What we are striving for in cancer care today is personalized medicine. So, if a patient with newly diagnosed NSCLC has an activating epidermal growth factor receptor (EGFR) mutation, we give that patient Tarceva (erlotinib), a tyrosine kinase inhibitor (TKI). Right? Well, yes -- but it doesn’t always work (the response rate is in the 70-75% range). Why not? We’re not sure, but it would be nice to learn why we don't see near a 100% response rate among patients with EGFR mutations, so that we can know to recommend other alternatives. On the other hand, if a patient does not have the mutation, we don’t really expect much in the way of a dramatic response. Again, only partly right. In fact, some studies show a significant minority of patients with normal (“wild type”) EGFR will have at least some response to Iressa (gefitinib) or Tarceva. So, what’s going on here? A recent study out of China looked at blood rather than tumor in 84 patients with advanced NSCLC who were all treated with Iressa. The analysis looked at naturally occurring single-nucleotide polymorphisms (SNPs, and pronounced “snips”), in the EGFR gene. A polymorphism, which basically means “many forms”, can be thought of as small part of a gene which has a different part (nucleotide) from the general population, but is otherwise normal, as best we can tell. A minority of individuals will have the different nucleotide (polymorphism) while the general population will have the same gene but with a different nucleotide. In these patients, it was found that having a particular SNP could predict for a better response to Iressa and longer survival than the rest of the group, while different SNP could predict for a poorer response and shorter survival with the drug. Still another SNP could predict for a greater risk of developing a rash, which we know is also associated with a benefit from these TKIs. This has also been seen with Tarceva.

This study did, however, have some drawbacks, admitted by the authors. The study involved only Han Chinese, was retrospective, small (only 84 patients), did not list the subtypes of the tumors, and did not look for EGFR mutations in the tumors. However, the study was somewhat enriched because 57% were female, 82% were non-smokers, and 80% had adenocarcinomas; taken together, these characteristics in a group would be expected to get some benefit from Iressa. In fact, only 40% of patients had progressive disease while on the drug. We look forward to other such studies, hopefully in larger populations, and which would be prospective. But this is an interesting start, and you may hear more about SNPs in the future. In the meantime, this work begins to provide a hint about why some patients do better or worse than others, even with the same EGFR mutation status.