lung cancer image

Maintenance Therapy: Continuation Versus Switch

Mark A. Socinski, MD

Mark A. Socinski, MD
University of Pittsburgh Medical Center
Pittsburgh, Pennsylvania

See Also: Discussion, References

Introduction

Several phase 3 trials have established the paradigm that a limited number of cycles (3-4) employing platinum-based chemotherapy in advanced non-small cell lung cancer (NSCLC) yields survival outcomes similar to longer durations (6 or more) of such therapy.1,2 These trials lead most oncologists to use 4-6 cycles in the first-line setting and certainly established the use of four cycles only as a standard for control arms of future clinical trials. The use of second-line therapy at the time of disease progression following platinum-based first-line therapy in advanced NSCLC has been shown to improve survival compared to either best supportive care (BSC) or placebo.3,4 Three agents are currently approved in the United States for use in this setting: docetaxel,4 pemetrexed,5 and erlotinib.3 One issue with delaying second-line therapy has been the "drop-off" of patients who never get second-line therapy presumably due to disease-related complications and erosion of their performance status (PS) precluding subsequent therapy. In phase 3 trials which report the rate of subsequent therapy following first-line therapy, this drop-off ranges from 30-50% of patients.6 Given the potential survival benefit of second-line therapies, new strategies were needed to insure exposure of greater number of patients to subsequent therapies.

Maintenance therapy in advanced NSCLC is an option for patients who have either SD or a CR/PR following four cycles of platinum-based therapy. The optimal approach (switch versus maintenance) is not clear and should be individualized for each patient based on their initial response to first-line therapy (radiographic and symptomatic), histology, toxicities (from first-line therapies) and preferences.

One such strategy has been the use of maintenance therapy. As the term implies, this strategy integrates known active agents earlier following first-line therapy or continues a component of the first-line regimen until documented disease progression. In general, the use of maintenance therapy has been confined to those patients who receive four cycles of therapy and have either stable or responding disease (i.e., non-progressors). The National Comprehensive Cancer Network (NCCN) has coined two terms with regard to maintenance therapy in this setting; switch or continuous. Switch maintenance entails switching to a maintenance drug that was not a component of the regimen used in the first-line setting while continuous maintenance entails the continuation of one of the drugs used in the initial regimen. Most maintenance trials have used progression-free survival (PFS) as the primary endpoint and overall survival (OS) as a secondary endpoint. With the exception of one trial,7 patients randomized to the control arms of these trials (BSC or placebo) received treatment at the discretion of their physician when they progressed and did not necessarily receive the drug used on the maintenance arm of the trial. Since the majority of the maintenance trials used erlotinib, pemetrexed or docetaxel all of which are approved for use as second-line agents (because they are either known (docetaxel and erlotinib) or believed (pemetrexed) to improve survival versus no treatment), there was an imbalance in the rates of exposure between the arms to these approved agents. Table 1 reviews the trials evaluating continuation maintenance.

TABLE 1. Phase III Trials of Continuation Maintenance in Advanced NSCLC

First Author No. of Patients Enrolled Chemotherapy Comparison Median PFS PFS HR Median OS OS HR
Chemotherapy
Brodowicz 8 352 Gemcitabine (N = 138)
BSC (N = 68)
3.6 mo
2.0 mo
NR
P<0.001
10.2 mo
8.1 mo
NR
P=0.172
Belani 9 519 Gemcitabine (N = 128)
BSC (N = 127)
7.4 mo
7.7 mo
1.09
p = 0.5
8.0 mo
9.3 mo
0.97
p = 0.8
Belani 11 401 Paclitaxel (N = 65)
Observation (N = 65)
8.9 mo
6.8 mo
NR
P=NS
17.5 mo
14.0 mo
NR
P=NS
Perol 10 834 Gemcitabine (N = 155)
Observation (N = 155)
3.8 mo
1.9 mo
0.55
p < 0.001
NR
NR
-
-
Paz-Ares* 12 939 Pemetrexed (N = 359)
Placebo (N = 180)
3.9 mo
2.6 mo
0.64
p = 0.0002
NR
NR
-
-
Monoclonal Antibodies
Sandler 13 878 Bevacizumab (N = 215)
Observation (N = NA)
6.2 mo
4.5 mo
0.66
p < 0.001
12.3 mo
10.3 mo
0.79
p = 0.003
Pirker 14 1125 Cetuximab
Observation
4.8 mo
4.8 mo
0.94
p = 0.3
11.3 mo
10.1 mo
0.87
p = 0.04
NA = Not Available
NR = Not Reported
NS = Not Significant
PFS = Progression-free survival
OS = Overall Survival
HR = Hazard Ratio
* Includes non-squamous histology patients only

