Saliva maintains oral health. Building on our past studies of saliva formation and its alteration during pathology, we previously developed novel approaches to treat salivary dysfunction using principles of gene therapy, as well as strategies to use normal salivary glands as a gene transfer target site for treating systemic single protein deficiency disorders (SSPDDs). The treatment of most head and neck cancer patients includes irradiation (IR). Salivary glands in the IR field suffer irreversible damage. Much of our effort this past year focused on a clinical gene therapy trial to restore salivary function to existing IR-damaged glands. Based on previous studies with a recombinant serotype 5 adenoviral (Ad5) vector encoding human aquaporin-1 (hAQP1), AdhAQP1, conducted in rats and miniature pigs, and extensive safety studies of the vector, we received an IND (BB-IND 13102) and approval to conduct a phase I clinical trial testing this vector in patients. The protocol for the trial, Open-label, dose-escalation study evaluating the safety of a single administration of an adenoviral vector encoding human aquaporin-1 to one parotid salivary gland in individuals with irradiation-induced parotid salivary hypofunction, received all required approvals and we began treating patients in 2008. We have completed the first two dose cohorts (4.8x10e7 and 2.9x10e8 vector genomes, vg, to a single parotid gland;3 patients/dose cohort), and treated one patient in the third dose cohort (1.3x10e9 vg/gland). All patients tolerated the vector and the related study procedures well. The first patient in the third dose cohort, however, was the only patient thus far to show an inflammatory response to vector delivery, i.e., developed a mild parotitis that resolved without medical intervention. This observation may indicate that we have approached an Ad5 vector dose to the parotid at which benefits are outweighed by side effects. A practical concern with using AdhAQP1 is that a conventional Ad5 vector unlikely will direct hAQP1 expression for more than 2-4 weeks. To develop a long-term strategy for hAQP1 gene transfer, we previously began to test the use of serotype 2, adeno-associated viral (AAV2) vectors in rodents, macaques and miniature pigs. This year we completed a study in irradiated miniature pigs using the AAV2hAQP1 vector. The study goal was to see if AAV2-mediated hAQP1 gene transfer would extend the restored salivary flow in these animals beyond that seen with AdhAQP1. Sixteen weeks after IR (20 Gy) salivary flow was decreased by 85-90%. AAV2hAQP1 administration at week 17 transduced only duct cells and resulted in a dose-dependent increase in parotid salivary flow to 35-40% of pre-IR levels after 8 weeks. Administration of either a control AAV2 vector or saline was without benefit, and led to further decreases in salivary flow. Little change was observed in clinical chemistry and hematology values after AAV2hAQP1 delivery. The findings suggest that delivery of AAV2hAQP1 to IR-damaged parotid glands may be useful in providing extended restoration of saliva output in previously irradiated patients with salivary hypofunction. We also evaluated the effect of IR on microvascular endothelial (MVE) cells in miniature pig parotid glands. Previously we showed in mice that damage to MVE cells in salivary glands occurs quite early after single dose IR. Similarly, after a single IR dose (25Gy) to parotid glands of 6 miniature pigs, local parotid gland blood flow rate decreased rapidly and remained below control levels throughout the 14 day study. Parotid microvascular density declined from 4-24 hours, and remained below control levels thereafter. The activity of both acid and neutral sphingomyelinase in parotid glands increased rapidly 4-24 hours post-IR, and then declined gradually. Also, the frequency of detecting apoptotic endothelial cell nuclei in glands followed comparable kinetics. Thus, similar to mice, miniature pigs also show marked damage to salivary gland MVE cells soon after IR. As described in many past annual reports, we have shown in multiple animal models (mice, rats, miniature pigs and rhesus macaques) that salivary glands are a potentially useful gene transfer target site for treating certain SSPDDs. This year we have extended these studies with two quite novel potential clinical applications. The first involves hypoparathyroidism. Two days after delivering an Ad5 vector encoding human parathyroid hormone, Ad.hPTH, to rat parotid glands, most secreted transgenic hPTH was detected in serum. Apparently, no vector escaped from the target tissue into the circulation, as QPCR biodistribution assays revealed vector copies only in parotid glands and not in liver or spleen. To show potential clinical applicability, Ad.hPTH (10e11 vp/gland) was administered to the parotid glands of parathyroidectomized (PTX) rats. Two days after transduction, high levels of hPTH (39.9 +/- 12.6 pg/ml) were found in serum, but no hPTH was detectable in saliva. Importantly, Ad.hPTH treatment led to normalization of serum calcium levels in the PTX rats and a significant increase in their urinary phosphorus/creatinine ratio, providing clear evidence of PTH action due to the parotid gland-produced transgenic hPTH. The second application studied involves diabetes mellitus. Glucagon-like peptide 1 (GLP-1) is a 37 amino acid peptide that is released into the circulation after a meal and acts as an incretin. Its biological half-life is quite short, 2-3 min, and it is inactivated by dipeptidyl-aminopeptidase IV (DPP IV) in the blood. An Ad5 vector encoding GLP-1 (Ad-GLP-1) was generated and initially shown able to direct biologically active GLP-1 production in vitro. The transgenic GLP-1 also was more resistant to degradation by DPP IV, due to an engineered Ala to Gly substitution at position 8. In vivo studies demonstrated that mice given the Ad-GLP-1 vector in their submandibular glands had serum levels of GLP-1 three-fold higher than those of mice transduced with a control Ad5 vector. In healthy fasted animals, serum glucose levels were similar between mice treated with Ad-GLP-1 and mice treated with a control vector. However, when challenged with a glucose tolerance test, Ad-GLP-1 treated healthy mice lowered their serum glucose levels significantly faster than the control mice. Next, we used a mouse model of drug-induced diabetes mellitus, administering alloxan, which destroys pancreatic beta cells. We found that the progression of hyperglycemia was significantly slowed in mice pre-treated with Ad-GLP-1 in their submandibular glands compared to mice pre-treated with the control vector.
Showing the most recent 10 out of 27 publications
