Patient Selection Important Issue in Targeted Rx of NSCLC

WASHINGTON—Targeted therapy shows promise in extending survival in non-small-cell lung cancer (NSCLC), but trial results are mixed and much further work needs to be done. One important next step is research on selecting patients according to the target protein, said Paul Bunn, MD, director of the University of Colorado Comprehensive Cancer Center in Aurora.

Speaking at the Cancer Research and Prevention Foundation's Lung Cancer Workshop III, Dr. Bunn suggested that routine use of tests to determine whether a patient's tumor is overexpressing the epidermal growth factor receptor (EGFR) could predict response to erlotinib (Tarceva), for example.

The issue is important because lung cancer survival rates are still low. In the past decade, the platinum chemotherapy drugs, along with paclitaxel, docetaxel (Taxotere), vinorelbine (Navelbine), and pemetrexed (Alimta), have all contributed to longer survival. But median survival is still a matter of months in advanced NSCLC. Now, in the era of targeted therapies, median survival may be creeping up another notch, but trial results have not been uniformly positive.

In ECOG 4599, the antiangiogenesis drug bevacizumab (Avastin) significantly improved median survival in advanced non-squamous-cell NSCLC. The drug, which targets the vascular endothelial growth factor (VEGF), increased survival by more than 2 months—a median of 12.5 months in the bevacizumab arm vs 10.2 months in the standard-care arm—resulting in a 21% decrease in the hazard rate of death, Dr. Bunn said. However, in the large ISEL trial with gefitinib (Iressa), the survival advantage for the drug, which targets the EGFR pathway, did not reach significance. Results were better in the major trial of erlotinib, which also targets the EGFR pathway. The BR.21 study showed a significant improvement in median survival—6.7 months with the drug vs 4.7 months with placebo—in the second- or third-line setting.

Following these results, researchers began to consider patient selection for the targeted therapies based on clinical features, such as sex and smoking history. Some differences in response were found: Women who were never smokers and had adenocarcinoma were more likely to respond to erlotinib, but the differences are not clear-cut, Dr. Bunn said, "so selecting patients based on clinical features is probably not a huge advance."

Selecting patients according to the status of the target protein is another option. The 2004 finding that patients with mutations in EGFR were more likely to respond to gefitinib triggered a spate of retrospective studies. While these studies have shown an association between the mutations and response to gefitinib and erlotinib, they have also shown that high levels of EGFR may make as much or more of a difference.

Measuring EGFR Status

There are various ways to measure EGFR status. But at least one study suggests, Dr. Bunn said, that patients who are EGFR positive by two different assays—FISH to measure gene amplification and IHC to measure the amount of protein on the cell surface—have impressively better survival rates than the general population of NSCLC patients given EGFR-targeted therapy. FISH and IHC are not standard for NSCLC patients, Dr. Bunn said, but at least one trial is now exploring this strategy prospectively.

Other trials of targeted therapies have tested other strategies, such as combining them with chemotherapy and with each other, but so far without much success, Dr. Bunn said.