Maintenance therapy in nonsmall-cell lung cancer

A new treatment paradigm

Abstract

Systemic chemotherapy with platinum-based regimens provides modest improvements in survival and quality of life for patients with advanced-stage nonsmall-cell lung cancer (NSCLC). Extended first-line chemotherapy with combination regimens for more than 4 to 6 cycles is not recommended because of cumulative toxicities and lack of proven advantage in survival with the increased duration of therapy. The early use of an anticancer agent as maintenance therapy after disease stabilization or maximal response with platinum-based regimens is, therefore, being recognized as a new treatment paradigm in NSCLC. Maintenance therapy can extend first-line treatment and provide an acceptable balance between efficacy and toxicity. The essential prerequisites for maintenance therapy include good tolerability, ability to administer extended cycles of therapy without cumulative toxicity, and an increase in the duration of progression-free survival. Pemetrexed has recently been shown to improve the median PFS in the maintenance setting. Molecularly targeted therapies with cytostatic properties and documented tolerability also have the potential to be effective as maintenance therapy to maintain tumor regression after an initial response to chemotherapy. Consequently, the role of erlotinib and other molecular targeted agents in this treatment setting is under active investigation in ongoing phase 3 trials. This could potentially establish a new standard on the clinical utility of molecular targeted agents as maintenance therapy for patients with advanced-stage NSCLC. Cancer 2009. © 2009 American Cancer Society.

Systemic chemotherapy with platinum-based regimens provides modest improvements in survival and quality of life (QoL) for patients with advanced-stage nonsmall-cell lung cancer (NSCLC). Several mechanistically distinct combination regimens have been developed and are in routine clinical use. However, the efficacy of standard platinum-based doublets for NSCLC has reached a plateau in the first-line therapy setting for advanced-stage disease.1 Recently, novel molecular targeted agents and the use of maintenance therapy have been investigated as new approaches to improve the outcome for patients with advanced stage NSCLC. Maintenance therapy is the continued administration of therapy after a specified number of treatment cycles once disease stabilization or maximum tumor response has been achieved; this could comprise1 or more of the agents used in the first-line therapy regimen or the selection of another noncross-resistant agent. Maintenance therapy may be continued for a defined period of time until either disease progression or unacceptable toxicity.

The terms “maintenance therapy” and “consolidation therapy” are not clearly defined in the literature and are frequently used interchangeably.2 Grossi et al define maintenance/consolidation chemotherapy as the prolongation of chemotherapy duration with the administration of additional drugs at the end of a defined number of initial chemotherapy cycles after achieving a maximum tumor response in an individual patient, whereas sequential chemotherapy is a treatment given after a noncross-resistant therapy for a defined number of cycles.2 However, in the context of NSCLC, “consolidation therapy” is a term generally used to describe continuation of therapy after concomitant chemotherapy and radiation for stage III locally advanced disease, while “maintenance therapy” refers to the treatment after standard systemic chemotherapy for advanced stage disease. Admittedly, universal consensus on definition of these terms has not been reached. For the benefit of this review, we adopted the definition by Grossi et al.

The initiation of maintenance therapy requires the lack of progressive disease with front-line therapy. A change in treatment regimen due to an unfavorable outcome with first-line chemotherapy (disease progression or unacceptable toxicity) is usually referred to as second-line therapy. However, if there is no evidence of tumor progression, ie, the patient demonstrates a complete response (CR), partial response (PR), or stable disease (SD), the immediate subsequent treatment regimen is usually referred to as maintenance therapy.

Rationale for Maintenance Therapy

Administering a different therapeutic regimen after first-line treatment has several theoretical advantages. The Goldie and Coldman hypothesis states that the proportion of chemotherapy-resistant cells in a tumor increases over time3; consequently, the early use of noncross-resistant treatment regimens may increase the probability of killing more cancer cells before resistance occurs. Patients who have an objective response or disease stabilization after initial chemotherapy may benefit from maintenance therapy when the tumor burden is low.4 Moreover, newer, third-generation cytotoxic therapies and targeted therapies are more effective than older treatments, have a good tolerability profile, and treatment can be extended for a longer period of time, suggesting they may be beneficial in the maintenance setting. The cytostatic and antiproliferative properties of biologic agents may continue to maintain the tumor responses achieved by first-line therapy (CR/PR or SD) and potentially provide a further increase in progression-free survival and/or overall survival (OS). On the basis of this, benefit is more likely expected from maintenance therapy with 1 or more agents that have different mechanisms of action from those of first-line cytotoxic therapies. Conversely, continuation of 1 of the active agents in the combination could be another way to sustain the benefit associated with front-line therapy. An agent without cumulative toxicity would be ideally suited for this.

