Like Charlie Brown always thinking that this time will be different when he tries to kick the football without Lucy pulling it away, we get lulled into thinking that we know the intuitive, obvious answer in medicine without really testing it, only to find that our assumption was wrong yet again. This time, the important new result comes from the annual meeting of the American Society for Therapeutic Radiation and Oncology (ASTRO), which is just ending in Miami. There, investigators from the Radiation Therapy Oncology Group (RTOG) released some important early results from a study, RTOG 0617, that will impact the way stage III (locally advanced) NSCLC is managed.

While in many older studies the dose of radiation was in the 60-61 Gray (Gy) range (Gy being the units of radiation dose) over 6-6.5 weeks, there has been a drift in routine practice to often higher doses, to 63-66 Gy pretty routinely and even up to 70-74 Gy in some places, and not just in clinical trials. This isn't really based on the proven value of a higher dose over a lower dose, but rather based on the concept that the 60-61 Gy level wasn't found to be the clear best dose in the past, but rather was what was considered to be more or less safe and feasible at the time, while there is really a dose-response effect beyond that. Then, over the past 10-15 years, certain centers have done research using more refined radiation techniques to deliver chest radiation up to doses in the range of 74 Gy or even higher, with concurrent chemotherapy. Since then, more and more radiation oncologists have followed, using conformal radiation techniques to routinely push what is perceived as a "standard dose" of chest radiation with concurrent chemotherapy, as many of us in the field came to view radiation to 60-61 Gy as potential under-dosing of treatment for concurrent chemoradiation today.

RTOG trial 0617 was designed to directly test high dose vs. lower dose (older standard) radiation with concurrent weekly carbo/Taxol (paclitaxel), followed by additional carbo/Taxol given every three weeks, for patients with locally advanced, unresectable NSCLC. It was also designed to test (in what is called a 2 x 2 randomized design) whether addition of the monoclonal antibody to EGFR Erbitux (cetuximab) improves outcomes when given to patients getting either the higher radiation dose or lower radiation dose plan.

This trial has actually been controversial from the beginning, because many clinicians felt it was unethical to conduct a trial that randomized patients with stage III unresectable NSCLC to 60 vs. 74 Gy with concurrent weekly carboplatin/paclitaxel. Why? Because many were already convinced that one or the other dose was clearly superior -- though people differed in their belief of which arm was unethical to be randomized to. Some people were wary about under-dosing radiation at 60 Gy, while others were concerned that treating to 74 Gy wasn't well tested enough to do this in a large multicenter trial setting, particularly with Erbitux added.

So the interesting news from ASTRO is that now that RTOG has enrolled 423 patients to the trial (a little over 500 were targeted in the study design, adjustable depending on a few parameters), the RTOG has now closed the high dose arm in the wake of analysis by the Data Safety Monitoring Board (DSMB), which reviewed the results thus far and concluded that there was no way for the higher dose arm to appear superior. This doesn't necessarily indicate that higher dose radiation is harmful, but that is often the trend when we see a DSMB closing a trial or an arm of a trial due to futility analysis before enrollment is complete. I expect to see that there is actually a worse outcome, whether statistically significant or just a trend in the wrong direction. Still, it could potentially be more or less the same and just not better. It could be far more toxic. It may be a particular problem to add Erbitux, especially with higher dose RT.

What can we safely conclude? Lower dose radiation remains the standard, and yet again we were wrong to presume that more is better, whether it's adding more chemo, or a targeted therapy to our current standards, or increasing the dose of radiation because it's feasible to do so.

It is also possible that escalating intensity of treatment may be more feasible in more specialized centers that tend to attract and treat disproportionately fit patients, but the wider experience in a broad population of patients and treating centers is associated with greater toxicity. Fortunately, we'll have more answers when the trial is completed (it will continue to randomize patients to chemo/radiation with or without Erbitux) and we see the actual data. In the meantime, it's another important reminder that it's not necessarily wise to follow the latest trend without actually testing that it's better than what was established as the best treatment for today.