmunity as observed in mice would be conserved in human beings.

Once pilot trials are finished and evaluated, the impetus should not fade away even in the absence of clinical efficacy or in the presence of a modest one. That would be the moment to try the best synergistic strategies defined in preclinical models, extending treatment to other diseases, that at some stage should include chronic viral infections resistant to conventional treatment.

We are hopeful that this road-map (Figure 8.2) will be followed in the near future. Preclinical data doubtlessly support clinical development.

Was this article helpful?

0 0

Post a comment