The setting of unresectable, stage IIIA or IIIB NSCLC (without a malignant pleural effusion) is currently one for which what we feel is best for the patient isn't necessarily something for which we have good evidence. For fit patients, there is a strong consensus that giving concurrent chemo with radiation provides a modestly but consistently higher cure rate than giving chemo and radiation sequentially. But that concurrent chemoradiation plan lasts for only 6-8 weeks, but whether there's more we should be doing, or what we should do, is entirely unclear.

As described in a prior post, several studies in the last decade have shown that about two cycles or 6-8 weeks of weekly chemotherapy along with about 60-66 Gray of radiation over 6-8 weeks is associated with the best survival results we've seen in unresectable, locally advanced NSCLC (somewhere in the 20% range long-term, and a median of about 16-18 months). There are two main approaches in North America for the chemotherapy. Some use the SWOG approach that showed very promising early results (prior post here), giving cisplatin and etoposide. The other very common alternative that is widely used in a community setting is weekly carboplatin/taxol. Until very recently had relatively little published experience to support it, but in the last few years now has been included in a few trials that demonstrate survival in the same ballpark as what we routinely see with cisplatin-based chemo: examples include RTOG trials such as described in a prior post and another abstract.

Cisplatin-based treatment is what nearly all of the prior pivotal studies had used, and many experts (myself included) look at the overall totality of evidence and feel that cisplatin is a little more efficiacious than its gentler cousin, carboplatin (see prior post). In a curative setting, I recommend the treatment that can provide perhaps a small incremental improvement in cure rate, so I've given cisplatin/etoposide when a patient is up to it. Even in my more frail patients, for whom a weekly therapy allows us to monitor how they're tolerating treatment from week to week before giving more chemo, I've commonly used a weekly cisplatin/navelbine approach that isn't actually well studied and published, but which has been shown to be feasible, and in my experience has been very well tolerated. I've been reluctant to give low-dose weekly carboplatin and taxol, largely out of concern that these medications at low weekly doses can enhance the radiation activity but may not kill micrometastatic cancer cells effectively. Remember that, as described in a prior post, the goal of treatment in this setting is to treat both the disease you can see and the potential for micrometastatic disease that is invisible but may be treated by effective chemo).

The biggest question now is what to do after the chemo and radiation are done. Although we have NO evidence that giving more treatment after the concurrent chemo/radiation improves survival (see prior post), I am like many other oncologists who are reluctant to just resign ourselves to the meager cure rates we currently achieve. "Consolidation chemotherapy", the additional chemo after the concurrent chemo/radiation, caught on as a standard partly because of the very encouraging survival in the early work but also because oncologists and their patients felt that more treatment would give a higher probability for cure. Now that the available evidence shows no benefit and only increased side effects, we're left with a situation in which the amount of chemo that is proven beneficial (2 cycles, or about 6-8 weeks) is less for stage IIIB NSCLC than would be routinely recommended to a patient with resected stage II NSCLC after surgery (3-4 cycles of platinum-based chemo, about 12 weeks). Frankly, that's unsettling.

The idea of giving "induction" chemo before surgery is an option, but the available evidence just doesn't show a benefit, and it may even be harmful (see prior post). We also have evidence that post-chemoradiation, "consolidation" taxotere doesn't improve outcomes (see prior post). Left without anything but ideas and good intentions, some oncologists stop after the 6-8 weeks, stating that we have no evidence that more is better, and that's true. Others give a couple of additional cycles of chemo, maybe cisplatin/etoposide, or carbo/taxol -- all reasonable thoughts. Since it became evident that taxotere was an option that was pretty convincingly not an improvement, I've often substituted three cycles of alimta for three cycles of taxotere. This isn't based on hard facts, but really a rationale and a hope to do better than our current standards offer. We know that alimta produced an essentially identical efficacy to taxotere in the setting of second line treatment for advanced NSCLC, but with a more favorable side effect profile overall (abstract here). Now, with strong evidence that alimta is not effective against squamous NSCLC (see prior post), I'd likely go with a couple of additional cycles of cisplatin/etoposide. Finally, our institution is also participating in the Stimuvax vaccine trial after chemo/radiation (as described in one more prior post), so I'm happy to offer this option that allows us to give a treatment that we'd anticipate to be minimally toxic and possibly effectve to increase the cure rates, but also acknowledging that there's no evidence yet that it adds beyond just observing patients after their chemo/radiation.

One final thing I'd mention is that I don't and would not be at all inclined to add a targeted therapy like tarceva or avastin into the mix for unresectable locally advanced NSCLC. Adding iressa as a maintenance therapy led to a quite significant decrease in survival compared to a placebo (see prior post), countering pretty much everyone's expectations that it would either help or have no significant effect. While that was in an unselected population and an EGFR inhibitor might lead to better results in a selected population like never-smokers or those with an EGFR mutation, I think that is a great research question, but I wouldn't give a treatment that has thus far only been shown to harmful for curable patients outside of a trial setting. Similarly, while avastin can improve survival in metastatic NSCLC, it can also lead to fatal complications, so don't ever add that for earlier stage patients outside of a clinical study.

In summary, about the only thing we can say with certainty about the patients with unresectable but potentially curable locally advanced NSCLC is that 6-8 weeks of concurrent platinum-based chemo with chest radiation is the cornerstone. We don't have any proof that adding more treatment before or after improves outcomes, but this is a setting in which the majority of both practicing general oncologists and lung cancer experts deviate from the evidence-based standards in hopes of improving outcomes from the humbling results we'd otherwise expect. None of us can be dogmatic that our intuition-based approach is definitely better than someone else's. This is an area in which we need to build on what we know, with newer agents in the next generation of clinical trials, in order to break beyond the current impasse.