Shortly after ASCO 2009, Dr. Pennell provided the highlights of the early report of the SATURN trial, conducted primarily in Europe, that randomized patients to maintenance tarceva (erlotinib) or placebo after four cycles of first line chemotherapy. The early report described a modest but statistically significant improvement in progression-free survival (PFS), but overall survival (OS) wasn't reported at ASCO. Frankly, the modestly favorable results in terms of PFS were overshadowed at ASCO by the maintenance therapy trial with alimta (pemetrexed), which showed a significant improvement in both PFS and OS. One of the more interesting presentations with new data at the World Conference on Lung Cancer in San Francisco a few weeks ago was an update of the SATURN trial results that included OS results. It revealed a significant improvement in OS, with a difference in median OS of one month (11 vs. 12 months, measured from the time of randomization, so patients had generally received about three months of treatment before then). The survival curves are shown below:

This was statistically significant, and the curves separate meaningfully, but I would say not as impressively as in the maintenance alimta trial. With a three month OS benefit overall and a 5 month improvement vs. placebo in patients with a non-squamous NSCLC tumor, the alimta trial captured the attention of far more people in the lung cancer world. However, one major shortcoming of the alimta trial is that only 19% of the patients on the placebo arm received alimta at the time of progression, and only about half of patients ever got any treatment we'd consider to be particularly effective second line therapy for lung cancer. Because of this, critics are right in saying that the alimta trial really isn't an ideal trial for testing the timing of alimta as maintenance as much as the difference between all patients on one arm getting alimta, vs. only 19% on the other arm. That's not exactly a fair test. This shortcoming in the alimta trial leaves an area for the SATURN trial to do better, but unfortunately it makes the same mistake. The presentation didn't give any understandable details of which agents people received after they progressed on the trial, but only 21% on the placebo arm ever got an EGFR inhibitor like tarceva. So this trial repeats the same mistake of testing the value of tarceva for everyone vs. tarceva for one fifth of the patients, and despite that handicap in favor of the immediate tarceva arm, the results aren't especially impressive. One other interesting point was that there wasn't a survival benefit on the maintenance tarceva arm in the patients with an EGFR mutation, the group of patients in whom you'd most expect to see a major difference favoring tarceva over placebo. In fact, there was a HUGE difference in PFS for tarceva in EGFR mutation patients, but that was erased when looking at overall survival. Why? Because the doctors were able to learn who had a mutation, and these patients essentially always received an EGFR inhibitor later. Apparently, these patients ended up doing just as well if they received tarceva after they progressed -- so no penalty for getting it as second line treatment instead of a maintenance therapy. And a clear example that if both patient groups actually get the same treatment, timing may not matter. My view of this trial is that while it's supportive of the idea that the benefit of maintenance or second line treatments can be very real, but all of these "maintenance" trials are greatly flawed in not really testing the timing of treatment as much as the difference between all patients getting an effective treatment vs. only some patients getting an effective treatment. And I must say that it really saddens me that only about 20% of patients in Europe are getting these therapies that we know improve survival in previously treated patients. The practice patterns in the US show that patients are more likely to get these treatments, but far too many patients are still missing out on them, some because they're followed off of all treatment and progress too much to safely receive more later. They therefore effectively missing a good opportunity to do better, and to me this group of patients is the strongest argument for shifting treatments earlier, whether that's chemo or tarceva.