Maintenance Therapy With Combination Regimens

The optimal duration of combination chemotherapy in advanced disease has been often debated because of the occurrence of cumulative toxicity and no proven advantages in clinical response with increased duration of chemotherapy. The guidelines published by the American Society of Clinical Oncology in 2004 state that first-line chemotherapy treatment should be limited to 4 cycles in nonresponsive patients and 6 cycles in responsive patients, because of the therapy-associated risk of cumulative toxicity.5 The 2008 National Comprehensive Cancer Network clinical practice guidelines also recommend a total of 4 to 6 cycles of first-line chemotherapy in patients with advanced disease.6 These guidelines were based on phase 3 trials that demonstrated no significant differences in survival or QoL between patients undergoing a short- or long-duration of combination chemotherapy.7, 8

Two randomized studies have addressed the issue of the duration of first-line therapy (Table 1). Socinski et al8 randomly assigned 230 patients with stage IIIB/IV disease to 4 cycles of carboplatin plus paclitaxel or continuation of therapy until disease progression. There was no benefit with continuation of chemotherapy beyond 4 cycles as noted by the similar OS, response rates (RR), and 1-year survival rate between the 2 treatment arms. There was a significant increase in neuropathy in patients in the continuous therapy study arm and no difference in QoL between the 2 study arms. A more recent study by Park et al9 also confirmed that the OS associated with 4 cycles of chemotherapy was not inferior to 6 cycles of chemotherapy for first-line treatment of advanced disease. A total of 452 patients were randomly assigned to receive either 4 or 6 cycles of third-generation, platinum doublet, first-line therapy. Despite an increase in progression-free survival with 6 cycles of chemotherapy, the 1-year survival rate (59% vs 62%) and RR were similar between 4 and 6 cycles of therapy. Significantly, patients who received 4 cycles of chemotherapy had less toxicity and regained their functional status more rapidly than patients assigned to receive 6 cycles.9

Table 1. Selected Phase 3 Trials Evaluating the Duration of First-Line Therapy

Study Author	Duration of First-Line Therapy	Efficacy			Toxicity	QoL Results
		ORR (CR+PR), %	PFS, mo	OS, mo
ORR indicates objective response rate; CR, complete response; PR, partial response; PFS, progression-free survival; OS, overall survival; QoL, quality of life; PD, progressive disease; NR, not reported. * Total patients enrolled/randomized. † Time to progression.
Socinski7 n=230*	Study arm A: 4 cycles of carboplatin+paclitaxel vs Study arm B: continuous treatment with carboplatin+paclitaxel until PD	Study arm A: 223 Study arm B: 24 P=.8	NR	Study arm A: 6.6 Study arm B: 8.5 P=.63	Overall rates of hematologic (neutropenia, thrombocytopenia, anemia) and non-hematologic similar on both arms; except Gr neuropathy (study arm A: 14%, vs study arm B: 27%, P=.02)	No statistical difference in QoL scores between 2 study arms
Park8 n=314*	Study arm A: 6 cycles of carboplatin+paclitaxel vs Study arm B: 4 cycles of carboplatin+paclitaxel	Study arm A: 42.4 Study arm B: 42.6	Study arm A: 6.2† Study arm B: 4.6†P=.001	Study arm A: 14.9 Study arm B: 15.9 P=.461	Overall rates of hematologic and non-hematologic similar on both study arms	No statistical difference between the 2 study arms until completion of 4 cycles of therapy. After 4 cycles, patients in study arm B showed improved role-functioning, reduced nausea/vomiting, sore mouth, and dyspnea (all P < .05) compared with study arm A.

