Oral Abstracts

ANGIOZYME, a synthetic ribozyme targeting the mRNA of VEGFR-1: clinical update

J. Wayne Cowens1, Vann P. Parker1, Roger Aitchison1, Gilad Gordon1, Patrice A. Lee1, Susan F. Radka1, T. Elise Jackson1, Michele Hurliman1, Nassim Usman1, Ernest C. Borden2, David E. Weng21Ribozyme Pharmaceuticals, Inc. Boulder, CO; 2Taussig Cancer Center, Cleveland Clinic Foundation, Cleveland, OH

ANGIOZYME was designed to test whether inhibition of angiogenesis through blockade of VEGF signaling can impact disease. ANGIOZYME is a synthetic ribozyme that selectively cleaves the messenger RNA for VEGFR-1. It contains chemically modified ribonucleotides, a small number of phosphorothioate linkages and an inverted sugar moiety which render it relatively resistant to the nucleases in biological fluids. ANGIOZYME has shown antitumor activity in five model systems in rodents {LLC-HM (lung), KM12L4a (colon), MCF-7 (breast), 4T1 (breast), ACHN (renal)}. This activity is manifested at doses of 30-600 mg/m2 through inhibition of angiogenesis, inhibition of the growth of the primary tumor and/or inhibition of the formation of metastases. Efficacy has been seen using IV, SC, and IP routes of administration. ANGIOZYME has been evaluated for toxicity in both rodents and primates with exposures up to 13 weeks. There was no effect on clinical, hematologic or biochemical parameters with the principal clinical finding of injection site reactions. In a single-dose Phase I study, ANGIOZYME was shown to have a bioavailability by subcutaneous injection of 90% in the 100 mg/m2 cohort. In a Phase I/II trial, 31 patients with metastatic cancer who had failed standard treatment received daily subcutaneous doses (30-300 mg/m2) of ANGIOZYME for up to 16 months. The principal toxicity was injection site erythema that occurred in most patients; these reactions were usually Grade 1, did not require treatment and did not lead to early terminations. Antibodies to ANGIOZYME were seen in four of five patients dosed longer than 100 days; however these antibodies were not associated with adverse events. Two patients (squamous cell cancer of the head and neck, melanoma) had minor responses. The pharmacokinetics were linear and did not show accumulation of ANGIOZYME with multiple dosing. Eight patients had tumor biopsies at least three weeks after initiation of treatment. Immunohistochemistry demonstrated the presence of ANGIOZYME in endothelial cells adjacent to the tumor. Phase II studies of ANGIOZYME (100mg/m2 subcutaneous daily) have been initiated in metastatic breast and colorectal cancer.

Pharmacokinetics in beagle dogs and cynomologus monkeys, of VNP20009, an attenuated Salmonella typhimurium for the treatment of cancer

C. Clairmont, J. Mao, L. Marshall, J. Pike, M. Sznol, J. DeGoes, K. Lee, I. King, J. Vodela1 and B. Almassian
Vion Pharmaceuticals, Inc. New Haven, CT; 1 Therimmune Reseach Corporation, Gaithersburg, MD

Background: VNP20009, a highly attenuated strain of Salmonella typhimurium, has been observed to preferentially accumulate in tumors following systemic administration, reaching tumor to normal tissue to ratios of > 1000:1 in mouse models. From data collected in phase I trials conducted in advanced cancer patients, VNP20009 is rapidly cleared from the blood following intravenous (IV) administration. One approach to optimizing the tissue colonization in species with rapid clearance, is to increase the exposure of tissues to VNP20009. Pharmacokinetic (PK) studies were conducted in both dogs and monkeys to determine the effect of various parameters: infusion time, steroids and dosing schedule on the systemic clearance of VNP20009.

Methods: VNP20009 was administered by IV injection as either a 30 minute, 2 hour or 4 hour infusion to cynomolgus monkeys or beagle dogs. Blood samples were taken and quantitatively cultured to determine real-time levels of VNP20009 measured as colony-forming units per ml (cfu/mL) of blood. Results: In both dogs and monkeys, VNP20009 reached high levels in the blood during infusion and rapidly cleared after the end of infusion. By extending the duration of the infusion to 2 or 4 hours, high levels of VNP20009 (104 to 105 cfu/mL) could be maintained over the duration of the infusion. In an effort to increase the dose level but maintain safety, the steroid dexamethasome was administered to counter the cytokine induction that follows a systemic dose of bacteria. Although steroids permit higher doses of VNP20009 to be given safely, the results demonstrated that steroids lower the peak VNP20009 blood levels. Administration of VNP20009 on d x 5 schedule in dogs demonstrated that daily doses at the MTD (3×108 cfu/kg) were tolerated and although the PK profile did not change, a larger overall dose can be safely administered without sacrificing peak blood levels.

Conclusion: Through the use of animal models with similar PK profiles to humans we evaluated various parameters in an effort to improve the PK profile of VNP20009. Our results indicate that multiple dosing and increased infusion time offer certain advantages (increased blood PK values, longer exposure) and should be further explored in the clinic.

