Erlotinib Is ‘Active, Well Tolerated’ in Pretreated NSCLC

NEW YORK—The investigational epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor erlotinib (Tarceva, also known as OSI-774) produced objective remissions and mild toxicity in stage IV non-small-cell lung cancer (NSCLC) patients who had undergone a number of prior chemotherapy regimens, according to results of a phase II study.

"There is no question that Tarceva is active and well tolerated in heavily pretreated NSCLC patients, all of whom had received previous platinum-based chemotherapy," said Philip Bonomi, MD, director of oncology, Rush Medical College, Chicago. "The responses are relatively durable, and survival is certainly very encouraging."

Rash and diarrhea were common but rarely severe; only a few patients required dose reductions due to adverse events, Dr. Bonomi said at the Mount Sinai School of Medicine Chemotherapy Foundation Symposium XX.

Response to erlo-tinib therapy appeared to correlate with rash; almost every responder developed rash vs only about half of nonresponders. "It may be that the rash has some predictive value in terms of response," he said.

The phase II NSCLC trial was based on preclinical studies showing that inhibiting the EGFR pathway reduces malignant cell proliferation. A high percentage (40% to 80%) of NSCLC tumors express EGFR. "When it first came out, people thought it would be a cytostatic drug, but this agent actually causes tumor regression," Dr. Bonomi said.

To determine the response rate in NSCLC after treatment with a platinum-containing chemotherapy regimen, investigators enrolled 57 patients with at least a low level of EGFR expression and relatively good performance status. Patients received erlotinib 150 mg/d orally, with an opportunity for dose reduction or escalation depending on toxicity. Response was confirmed at 6 weeks, and disease was assessed every 8 weeks.

The 57 study subjects included 23 men and 34 women (mean age, 62 years) who had received at least one prior chemotherapy regimen; all but 10 had received two or more regimens, and all but 2 had received a prior platinum. Median duration of treatment was 63 days (range, 12 to 843) and dose decreases were required in only three patients (5%).

Response Rate

Overall response rate was about 12%, including two complete responses (3.5%) and five (8.8%) partial responses. Another 22 patients (38.6%) achieved stable disease. "Roughly half of these heavily pretreated patients appear to have some benefit from treatment with Tarceva," Dr. Bonomi said. By multivariate analysis, the only significant predictor of response was time since last chemotherapy regimen; longer time since last regimen was associated with a higher response rate (P = .033).

Median overall survival was 8.4 months, which was "good" but not too unexpected. "Just as we see in secondary chemotherapy trials, the patients who survive that long are patients who have disease that is not progressing quite as rapidly," Dr. Bonomi said. "In fact, the 1-year survival rate is 40%, and that is as good as we see in front-line therapy."

Predictors of survival by multivariate analysis included performance status of 1 vs 2 (P =.044) and, again, time from initial diagnosis (P = .0001).

The latter predictor may seem obvious—that patients with more rapidly growing tumors tend to die sooner—but, Dr. Bonomi said, it is also possible that this tyrosine kinase inhibitor is exerting its effect preferentially in the population of tumors that do not grow as fast. "This is my own speculation," he said, "but it’s something to think about."

No Difference by Prior Chemo

Investigators also looked at survival by type of prior chemotherapy, but found no differential efficacy in the 47 patients who had two or more prior regimens, and no difference for the 14 who had prior platinum and prior docetaxel (Taxotere).

Rash occurred in three quarters of patients, but it was severe in only 2%; likewise, diarrhea was common (61%) but rarely severe (2%). One quarter of patients reported nausea, and 5% had dry eyes. Rash and diarrhea both occurred soon after starting therapy; median time to initial rash or diarrhea was 10 and 14 days, respectively.

All seven responders experienced rash, as did all but one of the 22 patients with stable disease. By contrast, rash was seen in only 15 of 28 patients with progressive disease (54%).

"There was at least a trend for survival by grade of rash," Dr. Bonomi said. "There really might be something to this, and I know additional studies are evaluating the relationship between rash and survival."

In Canada, a very large phase III study is underway now, comparing erlotinib 150 mg/d with placebo in 700 patients with stage IIIb/IV refractory NSCLC. The primary outcome measure is overall survival. Secondarily, investigators will report response rate, symptom benefit, quality of life, and other measures.

"If there is some survival advantage with this agent, I think this study is going to show it," Dr. Bonomi said.