Dose Finding in Human Trials of TLR9 Agonists: Induction of Interferon-α Inducible Genes in Blood Mononuclear Cells as a Measure of Biologic Activity of 1018 ISS.

In vitro studies of CpG oligodeoxynucleotide agonists of TLR 9, such as 1018 ISS, demonstrate a reproducible bell-shaped distribution of activity versus concentration for interferon-a production. Activity increased with increasing concentration to a maximum value, but rapidly declined at higher concentrations. A robust method of assessing the dose response of 1018 ISS in human trials is measurement of the induction of interferon-α inducible genes (IIG) in blood mononuclear cells obtained pre- and ∼ 24 hours post-administration using semi-quantitative RT-PCR. Expression levels were normalized to the housekeeping gene ubiquitin and the ratios of IIG expression pre and post 1018 ISS administration were calculated. Four small clinical trials have been performed in patients with several malignancies; B-cell lymphoma, colorectal cancer, sarcoma and glioblastoma. Three of the four involved dose escalation in an attempt to define an optimal subcutaneous dose. All trials involved coadministration of other therapy. The earliest sampling of the three IIG assessed in each of the 4 trials was studied for this abstract. Larger than 3 fold increases of IIG expression over predose levels was scored as elevated for that particular gene. >1/3 elevated IIG in an individual subject was scored as positive stimulation. The geometric mean of the ratios (GMR) of all three IIG was used as a composite value for response intensity in a subject. Dose finding studies went from 0.01 to 0.5 mg/kg in the earliest trial and 0.01–1.3 mg/kg in the later trials. A Phase 2 lymphoma trial administered a dose of 0.2 mg/kg to 23 follicular lymphoma subjects.

Results: A trial which administered GM-CSF into the same subcutaneous site showed much greater GMR at all doses than the other 3 studies and is considered separately. The other studies showed a dose dependent increase, both in GMR and fraction of positive subjects, from the lowest dose up to 0.5 mg/kg. (Table 1) Interestingly, only 1/2 subjects at the highest dose was positive and both had low GMR comparable to the lowest dose.

Conclusions: A bell-shaped dose response in both frequency of stimulated subjects and the intensity of stimulation of IIG is observed. If the biomarkers are correlated with the clinical activity of 1018 ISS, a narrow range of dose will produce optimal results and careful choice of dose is important. Further exploration of the upper end of the range is necessary to confirm whether the decline observed in 2 subjects is real. Coadministration of GM-CSF may augment the biological effect of 1018 ISS. Certain subjects remain refractory to stimulation.