SAN DIEGO--Genetically engineered bacteria have the potential to deliver
anticancer genes directly to a tumor site, according to four presentations
of preclinical data at the American Association for Cancer Research (AACR)
The new technology, known as TAPET (tumor amplified protein expression
therapy), uses genetically engineered Salmonella bacteria that have
been attenuated for virulence and infectivity, while retaining sensitivity
to antibiotics. Salmonella bacteria were chosen because they multiply
rapidly, can be easily modified genetically, and have been found to grow
under both aerobic and anaerobic conditions, such as occur within solid
These engineered bacterial strains are highly selective for tumor tissue
and expand within the tumor to levels 1,000 to 10,000 times greater than
found in normal tissue. For example, in one of the scientific presentations
at the AACR meeting, a genetically engineered nonpathogenic strain of Salmonella
typhimurium injected into tumor-bearing mice rapidly migrated to the
tumor site and achieved tumor-to-liver ratios of approximately 10,000:1.
Furthermore, delivery of the modified bacteria, in this and other mouse
models, led to significant tumor reduction and prolonged survival.
Tumor localization has been shown using TAPET strains in six tumor models,
including melanoma, breast, lung, colon, renal, and liver cancer.
The TAPET research program was initiated at Yale University in 1993
by scientists David Bermudes, PhD, Brooks Low, PhD, and John Pawelek, PhD,
and is currently being developed by Vion Pharmaceuticals, Inc. of New Haven,
"These modified bacteria have lost their ability to cause disease
elsewhere in the body but seem to retain it in the tumor and are able to
grow to very high numbers in the tumor," Dr. Bermudes said in an interview.
Although the genetically engineered bacteria by themselves have the
ability to slow tumor growth, he said, they cannot eliminate the cancer,
so Vion is developing a portfolio of TAPET organisms to deliver prodrug-converting
enzymes and/or cytokines to tumors.
Dr. Bermudes described research presented at AACR involving a Salmonella
strain engineered to express herpes simplex virus thymidine kinase
(HSV TK), an enzyme that converts the prodrug ganciclovir (Cytovene) to
its toxic phosphorylated form. When introduced into melanoma-bearing mice,
HSV TK-expressing Salmonella showed ganciclovir-mediated, dose-dependent
suppression of tumor growth and prolonged survival.
In another AACR presentation, Ellen Carmichael, PhD, associate director
of biology at Vion, said that TAPET bacteria have been engineered to deliver
E coli cytosine deaminase (CD), an enzyme that converts the prodrug
5-fluorocytosine (5-FC) to the toxic 5-fluorouracil (5-FU).
Tumor cells from mice injected with the CD-expressing bacteria showed
substantial levels of CD activity, whereas cells from other organs, such
as liver, had no detectable CD activity.
Given alone, the modified bacteria led to tumor reduction. Studies are
now underway, Dr. Bermudes said, to determine whether co-injection of 5-FC
further increases the antitumor effect of the CD-expressing bacteria.
"By converting 5-FC to 5-FU directly in the tumor, rather than
administering 5-FU systemically, hopefully, we will be able to reduce the
toxicity of 5-FU to other parts of the body," he said.
Addresses Vector Problems
Dr. Bermudes pointed out that the TAPET technology addresses several
problems that may be associated with gene therapy of cancer involving viral
or liposomal vectors.
First, whereas most viral vectors are delivered locally and thus can
only treat tumors locally, TAPET vectors
would be delivered systemically and would have the potential to find and
destroy clinically undetectable areas of metastatic disease disseminated
throughout the body.
At the AACR meeting, Ivan King, PhD, senior director of biology at Vion,
showed that attenuated Salmonella typhimurium inhibited the growth
of subcutaneously implanted melanoma and lung tumors, as well as lung metastases
A second problem with gene therapy is the development of a host immune
response, which reduces the effectiveness of subsequent treatments. To
avoid this problem, Vion is developing TAPET vectors in multiple serotypes.
Third, TAPET has the potential to overcome a major hurdle in the gene
therapy of cancer, ie, the inability to deliver adequate numbers of gene
copies to cancer cells. This is because TAPET bacterial vectors have been
shown to distribute and amplify uniformly within the tumor.
At the AACR meeting, Li Mou Zheng, PhD, associate director of biology
at Vion, reported that attenuated strains of Salmonella targeted
and amplified within tumors in mice. When examined by light microscopy,
Dr. Zheng found that the Salmonella distributed homogeneously within
the tumor tissues from the periphery to the center, suggesting that TAPET
vectors could potentially deliver the therapeutic proteins throughout the
Fourth, Vion is testing TAPET vectors that may express more than one
gene, while viral vectors are generally limited to the expression of a
Finally, since TAPET vectors remain fully sensitive to antibiotics,
they could be cleared from the body in the presence of an antibiotic at
any time, allowing greater therapeutic control.
Phase I human clinical trials of the TAPET technology are expected to
begin in 1998, according to Vion. "In our preclinical data, we have
been able to show efficacy in melanoma, lung cancer, and colon cancer models,
and it seems reasonable that the first clinical efforts in humans will
be in these cancers," Dr. Bermudes said.