Science Editor

The basic link between chronic inflammation and cancer is well known. But most of its details, as well as when immune activation is helpful and when it makes things worse in cancer, are not. Data presented at the Annual Meeting of the Association for Cancer Research this week, as well as a recent paper in Science, provided new insights into those links.

The usual estimate is that 20 percent of all cancer deaths occur in the context of inflammation and infection. But Michael Karin, professor of pharmacology at the University of California, San Diego, told the audience at Sunday's plenary session that "this is a gross underestimate." Karin's own estimate is that inflammation plays a direct role in at least 70 percent to 80 percent of cancers, and that "probably all solid malignancies are somehow connected to inflammation."

In his talk, Karin explored several facets of the connection between cancer and inflammation, focusing on the details if NF-kappaB is involved. NF-kappaB is probably the strongest transcriptional activator in mammalian cells when it is activated by infectious or inflammatory stimuli, via the kinase complex IKK.

IKK has two subunits, IKK-alpha and -beta, respectively. Karin said that "in most cases, activation of NF-kappaB is IKK-beta dependent," and spent the early part of his lecture showing data that in mice, IKK-beta is important for the development of colitis-associated cancer, and knocking it out can "completely annihilate" the cancer-promoting effects of cigarette smoke.

But when Karin's group knocked out IKK-beta in prostate epithelial cells in a mouse model of prostate cancer, "much to our disappointment, this had no effect whatsoever" on initial tumor development, progression or metastases. "Nothing that we looked at was impacted by the deletion of IKK-beta in prostate epithelial cells," he said.

Karin called the findings "quite disappointing, as well as surprising" - most prostate cancers, as they advance, have overactive NF-kappaB, and the assumption had been that blocking this activity would be a good strategy against the cancer.

But when Karin's group decided to inactivate IKK-alpha instead, that manipulation had "a major effect" on metastases, though not on development of primary cancer. Apparently, IKK-alpha affects metastasis development via the protein maspin in prostate cancer.

IKK-alpha also appears to promote metastases in breast cancer. In studies briefly presented by Karin in his plenary talk, as well as more elaborately during a poster session on Sunday, Karin and his colleagues showed that the repression of maspin via RANK-ligand, or RANKL, a protein better know for causing bone loss, also promoted lung metastases in mouse models of breast cancer.

A recent paper, in the April 10, 2009, issue of Science, came to another conclusion about NF-kappaB: that rather than trying to block components of the pathway, it might be a better approach to manipulate waves of the protein that apparently come and go in cells.

When NF-kappaB is activated in cells, it moves from the cytoplasm to the nucleus.

But such activation does not occur in the form of a smooth increase. Instead, stimulation tends to come in waves - and, in turn, cause waves of NF-kappaB. In their paper, the authors manipulated the length of time between those waves by pulsing cells with tumor necrosis factor alpha or TNF-alpha, and found that different intervals between stimulation led to different patterns of gene activation by NF-kappaB.

For example, while a single five-minute pulse of TNF-alpha activated the chemokine RANTES, the same amount of TNF-alpha given via pulses interspersed at 200-minute intervals failed to do so.

The authors concluded that the timing of stimulation is likely to be functionally important, and could be targeted therapeutically - findings that, they said in a press release announcing the findings, "both aid and challenge drug designers."

Also in AACR news:

• Abraxis BioScience Inc., of Los Angeles, presented 10 abstracts about its breast cancer drug Abraxane (paclitaxel protein-bound particles) and its pipeline of nanoparticle albumin-bound products. The Abraxane studies evaluated the drug in combination with anti-VEGF agents and explored the angiogenic role of the SPARC biomarker, which has been associated with a positive Abraxane response. Other posters focused on nanoparticle albumin-bound rapamycin and various combination approaches.

• Array BioPharma Inc., of Boulder, Colo., presented data showing that its KSP inhibitor ARRY-520 achieved a 100 percent response rate in hematological cancer models. A separate study showed that tumors relying on MCL-1 are most likely to respond to ARRY-520. Array also presented data on checkpoint kinase 1 (Chk1) inhibitors, a new class of drugs that may bolster the efficacy of chemotherapy.

• AVEO Pharmaceuticals Inc., of Cambridge, Mass., presented preclinical data showing that the triple VEGF receptor inhibitor, AV-951, was active in models of hepatocellular carcinoma. The drug is in Phase II for renal cell carcinoma and Phase Ib for several other types of cancer.

• Avila Therapeutics, of Waltham, Mass., presented preclinical studies showing that the Btk inhibitor AVL-291 and pan-ErbB inhibitor CNX-222 completely and selectively silenced their targets. Other data showed that Avila's covalent technology can be used to measure the silenced proteins, which may serve as biomarkers to guide treatment.

• BioAlliance Pharma SA, of Paris, said preclinical results of its oral irinotecan nanoparticle formulation in advanced colorectal cancer showed that the product was well tolerated when administered to animals. In animals bearing colorectal tumors, it was shown to induce a similar tumor growth inhibition to the one observed with intravenously administered irinotecan, with a nearly 80 percent tumor growth inhibition.

• Calistoga Pharmaceuticals Inc., of Seattle, presented data on its p110 delta selective PI3K inhibitor CAL-101 and its p110 beta selective PI3K inhibitor CAL-120. Preclinical studies showed the efficacy of CAL-101 in hematological malignancies and CAL-120 in solid tumors.

