Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner
Report on the Relative Efficacy of Oral Cancer Therapy for Medicare Beneficiaries Versus Currently Covered Therapy

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

Part 1. Gefitinib and Erlotinib for Non-Small Cell Lung Cancer (continued)


The search strategy yielded 299 articles. The selection process is described below:

Identified by search strategy (N = 299)

     |------ Excluded based on review of abstract (N = 98)

Included based on review of abstract (N = 199)

     |------ Excluded based on full-text review (N = 86)
     |          4 cases selected on AE
     |          12 not phase II-III for efficacy
     |          3 abstract superseded by published article
     |          10 no new data reported
     |          30 no primary or original data (review article)
     |          3 wrong disease
     |          10 wrong drug
     |          12 wrong outcome
     |          1 not possible to obtain a copy of publication

Included in full-text review and evidence tables (N = 114)

The 114 included studies are comprised of 83 full reports, 30 abstract-only publications, and 1 citation in the grey literature (press releases, etc.). Study designs included 8 phase III controlled clinical trials, 14 phase II uncontrolled clinical trials, and 92 studies of other designs. The majority of the studies of other designs were retrospective series of patients receiving gefitinib under AstraZeneca's pre-approval Expanded Access Program (EAP); these provided data on adverse events and predictors, but were not included in the assessment of efficacy.


First-line treatment combined with standard chemotherapy

There are four completed, large, randomized controlled phase III studies that compare epidermal growth factor receptor tyrosine kinase (EGFR-TK) inhibitors in combination with standard chemotherapy against standard chemotherapy alone for patients with previously untreated advanced non-small cell lung cancer (NSCLC) (Table 2).6-9 Two of these used gefitinib6,7 and two, which have only been presented in abstract form, used erlotinib.8,9 The patient population and design of the studies were nearly identical except the studies with gefitinib used two different doses, 250 mg and 500 mg daily, in a three-arm study, while the erlotinib studies used a single dose. The chemotherapy used in the paired studies with each drug was standard combination chemotherapy of either carboplatin and paclitaxel in the North American study,6,8 or cisplatin and gemcitabine in the study done outside North America.7,9

None of the four studies showed a statistically significant difference in overall response rate (CR + PR) (Table 3), time to treatment failure (or progression-free survival), or overall survival (Table 4) between the EGFR-TK arm(s) and the placebo arm. Unpublished subgroup analysis of one study8 by smoking status showed an increase of median survival from 10 months in the placebo group to 22 months in the erlotinib-containing combination arm among the 8 percent of patients in the study who had never smoked. We are not aware of results of a similar analysis in the other three studies.

In addition to the phase III studies, a single, small, uncontrolled phase II study of docetaxel 75 mg/m2 plus gefitinib 250 mg daily in patients at least 70 years of age reported a response rate of 50 percent.10

First-line treatment as single agent

Eleven uncontrolled studies of gefitinib or erlotinib as a single agent in previously untreated patients with advanced NSCLC were identified (Table 5), seven of which11-17 were in unselected patients. Two of the studies were restricted to patients with bronchoalveolar carcinoma (BAC) histology,18,19 one included only patients age 70 and greater,20,21 and the remaining study required patients to have poor performance status (PS 2-3).22 The trials in unselected patients included a total of 293 patients and reported response rates ranging from 5 percent to 61 percent but with most falling between 25 percent and 33 percent (Table 6). The two studies in patients with BAC histology included patients who had been previously treated, but only one of these studies18 has yet reported the results separately for these two groups.

The overall response rate in the two studies of BAC with gefitinib 500 mg daily and erlotinib 150 mg daily was 18 percent and 24 percent, respectively. In the study of gefitinib 500 mg daily, patients without prior treatment had a response rate of 21 percent compared to 10 percent in previously treated patients; however, the overall survival of the two groups was similar. In the study restricted to previously untreated elderly patients with advanced NSCLC, a response rate of 12 percent was reported.21 Finally, in a single study (18 evaluable patients) of patients with poor performance status, no responses were observed.22

Survival data among all groups of studies show median survival of 10 to 12 months; some studies reported 1-year survival which ranged from 43 to 76 percent. Notably, compiled data from the EAP showed a median survival of 6 months and 1-year survival rate of 29.7 percent.23

Second- or third-line treatment

Two randomized, placebo-controlled trials of single agent EGFR-TK inhibitors in previously treated patients have been completed and reported in preliminary form (Table 7). Both studies provided best supportive care to patients on both arms. In the BR21 trial, patients received either erlotinib 150 mg daily or placebo.24,25 As expected based on the earlier phase II study of erlotinib, the response rate (complete response plus partial response) was 9 percent versus < 1 percent in the placebo arm (Table 8). The median survival of 6.7 months in the erlotinib arm was 2 months longer than the placebo arm (Table 9). Progression-free survival and overall survival also significantly favored the erlotinib arm, with hazard ratios of 0.61 and 0.70, respectively.