Similar results were noted by Smith and co-workers7 in a randomized trial comparing 3 versus 6 cycles of mitomycin, vinblastin, and cisplatin chemotherapy in patients with advanced NSCLC. No difference in OS or 1-year survival rate was noted with 6 cycles of chemotherapy, and RR and time to disease progression were identical between the 2 treatment arms.

These studies demonstrate that continuation of combination chemotherapy beyond a defined number of cycles did not provide additional benefit, but it was associated with higher toxicity. Studies of combination regimens beyond 6 cycles had a primary endpoint of OS; however, the trials failed to show significant improvements in OS and resulted in cumulative toxicities, leading to a discontinuation of further clinical investigation. With the availability of newer agents that have demonstrated efficacy beyond the first-line setting for advanced NSCLC (docetaxel,10 pemetrexed,11 and erlotinib12), some oncologists believe that it may be prudent to limit first-line cytotoxic therapies to 4 cycles and continue treatment with a single agent as maintenance therapy to sustain the benefit with first-line therapy.13

Single-Agent Maintenance Therapy

Maintenance therapy is increasingly being recognized as a new treatment paradigm that can extend first-line treatment and provide a more acceptable balance between efficacy and toxicity. The essential prerequisites for novel maintenance therapy should include good tolerability, ability to administer extended cycles of therapy without cumulative toxicity, and proven efficacy. Data from recent studies with single-agent maintenance therapy have provided validation for this new approach.

Maintenance therapy with chemotherapeutic agents

Third-generation chemotherapeutic agents, such as paclitaxel, gemcitabine, vinorelbine, and docetaxel, have been evaluated as single-agent maintenance therapy (Table 2). Two studies have evaluated paclitaxel as maintenance therapy after first-line chemotherapy with carboplatin and paclitaxel.14, 15 In addition to good tolerability, maintenance therapy with paclitaxel was associated with a longer time to progression and increased median OS. However, neither of the 2 studies was designed for primary evaluation of the role of maintenance therapy. In 1 study, maintenance therapy with paclitaxel was administered to patients in both treatment arms, after the completion of the first-line carboplatin and paclitaxel regimen. Therefore, the role of paclitaxel as maintenance therapy is yet to be established in a phase 3 study.

Table 2. Phase III Trials of Maintenance/Consolidation Therapy With Chemotherapeutic Agents

Study/Author (Total Patients Enrolled/Randomized)	First-Line Therapy	Maintenance Therapy	Efficacy			Most Frequent Grade 3/4 AEs During Maintenance Therapy (% of Patients)
			ORR (CR+PR), %	PFS, mo	OS, mo
ORR indicates objective response rate; CR, complete response; PR, partial response; PFS, progression-free survival; OS, overall survival; AE, adverse event; pac, paclitaxel; gem, gemcitabine; vin, vinorelbine; doc, docetaxel; pem, pemetrexed; NR, not reported; NS, not significant. * This trial continued paclitaxel as maintenance therapy in both study arms following first-line therapy with carboplatin and paclitaxel; ORR and OS data reported are from the combined first-line + maintenance therapy regimen and time to progression data are from the maintenance phase of trial. † Time to progression.
Belani (N=444)14	Carboplatin + weekly pac vs carboplatin every 3 weeks + pac	Pac*	Weekly pac: 27.6; every 3 weeks pac: 19.2; P = .037	Weekly pac: 8.3†; every 3 weeks pac: 7.3^¤	Weekly Pac: 9.7; every 3 weeks pac: 10.7	Neutropenia (2.1), fatigue (2.8), neuropathy (2.1), arthralgia (2.1), abdominal pain (2.1), dyspnea (2.1)
Brodowicz (N=352)3	Cisplatin-gem	Gem vs placebo	Gem: 50.7; placebo: 45.6; P = NS	Gem: 3.6†; placebo: 2.0†; P < .001	Gem: 10.2; placebo: 8.1; P = NS	Neutropenia (14.9), alopecia (4.3), anemia (2.6), leukopenia (2.3)
Westeel (N=573)15	Mitomycin-ifosfamide-cisplatin	Vin vs placebo	Vin: 57.8	Vin: 5; placebo: 3; P = NS	Vin: 12.3; placebo: 12.3; P = NS	Leukopenia (46), infection (13), anemia (9), neuropathy (7)
Fidias (N=552)17	Cisplatin-gem	Doc (maintenance vs salvage)	Maintenance doc: 42.5; salvage doc: 9.9	Maintenance doc: 6.5; salvage doc: 2.8; P < .0001	Maintenance doc: 11.9; salvage doc: 9.1; P = NS	Neutropenia (26), febrile neutropenia (2.8)
Ciuleanu (N=663)19	Platinum-doublet	Pem vs placebo	NR	Pem: 4; placebo: 1.97; P < .00001	Pem: 13; placebo: 10.2; P = NS	Fatigue (4.3), anemia (2.7), neutropenia (2.7)

