I hadn't meant to get into a discussion of the treatments for specific cancers, or how we get the new drugs to lung cancer patients in a timely fashion, but in response to Charmaine's poignant comments, here goes with my take on this area: 1) The prognosis of lung cancer plateaued in large part because the older chemotherapeutic agents, on which we spent decades of (relatively) unsuccessful work, failed to push the needle very far. I don't think we ignored lung cancer so much as nothing we tried worked very well; 2) In contrast to a number of diseases where we have been relatively more successful (CML, GIST, Breast Cancer) we did not have a very good handle on what drove the growth of NSCLC; 3) We are finally beginning to understand the biologic drivers of some of these cancers (EGFR and EML4-ALK, for instance), and fascinatingly the best results have often been in non-smokers such as Charmaine, probably because smokers have a different set of mutations.

Access to new drugs is always a tough issue. Until a drug is approved by the appropriate regulatory bodies (both in the US and abroad), pharmaceutical companies have little real incentive to produce large quantities of the drug, and as Charmaine rightly points out, even within clinical trials there tend to be significant limitations on who is eligible for a new agent. The latter issue may be changing somewhat: the recent crizotinib trial that enthralled us all at the June ASCO meeting allowed patients to have had several prior regimens, and indeed 41% of the NSCLC patients entered in the trial had received 3 or more prior regimens, something one may be able to do with a relatively safe and highly targeted drug.

We encourage all patients and physicians to participate in clinical trials, because we can imagine no other way to get to a cure for cancer than through such trials: we quite simply need information on efficacy and safety before we expose the general population of patients to drugs. At the same time, when a drug looks promising, it is certainly in everyone's best interest to get the drug to patients in a rapid fashion. One way of doing this prior to drug approval is through Expanded Access Programs, available for several drugs. In the past this option was cumbersome, little used, and poorly understood by practicing oncologists. The FDA has recently issued new guidelines regarding what are called "Individual Patient IND's" that make up the basis for expanded access, and working with the FDA we created a module for ASCO University that explains in a straightforward fashion what is required of a physician. This went online at our ASCO University website just recently (http://university.asco.org/ExpandedAccess ), and we hope it will allow patients like Charmaine to experience the benefits of active new agents.

We also, as Charmaine suggests, need to re-think clinical trial design in the era of biomarker-driven targeted therapy. I know that this is an active area of thought in the clinical research community, at the NCI, and at the FDA. In Charmaine's case, the lack of readily measurable disease sounds like a barrier to study entry as well. There are many barriers to trial entry, and like a growing number of clinical trialists I would prefer our trials to be more representative of the general population of cancer patients, with fewer exclusion criteria. At the same time, we still need to be rigorous in judging drugs; we have a responsibility to both current and future patients. Finding the right balance isn't easy, and certainly will require regular review.  My hope is that the problem of "unmeasurable disease" will ultimately have a technologic fix, in the form of improved imaging techniques and serum biomarkers that  improve both our ability to measure tumor volume and predict response.