Multiple subgroup analyses indicated a survival advantage in every group examined except smoking status and possibly EGFR expression. The survival benefit among non-smokers (21 percent of subjects) was significantly greater than among the current or former smoker subgroup. Although survival was better among the smoker subgroup in the erlotinib arm, this advantage did not reach statistical significance. The availability of appropriate tumor tissue from only one-third of study subjects limited the analysis of EGFR expression correlation with survival, which suggested the possibility of a greater survival benefit in the EGFR-positive group.

In contrast, a larger study of gefitinib in previously treated patients, which has been reported only as a press release from the pharmaceutical sponsor,26 failed to demonstrate a statistically significant advantage for the gefitinib-treated group (hazard ratio = 0.89, p = 0.11). Subgroup analysis suggested survival advantage in never smokers and East Asian patients, but not among those with adenocarcinoma. No quantitative estimates are available in the limited reporting in the press release; further analyses and more detailed reporting are expected.

Of six prospective phase II studies of EGFR-TK inhibitors in patients with advanced, previously treated NSCLC, four used gefitinib and two erlotinib (Table 7). Patients in two gefitinib studies were randomized to either 250 mg or 500 mg daily, while the third used only the 250 mg dose. Response rates were not significantly higher in the 500mg arms (Table 8). For the 250mg dose arms, objective response rates were 12 percent and 18.5 percent in the randomized dose comparisons, and 5 percent, 10 percent and 12 percent in the other phase II trials. Median survival was 7 and 8 months in the two larger studies, compared to 5 and 4 months in the others (not reported in one). The lower response rates in the latter studies suggest that patient selection may contribute to observed tumor response (Table 9).

The two phase II studies of erlotinib found objective response rates of 10 percent and 12.3 percent.27,28 One erlotinib study required patients to have tumor samples that expressed EGFR, while the other, like the phase II gefitinib studies, did not. In this study of 57 patients, 27 a response rate of 12.3 percent and median survival of 8.4 months was similar to those in the larger gefitinib studies.

Quality of life

Quality of life data were reported for some of the clinical trials of EGFR-TK inhibitors (Table 10). The only controlled trial demonstrated a significant average improvement in quality of life symptom measures associated with erlotinib versus placebo.24,25 Other data on erlotinib and all data on gefitinib are uncontrolled.29-31

No studies directly compared quality of life outcomes associated with EGFR-TK treatment to those associated with cytotoxic chemotherapy; however, two controlled trials suggest that improvements in quality of life measures can also occur with pemetrexed or docetaxel regimens.32,33 However, while the majority of patients in the pemetrexed and docetaxel arms were stable or improved, the proportions worsening (33 percent and 27.9 percent) exceeded the proportions that improved (21.2 percent and 21.6 percent) in each arm.32 Based on these data, this study would likely not show average improvement in quality of life.

Shepherd et al.,33 using the EORTC QLQ-C30 instrument with the LC13 lung cancer module, a disease-specific quality of life measure, found a trend toward less deterioration in quality of life measures associated with docetaxel than with best supportive care; however, this effect was not statistically significant. The sparse data on quality of life outcomes, the wide variety of quality of life measures, and the lack of direct comparative trials makes the comparison between gefitinib and docetaxel in terms of quality of life outcomes uncertain.

Efficacy data are available for studies of other cytotoxic chemotherapies in previously treated patients with NSCLC (Table 11). Docetaxel has been shown to have survival advantage over best supportive care33 and over vinorelbine or ifosfamide,34 with identical survival outcomes to pemetrexed.32 No prospective clinical trial data (phase III or phase II) are available for any treatments beyond second-line. Massarelli, et al.35 describe similar survival associated with various third- and fourth-line chemotherapy regiments in a small retrospective study.

These trials of traditional chemotherapy have similar inclusion criteria and similar demographic characteristics to Shepherd et al.25 (erlotinib versus placebo), but tend to have fewer patients with PS > or = 2 (ranging from 11 percent to 24 percent versus 35 percent) and fewer patients with > or = 2 prior chemotherapy regimens (ranging from 25 percent to 35 percent versus 49 percent). The ISEL study (gefitinib versus placebo) is not reported in sufficient detail to evaluate the comparability of the patient populations. The differences noted might lead one to expect slightly poorer survival in the EGFR-TK trials as compared to these second- and third-line chemotherapy trials; however, such indirect comparisons are subject to biases (due to unmeasured or unreported factors) that may be greater than the effects of the variables noted.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care