In fact, the researchers report in the New England Journal for June 24th, the trial was terminated early because progression-free survival, the primary end point, was seen to be significantly longer with gefitinib at the interim analysis.
Dr. Akira Inoue at Tohoku University Hospital in Sendai and colleagues in the North-East Japan Study Group tested cytologic or histologic specimens from patients with advanced non-small-cell lung cancer for the presence of EGFR mutations sensitive to the tyrosine kinase inhibitor gefitinib. They used a PCR method they had developed specifically for the rapid detection of EGFR mutations.
Patients older than 75 years and those previously treated with chemotherapy were excluded. The investigators originally intended to enroll 320 patients, and planned an interim analysis to be performed 4 months after enrollment of the 200th patient.
Eligible patients were randomized to receive either gefitinib, 250 mg orally per day, or paclitaxel and carboplatin. Gefitinib was administered until disease progression, or the occurrence of intolerable side effects or withdrawal of consent; chemotherapy was administered for at least three 3-week cycles.
A total of 230 patients were enrolled before the trial was stopped. The interim analysis revealed that median progression-free survival was 10.4 months in the gefitinib group and 5.5 months in the chemotherapy group, yielding a hazard ratio for death or disease progression of 0.36 (p<0.001).
Similar progression-free survival times were seen on the final analysis 7 months later.
Rates of 1-year progression-free survival were 42.1% and 3.2% with gefitinib and chemo, respectively. Corresponding rates at 2 years were 8.4% and 0%.
Overall survival was a secondary endpoint of the trial, and did not differ significantly between the two groups. Median survival times were 30.5 months with gefitinib and 23.6 months with carboplatin-paclitaxel (p=0.31).
After the trial was discontinued, nearly all patients who had completed chemotherapy received gefitinib as second-line therapy. The response rate was 58.5%
Dr. Inoue and colleagues note that patients treated with gefitinib should be carefully monitored, as the agent can cause interstitial lung disease. The incidence of this side effect seems to vary between countries; in the US it is about 0.3%, in Japan it is 4%-6%.
Asked by email if certain non-small-cell lung cancer patients are more likely than others to have gefitinib-susceptible tumors, Dr. Inoue replied, “Adenocarcinoma, non-smoking history, female, Asian patients are likely to have EGFR mutations; however, I think it is more important to detect such mutations among NSCLC patients who are unlikely to have mutations.”
Therefore, he stated, “I recommend all NSCLC patients (should) be tested for EGFR mutations, although such mutations are mostly found in adenocarcinoma.”
However, histologic typing is not always accurate, Dr. Inoue commented, and shouldn’t be relied on to guide EGFR testing. Furthermore, “Even in NSCLC patients with squamous cell carcinoma or large cell carcinoma, if the mutation was detected, they could benefit from EGFR-TKI.”
N Engl J Med 2010;362:2380-2388.