Westeel and colleagues compared maintenance vinorelbine therapy with observation in patients with advanced NSCLC who had responded to first-line treatment with mitomycin-ifosfamide-cisplatin. OS and progression-free survival did not differ between the 2 study arms, and the toxicity of maintenance vinorelbine appeared greater than previously reported with vinorelbine treatment in chemo-naive patients.16 These findings are not totally surprising because vinorelbine is not associated with any anticancer activity in NSCLC patients who progressed with first-line platinum-based chemotherapy.

In another randomized study, Brodowicz and coworkers reported benefit with gemcitabine as maintenance therapy in patients demonstrating a response to, or disease stabilization with, 4 cycles of cisplatin and gemcitabine. Although no significant differences were noted in the RR or response duration between gemcitabine maintenance and observation, the median progression-free survival was significantly prolonged. However, the difference in median OS did not reach statistical significance (13 months vs 11 months, P = .19). Sixteen percent of patients with disease stabilization or response to therapy were excluded from randomization for various reasons and treatment delays were noted in 22% of patients randomized to gemcitabine maintenance therapy. Despite these limitations, this study validates the suitability of a well-tolerated single agent as maintenance therapy.

Docetaxel is approved for both first- and second-line treatment of advanced NSCLC.17 A phase 3 study by Fidias et al evaluated the efficacy and safety of docetaxel as maintenance (early) therapy versus docetaxel as salvage (delayed) therapy in patients who had received 4 cycles of carboplatin and gemcitabine (Fig. 1). Patients who received docetaxel as maintenance therapy experienced a prolonged progression-free survival and better overall RR, but only a trend toward improvement in survival (11.7 months vs 9.1 months) compared with patients who received docetaxel as salvage therapy. The toxicity profile of both docetaxel study arms was comparable and no difference in patient QoL was observed between the treatment arms.18 A potential flaw in the design of this trial required computed tomography (CT) evaluations only every 3 months on the delayed docetaxel study arm, and the lack of survival advantage could be at least, in part, because of the finding that approximately 40% of the patients randomized to the delayed docetaxel study arm did not receive the planned therapy for various reasons. This study also illustrates the difficulties in designing a study to evaluate maintenance therapy because the drop-out by itself could impact the outcome of the study.

Figure 1. Depicted is the trial design of the phase 3 trial of immediate versus delayed docetaxel after first-line therapy with gemcitabine and carboplatin in advanced NSCLC.17 NSCLC indicates nonsmall-cell lung cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; AUC, area under the concentration-time curve; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; BSC, best supportive care; OS, overall survival; PFS, progression-free survival.

Pemetrexed, a multitargeted antifolate compound, is currently approved for second-line therapy of advanced NSCLC19 and has recently gained approval in the European Union and the United States for first-line therapy in combination with cisplatin for patients with nonsquamous histology. A recent phase 3 study that evaluated the role of pemetrexed as maintenance therapy in pemetrexed-naive patients demonstrated significant improvement in progression-free survival when compared with placebo (Fig. 2).20 Patients with a response or SD after 4 cycles of platinum-based, first-line chemotherapy that did not contain pemetrexed were randomized to receive either pemetrexed or placebo as maintenance therapy. Patients taking pemetrexed had a significantly longer progression-free survival than those taking placebo (4.04 months vs 1.97 months; P < .00,001). The benefit of pemetrexed was limited to patients with nonsquamous histology according to a preplanned subgroup analysis. Preliminary results also favored a trend toward improvement in OS with pemetrexed as maintenance therapy (13 months vs 10.2 months; P = .06) and the final results are awaited. Results from the analysis of survival data by histology were similar to those of the general population: prolonged median OS with pemetrexed compared with placebo in patients with adenocarcinoma or nonsquamous histology, and shorter OS with pemetrexed in patients with squamous cell histology. Pemetrexed treatment was associated with greater hematologic toxicity than placebo, but the toxicity was considered mild. This study suggests that pemetrexed is clinically ineffective as maintenance therapy in patients with squamous cell carcinomas. One criticism of this trial is that about 50% of the patients in the placebo study arm never received additional therapy upon progression, and of those who did receive additional therapy, only a minority received drugs known to improve OS. A concern, therefore, is that this trial simply tested whether pemetrexed is an effective second-line agent and did not address the question of timing of therapy, ie when pemetrexed should be administered.

