lung cancer image

EGFR TKIs in the Maintenance Setting

David R. Spigel, MD

David R. Spigel, MD
Sarah Cannon Research Institute
Nashville, Tennessee

See Also: Discussion, References

Introduction

For unselected patients with newly diagnosed advanced non-small-cell lung cancer (NSCLC), the median overall survival (OS) is 8-10 months.1-2 Platinum-based chemotherapy is standard treatment in the first-line setting where the median progression-free survival (PFS) is 4-6 months. Historically, extending first-line therapy for patients who are benefitting from treatment has been limited by toxicity, and has not been shown to improve response rates or survival.3-6 Patients who are still candidates for additional therapy routinely receive single agent chemotherapy or the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs), erlotinib or gefitinib, as second-line treatment at the time of disease progression. The median PFS for patients who are treated in the refractory setting is 2-3 months.3-4

Extending, or maintaining, first-line treatment beyond 4-6 cycles for those patients who have responding or stable disease and no unacceptable toxicity can potentially prolong disease control and survival.

EGFR TKIs and Refractory NSCLC Erlotinib and gefitinib are oral EGFR TKIs approved for the treatment of advanced NSCLC. These agents have been extensively studied in all lung cancer treatment settings in both unselected and selected patients. The pivotal trial that led to the Food and Drug Administration's approval of erlotinib in the United States was BR21.3 Erlotinib was shown to improve OS compared with placebo in patients with refractory NSCLC, regardless of histology. Gefitinib's approval was based on positive results from phase II studies where it was used as a single agent in patients with refractory NSCLC.5-6 Gefitinib was associated with low response rates (RRs) and symptom improvement. Both agents are used worldwide, however, gefitinib's use in the U.S. is restricted to patients who have benefitted from gefitinib treatment, after a randomized phase III trial of 1,692 patients failed to demonstrate an OS advantage for gefitinib compared with placebo (HR .89, p=.087).7

Table 1. EGFR TKIs in the Maintenance Setting: In Combination with, and Following, First-Line Chemotherapy

Study Treatment Arms N Findings
INTACT-1 8 Cisplatin
Gemcitabine
Gefitinib
1093 No advantage for gefitinib in terms of: RR, TTP, or OS
Cisplatin
Gemcitabine
Placebo
INTACT-2 9 Carboplatin
Paclitaxel
Gefitinib
1037 No advantage for gefitinib in terms of: RR, TTP, or OS
Carboplatin
Paclitaxel
Placebo
TRIBUTE 10 Carboplatin
Paclitaxel
Erlotinib
1059 No advantage for erlotinib in terms of: RR, TTP, or OS*
Carboplatin
Paclitaxel
Placebo
TALENT 11 Cisplatin
Gemcitabine
Erlotinib
1172 No advantage for erlotinib in terms of: RR, TTP, or OS*
Cisplatin
Gemcitabine
Placebo
RR: Response Rate
TTP: Time to Progression
OS: Overall Survival
* There was an OS advantage in favor of erlotinib in a subset of patients who were nonsmokers

EGFR TKIs in the Maintenance Setting

In Combination with, and Following, First-Line Chemotherapy

An early goal in the development of EGFR TKIs was to demonstrate that these agents could enhance the treatment benefits of modern platinum-doublet chemotherapy and extend, or maintain, disease control beyond 4-6 cycles of chemotherapy. Multiple randomized trials were designed to study these agents in combination with, and following, chemotherapy.(Table 1) INTACT-1 was a randomized phase III study comparing cisplatin, gemcitabine, and gefitinib (2 dose levels) with chemotherapy alone in 1,093 previously untreated patients with advanced NSCLC.8 All patients received up to 6 cycles of combination therapy followed by maintenance gefitinib/placebo until disease progression. The study failed to demonstrate any advantage for gefitinib in terms of RR, time to progression (TTP), or OS. INTACT-2 was designed to compare carboplatin, paclitaxel, and gefitinib (2 dose levels) with chemotherapy alone in 1,037 patients treated in the first-line setting.9 Similarly, this trial failed to demonstrate an advantage for gefitinib in terms of RR, TTP, or OS. Erlotinib was studied in a randomized phase III trial (TRIBUTE) in 1,059 patients with newly diagnosed advanced NSCLC.10 Patients were randomized 1:1 to chemotherapy (carboplatin + paclitaxel) and erlotinib or to chemotherapy and placebo. Erlotinib/placebo was continued until disease progression following up to 6 cycles of chemotherapy. As with gefitinib, erlotinib did not improve RR, TTP, or OS. However, in a small subset of patients who were never smokers, erlotinib improved survival (22.1 v. 10.1 months, HR .49, p=.01). Erlotinib has also been studied in combination with cisplatin and gemcitabine in the TALENT trial.11 This phase III trial enrolled 1,172 patients with newly diagnosed advanced disease. There were no differences in OS, RR, TTP, or quality of life. However, as in TRIBUTE, there was a small subset of never smokers who experienced improvement in OS with erlotinib.

