Non-resectable metastatic non small cell lung cancer is a uniformly fatal illness. Patients may respond temporarily to treatment but in virtually all cases the disease will eventually progress. Therapy, depending on the extent of disease, is a combination of chemotherapy and radiation.
Erlotinib (Tarceva), a epidermal growth receptor inhibitor (HER1/EGFR), was recently approved by the FDA as maintenance therapy for the disease in patients with stable disease after treatment with platinum based therapy. The approval was based on the results of the Saturn study which was reported at the 2009 ASCO meeting.
The study showed:
Response rate was 12% with E versus 5% with P (platinum based treatment). Disease control rate (complete response + partial response + stable disease >12 wks) was
40.8% with E versus 27.4% with P (p<.0001). OS (Overall survival) data are not yet
mature.
Further analysis released in August 2009 showed:
The study showed that patients with NSCLC treated with Tarceva had a 23 percent improvement in overall survival compared with patients who received placebo (hazard ratio=0.81; p-value=0.0088). The hazard ratio, which assesses risk in the overall trial population, is widely recognized as the best measure of overall benefit in large randomized clinical trials. A hazard ratio of less than one for survival indicates a reduced risk of death. The median survival (a single point estimate of benefit) for patients receiving Tarceva was 12 months versus a median survival of 11 months for patients receiving placebo. (medicynical emphasis)
Medicynical note: The yearly cost of Tarceva is in the range of $30,000-60,000/year. This according to the Saturn study buys an improvement in median survival of 1 month. Added to this are the cost of the initial platinum based chemotherapy, radiation if given, doctor’s fees and imaging costs.
Is it cost effective to spend over $30,000-60,000 or more depending on length of treatment for a median survival improvement of 1 month? If not covered by insurance would you pay for this drug? Would you expect someone else to pay for it for you?
Medicynic 
Statistics are funny things. The ones quoted above make it sound as if Tarceva adds a month of living to most patients, yet costs a fortune. The truth is that it can add years to the life of some patients, and not even a day to the lives of many more. Most people (9 out of 10, I believe) are not helped by Tarceva, but doctors will try it for a few months to see if it stops the cancer. These people skew the statistics. Some other people, like me, have a particular genetic mutation that makes Tarceva a perfect choice. It stops the cancer without making you sick and weak. It does have side effects, but they are manageable. I have been living a full, productive life for several years because of this drug and I am grateful that it exists. The way to trim medical costs may be to try to determine ahead of time who will benefit from which drug. Right now, doctors usually try one after another, hoping that one will work. Tarceva was the third type of chemo I took. My docs asked if I would allow them to do a genetic test that would help them identify people who are likely to be helped by Tarceva. The tests showed I was a good candidate, and I was.
The problem with Tarceva and the other new targeted drugs is that their cost/year approximates and in many instances exceeds the average and median yearly income in the U.S. As you note the outcomes are spotty and many of those, if not most treated get little or no benefit–while some do quite well.
Our health care non-system is literally bankrupting individuals and the country. How does a system pay for such advances? Control costs? Do we need to change? How?
I absolutely agree about health costs bankrupting people. As for controlling costs, as a frequent consumer of medical services I am appalled at the number of times I am asked to repeat the same information (meaning someone has to be paid to write and store it somewhere) to three or four people in a single facility; am offered a phenomenally expensive monitoring test that has little benefit to ANYONE; or had providers give a blank look when I asked what something costs. Add to this the mountains of disposable stuff that is thrown away never having been used but having been opened; the duplication of super-expensive imaging machinery at facilities only a few miles apart; plus of course the cost of medications that are horribly expensive and are given to anyone who might be helped no matter how unlikely this might be.
I have had more than one medical professional ask, “What’s the matter? Don’t you have insurance?” when I’ve questioned whether a procedure is necessary. The suggestion is that insurance money isn’t real money. That attitude is only one that has to change.