Several weeks ago, I described the results of a survey I sent out to several colleagues who are lung cancer experts around the country, asking how they would manage a case of a newly diagnosed Caucasian never-smoking patient with advanced NSCLC, adenocarcinoma, and asymptomatic subcentimeter brain metastases, treated with whole brain RT before starting systemic therapy.

I then asked these same experts the question but now with the patient being an ex-smoker with a squamous NSCLC tumor. The basic sketch is that it's a 61 year old Caucausion woman with a very good performance status, no limitation on health or motivation for treatment, and a metastatic NSCLC, squamous subtype . As in the prior case, she had treated asymptomatic brain metastases and was presenting for consideration of first line treatment. What would various experts recommend as a plan for the next few months?

One major difference was that, unlike the never-smoker with an adenocarcinoma, very few experts now favored sending tissue for an EGFR mutation. The vast majority had favored checking this in a never-smoker, who might be considered for up-front EGFR inhibitor therapy, but when the case was an ex-smoker with a squamous tumor, only 6 of 20 were interested in any molecular testing: 3 for KRAS mutation, 1 for EGFR IHC before considering erbitux (cetuximab), and two who still wanted to check for an EGFR mutation. Most were inclined to make a treatment recommendation without any molecular testing.

Not surprisingly, everyone recommended a platinum doublet, and in this case the clear winner was a combination with gemcitabine, recommended by 14 of 20 respondents. I presume that this was related to the relatively more favorable results with cisplatin/gemcitabine in the "JMDB" trial vs. cisplatin/pemetrexed among patients with squamous tumors. Perhaps the much lower incidence of hair loss with gemcitabine than a taxane was also a factor, but I only asked them what they would recommend, not why. The expert responses were evenly split between cisplatin and carbo with gem, and 3 of these 14 would add erbitux.

Another 3 people recommended the regimen of cisplatin/navelbine (vinorelbine) with erbitux from the FLEX trial, and one recommended cisplatin/taxotere (docetaxel) with erbitux. The three remaining respondents favored carbo/taxol (paclitaxel), without erbitux.

As in the first case of a patient with a metastatic lung adenocarcinoma, the question of whether to give 4 or 6 cycles of first line therapy was split pretty much straight down the middle, and many people hedged a bit and said it depended on how the patient was tolerating it. Even though I forced them to describe a plan for what to do several months later, the reality is that we never do this. In real life, our treatment recommendations are modified by how a patient is tolerating the treatment and how the cancer is responding. This is how it should be, I think, rather than making hard rules for a long-range strategy without considering the new information that you learn from working with a patient.

After this time, everyone who would include erbitux would continue it, although I must admit that I struggled with that (I actually said cis/gem/erbitux first line, though there are many good choices), since weekly maintenance therapy isn't very feasible for a non-curative therapy, in my mind. Among those who were giving chemo alone, three favored an immediate transition to taxotere, while most others favored taxotere after progression. Many mentioned the EGFR inhibitor tarceva (erlotinib) as a feasible option as well, though it would depend on how well or poorly the cancer responded to prior chemo.

I'm not very surprised by these results but didn't expect to see 70% of respondents favor a platinum/gemcitabine combination. I think it's a good choice and favored it myself, but we don't have any evidence that a platinum/gemcitabine combination is superior to a platinum/taxane or platinum/vinorelbine combination in any direct comparison. Not surprisingly, erbitux certainly was of greater interest in an avastin-ineligible population, but it was still recommended by only a minority of experts.

I was impressed to see so much cisplatin use, which I do slightly favor in especially fit patients, but it certainly suffers from an image problem. Still, I do believe that the overall evidence suggests that cis is superior to carbo, and that while it's a palliative setting, there is some evidence that it provides the same degree of relative survival benefit as erbitux or even avastin, though without the marketing muscle behind it. I think that's what we were seeing when 10 of 20 lung cancer experts recommended cisplatin for a patient with advanced NSCLC.

I also think it's interesting that taxotere was clearly favored as a second line agent. That's where the evidence is, but this agent also isn't relished by the majority of oncologists for this setting. Perhaps that will change as alimta becomes eliminated as an option for patients with squamous NSCLC.

And there is still plenty of room to debate whether second line treatments should be moved up to immediate "maintenance" therapy or not. The folks I surveyed were still clearly more inclined to either continue erbitux or wait to see progression before starting second line therapy.

I think it's interesting and that the standards are still evolving. Overall, there's still plenty of room for individualizing based on patient preference and physician style.