Erbitux (cetuximab) is a monoclonal antibody to EGFR, and it's actually made from a protein that is part mouse and part human (called a chimeric protein, named for the mythologic creature chimera that was composed of multiple parts from different animals). It's uncommon but not rare for patients to have an allergic reaction to this protein, and in most large national and international studies show rates of hypersensitivity reactions (HSRs) in the 1-3% range. In severe cases, these reactions can be very serious, causing low blood pressure, fainting, wheezing, and shortness of breath; they can even be fatal. While that is a small but real risk, wht is fascinating and especially scary is that there is an area in the southwest US -- including Tennessee, the Carolinas, northern Georgia, and perhaps other areas -- in which about 20-25% of patients develop these reactions.

Oncologists in these areas had noted over the past few years that they seemed to have a higher than expected rate of these complications, but the problem was highlighted in real terms in a 2007 article in the Journal of Clinical Oncology (abstract here) that retrospectively reviewed the experience of 88 patients on clinical trials and 55 patients off of clinical trials who received erbitux at one of three institutions in the region: Sarah Cannon Cancer Center and Vanderbilt-Ingram Cancer Center in Nashville, TN, and the University of North Carolina at Chapel Hill. They found a 22% rate of moderate to severe HSRs, ten times what we see in the rest of the country or world. These reactions occurred at the time of the first infusion but rarely afterward if someone did well the first time. They were more commonly seen in patients with a history of allergies, and it was interesting to see that reactions were more commonly seen in patients with NSCLC than other tumor types. The association of this type of reaction with lung cancer more than colon cancer is puzzling and suggests that there may be some correlation with a tobacco-related antigen, but we still need to learn more about this.

Another report was published in the New England Journal of Medicine (abstract here) in which the investigators looked at the serum of multiple patients who had received erbitux and had reactions and compared their serum proteins to the findings in patients who didn't have reactions. They also assessed the serum proteins of other patients who hadn't received erbitux but lived in the same area of the southeastern US, and also to blood from cancer patients or normal subjects from Boston and California. Among the 25 of 76 erbitux recipients who developed an HSR, 17 were found to have a specific allergy-related protein (immunoglobulin E, or IgE, to a specific antigen, which is an immune reaction inducing snip of protein)) in their blood BEFORE starting treatment. This particular IgE was seen in only 1 of 51 patients who didn't develop a reaction. Interestingly, this IgE protein was seen in 21% of the other people from the same area, but it was only present in 1% of the samples from Boston and 6% of the serum samples from California. The current conclusion is that this reaction is related to some plant antigen in the southeast that leads to a pre-existing allergy to erbitux.

How serious is this? I recently spoke about it today with two colleagues from that part of the country -- one in Memphis and another at Duke. My colleague from Duke said that they simply do not use erbitux there, at least not for lung cancer, since the perceived risk is so high and the benefit is felt to be so modest. The oncologist from Memphis said that severe reactions are so common that the doctors and nurses at their cancer center have become experts at running "codes", the emergency events where patients become unresponsive, lose a measurable blood pressure and pulse, etc. -- so that they're now more experienced than the folks in the ICU.

As we consider whether erbitux has a place in lung cancer, it's important to keep in mind that, like avastin, there is the potential for serious and even fatal side effects, and in some regions the risk may be unacceptably high. While the identification of a marker leads us to the possibility of a screening test being developed to identify the patients at high risk for a serious reaction, at the present time we may well develop very different regional practice patterns based on distinct differences in the risk/benefit balance with erbitux.