With the recent publication of the Eli Lilly-sponsored phase III trial of immediate versus delayed Taxotere (docetaxel) after the completion of first-line chemotherapy in patients with advanced NSCLC (abstract of paper by Fidias and colleagues here), I think the time has come to critically evaluate this as a potentially practice-changing concept. Call it whatever you want: maintenance therapy, sequential treatment, or early/immediate second-line therapy, but whatever you do, don’t call it a fluke. The idea behind maintenance is simple: after finishing your initial 4-6 cycles of platinum-doublet chemotherapy, you move immediately on to more treatment rather than stopping and waiting for the cancer to progress. In the last year and half, we have seen four major phase III trials test some version of maintenance treatment with chemo or Tarceva, and all have generally supported a benefit. The Fidias paper is the first of these to be published, so let’s take a closer look at the design: (Click to enlarge) In this trial, 566 previously untreated patients with metastatic NSCLC were all initially treated with carboplatin and gemcitabine. Those that finished four cycles without progression were randomly assigned to either receive Taxotere immediately and continue it for a maximum of 6 cycles, or to be followed and receive Taxotere once their cancer started to progress. The trial was designed to be large enough to see a 4 month improvement in overall survival (OS), which in retrospect seems ridiculously ambitious. To give you a comparison, the ECOG 4599 trial, which led to the approval of Avastin in lung cancer, only showed a 2 month improvement in OS. In any case, the trial did not quite make the 4 month improvement, missing it by a little over a month (OS 12.3 months for immediate Taxotere versus 9.7 months for delayed; p=0.085), but it did have an impressive improvement in progression-free survival (PFS; 5.7 versus 2.7 months; p=0.0001), which is a measure of patients still alive without progressive cancer. Essentially what this means is that patients who started immediately on Taxotere delayed the progression of their cancer by 3 months compared to those who waited, and lived an average of almost 3 months longer. This is a pretty clinically significant improvement, although there are some important things to point out about this trial. First, only 309 patients (out of the original 566) made it to the randomization, so these were already the patients who were doing very well after first-line chemo. Second, 37% of the patients (58 of 156) who were designated to receive the delayed Taxotere never got it (as opposed to only 8 patients in the immediate arm), about half of them because their cancer progressed so much that they were unable to get more treatment. So it is unclear if the benefit of immediate Taxotere was truly better effect, or just the results of most of the patients in the immediate Taxotere group actually getting the chemo.

Maintenance is actually an old idea, and has been tested many times in the past unsuccessfully, making the new results somewhat of a surprise. When prior trials have tested fewer cycles of chemotherapy versus more cycles of chemotherapy, almost invariably there has been no difference in survival with less toxicity seen with shorter treatment. For this reason, the standard strategy for NSCLC patients has been to give 4 cycles of chemotherapy and then stop. The difference between those trials and the new ones is probably in the agents used, which have been the most active second-line agents for NSCLC (i.e. Taxotere, Alimta, and Tarceva (with or without Avastin)). A second difference has been in the use of a different agent for maintenance than that used in first-line, something called “non-cross resistant” therapy. The theory behind it is that cancer eventually develops resistance to any chemotherapy, and by switching to a new agent one can possibly delay this resistance. The other three trials that have been completed in the last year were 1) the maintenance Alimta versus placebo trial; 2) the SATURN trial which tested maintenance Tarceva or placebo; and 3) the ATLAS trial, which tested the combination of Tarceva or placebo with maintenance Avastin (SATURN and ATLAS have been the topics of recent posts by Dr. West). Only the results of the Alimta trial have been presented, which showed a 2 month improvement in PFS (see below) and almost 3 month PFS benefit in patients with non-squamous histology (4.4 versus 1.8 months). It has also been announced that the other two met their endpoints of improved PFS, and my guess is that the 2-3 month improvement in PFS will continue to hold for all of the trials since all three agents have about the same activity in second-line NSCLC patients. So what do we, as practicing oncologists, do with all of this new information? Remember that maintenance chemo is not currently standard of care treatment (other than Avastin, which could be the topic of an entire post in itself), so adopting it would represent a major change in how we treat new NSCLC patients. I consider myself somewhat conservative when it comes to adopting new treatments before adequate evidence is available, but this is beginning to make be a believer. I do think that the benefit from these trials is essentially in making sure that most people receive active second-line treatment, rather than some hypothetical biologic benefit from treating the cancer earlier rather than later, but regardless of mechanism it seems to be a real effect. Since many patients who take a break after first-line chemo never get to the second-line, this is the best way to expand the pool of patients who get treatment that may be beneficial. The opposite side of the coin, though, is that most patients are pretty beaten-up after 4-6 cycles of platinum doublet chemo. Patients in clinical trials like the ones I talked about here tend to be younger and healthier than the typical patients Dr. West and I see in clinic. It would be hard to tell every 70 year old patient, exhausted after successfully finishing 4 cycles of platinum doublet chemo, that they need to immediately start a new treatment with no break. It is no great service to our patients to trade time for suffering. What I think these results should do is allow us at least bring up the option of maintenance with our patients, and give us justification to use it when it seems appropriate. I may also start following my patients who are on a chemo break a little closer, and broach the topic of second-line treatment earlier than before. Please remember this is just my personal opinion, and does not represent the recommendation of any institution or of the oncology community at large.