A Phase 2 Trial To Evaluate the Rate of Immune Response Using Recombinant Idiotype for Treatment of Follicular Non-Hodgkin’s Lymphoma

immerman J1, Levy R1, Vose J2, Czerwinski D1, Ingolia D3, Denney D3, Kunkel L31Stanford University Medical Center, Stanford, CA; 2University of Nebraska Medical Center, Omaha, NE; 3Genitope Corporation, Redwood City, CA

Background: Tumor-specific variable regions of the clonal immunoglobulin (Idiotype or Id) expressed by B cell Non-Hodgkin’s Lymphoma (NHL) can be exploited as a target for active immunotherapy. Studies immunizing follicular NHL patients (pts) with Id protein derived from tumor-myeloma hybridomas have shown induction of specific cellular and humoral anti-Id immune responses (IR) that correlate with improved disease-free and overall survival. Tumor regressions and molecular complete remissions following immunization have also been observed. These results provide rationale for further study, however, isolation of Id by traditional hybridoma methods is inefficient and wide application is limited.

Objective: A phase 2 study was designed to assess the ability of recombinant Id produced by a novel expression technology, termed HiGET™, to elicit anti-Id IR in follicular NHL patients in 1st clinical remission. Methods: Patient specific Id genes are PCR amplified from small numbers of tumor B-cells, cloned into plasmid vectors, and transfected into mammalian cells that yield stable sources of recombinant Id protein. Conjugation to the carrier keyhole limpet hemocyanin (KLH) yields the final product, Id-KLH. Treatment consists of sub-cutaneous (s.c) immunizations of Id-KLH on day 1, along with GM-CSF (s.c. days 1-4). Immunizations are administered every 4 weeks for 4 doses, with a 5th immunization at week 24. Results: Twenty-one of 22 pts immunized are evaluable for humoral and 16 are evaluable for cellular anti-Id IR. Overall, 13 of 21 pts developed specific IR: 7 humoral, 2 cellular, and 4 both. IR were seen in both CR/CRU patients (9/13) and PR patients (4/8). The overall 62% IR rate compares favorably with the 50% rate seen in the previous studies using hybridoma-derived Id proteins. The most common side effects are mild and moderate injection site reactions and flu-like symptoms. This study demonstrates that recombinant Id immunotherapy can induce specific humoral and cellular IR. A phase 3 randomized trial utilizing recombinant Id in follicular NHL pts following 1st clinical remission is underway at 22 sites in North America to determine the efficacy in improving disease-free and overall survival.

Shikonin, a small molecule compound from Chinese herbal medicine with anti-inflammatory and anti-HIV-1 activity, is an inhibitor of multiple chemokine receptors

Xin Chen1, Lu Yang2, Jim Turpin2, Joost J. Oppenheim1 and O.M.Zack Howard11 Laboratory of Molecular Immunoregulation, Division of Basic Sciences, National Cancer Institute-Frederick Cancer Research and Development Center, Frederick, Maryland 21702-1201 and 2 Infectious Disease Research Department, Southern Research Institute, 431 Aviation Way, Frederick, MD 21701

Zicao, a commonly used traditional Chinese medicine containing shikonin, exhibits a variety of biological activities including inhibiting HIV-1 infection. Our previous data showed that shikonin selectively blocked chemokine ligands binding to CCR1 receptor. In this study, we further characterized the anti-chemokine function of shikonin. At non-cytotoxic concentration, shikonin inhibited monocyte chemotatic response to a variety of CC chemokines (MCP-1, MIP-1α, RANTES), CXC chemokine (SDF-1α ) and classic chemoattractants (fLMP and C5a) that use numerous receptors, in a dose-dependent fashion. The IC50 of RANTES-induced monocytes migration was under 10-7 M. This inhibition was not reversible. Monocyte calcium mobilization induced by RANTES, MCP-1 and C5a was attenuated by shikonin in a time-dependent fashion. Shikonin did not significantly influence CCR1 receptor cell surface expression on HEK/CCR1 cells as shown by FACS assay, but reduced CCR5 receptor expression on HEK/CCR5 cells in a dose-dependent manner with an IC50 of approximately 5´10-6 M. Furthermore, treatment by shikonin synergistically enhanced the reduction of CCR5 surface expression induced by RANTES. In a study of human PBMC infected with HIV-1 (ROJO), shikonin inhibited HIV reverse transcriptase acitivity in a dose-dependent fashion with an IC50 of 1.261´10-7 M. The activity was not based on a toxic effect as IC50 of human PBMC viability under the same conditions was 1.993´10-6 M. Furthermore, shikonin inhibits CCR5 antibody binding to the receptor. Interference with both HIV co-receptors and chemokine receptor signal transduction contribute to the anti-HIV activity of shikonin. Overall, these results suggest that shikonin is a pan-chemokine receptor inhibitor by blockade of chemokine receptor function and downstream signaling.