The five chemotherapy trials shown have evaluated gemcitabine,8-10 paclitaxel,11 and pemetrexed (in non-squamous histology only) 12 and have yielded variable results. In general, there has been a more consistent result on PFS compared to OS although mature OS results for the two trials reported most recently10,12 are not yet available. The two trials using the monoclonal antibodies bevacizumab13 and cetuximab14 did not isolate their role as maintenance drugs. Since these biologic agents were given with the initial chemotherapy it is unclear how much of the PFS and OS benefit seen is related to their use with chemotherapy versus their use as maintenance agents.

TABLE 2. Phase III Trials of Switch Maintenance in Advanced NSCLC

First Author No. of Patients Enrolled Chemotherapy Comparison Median PFS PFS HR Median OS OS HR
Chemotherapy
Westeel 15 573 Vinorelbine (N = 91)
Observation (N = 90)
5 mo
3 mo
0.77
p = 0.11
12.3 mo
12.3 mo
1.08
p = 0.6
Fidias 7 566 Immediate Docetaxel (N = 153)
Delayed Docetaxel
(N = 156)
5.7 mo
2.7 mo
NR
p = 0.0001
12.3 mo
9.7 mo
NR
p = 0.08
Ciuleanu 16 NA Pemetrexed (N = 441)
Placebo (N = 222)
4.0 mo
2.0 mo
0.60
p < 0.0001
13.4 mo
10.6 mo
0.79br /> p = 0.012
Non-squamous 481 Pemetrexed
Placebo
4.4 mo
1.8 mo
0.47
p < 0.0001
15.5 mo
10.3 mo
0.70
p = 0.002
Squamous 182 Pemetrexed Placebo 2.4 mo
2.5 mo
1.03
p = 0.8
9.9 mo
10.3 mo
1.07
p = 0.6
EGFR TKIs
Cappuzzo 17 1949 Erlotinib (N = 437)
Placebo (N = 447)
2.9 mo
2.6 mo
0.71
p < 0.0001
12.0 mo
11.0 mo
0.81
p = 0.008
Perol 10 834 Erlotinib (N = 155)
Placebo (N = 155)
2.9 mo
1.9 mo
0.82
p = 0.002
NR
NR
-
-
Kabbinavar 18 1145 Bevacizumab / Erlotinib
(N = 370)
Bevacizumab / Placebo
(N = 373)
4.8 mo

3.7 mo
0.71

p = 0.0006
14.4 mo

13.3 mo
0.92

p = 0.5
Takedav 19 604 Platinum doublet / Gefitinib (N = 300)
Platinum doublet (N = 298)
4.6 mo

4.3 mo
0.68

p <= 0.001
13.7 mo

12.9 mo
0.86

p = 0.11
Gaafar 20 173 Gefitinib (N = 86)
Placebo (N = 87)
4.1 mo
2.9 mo
0.61
p = 0.002
10.9 mo
9.4 mo
0.81
p = 0.2
Zhang 21 296 Gefitinib (N = 148)
Placebo (N = 148)
4.8 mo
2.8 mo
0.42
p =< 0.0001
18.7
16.9
0.84
P=0.26
NA = Not Available
NR = Not Reported
NS = Not Significant
PFS = Progression-free survival
OS = Overall Survival
HR = Hazard Ratio

Table 2 reviews the trials evaluating switch maintenance. The three chemotherapy trials have evaluated vinorelbine,15 docetaxel7 and pemetrexed.16 With regard to the latter trial,16 no histologic restrictions were used regarding eligibility for this trial although as shown in Table 2 the benefit for pemetrexed as a switch maintenance drug was seen only in those patients with non-squamous histology. The results of the two trials using docetaxel and pemetrexed (both approved as second-line agents) were largely consistent with each other with regard to both the difference in PFS and OS.

