July 24, 2012 — Navidea Biopharmaceuticals Inc., a specialty pharmaceutical company focused on precision diagnostic radiopharmaceuticals, announced that clinical data from a head-to head study comparing its amyloid imaging candidate, AZD4694, to the benchmark amyloid imaging agent, 11C-PiB (PiB), was presented at the Alzheimer's Association International Conference (AAIC) in Vancouver, Canada by Professor Christopher Rowe, MD, FRACP, Director of the Department of Nuclear Medicine and Centre for PET at Austin Health, Melbourne, Australia.
In addition, on Friday, July 13, 2012, Navidea received approval for its Phase 2 AZD4694 protocol from the New England Institutional Review Board (IRB), a centralized IRB that oversees the approval of clinical protocols for investigational drugs for multiple research organizations. Trial enrollment is expected to commence in the coming months.
Rowe’s talk was entitled, “Head to head comparison of 11C-PiB and 18F-AZD4694 for A? imaging in ageing and dementia.” The presentation highlighted new results from a blinded-reader analysis which examined imaging characteristics such as binding kinetics, standard uptake value ratios (SUVR) in three time intervals, and non-specific white-matter retention for these agents obtained in the same subjects. The two agents, 11C-PiB and 18F-AZD4694, were employed to sequentially image the brains of forty-five participants (25 healthy elderly controls, 10 subjects with Mild Cognitive Impairment, 7 subjects with probable Alzheimer’s Disease and 3 subjects with fronto-temporal dementia). The mean age of the participant groups was about 74, except that the three fronto-temporal dementia patients averaged about 68 years old. Cognitive impairments in the patient groups showed mean scores on the Mini-Mental State Examination of 24 to 27 and were within the mild to moderate range.
As previously reported, quantitative measures, such as SUVR, of 18F-AZD4694 binding to cortical amyloid plaques are essentially identical to 11C-PiB. A very tight performance correlation was observed (r=0.98, p<0.0001; slope 0.95). Visually, images obtained in the same patient with the same scan times, the same data processing and the same display scales, were identical. 18F-AZD4694 had comparable binding kinetics and dynamic range of SUVR to the benchmark 11C-PiB as well as similar high values of cortex to white matter ratios. Using the PiB as reference, the sensitivity, specificity, and overall accuracy within the 40 to 60 minute readings ranged from 94 percent to 98 percent. Additionally, confidence levels from the blinded readers averaged 93 percent using 40 to 60 minute readings, indicating a high degree of confidence in the interpretability of the images within this time range.