In the Keynote Lecture at the Targeted Therapies in the Treatment of Lung Cancer Conference, Dr. Joseph Nevins from Duke’s described emerging approaches for how to personalize cancer management, including who to treat, and how to treat.  I described in my last posts his efforts to refine our understanding of which patients with early stage, resected lung cancer are at higher or lower risk for cancer recurrence, and therefore better or less well served by receiving adjuvant chemotherapy.  His lab is also studying how gene profiles of cancers can help us refine how to treat.

   Herceptin, also known as trastuzumab, is a monoclonal antibody against HER-2/neu, a protein over-expressed in a subset of breast cancers.  If given to an unselected population of breast cancer population, it would have a response rate of less than 10% and would potentially not have been recognized as a valuable and active drug in treating breast cancer.  On the other hand, when given to an appropriately selected population of patients with over-expression of HER-2/neu, it has a response rate of 35-50% and emerges as a remarkably useful drug for breast cancer.

    Thus far in lung cancer, we have rarely individualized our treatment recommendations, and even our targeted therapies have been given in an “untargeted” way.  We have just started to identify patients, using EGFR mutations and/or clinical variables like smoking status or tumor subtype, to identify some patients as better or worse candidates for tarceva or avastin.  But these are very rudimentary first steps.  Dr. Nevins and his colleagues at Duke are looking at an array of genes to identify classes of different tumors with certain patterns of genes turned on and off (abstract here):

 (click to enlarge)

Nevins and his group are working to reliably identify for an individual tumor which targeted pathways are turned “on” or “off”, therefore determining what is driving the cancer and helping to predict which targeted therapies will be more or less likely to work against it.  As we develop more and more molecularly targeted therapies against not only EGFR and angiogenesis but also many other important molecules with regulatory functions within the cell, these gene profiling approaches can help us understand how to enrich with the right population most likely to respond to that therapy. 

   Dr. Nevins and his group are also conducting research in which they can identify from gene signatures which chemo drugs are more or less likely to work for a particular person’s cancer (abstract here).  With the growing ability to predict which chemo and targeted therapies are most likely to work against a cancer, Nevins is working with investigators from the cancer cooperative group CALGB to do a “window of opportunity” trial for patients with resectable, early stage NSCLC.  Still in development, the plan will be to obtain a biopsy, do gene expression analysis on the tumor to predict which chemo drugs and targeted therapies will work best against the cancer, give those treatments for several weeks, then perform surgery and assess the response to that pre-operative therapy.  After surgery, adjuvant chemotherapy can be given as per current standard recommendations.  With this approach, it is hoped that patients can receive the best of current treatment approaches with the addition of a potentially more effective early treatment that can be directly tested.

  These approaches haven’t been tested as thoroughly as their approaches of identifying the high risk and low risk early stage lung cancer patients, but they’re right at the center of our research for the next few years.  Not only do we want new treatments, we also want to be able to identify which drugs will work best for which patients.  Such improvements could lead to a new chapter in cancer treatment.