TRAIL Induced Apoptosis of Human Melanoma is Regulated by Smac/DIABLO Release from Mitochondria

Xu Dong Zhang, Xi Yi Zhang, Christian P. Gray, Tam Nguyen and Peter Hersey
Department of Oncology and Immunology, Newcastle Mater Misericordiae Hospital, Newcastle, N.S.W., Australia

Background: We have previously shown that the level of expression of death receptors for TRAIL, in particular TRAIL-R2 appeared an important determinant of TRAIL-induced apoptosis in melanoma cell lines. However, TRAIL induces low levels of apoptosis in some melanoma cell lines despite relative high levels of TRAIL-R2 expression on the cell surface. Objective: To elucidate the intracellular mechanisms that regulate the sensitivity of melanoma cells to TRAIL induced apoptosis.

Methods: We examined the multiple checkpoints in the TRAIL-initiated apoptotic signalling pathway by flow cytometry, western blot, immunofluorescence staining. c-DNA for XIAP, Bcl-2 and Smac/DIABLO were transfected into TRAIL-sensitive or TRAIL-resistant melanoma cells, respectively, to confirm their role in regulation of TRAIL-induced apoptosis. Results: We identified four melanoma lines with high TRAIL-R2 expression but low sensitivity to TRAIL induced apoptosis. The lines were shown to have similar levels of TRAIL-induced activated caspase-3 as the TRAIL sensitive lines but substrates downstream of caspase-3 (ICAD and PARP) were not degraded in the insensitive cell lines. This appeared to be due to inhibition of caspase-3 by XIAP in that XIAP was bound to activated caspase-3 and transfection of XIAP results in inhibition of TRAIL induced apoptosis. Reduction of XIAP levels by pre-treatment with Actinomycin D or by specific inhibition of XIAP by overexpression of Smac/DIABLO in the resistant melanoma cells was associated with the appearance of catalytic activity by caspase-3, increased TRAIL-induced apoptosis and reduced XIAP levels. After exposure to TRAIL, XIAP levels were markedly reduced in the TRAIL sensitive compared to resistant lines. This appeared to be associated with Smac/DIABLO release from mitochondria as this release was greater in TRAIL sensitive compared to resistant lines and had similar kinetics to downregulation of XIAP levels. Furthermore, overexpression of Bcl-2 inhibited Smac/DIABLO release and downregulation of XIAP levels. These results indicate that inhibition of caspase-3 by XIAP appears an important mechanism for protection of melanoma cells from TRAIL-induced apoptosis and provide evidence that TRAIL induced Smac/DIABLO release from mitochondria is a key regulator of TRAIL induced apoptosis of melanoma.

The proteasome inhibitor PS-341 sensitizes tumor cells to TRAIL-mediated lysis

TJ Sayers1, A. Brooks1, N. Seki2, A. Raziuddin1, W. Ma1, B. Blazar3, PJ Elliott4, and W. Murphy11SAIC, 2LEI -NCI Frederick, MD, 3 University of Minnesota and 4Millennium Pharmaceuticals

TRAIL has received much interest as a promising agent for cancer therapy, since it promotes apoptosis in many tumor cells, yet has little effect on most normal cells. The proteasome inhibitor PS-341 also induces apoptosis in many tumor cells, and its inhibition of NFκ-B may underlay some of its pro-apoptotic activities. Since TRAIL also activates NFκ -B in many cells, we combined PS-341 with TRAIL in an attempt to amplify the apoptotic effects of TRAIL. In overnight (18h) assays combinations of PS-341 and TRAIL were much more effective than either agent alone in promoting apoptosis of a murine myeloid leukemia C1498, a murine renal cancer RENCA and a human breast cancer MDA-341. Optimal doses of PS-341 were in the range of 1-100nM and the optimal amounts of TRAIL also varied depending on the individual cell line. Apoptosis in all tumor cell lines was blocked by the general caspase inhibitor ZVAD-FMK. Western blotting analysis showed that doses of PS-341 which sensitized C1498 to TRAIL-mediated lysis also significantly decreased levels of the anti-apoptotic protein cFLIP, whereas levels of bcl-2 and bax were unaffected. In a bone marrow purging model , lethally irradiated B6 mice received bone marrow (3X106 cells) containing 10,000 C1498 cells, or bone marrow plus C1498 cells which had been treated overnight with PS-341 or TRAIL. Treatment of bone marrow cells (BMC) overnight with 10nM PS-341 did not result in significant myelosuppression, as assessed by CFU-c colony formation in soft agar in vitro or hematopoetic reconstitution experiments in vivo. Mice receiving untreated or TRAIL- treated bone marrow plus C1498 all succumbed within 25 days due to extensive tumor growth, whereas treatment with PS-341 alone significantly increased survival times. These data suggest that PS-341 exerts direct antitumor effects in vivo, and can possibly be used in combination with TRAIL to enhance TRAIL-mediated apoptosis effects on C1498 tumor cells without significant toxicity on normal bone marrow cells. Current studies are examining whether the combination of PS-341 with TRAIL may result in greater antitumor effects in vivo.