Figure 2. Depicted is the trial design of the phase 3 trial of pemetrexed versus placebo as maintenance therapy for advanced NSCLC.19 NSCLC indicates nonsmall-cell lung cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; CR, complete response; PR, partial response; SD, stable disease; BSC, best supportive care; PD, progressive disease; PFS, progression-free survival; OS, overall survival.

Meta-analysis of maintenance chemotherapy

Recently, the results of a meta-analysis that studied all the trials published in the literature for maintenance therapy of NSCLC were reported.21 Randomized trials that were included in this systematic review evaluated 1 of 3 treatment design regimens: 1) a defined number of cycles of chemotherapy versus continuation until disease progression, 2) a defined number of cycles of initial chemotherapy versus a greater number of cycles of the same chemotherapy, or 3) a defined number of cycles of initial chemotherapy versus the same initial chemotherapy followed by additional cycles of different chemotherapy. Overall, administration of a higher number of treatment cycles was associated with a significant improvement in progression-free survival (hazard ratio [HR], 0.78; 95% confidence interval [CI], 0.72-0.86; P < .001), but the impact on OS was not significant (HR, 0.94; 95% CI, 0.87-1.01; P = .10). The benefit was at the expense of increased adverse events that could impair patient QoL.21 When the investigators included the study of maintenance pemetrexed by Ciuleanu et al20 to their analysis, the improvement in OS reached statistical significance (HR, 0.92; 95% CI, 0.86-0.99; P = .03). Taken together, the data provide support to the notion of maintenance therapy in advanced-stage NSCLC and the benefit has to be balanced with potential increase in toxicity. It is important to note that the continued administration of cytotoxic therapies can present with cumulative toxicities, and the incidence of this has been limited in maintenance analyses.

Maintenance therapy with molecularly targeted agents

The clinical utility of targeting biologic processes and signaling pathways as treatment options in NSCLC has led to the development of biologic agents directed against the epidermal growth factor receptor (EGFR) and vascular endothelial growth factor signaling pathways.22, 23 Erlotinib, an inhibitor of the EGFR tyrosine kinase, is approved for the treatment of advanced NSCLC after progression with 1 or 2 prior chemotherapy regimens. Bevacizumab is approved in combination with carboplatin and paclitaxel for advanced nonsquamous NSCLC in the United States, and in combination with platinum-based therapy in the European Union. On the basis of the established activity of these agents in NSCLC and the well-tolerated toxicity profile, they are potentially ideal candidates for maintenance therapy.12, 24 The cytostatic properties of targeted therapies could work effectively in patients with minimal disease and maintain tumor regression after an initial response to chemotherapy. Table 3 summarizes the data from existing and ongoing comparative trials.

