The emergence of targeted therapies provides a goal of treating the cancer more selectively, thereby minimizing side effects, while hopefully achieving results as good as or better than standard chemotherapy. Although this is important in the entire population of cancer patients, this is a particularly welcome benefit in patients who may be reluctant to or not healthy enough to receive standard chemotherapy. As I mentioned in my last post on the association of age and benefits of chemotherapy, chronological age is not nearly as important as performance status, at least up until around age 80, where we have very few patients on clinical trials to help tell us what to expect. Regardless, there have been several studies of tarceva/erlotinib in older patients, and other trials specifically looking at patients with marginal performance status regardless of age, asking whether we can do as well or better with tarceva as with conventional chemotherapy in these patients.

Dr. Jackman and colleagues from Dana Farber Cancer Institute in Boston presented an important trial of an initial group of 66 patients age 70 or older to receive tarceva at 150 mg daily as their initial treatment for advanced NSCLC (abstract here). The vast majority of these patients had adenocarcinomas with or without BAC features, so likely there was some particular selection for patients felt most likely to do well with tarceva. In any event, they found a response rate of 12% and a "disease control rate", the combination of significant shrinkage and stable disease, of 60%. THe duration of stable disease averaged about 6 months. Toxicity was as expected, with a rash seen in 81% of patients, diarrhea in 69%, and 3 cases and 1 death from inflammation in the lung called interstitial lung disease, a rare (seen in about 0.5%-1.5% of people treated, and about 1/3 of cases are fatal) but reported side effect of tarceva or iressa. In an update of the study, now with 80 patients, it was reported that patient quality of life as measured on a validated survey was improved among treated patients who had stable disease or a response to treatment.

This year, however, we also learned about a trial of 103 patients selected not by age but by performance status. In this trial by Rogerio Lilenbaum in Miami, who has a particularly expertise in lung cancer management in older and sicker patients, the question was whether standard two-drug chemo with carboplatin/taxol would compare favorably or unfavorably to tarceva as a single agent in patients with a marginal performance status of 2, which corresponds to being symptomatic and spending extra time in bed, but not more than half of usual waking hours. These are patients for whom chemotherapy has appeared helpful but still somewhat debatable. So the trial was designed as follows:

(click to enlarge)

Although this trial wasn't large enough to make any major conclusions, the results strongly suggested that the patients who started with chemo did better, in terms of higher response rate, longer time before progression, and better overall survival, as shown in the table below.

However, as luck would have it, the tumor tissue they had from a subset of patients in the trial showed that ALL of the tumors with EGFR mutations (which seem to predict for doing better with EGFR inhibitiors) were in patients assigned to the chemo arm, and ALL of the tumors with RAS mutations (which seem to predict for doing poorly with EGFR inhibitors) were in patients assigned to tarceva.

It is also potentially important to note that the trial gave tarceva to patients after progressing on chemo, but patients starting with tarceva didn't tend to get chemo after progressing. And the curve below shows that it appears that the patients who received tarceva after chemo did better than other people on either arm of the study:

So what can we take from this study? It's not big enough to give us any ironclad answers, but to me it suggests that chemo is still a better choice for patients with a marginal performance status, at least UNSELECTED patients. I've already discussed how patients clinically selected by factors like never-smoking consistently do very well with EGFR inhibitors, so selecting patients for initial EGFR inhibitor therapy based on never-smoking could produce very different results. And I'll write an upcoming post about very promising results when giving EGFR tyrosine kinase inhibitors to patients with EGFR mutations or other molecular factors suggesting a high likelihood of doing well with them. So appropriately selected patients may do quite well getting EGFR inhibitors like tarceva early, but based on the results of the trial I reviewed above, the broader population of sicker patients may do better with conventional chemotherapy.