There are two widely tested epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) -- iressa (gefitinib) and tarceva (erlotinib). As discussed in my summary of their history (posts here and here), iressa was the first out of the gates, but it failed to demonstrate a significant survival benefit compared with a placebo in previously treated patients with advanced NSCLC, while tarceva did show a benefit. Nevertheless, iressa was approved by the FDA for about a year, based on preliminary promise and no alternatives, before the trial with tarceva became positive. By the time the tarceva trial came out and it was approved by the FDA, gefitinib was relegated to a secondary role in the US, but it has always fared better in Asia, where it continues to be widely used.

As shown above, these two agents are very similar but not identical, leading doctors and patients to question whether there might be a benefit to following one with another. Practically, with iressa the first one available and then restricted for new prescriptions after that, this has generally meant trying tarceva after progression on iressa. So what do we know? First, it's certainly possible to have a response to tarceva after progressing on iressa. A few years ago, Dr. David Garfield, a friend in the Denver area with a major interest in bronchioloalveolar carcinoma, wrote a case report of an elderly man with a marginal performance status who responded to dose-reduced tarceva after showing no response at all on iressa (the report doesn't specify whether he had real progression or just failed to respond at all and switched to his next treatment). On the other hand, this wasn't a common event, as described in a comment (here) that noted no benefit in five patients treated with tarceva after iressa. Two other reports have been published in the last couple of years that provide some insight with slightly larger numbers, in that at least you can count the subjects with the digits on one hand. Last year, a group from Korea described their experience of treating 21 fairly young patients (median age 56) with tarceva after prior progression on iressa (abstract here). As expected in a group of lung cancer patients in Asia, this group predominantly had adenocarcinoma tumors, and 52% were never-smokers, but many had received 2-3 prior treatments, and the majority of the patients had a performance status of 2-3 (definitely marginal to frail). They reported that two patients actually had partial responses and four had stable disease for at least 90 days, and that the patients who showed stable disease or better with iressa tended to do better with tarceva. On the other hand, the patients with EGFR activating mutations didn't do particularly well, presumably because their progression on iressa had been associated with a new, secondary mutation causing acquired resistance that also rendered the patients unresponsive to tarceva as well. Another reported series with similar conclusions came out of Singapore (abstract here). This one included 14 patients who had all received iressa followed by later tarceva after progression, nearly all never-smokers of Chinese descent with an adenocarcinoma, and predominantly younger women (median age 56). They also checked for EGFR mutations and found them in 8 of 14 patients. Five patients were noted to benefit, including two with good responses and three with stable disease, although two of the patients progressed within two months. Here is one particularly dramatic example of a response to tarceva after iressa: All of these patients who achieved disease control with tarceva after iressa had EGFR mutations. So putting all of these results together, we're still talking about only 41 patients, the vast majority Asian never-smokers. My overall impression of iressa vs. tarceva is that the former is considerably less effective at standard doses than the latter, so it's believable to me that you could squeeze some benefit from giving the more potent EGFR inhibitor after progression on a less potent one. They do have subtle differences in shape, so it's possible that this would be more effective than just increasing the dose of iressa. Thus far, it doesn't appear that results with tarceva are clearly improved by just increasing the dose of tarceva if someone isn't getting a benefit. I also think it may be important that these results are in a population that has shown remarkable results with EGFR inhibitors overall, and I wouldn't presume that the benefit would be as likely in a Caucasian smoker population. Overall, there is certainly some proof of principle that the EGFR inhibitors aren't the same, but I think that the greatest appeal is in giving tarceva after iressa, rather than iressa after tarceva. Moreover, I think the results could be less favorable in lung cancer population more typical for North America. I've been searching for reports of other potential options for modifying and improving results after "acquired resistance" to an EGFR inhibitor, like adding avastin or another treatment, but thus far I haven't found any results. If someone has a favorable story, I'd love to be able to offer a firmer answer and not just some ideas to consider.