Table 3. Phase III Trials of Maintenance Therapy With Targeted Agents

Study/Author (Total Patients Enrolled/Randomized)	First-Line Therapy	Maintenance Therapy	Efficacy
			ORR (CR+PR), %	PFS, mo	OS, mo
ORR indicates objective response rate; CR, complete response; PR, partial response; PFS, progression-free survival; OS, overall survival; car, carboplatin; pac, paclitaxel; bev, bevacizumab; cis, cisplatin; gem, gemcitabine; NS, not significant; NR, not reported; HR, hazard ratio. * These trials continued bevacizumab as maintenance therapy following first-line therapy with a platinum-doublet and bevacizumab. Efficacy data reported are from the combined first-line therapy + maintenance therapy regimen. † Results are pending for SATURN and ATLAS.
E4599/Sandler (N=878)23	Car-pac-bev	Bev*	Car-pac-bev: 35; car-pac: 15; P < .001	Car-pac-bev: 6.2; car-Pac: 4.5; P < .001	Car-pac-bev: 12.3; car-pac: 10.5; P = .003
AVAiL (BO17704)/ Manegold (N=1043)24, 25	Cis-gem-bev (7.5 mg/kg and 15 mg/kg)	Bev*	Cis-gem-bev (7.5): 34; cis-gem-bev (15): 30; cis-gem: 20	Cis-gem-bev (7.5): 6.7; cis-gem-bev (15): 6.5; cis-gem: 6.1; P < .05	Cis-gem-bev (7.5): 13.6; cis-gem-bev (15): 13.4; cis-gem: 13.1; P = NS
WJTOG/ Hida (N=603)	Platinum-doublet	Gefitinib	NR	HR 0.68 (95% CI, 0.57-0.8; P = .03)	No improvement P = NS
SATURN trial (planned N=850)39	Platinum-doublet	Erlotinib vs placebo	Primary endpoint: PFS
			Secondary endpoints: OS, biomarkers, safety†
ATLAS trial (planned N=1150)41	Platinum-doublet + bev	Erlotinib/bev vs placebo/bev	Primary endpoint: PFS
			Secondary endpoints: OS, ORR, safety†

Two recent phase 3 trials of bevacizumab in combination with platinum-containing chemotherapy included the continued use of bevacizumab as maintenance therapy after initial response or disease stabilization to first-line therapy.24-26 The objective for both these studies (ECOG 4599 and AVAiL) was to evaluate bevacizumab plus chemotherapy as a first-line treatment option for patients with advanced NSCLC; patients then continued bevacizumab until disease progression. Both studies met their primary endpoints: bevacizumab treatment significantly improved patient OS in the ECOG 4599 study; the AVAiL study demonstrated that patients who received bevacizumab therapy had a significant improvement in progression-free survival. The AVAiL study analyzed OS as a secondary endpoint and the median OS in both bevacizumab study arms was similar to chemotherapy alone.

The EGFR tyrosine kinase inhibitors (TKIs), erlotinib and gefitinib, did not improve OS when given in combination with platinum-containing chemotherapy regimens in the first-line setting.27-30 The lack of benefit when EGFR TKI and cytotoxic platinum-based regimens are given in combination is hypothesized to be because of the ability of EGFR TKIs to block the effects of chemotherapy by causing G1 cell cycle arrest, thereby reducing the efficacy of platinum compounds.31 Thus, the effects of a chemotherapy and EGFR TKI combination may be sequence-dependent.32, 33 This hypothesis is supported by the results of the FAST-ACT trial, a placebo-control randomized phase 2 study of 150 unselected patients from Asia and Australia using gemcitabine and carboplatin on Day 1 and Day 8 subsequently followed by erlotinib from Days 15-28. All patients received erlotinib or placebo as maintenance therapy.34 Tumor RR was 37% versus 24% in favor of the sequential erlotinib study arm. Median progression-free survival was 7.2 months with erlotinib versus 5.5 months with placebo, HR = 0.57 (95% CI 0.38-0.84), and P = .005. Treatment with erlotinib was also associated with increased duration of response, nonprogression rate at 16 weeks and TTP. The high tumor RR suggested the benefit of sequential administration, while the prolonged progression-free survival may be explained by sequential administration with the combination and maintenance therapy with EGFR TKI. The observed benefit is unlikely to be solely related to the Asian ethnicity with higher incidence of EGFR mutation because this is an unselected population comprising males, smokers, and nonadenocarcinomatous tumors. A phase 3 trial being initiated in 2009 will evaluate this treatment regimen. This study allows for a novel approach to treatment using intercalated doses of EGFR TKI with systemic chemotherapy for patients with advanced-stage NSCLC.

