I reviewed some of the differences in the biology and clinical behavior of never-smoker lung cancers vs. the much more common lung cancer seen in current or former smokers. The main reason it is worth discussing is that there appear to be important differences in how never-smokers with NSCLC respond to some treatments, particularly EGFR tyrosine inhibitors like Tarceva, or Iressa previously. Among the first reports of this came from Memorial Sloan Kettering Cancer Center in NYC, which reviewed their experience with single-agent iressa several years ago and found that never-smokers were much more likely to demonstrate a significant response to iressa than current or former smokers (36% vs. 8% response rate, shrinking the cancer by 50% or more of the total measured volume). Along with bronchioloalveolar carcinoma histology, smoking status was among the most important variables in predicting response to iressa (abstract here).

As we learned about EGFR receptor mutations, first described in 2004 and associated with a high likelihood of response to EGFR inhibitors, we got some explanation for why never-smokers seemed to do well. In fact, the initial report by Dr. Lynch in the New England Journal of Medicine described 9 patients with excellent responses to gefitinib, of whom 6 were never-smokers. A study of tumor tissue out of Memorial Sloan Kettering showed that 7 of 15 lung tumors from never-smokers carried an EGFR mutation, compared with only 4 of 81 from smokers. Another study of patients who had surgery for early stage lung cancer in Japan showed that among 154 resected tumors (36% were from never-smokers, really highlighting how common never-smoker lung cancer is in Japan!). Since then, multiple trials have shown that never-smokers are significantly more likely to have EGFR mutations than smokers, and are also significantly less likely to have mutations in the gene k-ras (rhymes with mass) that in study after study are associated with no benefit from EGFR inhibitors. There are many controversial issues in lung cancer, but this is not one of them. From all over the world, the studies all show the same thing.

Another important report by Pham and colleagues came out of Memorial Sloan Kettering recently (I have to tip my hat and say they've really been moving the field of EGFR biology very well) that showed that IT'S NOT ALL OR NOTHING. While 51% of their never-smoker lung tumors had an EGFR mutation, 19% of their former smoker lung tumors had the mutation, vs. only 4% from current smokers. And the longer it had been since patients quit smoking, the higher the likelihood of having an EGFR mutation. In fact, more than a third of the patients who had quit smoking more than 25 years before diagnosis had an EGFR mutation, as shown in the following slide/figure:

So patients who quit decades ago may behave more like a never-smoker than a current smoker.

So how well do the EGFR inhibitors work in never-smokers? In the NCI-Canada trial BR.17 that led to the FDA approval of tarceva in the US, the group of previously treated patients that had the greatest benefit were never-smokers, who had a more than doubling of survival compared to placebo. That was a positive trial for an EGFR inhibitor, but several trials have been negative overall, including a trial of iressa alone vs. a placebo (ISEL, described further below), and a couple of trials of chemo (with two different types of chemo) with or without tarceva, and in those negative trials, the never-smokers still had a significant benefit. Here's the results for the Iressa Survival Evaluation in Advanced Lung Cancer, or ISEL, trial, showing negative results overall but a significant survival benefit in never-smokers in the setting of prior treatment:

(click to enlarge)

One of the negative trials of chemo with and without tarceva in previously untreated NSCLC patients with advanced disease is known as TRIBUTE (abstract here). This trial showed no real hint of benefit overall for chemo with tarceva at the same time, but the never smokers did show a significant improvement in both overall survival and time to disease progression, as shown below:

(click to enlarge)

In fact, the dramatic improvement in survival in never-smokers in this trial has led some people to recommend giving chemo with tarceva at the same time in never-smokers, including many of the folks at Memorial Sloan Kettering. On the other hand, I and several other experts in the field feel that this is a mistake. Why? Because overall the results show that standard chemo with tarceva at the same time may actually interfere with each other. I'll discuss the topic of chemo and EGFR inhibitors at the same time in another post, because it's a whole topic. But if you take a look at the curves again, I'll add arrows where I would say the two arms really separated to show a major benefit from tarceva, and it's at around 5-6 months after the start of the trial:

(click to enlarge)

Why does that matter? Because patients received a maximum of 6 cycles of chemo, which ended about 5 months into the trial, and then were on maintenance chemo or placebo alone. And the benefit really looks like it happens after the chemo is done. I believe that never-smokers got the benefit from chemo and tarceva together DESPITE the chemo, because they get so much benefit from tarceva that they can even withstand the challenge of chemo and tarceva concurrently. That is still totally controversial, but that's my view of the evidence. Importantly, there is a trial being done by the CALGB (Cancer and Leukemia Group B, a national cancer cooperative group) that is testing carbo/taxol along with tarceva against just tarceva alone in never- or light former smokers (defined as having quit at least a year ago and less than 10 "pack-years", the combination of number of packs smoked per day by number of years smoking). The schema for the trial is shown here:

(Click to enlarge)

One thing we really don't know yet is whether never-smokers respond very well to chemo as well. There are just a few studies that have looked at that, and some never even got published. Overall, they do suggest that never-smokers may do better with standard chemo than other patients (although the never-smokers on the TRIBUTE trial who got chemo alone did no better than we'd expect for the overall NSCLC population with this combination). I've definitely seen some very impressive results with regular chemo in never-smokers. So although I would strongly consider starting with tarceva in a never-smoker (as a single agent, not along with chemo), I and many others in the field also think it's very appropriate to start with chemo (with avastin, if the patient is eligible) and then move to tarceva as the definite next treatment if we didn't start with that. I'll also mention that I'll be leading a national trial of tarceva along with avastin in never-smokers, which will be available as a first-line treatment or for later treatment if the patients have not received tarceva or avastin before.

I know this is long and complicated. Sorry about that, but there's a lot of ground to cover. And the main point is that never-smokers are the most consistently identified clinical subgroup to demonstrate major benefits from EGFR inhibitors, but we don't know if they would also do better than other patients with other types of treatment, because we really haven't looked yet. But for the first time we're starting to see trials looking just at never-smokers. And that can only lead to better knowledge and perhaps new treatment options.