Gene therapy offers promising approach to tackle pancreatic cancer

Jul 10 2007

A molecularly engineered therapy selectively embeds a gene in pancreatic cancer that shrinks or eradicates tumors, inhibits metastasis, and prolongs survival with virtually no toxicity, researchers from The University of Texas M. D. Anderson Cancer Center report in the July 9 edition of Cancer Cell.

"This vehicle, or vector, is so targeted and robust in its cancer-specific expression that it can be used for therapy and perhaps for imaging," notes senior author Mien-Chie Hung, Ph.D., professor and chair of M. D. Anderson's Department of Molecular and Cellular Oncology.

The researchers call the system a versatile expression vector - nicknamed VISA. It includes a targeting agent, also called a promoter, two components that boost gene expression in the target tissue, and a payload - in this case a gene known to kill cancer cells. It's all packaged in a fatty ball called a liposome and delivered intravenously.

Researchers are working with M. D. Anderson clinicians to move the system, developed and tested in mouse models of pancreatic cancer, to a Phase I clinical trial.

"This looks like a promising approach to gene therapy for pancreatic cancer and we are working to bring it to a clinical trial," says James Abbruzzese, M.D., professor and chair of the M. D. Anderson Department of Gastrointestinal Oncology.

He estimates that it will take between one and two years to complete U.S. Food and Drug Administration requirements for a Phase I trial. Abbruzzese's clinicians are working with Hung's group to compile preclinical information for FDA review.

About 37,000 cases of pancreatic cancer are diagnosed annually in the United States. Early diagnosis is extremely difficult, so the disease is often discovered at a late stage after it already has spread, or metastasized. Fewer than 4 percent of pancreatic cancer patients survive five years after diagnosis, one of the lowest cancer survival rates.

In a test of the therapy against two aggressive lines of pancreatic cancer in two different types of mice, researchers loaded the VISA system with a mutant version of a gene named Bik, which expresses a protein that naturally forces cancer cells to kill themselves. The team created the more lethal mutant and named it BikDD.

Untreated control mice in both experiments all died within 40 days. Mice treated with the mutant gene delivered via a less-targeted viral promoter driven expression system employing cytomegalovirus (CMV) all died within 90 days, most much earlier. In both trials, the VISA-BikDD mice lived longer, with at least half surviving for 14 months with no detectable sign of cancer recurrence.

One test involved a pancreatic cancer line that spreads swiftly. In that case, live imaging showed that in control mice the cancer spread to the liver, spleen, kidneys, bladder, lungs, bone and intestines. Mice treated with the CMV BikDD showed only a few small tumors in nearby organs. There were no detectable metastases in mice treated with the VISA-BikDD combination, the authors report.

"All cancer drugs have some toxicity and so cause side effects, which affects the dose that can be administered," Hung says. The team looked for signs of acute systemic toxicity and of pancreatitis, a dangerous inflammation of the pancreas. While the cytomegalovirus-BikDD showed signs of both toxicities, the authors report "VISA-BikDD produces virtually no systemically acute toxicity or acute pancreatitis."

Abbruzzese says any clinical trial will advance under a National Cancer Institute Specialized Programs of Research Excellence (SPORE) grant in pancreatic cancer. SPORE awards are designed to translate scientific findings into the clinic.

"There are no good options for pancreatic cancer patients now," says Abbruzzese, "That's why we are trying new approaches such as this one as part of SPORE."

Hung's team set out to address two obstacles to gene therapy for pancreatic cancer: lack of a targeted gene expression system that hits the cancer hard but spares normal tissue, and the need for an animal model that allows monitoring of tumor growth and metastasis without having to kill and dissect the animal.

They solved the monitoring problem by modifying two lines of pancreatic cancer to stably express firefly luciferace enzyme - the bioluminescent agent that lights up fireflies. This allowed live imaging of the mice to monitor cancer progression.

The researchers already had developed several components of the gene expression system. Led by first author Xiaoming Xie, Ph.D., a postdoctoral fellow and now instructor in Hung's department, the team used those components and pieces developed elsewhere to painstakingly build what became C-VISA-BikDD.

They used a gene-based targeting promoter called CCKAR known to be active in pancreatic cancer but not in healthy pancreatic tissue. The problem with pancreatic-cancer-specific gene promoters such as CCKAR, Hung says, is that they select narrowly for cancer cells but are not highly active once they get there. The success of gene therapy depends on efficient gene expression after delivery.

A system using a viral promoter such as CMV has the opposite problem: strong activity but indiscriminate delivery of the therapeutic gene to healthy tissue, which increases toxicity. The team's goal was to retain CCKAR's selectivity while boosting its activity to levels seen in CMV promoter-driven therapy.

To strengthen activity, the team added two key components: a regulatory gene sequence known as WPRE and the two-step transcriptional amplification system (TSTA) known to amplify activity of the prostate-specific antigen (PSA) promoter.

This enhanced C-VISA package increased gene activity levels by 67.6 percent to 287 percent over the CMV-driven therapy in three lines of pancreatic cancer. Lab experiments showed gene activity was limited to pancreatic cancer cells and was minimal or absent from other cancers and three lines of normal pancreatic, lung and breast cells.

The next step was to load the vector with a weapon. The team packaged its mutant version of the Bik gene along with C-VISA in a liposome. For comparison, the liposomal gene also was attached to the CMV vector.

They tested both systems in the lab against all 13 available human cell lines of pancreatic cancer available worldwide. BikDD killed cells in all lines, with the VISA-delivered gene destroying more cells in all but one case. The C-VISA system spared healthy tissue while CMV-BikDD destroyed healthy cells. Subsequent tests compared the two methods in the mouse models.

Hung's team continues research on gene expression vehicle. "VISA is versatile enough that if you change the promoter, you could target other cancers or even other diseases," Hung says.

Research was funded by Abbruzzese's pancreatic cancer SPORE grant from the NCI, M. D. Anderson's NCI center support grant, a grant from the Topfer Research Funds, and an AstraZeneca Fellowship.

M. D. Anderson has filed for a patent on VISA-BikDD, which has been licensed to Alchemgen Therapeutics Inc. (ATI). ATI partially supports Hung's research. Hung and M. D. Anderson own stock in the company. These arrangements are managed by M. D. Anderson in accordance with its conflict of interest policies.

Co-authors of the Cancer Cell paper with Hung, Abbruzzese and Xie are: Weiya Xia, Hsu-Ping Kuo, Yuanfang Liu, Zheng Li, Qingqing Ding, Su Zhang, Bill Spohn, Yan Yang, Yongkun Wei, and Jing-Yu Lang, all of the Department of Molecular and Cellular Oncology; and Zhongkui Li, Douglas B. Evans, and Paul J. Chiao, all of the Department of Surgical Oncology. Kuo and Hung also are affiliated with the Graduate School of Biomedical Sciences, operated jointly by M. D. Anderson and The University of Texas Health Science Center at Houston.

http://www.mdanderson.org/