The general approach to NSCLC is in transition right now, as the line between first and second line therapy are becoming increasingly blurred. A few years ago, the clear standard was that we usually stop first line chemo after four to six cycles, then follow a patient clinically and radiographically until they show evidence of progression, at which time we’d start second line treatment. But now, a growing proportion of our standard treatment protocols include a maintenance phase of ongoing treatment with a targeted therapy after 4-6 cycles, usually of platinum-based chemo with that same targeted therapy. So while we don’t have established proof of the value of maintenance therapy, it’s most common to continue avastin (bevacizumab) as a single agent after 4-6 cycles with platinum-based doublet chemo. While erbitux (cetuximab) is less clearly established and less commonly used, the trial that demonstrated a survival benefit also included a maintenance phase. Erbitux, however, is a more practically challenging situation than with avastin because erbitux is a weekly treatment, not especially convenient for ongoing maintenance therapy. At the same time that we have first line treatment extending out beyond 4-6 treatments until progression, there are studies moving second line treatment earlier, so that there is a seamless transition to starting right after first line treatment ends, usually after 4 cycles. An initial study with taxotere (docetaxel) IV every three weeks after 4 cycles of carboplatin/gemcitabine that tested immediate vs. delayed second line therapy (starting at the time of progression) showed a highly positive difference in progression-free survival favoring immediate second line taxotere, and a strong trend toward superior overall survival (see prior post for details). While that trial raised the attention of many oncologists to this question, it didn’t lead to a sea change in how we manage patients. Most of the other experts I spoke with agreed that we’d like to see another trial show a similar result, which is what we got this year with a trial of alimta (pemetrexed) vs. placebo IV every three weeks after four cycles of any of several platinum-based doublets. As highlighted in a prior post, this study showed results that I would consider remarkably similar to the taxotere trial, again with a highly significant improvement in progression-free survival and a nearly statistically significant improvement in overall survival (p = 0.06) that was nearly three months in absolute terms. The key shortcoming of the trial was that only half of the patients randomized to placebo went on to receive any second line therapy, since there are many parts of Europe that don’t consider it a clear standard of care, and that’s where the study was primarily completed. Even with that important caveat, I would consider the results to be so compelling that it merits a change in my practice. What’s more, the results with alimta were limited to the patients with non-squamous NSCLC, who actually had a 5 (!) month improvement in overall survival, while the squamous patients actually had a detriment in their survival on the alimta arm (hence the change in the FDA label that removes the approval of alimta for patients with squamous NSCLC).

For the EGFR inhibitors, we’re largely waiting on results, though I suspect the value of maintenance or immediate second line NSCLC is going to be very similar to what we see with chemo. A Japanese trial showed essentially comparability of switching immediately from 3 cycles of chemo to iressa, compared with 6 cycles of chemo followed by a choice of treatment (abstract here), and an improvement in survival among never-smoking adenocarcinoma patients who received early iressa. But the trials that will help establish or deny a role for immediate second line tarceva are being completed now by Genentech, one trial in Avastin-eligible patients, and the other in avastin-ineligible patients: (click to enlarge) Finally, there’s the question of how to combine targeted therapy and chemotherapy. The ATLAS trial, above, will be looking at avastin maintenance alone vs. adding tarceva to avastin maintenance. In addition, the Eastern Cooperative Study Group (ECOG) is going to compare maintenance avastin to maintenance alimta or the combination of avastin and alimta after four cycles of carbo/taxol/avastin: The trial above won’t answer the question of whether there’s really a value from maintenance avastin, and some oncologists don’t use it routinely. My interpretation of all of the evidence put together is that although the evidence in favor of transitioning straight from first line to second line isn’t perfect, I find it to be consistent and helpful enough to be inclined to recommend it in certain cases. It’s important to highlight the many situations in which I don’t think early second line therapy necessarily applies: 1) Patients who progress or have prohibitive side effects before the end of four cycles of chemo 2) Patients already getting maintenance avastin or erbitux 3) Patients with a very slow growing, minimally symptomatic cancer 4) Patients responding well on a treatment like tarceva (continues until progression) The studies of planned early second line therapy, such as the ones with taxotere and alimta described above, both had about half of the patients come off the study before getting through four cycles of chemo without progression or side effects. I suspect that another 20-30% of patients are going to already be on maintenance avastin or erbitux, have slowly progressing cancer, or are responding well on tarceva. But for the significant minority of patients who don’t fit in any of these other categories, I now favor moving straight from first line to second line treatment. Although the benefits have been shown with chemo and not yet with tarceva, I strongly suspect that the same benefit will apply with any effective second line therapy. My approach is to use whichever second line therapy would be optimal. I don’t think that it’s necessary for every patient who proceeds through 4-6 cycles of chemo to move immediately to second line treatment with no break – a vacation is a terrific idea (Hawaii? What's not to like?). But the strength of the evidence leads me to recommend that many patients need to be on some form of treatment most of the time to reduce the risk of rapid progression that could lead to a such a decline that they can't benefit from more treatment. I believe that some form of ongoing therapy for the patients without particuarly indolent lung cancer can keep them from losing the opportunity for salvage therapies that are often beneficial but require a certain level of performance status and constitution. We'll see more work in this field, so my conclusions and treatment recommendations may change as new information comes in.