Data from the erlotinib phase 3 trials, TRIBUTE and TALENT, suggested that erlotinib therapy may be effective as maintenance monotherapy after platinum-based chemotherapy.27, 28 In the TRIBUTE study, for the 861 patients who survived beyond 4 months (408 patients randomized to erlotinib and 453 randomized to placebo), the median OS with erlotinib treatment was 13.6 versus 12.2 months with placebo (P = .04). Of the 740 patients who survived beyond 6 months (erlotinib 348; placebo 392), median survival with erlotinib was 15.4 versus 13.8 months for placebo (P = .007).27 Likewise, in TALENT, patients treated with erlotinib for more than 150 days showed increased response duration with erlotinib compared with placebo.28 Somewhat similar results were observed in a retrospective subgroup analysis of the INTACT-2 trial of carboplatin plus gemcitabine with or without gefitinib as first-line therapy for NSCLC. Although no survival benefit was reported in the treatment arm of this trial, there was a trend toward longer survival in the subgroup of patients with adenocarcinoma in the experimental study arm, which included maintenance gefitinib.29 These data are suggestive of TKI efficacy as maintenance therapy for selected patients and warrant prospective evaluation of standard chemotherapy followed by maintenance targeted therapy for treatment of advanced NSCLC.

A randomized phase 3 trial conducted by the West Japan Thoracic Oncology Group evaluated the efficacy of gefitinib maintenance therapy after platinum-doublet chemotherapy in previously untreated patients with advanced disease. Eligible patients were randomized to receive either 3 cycles of chemotherapy followed by gefitinib maintenance therapy or 6 cycles of chemotherapy. Gefitinib maintenance therapy was associated with a significant improvement in progression-free survival duration (HR, 0.68; 95% CI, 0.57-0.80; P < .001) but not in OS (P = .10). A prespecified analysis of OS by subgroup showed a significant improvement in OS (HR, 0.79; 95% CI, 0.65-0.98; P = .03) with gefitinib maintenance in patients with adenocarcinoma histology.35 These results demonstrate the potential clinical benefit of gefitinib maintenance therapy in clinically selected patient groups. However, a recent phase 3 trial that prospectively evaluated maintenance treatment with gefitinib after chemoradiotherapy in patients with surgically unresectable, locally advanced stage III disease demonstrated that gefitinib was associated with a statistically significant decrease in patient survival, compared with placebo.36

The efficacy of erlotinib monotherapy has been demonstrated in patients with NSCLC in second- or third-line settings12; patients were not selected based on molecular or histological phenotypes. The efficacy of erlotinib appears to be comparable to other agents used second-line,37 but this has not yet been established in direct comparative studies. Erlotinib's favorable toxicity profile, single-agent activity,12 good QoL profile,38 and cost-effectiveness39 make it an attractive choice for use as maintenance therapy for advanced NSCLC. Erlotinib is associated with skin toxicities, however their impact is relatively mild and readily manageable over time.40 The SATURN (Sequential Tarceva in Unresectable Lung Cancer) phase 3 clinical trial is determining whether erlotinib is effective as maintenance therapy in advanced NSCLC.41 In this multicenter, double-blind, randomized study, patients (n = 850) with advanced (stage IIIB/IV) NSCLC were randomized to receive either erlotinib (150 mg/day) or placebo, after documented disease control (CR/PR/SD), after 4 cycles of standard platinum-based chemotherapy (Fig. 3). Treatment is continued until disease progression, unacceptable toxicity, or death. The primary endpoint of SATURN is to determine whether administration of maintenance erlotinib after standard platinum-based chemotherapy increases progression-free survival when compared with placebo in the overall population. The coprimary endpoint is to determine whether erlotinib administration results in improved progression-free survival compared with placebo in patients who are EGFR protein expression (by immunohistochemistry) positive. Secondary endpoints include OS between the 2 treatment arms for all patients, for patients who are EGFR protein expression positive, and for patients who are EGFR protein expression negative. A recent announcement was made by the sponsor that the study met its primary endpoint of erlotinib significantly extending patient progression-free survival.42 Full presentation of the data are awaited. The ATLAS trial is evaluating maintenance treatment with bevacizumab with or without erlotinib.43 Patients who do not demonstrate disease progression upon completing 4 cycles of chemotherapy (paclitaxel/carboplatin) plus bevacizumab are randomized to receive either maintenance bevacizumab alone or bevacizumab in combination with erlotinib. The primary endpoint of this trial is progression-free survival and secondary endpoints include OS and safety. Interim analysis was performed on the 1157 enrolled patients and the study was prematurely stopped because of the significant improvement in progression-free survival.44 (Media Release Basel, Feb 3, 2009).

