Infused throughout the website is a constant recognition that “patients are different”, but while we know this intuitively, we’re really not moved to a point of individualizing treatment on the basis of this.  There are many lines of clinical research that are moving in that direction, and one of the key elements is pharmacogenomics, the study of the genetic underpinnings of the differences among people in response to a medication, both in terms of response and side effects.  The slides for this post are stolen appropriated from Dr. David Gandara, one of the true leaders in the field of lung cancer, who heads the Lung Cancer Program at the University of California at Davis, as well as the Lung Cancer Committee for the Southwest Oncology Group (SWOG).  He is also the new President Elect of the International Association for the Study of Lung Cancer.  He and his collaborators have been adding to our understanding of this field as it relates to lung cancer.

   Whether we’re talking about conventional chemotherapy, EGFR inhibitors, or just narcotics, it’s clear that there are major differences in how different patients respond.   Although we’re on the cusp of tailoring treatments based on type of NSCLC (squamous vs. adeno, or BAC, for instance), and smoking status, and perhaps even racial background, our general practice has historically been to use a one size fits all approach:

(Click on image to enlarge)

In fact, though, we know that some patients have a very nice response to chemo, or a targeted therapy like tarceva, while others have a minimal response or actually progress on treatment.  We also know that some people tolerate chemo very well, while others terrible side effect problems.  We could really improve the field if we could identify, using molecular profits, the best vs. the worst responders and the patients who were destined to develop terrible treatment side effects before we actually give the therapy:

By doing that, we could separate out those patients who would have a poor response or bad toxicity and give them a different treatment instead:

    At the present time, we routinely use squamous vs. adenocarcinoma histology, as well as smoking status, for clinical decision-making, but these are clinical and not molecular profiles.  We have several examples of molecular profiles that are very promising and are beginning to be employed but are not yet considered standard approaches for routine management.  Examples include ERCC1 for assessing resistance to cisplatin (post here), gene signatures to predict which patients are responsive to various chemo agents (post here), EGFR FISH in predicting improved outcomes on cetuximab (post here), or resistance to EGFR inhibitors as predicted by ras mutations (post here).  

    Another potential implication of this work is that there may be important differences in how various treatments work in different populations.  EGFR inhibitors may be considerably more effective in studies from Japan than ones from the US or Europe.  Chemo may work differently in some populations than in others.  But as more and more trials are done in Asia, Europe, and elsewhere around the world, the study of pharmacogenomics reminds us that it is important to consider whether we can generalize conclusions across continents and different races.  We’ll turn to such questions next.