Carcinoembryonic antigen (CEA) was one of the first oncoproteins discovered a half century ago. The protein is expressed in a subset of most epithelial cancers, including non-small-cell lung cancer, colorectal, pancreas, breast, gastric and medullary thyroid cancers. Globeimmune Inc. estimates that CEA is overexpressed in approximately 500,000 new cancer cases in the U.S. each year.

"It's in close to 100 percent of colon cancers and somewhat lower percentages of the other [epithelial cancers]," Timothy Rodell, Globeimmune's president and CEO, told BioWorld Insight.

Beyond being expressed in tumors – certainly a necessity for a tumor target – CEA has some other characteristics that make for a good tumor target. As the name implies, CEA is expressed during embryonic development, but the protein is hardly expressed in normal adult tissue, making it easy for drugs that target it to distinguish between tumor cells and normal tissue.

In fact, in 1996, Immunomedics Inc., of Morris Plains, N.J., gained FDA approval to market CEA-Scan as a radiographic marker for CEA-positive tumors. CEA-Scan is a monoclonal antibody fragment labeled with technetium-99m that can be read with a computed tomography scan. (See BioWorld Today, July 2, 1996.)

And some clinical trials have used serum CEA as a marker for tumor burden since some of the protein is released into the bloodstream.

Unlike some other markers that are found on tumor cells, CEA actually has a hand in tumorogenesis – it's believed to promote cell adhesion, which is important for metastasis, and has anti-apoptosis activity – so tumors should have a hard time becoming resistant to drugs simply by deleting the CEA protein.

FIRST ATTEMPTS

"Cancer vaccines have a checkered history," Rodell said, noting that some of the platforms haven't been particularly potent, some were tested in later-stage patients with large bulky tumors that may not be cleared easily by immune cells, and others stimulated antibody production, which can't get inside tumor cells where some antigens reside.

Some or all of those issues have doomed previous CEA-targeting immunotherapies to failure.

In 2002, South San Francisco-based Titan Pharmaceuticals Inc.'s Ceavac failed to improve survival in metastatic colorectal cancer patients compared to placebo. Ceavac was a monoclonal antibody designed to stimulate an immune response toward CEA-expressing cells. (See BioWorld Today, Dec. 12, 2002.)

Likewise, Panvac-VF, developed by now-defunct Therion Biologics Corp., failed to improve overall survival compared to palliative chemotherapy in patients with pancreatic cancer that had failed first-line treatment. Panvac-VF was a live virus vaccine based on a poxvirus engineered to contain the gene for human CEA protein. (See BioWorld Today, June 29, 2006.)

BETTER TARGETING

Immunotherapy platforms have gotten better over the last few years, spurring some companies to dust off CEA as a potential target.

Globeimmune, of Louisville, Colo., developed a CEA-targeting immunotherapy called GI-6207 using the company's Tarmogen platform.

A Tarmogen contains a heat-inactivated yeast expressing a target protein. Once inside the patient, dendritic cells engulf the Tarmogen. The proteins, including the cancer target, are chopped up and presented to T cells. Those activated T cells then search for other cells expressing the antigens presented by the dendritic cells.

In a phase I trial in CEA-positive advanced malignancies, five patients saw their disease stabilize. "Interestingly, in all five of those, their CEA levels stabilized or came down as well, so you've got a marker for what's going on in the tumor," Rodell said.

A phase II trial in patients with medullary thyroid cancer (MTC) being run at the National Cancer Institute is ongoing, with data expected in the second half of 2016. Patients will either be treated with GI-6207 for up to two years or observed for six months and then administered GI-6207 for up to two years. The trial will measure changes in calcitonin levels, a tumor marker that correlates with tumor size in MTC, after six months.

Rodell said MTC was chosen rather than the other CEA-expressing cancers because "there really isn't much out there" for patients with asymptomatic metastatic MTC. Cometriq (cabozantinib, Exelixis Inc.) and Caprelsa (vandetanib, Astrazeneca plc) are both approved for advanced MTC, but, given their toxicity, the drugs generally aren't prescribed until late in progression of the disease.

Last week, Celgene Corp. exercised its option to license GI-6207. Globeimmune will receive a $1.9 million option exercise payment and is eligible for regulatory and sales milestones. If the drug is approved, Globeimmune will receive royalties on worldwide sales of GI-6207. (See BioWorld Today, May 18, 2009.)

As part of the companies' 2009 collaboration and option agreement, Summit, N.J.-based Celgene had until after data from the phase II trial in MTC were available to make a decision about whether to license GI-6207, but decided to license the drug early. Brian Gill, Celgene's vice president of global corporate communications, declined to comment on what motivated the company to invoke its option earlier than required.

Celgene's experience with the platform could have had something to do with the decision to license the drug early. As part of the partnership, Celgene already has rights to a drug developed by Globeimmune, a Tarmogen called GI-6301 targeted at brachyury protein.

Roche AG also has a CEA-targeted drug in the clinic, RG7813. The molecule is a immunocytokine combining an engineered cytokine (IL2v) with an antibody against CEA.

Basel, Switzerland-based Roche is testing RG7813 in phase I trials both as a monotherapy and in combination with atezolizumab, its anti-PD-L1 immunotherapy designed to interrupt the PD-1/PD-L1 interaction that tumors use to signal to immune cells that they shouldn't be attacked.

CAR T is another obvious immuno-oncology platform where targeting CEA might be useful, although the technology hasn't proved as efficacious in solid tumors as it has in blood cancers.

A search of clinicaltrials.gov revealed a phase I trial being run by Southwest Hospital in China testing Anti-CEA-CAR-T in patients with relapsed or refractory CEA-positive lung cancer, pancreatic cancer, gastric cancer, breast cancer and colorectal cancer.