Figure 3. Depicted is the trial design of the phase 3 SATURN trial of erlotinib as maintenance therapy for advanced NSCLC.39 NSCLC indicates nonsmall-cell lung cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; EGFR, epidermal growth factor receptor; IHC,; PFS, progression-free survival; OS, overall survival.

These studies may lead to a new standard on clinical utility of molecular targeted agents as maintenance therapy.

Assessment of Efficacy: Is Progression-Free Survival a Valid Endpoint in the Maintenance Setting?

In general, for phase 3 studies, OS is considered the most reliable study endpoint and is usually the preferred endpoint for regulatory approval.45 However, the use of second- and third-line therapies could potentially confound the effect of a first-line or maintenance regimen on OS in the treatment of advanced NSCLC. It is difficult to control for poststudy therapy in the maintenance and nonmaintenance study arms; outcomes with subsequent therapies may be influenced by many clinical and biological factors. This is increasingly relevant with the recent availability of multiple active agents for advanced NSCLC and for other tumor types. For example, in metastatic breast cancer, the addition of bevacizumab to chemotherapy resulted in a robust improvement in progression-free survival, but without a distinct survival advantage.46 These results were confounded by subsequent lines of therapy, including bevacizumab combined therapy. Likewise, in the IPASS trial comparing gefitinib to carboplatin/paclitaxel as first-line treatment for advanced NSCLC, gefitinib treatment resulted in a superior progression-free survival; however, the OS between the treatment arms was similar, potentially because of the 40% crossover rate in both treatment arms.47

Progression-free survival assessment is generally based on objective and quantitative tumor assessment; however, various frequencies of tumor evaluation and differences in interpretation of radiologic images that define progression could result in inconsistent progression-free survival measurement.48 To ensure accurate progression-free survival measurement, a strict schedule of efficacy assessment and patient follow-up should be implemented. In large clinical trials, tumor assessment is usually carried out every 6 weeks and to further increase the confidence in progression-free survival assessment, an independent review of radiographic images is frequently performed. However, progression-free survival does not account for symptomatic progression and the risk/benefit of progression-free survival depends on the agent administered. Supplementary information including QoL and symptom improvement may add value to the prolongation of progression-free survival.

In NSCLC, approximately one third of patients who receive first-line therapy do not get any subsequent therapy for a variety of reasons. The window of opportunity to confer a positive impact may, therefore, lie in the use of a maintenance regimen before disease progression. Anticancer agents that are cytostatic to tumor cells can delay tumor progression and lead to prolonged symptom-free periods. These agents, thus, aid in the maintenance of the disease-free status, and in such cases, progression-free survival reflects the duration of control of the disease process and a possible improvement in patient QoL. For all of these reasons, we believe that the use of progression-free survival represents a clinically meaningful endpoint for evaluation of agents as maintenance therapy for advanced-stage NSCLC.

Summary

Data from multiple recent studies have demonstrated a role for maintenance therapy after combination chemotherapy in advanced NSCLC. This has paved the way for a new treatment paradigm for patients with advanced-stage NSCLC. Extended first-line chemotherapy with combination regimens for more than 4 to 6 cycles is not recommended because of cumulative toxicities and the lack of proven advantages in survival with the increased duration of therapy. The early use of an effective and well-tolerated anticancer agent as maintenance or consolidation therapy after disease stabilization or maximal response with platinum-based regimens appears beneficial in advanced disease. Pemetrexed with its favorable toxicity profile as maintenance therapy can be a useful strategy to improve the outcome of patients with advanced-stage nonsquamous NSCLC. Because QoL data with newer, third-generation cytotoxics as maintenance therapy is not yet available, the clinical benefit with these agents has to be balanced between the improvements seen in progression-free survival and the slight increase in toxicity. The role of erlotinib and other molecular targeted agents in the maintenance setting is under active investigation and could potentially prove to be a new standard treatment option.

Conflict of Interest Disclosures

Tony S. K. Mok has received honoraria for speaking engagements and consultancy services from AstraZeneca, Roche, Eli Lily, Pfizer, Merck, and Bristol-Myers.