These 4 phase III trials failed to demonstrate any advantage for EGFR TKIs in unselected patients when combined with, and following, first-line chemotherapy. These findings were unexpected given the positive benefits of chemotherapy alone and the single agent disease control rates with erlotinib and gefitinib in unselected patients with refractory disease. One potential explanation for these findings was the antagonistic relationship of EGFR TKIs and chemotherapy.12 Cell-cycle arrest induced by EGFR TKIs in EGFR wild-type tumors, makes cells less vulnerable to chemotherapies which depend on cell-cycling for activity. However, a more compelling explanation for these negative trials is that EGFR TKIs work best in patients selected for the presence of EGFR activating mutations.13-14

Use as First-Line Single-Agents until Progression

More recently, these agents have moved into the first-line treatment setting in patients with known EGFR mutations. The high single-agent efficacy of EGFR TKIs in patients selected for mutations provided strong rationale to move these agents into the first-line single-agent setting. Erlotinib and gefitinib have each been compared to chemotherapy in patients newly diagnosed with advanced disease in multiple randomized trials.15-21 EGFR TKIs were administered daily and maintained until disease progression or unacceptable toxicity. EGFR TKIs lead to superior PFS and RR compared with first-line platinum doublet chemotherapy in these selected patients. Importantly, improvements in OS have been difficult to demonstrate. This is likely confounded by the crossover of patients on placebo to EGFR TKI treatment which is widely available as post-study treatment.

Another approach to maintaining disease control achieved with first-line chemotherapy has been to switch patients after 4-6 cycles to a new regimen: switch maintenance.

Switch Maintenance Following First-Line Chemotherapy

EGFR TKIs have also been studied in the maintenance setting following first-line chemotherapy. Patients who have achieved a response or stable disease with first-line chemotherapy are switched to a new agent to extend or maintain disease control. SATURN was a randomized phase III trial designed to study the role of erlotinib in patients who had achieved disease control following 4 cycles of platinum-doublet chemotherapy.22 (Table 2) In this trial 889 patients were randomized 1:1 to erlotinib 150 mg orally daily or to placebo until disease progression or excessive toxicity. The trial's co-primary endpoints were PFS in all patients, and PFS in those patients with EGFR-immunohistochemistry-positive cancers. The switch to erlotinib maintenance resulted in improved PFS in both the overall cohort (12.3 v. 11.1 weeks, HR 0.71, p<.0001), and in the EGFR-expressing cohort (12.3 v. 11.1 weeks, HR 0.69, p<.0001). Maintenance erlotinib also improved overall survival (HR .77, p=.024), a noteworthy finding in an unselected group of patients. Moreover, in a planned subset analysis of patients with EGFR mutations, the HR for PFS was 0.10 in favor of erlotinib (p<.001). An analysis for OS did not show an advantage for erlotinib in this small subset, however, this was likely due to the crossover effect of patients on the placebo arm who subsequently received erlotinib off-study.

Table 2. EGFR TKIs in the Maintenance Setting: Switch Maintenance Following First-Line Chemotherapy
Study Treatment Arms N PFS OS Placebo Crossover to EGFR TKI at PD (%)
SATURN 22 Erlotinib 884 12.3 weeks
HR .71
p<.0001
12 months
HR .77
p=.024
-
Placebo 11.1 weeks 11 months 21%
ATLAS 23 Erlotinib + Bevacizumab 768 4.8 months
HR .72
p<.0012
15.9 months
HR .90
p=.269
-
Placebo + Bevacizumab 3.7 months 13.9 months 55%
IFCT-GFPC 0502 24 Erlotinib 464 2.9 months*
HR .83
p=.002
HR .91 2nd-Line Pemetrexed mandated: 55% received
Gemcitabine 3.8 months*
HR .51
P<.0001
HR .86 72% received
Observation 1.9 months   60% received
EORTC 08021-ILCP 01/03 25 Gefitinib 173** 4.1 months
HR .61
P=.002
10.9 months***
HR .81
P=.20
-
Placebo 2.9 months 9.4 months NR
INFORM 22 Gefitinib 296 4.8 months
HR .42
P,.0001
18.7 months
HR .84
P=.26
-
Placebo 2.6 months 16.9 months 32%
PD: Progressive Disease
PFS: Progression-Free Survival
OS: Overall Survival
* Compared with the control arm
** Trial Closed Early
*** Primary Endpoint
NR: Not Reported

