A Southern mystery

In 2004, strange things were happening when people living in the Southern United States received Erbitux, an anticancer drug. After Erbitux was approved, the first three patients that oncologist Bert O'Neil treated at the University of North Carolina, Chapel Hill, had severe anaphylactic reactions. "One fell out of their chair," passing out as blood pressure plummeted. "It alarmed us."

"I was quite upset," says research oncologist Christine Chung, when her patient with head and neck cancer had a severe reaction to the drug. "This was a young man and a last ditch effort" to gain a little more time for this patient, who could no longer take the drug due to the reaction. He later died.

There is always a risk of these types of reactions, says David Mauro, a cancer researcher with Bristol-Myers Squibb, which markets Erbitux (cetuximab) in partnership with ImClone. "That we saw [the reactions] with Erbitux was not a surprise." The overall incidence of severe reactions to infusions of the drug when it was launched was 3%; in several mid-Southern states, such as Tennessee, North Carolina, and Arkansas, however, the incidence was much higher - about 20%. A few people died from the reaction, Mauro says.

The company wasn't about to give up on what it deemed "a very effective anticancer agent," says Mauro. According to clinical trial data, Erbitux can extend survival in head and neck cancer and advanced colorectal cancer by several months.

Bristol-Myers Squibb checked batches of the drug and reviewed the infusion methods. Nothing. Conversations at an American Society of Clinical Oncology meeting led to a collaboration between O'Neil, Chung, and others to discover what was happening in the South. "We all asked around," Chung says, "and did the study when we had enough [serum] samples to prove our point that the patients in the South carry preexisting antibody to cetuximab."

Cetuximab is made in a mouse cell line, unlike most other monoclonal antibodies, which are derived in a Chinese hamster ovary cell line. But, patients weren't reacting to a mouse antigen. The researchers discovered that the mouse cell line expresses a sugar, galactose-α-1,3-galactose, which previous papers reported could cause rejection in organ transplants (Immunol Cell Biol, 83:674-86, 2005; Tissue Antigens, 68:459-66, 2006).

ImmunoCAP testing showed that more than 95% of IgE antibodies in serum from patients' blood, collected before infusion, became bound to the portion of the cetuximab heavy chain that carries galactose-α-1,3-galactose. "There is some cross-reacting antigen with very similar structure to a very small portion of cetuximab," Mauro says, and this cross-reaction likely underlies patients' symptoms.

Of 25 patients who showed a hypersensitivity reaction to cetuximab, 17 had IgE antibodies to the monoclonal antibody in blood collected before the drug was infused; this was true for just one of 51 patients who had no adverse reaction. Eight patients who did not test positive for the IgE antibodies also had less spectacular immune reactions (N Engl J Med, 358:1109-17, 2008).

In serum samples collected in Tennessee, 20% were positive for the IgE antibody, compared to 6% from California and 0.6% in Boston. What might be unique about the South, that makes people more likely to carry this antibody? Could it be an odd fungal infection, histoplasmosis, which is prevalent in the region? Tick bites? Some other endemic culprit?

"We've looked at every indigenous mold or spore, and so far we are not sure" why Southern Americans are so susceptible, Mauro says. O'Neil and Chung say they continue to use the drug in Southerners, but admit that other doctors may have stopped. For now, Bristol-Myers Squibb is working with Phadia, based in Uppsala, Sweden, to develop and commercialize a diagnostic test for the IgE antibodies that cross-react with Erbitux. Mauro predicts that the test will be available within a year.