• Celator Pharmaceuticals Inc., of Princeton, N.J., presented data on its liposomal formulations of cytarabine/daunorubicin (CPX-351) and irinotecan/floxuridine (CPX-1), both of which are in Phase II. Preclinical studies showed that CPX-351 improved survival compared to its components in leukemia models, while CPX-1 demonstrated synergy with Avastin (bevacizumab, F. Hoffmann-La Roche Ltd.) and Erbitux (cetuximab, Eli Lilly and Co.) in colorectal models.

• Curis Inc., of Cambridge, Mass., presented preclinical data showing CUDC-305, a novel Hsp90 inhibitor, demonstrated potent antitumor activity in preclinical non-small cell lung cancer models, including those containing mutations conferring resistance to marketed drugs. It also demonstrated tumor regression in breast, colorectal and glioblastoma preclinical cancer models as a single agent or in combination with other drugs or standard chemotherapeutics. The compound exhibited significant brain penetrability and also significantly enhanced survival in preclinical intracranial tumor models. Another abstract demonstrated that CUDC-305 induced tumor regression in a subcutaneous xenograft model for glioblastoma.

• Cytokinetics Inc., of South San Francisco, presented data showing that in vitro exposures of GSK-923295 were predictive of in vivo efficacy in a murine xenograft model. In addition, GSK-923295 accumulated in xenograft tumor tissue, which may enhance antitumor activity in that species. Another abstract concluded that GSK-923295 inhibits cell growth and induces cell apoptosis in sensitive breast cancer cells. In addition, basal B subtype breast cancer cell lines are more sensitive than luminal subtype or nontransformed human mammary epithelial cultures.

• Esperance Pharmaceuticals, of Baton Rouge, La., presented positive results from the a preclinical program for EP-100, the first candidate from the company's Cationic Lytic Peptide platform technology. EP-100 is a targeted membrane-disrupting peptide designed to selectively target luteinizing hormone-releasing hormone receptors. Results demonstrated that EP-100 was fast acting, destroying LHRH receptor overexpressing OVCAR-3 cells within an hour compared to the unconjugated CLIP 71, which was less cytotoxic and was slow acting (p < 0.005). A Phase I trial in patients with solid cancerous tumors is planned for later this year,

• Geron Corp., of Menlo Park, Calif., said imetelstat (GRN163L), now in six clinical trials examining the safety, pharmacokinetics and pharmacodynamics of the drug, alone or in combination with standard therapies, in solid tumors, chronic lymphoproliferative diseases, multiple myeloma, lung and breast cancers, continued to show telomerase inhibition not only in the bulk tumor fraction of patients' bone marrow, but also in the stem cell-containing fraction. Imetelstat is a short-chain oligonucleotide that binds with high affinity and specificity to the catalytic site of telomerase, resulting in competitive inhibition of enzyme activity.

• Lorus Therapeutics Inc., of Toronto, presented positive data for its lead clinical drug candidate LOR-2040, formerly known as GTI-2040, combined with Ara-C in acute myeloid leukemia. The results of the pharmacodynamic studies showed that R2 levels in leukemia cells were decreased following treatment with LOR-2040 alone or in combination with Ara-C that resulted in a significant decrease in levels of intracellular dNTPs. Analysis of the bone marrow samples from acute myelogenous leukemia patients treated with LOR-2040 and high-dose Ara-C showed reduced levels of dNTPs had higher amounts of Ara-CTP, which is the cytotoxic product of Ara-C.

• Micromet Inc., of Bethesda, Md., presented data showing that its BiTE antibody MT110 can eliminate cancer stem cells. Data from a study of MT110 in mice showed that the BiTE antibody eliminated even large numbers of implanted cancer stem cells, resulting in a complete inhibition of tumor growth and survival, while untreated animals with implanted cancer stem cells experienced tumor formation and death. In a separate presentation, nonclinical data that indicated that Micromet developed BiTE antibodies from the EGFR-specific monoclonal antibodies Erbitux (cetuximab) and Vectibix (panitumumab) were highly active against KRAS- and BRAF-mutated human colorectal cancer cell lines. The data showed that the company's BiTE technology can overcome significant limitations of conventional monoclonal antibodies, the company said.

• Nektar Therapeutics, of San Carlos, Calif., presented data demonstrating that NKTR-105, a novel PEGylated form of docetaxel, has superior antitumor activity, as well as greater and sustained exposure in plasma and tumor tissue, over docetaxel in preclinical models. In a comparative study of NKTR-105 and docetaxel, NKTR-105 treatment significantly delayed tumor growth as compared to docetaxel (122 percent vs. 48 percent), which was associated with greater and sustained tumor exposure to docetaxel after NKTR-105 administration. Treatment with NKTR-105 increased docetaxel concentrations in tumors (0.4 to fourfold higher) and resulted in an improved time-concentration profile (twofold greater) as compared to docetaxel in a non-small-cell lung (H460) mouse xenograft model.

• Onconova Therapeutics Inc., of Newtown, Pa., said Phase I data from more than 125 patients with solid tumors treated with ON 01910 in eight different clinical protocols indicated a good tolerability profile. Onconova is conducting additional single-agent trials and new combination therapy clinical trials of ON 01910. The compound has shown broad-spectrum antitumor activity against both solid tumors and hematological malignancies and in nonclinical studies has demonstrated synergistic activity when combined with several classes of conventional chemotherapeutic agents.

• Oxigene Inc., of Waltham, Mass., reported data showing that OXi4503 has pronounced antitumor activity when administered as a single agent or in combination with chemotherapy in acute myelogenous leukemia xenograft models. Results showed that the antitumor activity of OXi4503 against HL60 xenografts was superior to that observed for fosbretabulin, and that HL60 AML xenografts are sensitive to OXi4503 administered as a single agent. In addition, the combination of OXi4503 with araC increased antitumor activity with no apparent increase in toxicity.