As I noted in prior posts on the subject of unresectable stage III NSCLC, there is a general consensus that overlapping chemo and radiation is associated with better cure rates for this stage of locally advanced NSCLC than doing one followed by the other. At the same time, however, the overlapping, or concurrent chemo and radiation approach is associated with more challenges in terms of side effects, particularly esophagitis, as well as greater drops in blood counts, and potentially more inflammation in the lungs, or pneumonitis. The approach that I have generally advocated in the last few years, at least for patients fit enough to pursue it, is the concept of concurrent cisplatin-based chemo with concurrent chest radiation, followed by consolidation ("chaser") chemo with a different agent than what the patient received with the radiation.

This approach is based on some really pretty much unprecedented results in the SWOG trial 9504 that I've already described. 83 eligible patients received cisplatin/etoposide with concurrent chest irradiation for about 6 weeks, and then after a three week break started taxotere every 3 weeks. To go on the trial, you needed to not only have stage IIIB NSCLC (without a malignant pleural effusion) and be reasonably fit, you needed to have quite favorable breathing tests (pulmonary function tests) showing a good lung reserve in case there was significant damage from treatment. Many patients in the general world don't meet such stringent requirements. But the trial was hugely influential because nearly 1/3 of the patients on the trial remained without evidence of disease progression three years later (published abstract here), and even with longer follow-up appeared to do remarkably well (updated results here), with about twice as many patients as long-term survivors compared to what we'd expect historically.

(click to enlarge)

Some critics of this approach said that these better results could be because the trial accepted only particularly fit patients with very good lung function, and/or this new trial was done in an era where PET scans are now available to detect metastatic disease that we previously missed in patients who were called stage III NSCLC but actually had stage IV disease we missed ("stage migration" as we get more sensitive tests for finding advanced disease). With stage migration, you can see the Will Rogers Phenomenon, where patients with lower stage disease seem to do better just because you've culled out the patients destined to do more poorly, and the patients with advanced disease also seem to do better because you added patients with minimal, barely detectable metastatic disease.

The larger SWOG study 0023 (see details in prior post) also suggested that patients who received this combination of chemo and radiation followed by consolidation taxotere did unusually well, although this trial gave maintenance Iressa to many patients who actually did worse than other patients. Because everyone received the same chemo and radiation, it really didn't demonstrate the value or lack thereof for the consolidation taxotere.

One trial that did isolate the value of the taxotere is a Hoosier Oncology Group (or HOG) trial (designated LUN 01-24) in which all of the patients received the same cisplatin and etoposide with radiation, but half of the patients received further chemo with taxotere afterwards, while the other half were just watched carefully:

Unlike the SWOG trials, patients with less robust lung function were permitted to go on this study. Unfortunately, we don't have any survival results yet, and in fact the HOG trial closed a bit early, after declaring that it could not possibly detect a statistically significant survival benefit. Now, this may sound like there could be no value to the taxotere, but the trial was designed to detect an unrealistically huge difference in median survival between the two arms of 11 months (even 3-4 months would be a meaningful improvement, but the trial is too small to find a realistic but still statistically significant difference). So even though it won't be statistically significantly positive, we need to see in the next year or so whether there is any evidence of a benefit, perhaps an advantage of a few months. Right now all we know is how well this approach was tolerated, which was reported as quite challenging by the investigators from the HOG (abstract here). Briefly, only 163 of their 241 patients enrolled on the trial have gotten through the first phase of chemo and radiation and get to the point of randomization between consolidation taxotere and close observation without having progression or some other problem before that. Of the 76 patients randomized to taxotere, 7% of patients ended up getting no chemo, 34% received just a single cycle, 30% received two cycles, and only 29% received the full three cycles as intended. There were four deaths from treatment on that arm, problematic decreases in blood counts, and 20% of patients were hospitalized:

So that's pretty tough. Were the results worse than we might have expected because these patients were less cherry-picked than in the earlier SWOG trials? Actually, no, because the SWOG trials also had the same risk of dying on treatment of about 5%, and a high risk of a low blood counts. The differences could be just whether people focus on the danger of the treatment or the benefits if you get through it. And right now we need the survival results for the HOG trial, to see whether there's any suggestion that the added side effects of the taxotere are worth it in terms of better survival, whether statistically significant or not.

In the meantime, there are ways to reduce the risk of problems. I routinely give white blood cell growth factor support, specifically neulasta, after the taxotere in order to minimize the risk of infectious complications in the setting of a low white blood cell count, and I routinely administer a red blood cell growth factor like aranesp or EPO/erythropoietin to anemic patients in order to reduce the potential need for red blood cell transfusions. I also warn patients about increased shortness of breath or coughing that may indicate worsening pneumonitis on taxotere, and I follow patients' oxygen saturations and often repeat chest x-rays frequently to ensure that no new inflammation is developing in the chest. If so, I stop taxotere rather than push to a point that may be too risky.

In the end, this aggressive treatment plan is a double-edged sword, and we need to thread a needle between pushing hard against the cancer and harming patients. Treatment plans often need to be modified as patients proceed through therapy. Whether the SWOG approach will prove to be a major advance or is really largely an effect of selecting healthier and better staged patients is still an open question, but without a better alternative, this is the strategy I pursue for appropriate candidates.