The most robust data is seen in the six trials that evaluated either erlotinib or gefitinib (epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs)) as switch maintenance.10,17-20 Focusing on hazard ratios (HR), the effect of these agents on both PFS and OS has been consistently in favor of the EGFR TKI (HR ranging from 0.42 to 0.82 for PFS and 0.81 to 0.92 for OS. These trials are somewhat heterogeneous with regard to patient characteristics: three trials10,17,20 were conducted in primarily Caucasian populations, two19,21 in Asian populations and one18 included only bevacizumab-eligible patients. These are important considerations as ethnicity is an issue with regard to the frequency of sensitizing mutations in the EGFR gene which are associated with remarkable sensitivity to EGFR TKIs. As an example, Cappuzzo et al17 showed that the HR for PFS in patients with EGFR mutations receiving erlotinib (compared to placebo) was 0.10 (95% CI, 0.04-0.25, p<0.0001) compared to a HR of 0.78 (95% CI 0.63-0.96, p=0.0185) for those patients with wild type (wt) EGFR. Zhang et al21 showed a similar a similar effect in a Chinese population with a HR of 0.17 in the EGFR mutation-positive patients compared to 0.86 in the EGFR wt Chinese patients. However, the rate of EGFR mutations in the largely Caucasian population reported by Cappuzzo et al17 was 11% compared to 38% in the Asian population reported by Zhang et al21 likely accounting for the much lower HR for PFS seen in the Chinese trial. The important point is that the PFS and OS advantage seen in the Cappuzzo trial is largely driven by the EGFR wt patients who get a significant yet more modest benefit from maintenance EGFR TKI compared to the EGFR mutant population.

Taken in its entirety, the data summarized in Tables 1 and 2 argue for switch maintenance rather than continuation maintenance. Of interest, the OS benefit of switch maintenance with both pemetrexed16 and erlotinib17 seemed to be greater in those patients who had stable disease (SD) following their initial four cycles rather than those patients who had either a complete or partial response (CR/PR). Only one drug (pemetrexed) has been evaluated in both continuation and switch maintenance strategies.12,16 In the pemetrexed continuation maintenance trial,12 the PFS benefit was greater in those patients who experienced a CR/PR to their initial four cycles compared to those with SD. No OS data is yet available from this trial which will be very informative with regard to shaping thoughts about the optimal strategy in individual patients. One could interpret the findings above to suggest once again the important aspect of maintenance is the exposure of patients to more lines of active therapy. If patients do not have good response to their initial four cycles of first line therapy (i.e., SD), switching to a maintenance agent appears to be more impactful on OS compared to those patients who have a more robust CR/PR. However, the impact of continuation pemetrexed maintenance, at least on PFS, seemed to be greater in those patients who had a CR/PR compared to those with SD.12 The two trials employing switch maintenance with pemetrexed16 and erlotinib17 lead to the approval by the Food and Drug Administration (FDA) of both these agents as maintenance drugs in advanced NSCLC. There were no new safety signals noted with the use of either agent in this setting. This is not surprising given that both drugs are approved for use in the second-line setting. The available data on patient-reported outcomes suggest no detrimental effect on quality-of-life (QoL) compared to placebo.2 This is reassuring but needs to be considered in the context of treatment continuation and all the factors (increased toxicity, costs, inconvenience, etc) that prolonged therapy comes with compared to offering the patient a treatment break.

Conclusion

Maintenance therapy is an effective strategy improving both PFS and OS with acceptable toxicity and should be considered a standard of care and discussed as a therapeutic option in the appropriate patient with advanced NSCLC.

In summary, maintenance therapy in advanced NSCLC is an option for patients who have either SD or a CR/PR following four cycles of platinum-based therapy. The optimal approach (switch versus maintenance) is not clear and should be individualized for each patient based on their initial response to first-line therapy (radiographic and symptomatic), histology, toxicities (from first-line therapies) and preferences. The impact of maintenance therapy on PFS and OS is significant; however, delivery of second-line therapy upon progression may accomplish the same OS benefit.1,2,6 The advantage of maintenance therapy is that more patients will be exposed to active agents due to their immediate use whereas a watch and wait strategy will results in a substantial minority of patients being unfit for second-line therapy most often due to disease-related complications. The data summarized above clearly show that maintenance therapy is an effective strategy improving both PFS and OS with acceptable toxicity and should be considered a standard of care and discussed as a therapeutic option in the appropriate patient with advanced NSCLC.

Back to top