Erlotinib was also studied in a maintenance setting following first-line chemotherapy in the ATLAS trial.23 This was a phase III trial designed to study the combination of erlotinib and bevacizumab or placebo and bevacizumab following first-line platinum-doublet chemotherapy and bevacizumab in patients with nonsquamous NSCLC. Maintenance erlotinib and bevacizumab improved PFS (4.8 v. 3.7 months, HR .72, p=.0012) compared with bevacizumab alone, however did not improve OS (a secondary endpoint).

Perol et al randomized 464 patients to observation alone or maintenance gemcitabine or erlotinib following 4 cycles of first-line cisplatin and gemcitabine in a three-arm randomized phase III trial.24 All patients received pemetrexed at progression. This was the only trial to mandate the choice of therapy at the time of progression. PFS was significantly improved by gemcitabine (HR .51, p<.0001), and erlotinib (HR .83, p=.002) compared with observation alone. However, OS was not improved with either maintenance gemcitabine or erlotinib.

Gefitinib has also recently been studied in the maintenance setting by Gaafar et al.25 One-hundred seventy-three patients with stable disease following 4 cycles or platinum-based chemotherapy were randomized to gefitinib 250 mg orally daily or placebo until disease progression or excessive toxicity. The study's primary endpoint was OS. This trial was closed early due to poor accrual. There was no advantage for gefitinib in the OS analysis (HR .81, p=.2), however, PFS was improved (HR .61, p=.002) in favor of gefitinib.

Finally, a randomized phase III trial looking at the role of maintenance gefitinib in Chinese patients following disease control with chemotherapy was recently reported.26 The INFORM trial randomized 296 patients to maintenance gefitinib or placebo. Gefitinib improved PFS (4.8 v. 2.6 months, HR .42, p<.001), but did not improve OS (18.7 v. 16.9 months, HR .84, p=.26). Thirty-two percent of patients on the placebo arm crossed-over to receive second-line targeted therapy.

One criticism of this approach to treatment is that it is simply moving second-line therapy up earlier in a patient's plan of care. Earlier treatment should improve PFS compared with no therapy, thus the positive PFS results from these trials are expected. Whether these patients could do just as well if they waited to receive treatment only at progression is unclear. SATURN is the only trial to demonstrate a survival advantage, but only 21% of patients in the placebo cohort went on to receive post-study EGFR TKI treatment.

Conclusion

The consensus role for EGFR TKI use today is as first-line treatment in patients with EGFR activating mutations. Switch maintenance use of erlotinib, and EGFR TKI use in for refractory disease, are settings that will likely increasingly be limited to unselected patients and patients with EGFR wild-type tumors.

Extending, or maintaining, first-line treatment beyond 4-6 cycles for those patients who have responding or stable disease and no unacceptable toxicity can potentially prolong disease control and survival. The single agent activity in the refractory setting, oral dosing, and acceptable safety profiles of the EGFR TKIs make these agents ideal candidates for maintenance use. However, randomized studies evaluating these agents in the first-line setting in combination with, and following, chemotherapy failed to demonstrate any advantages in terms of RR, PFS, or OS in unselected patients. Conversely, trials comparing these agents with chemotherapy in the first-line setting in patients with EGFR mutations have consistently demonstrated improvements in PFS. EGFR TKIs can be used to maintain disease control for extended cycles in the majority of these selected patients. Recently, another approach to maintaining disease control achieved with first-line chemotherapy has been to switch patients after 4-6 cycles to a new regimen (i.e. switch maintenance). Five randomized trials using EGFR TKIs in this setting have proven to extend PFS in unselected patients. However, only one trial with erlotinib demonstrated that this strategy also improved OS. Survival analyses from the other studies may have been confounded by the crossover effect of patients receiving EGFR TKIs as subsequent treatments at the time of first progression.

EGFR TKIs have been extensively studied in multiple lung cancer treatment settings. American Society of Clinical Oncology (ASCO) and National Comprehensive Cancer Center (NCCN) guidelines now support maintenance erlotinib in unselected patients who have benefitted from first-line chemotherapy.27-28 However, the consensus role for EGFR TKI use today is as first-line treatment in patients with EGFR activating mutations, which is also endorsed by these guidelines. Switch maintenance use of erlotinib, and EGFR TKI use in for refractory disease, are settings that will likely increasingly be limited to unselected patients and patients with EGFR wild-type tumors.

Back to top