Maintenance Chemotherapy for Nonsmall Cell Lung Cancer

Summary of Key Points

Maintenance therapy offers the possibility of continued active treatment to delay disease progression and symptom deterioration and, more importantly, improved overall survival of patients with advanced nonsmall cell lung cancer (NSCLC) already treated with induction chemotherapy.
The target population for maintenance are patients who achieved objective response or disease stabilization with induction chemotherapy with minimal cumulative toxicity.
Meta-analyses and patients’ preference support the use of maintenance in advanced NSCLC.
Maintenance chemotherapy doesn’t impair quality of life nor generate an additional cost compared with the benefits achieved.
Excluding targeted agents with known driver mutations, no predictive biomarkers are available to select better candidates for maintenance therapy with chemotherapy.
In this setting new treatment opportunities with immunotherapy and other agents represent a research priority.

The quest to control NSCLC has been long and remains frustrating, but since the 2000s, substantial advances have been made in therapeutic options for patients with this disease. The most striking advances have been in treatments linked to the identification of molecular changes acting as so-called drivers of malignancy for many patients, but optimization of the delivery of chemotherapy, particularly with maintenance therapy, has also led to improved survival for patients.

The concept of maintenance therapy was initially rejected after several studies published in the early 2000s demonstrated that continuation of a platinum-based doublet beyond four cycles did not result in a significant survival advantage but did cause progressive toxicity. Around the same time, studies demonstrated the efficacy of second-line chemotherapy, most notably with docetaxel. The overall interpretation of these results led to the standard treatment paradigm of treatment with a platinum-based doublet for four cycles (six for patients who had response) followed by a so-called treatment holiday until the time of progression, at which point standard second-line chemotherapy was offered. The widespread belief was that patients benefited from a break from chemotherapy and that close surveillance would provide the opportunity for patients to receive beneficial future treatments.

This approach began to be questioned, however, with the development of new agents. For example, some new chemotherapy drugs, such as pemetrexed, could be given on a continuous basis with a lower risk of long-term toxicities such as neuropathy, which had limited the long-term use of other agents, such as the taxanes. In addition, the era of targeted agents began, and almost all of these agents (such as bevacizumab, erlotinib, and gefitinib) are administered continuously until progression.

The maintenance approach is currently defined as continuation or switch maintenance treatment. With continuation maintenance, one or two of the agents administered as part of a first-line combination regimen are continued beyond the four to six cycles. This concept is not completely new because it was extensively investigated in several trials starting in the 1980s, but it was only in 2006, with the licensing of bevacizumab, that an approved drug was available in a continuous maintenance setting. More recently, strongly positive data with pemetrexed given as continuation maintenance therapy after four cycles of a platinum-based doublet further contributed to a change in the treatment paradigm. Less compelling data support the use of gemcitabine as continuous maintenance therapy.

The concept of switch maintenance is more recent and is based on switching to an alternative agent (i.e., one that was not part of the first-line regimen) after completion of four to six cycles of doublet chemotherapy in the absence of disease progression. Definitive data support the use of pemetrexed and erlotinib, and less robust data are available for docetaxel. It could be argued that such an approach may be simply considered early initiation of second-line treatment. Although the agents investigated in this setting, namely pemetrexed, erlotinib, and docetaxel, are indeed all approved agents for standard second-line therapy, their use for patients who had an objective response or disease stabilization after completion of first-line chemotherapy is biologically different from their use after disease progression. The term early second-line treatment is therefore inaccurate and should not be used.

The results of the Sequential Tarceva in Unresectable NSCLC (SATURN) and JMEN trials were the true impetus for maintenance therapy, which led to guidelines in support of maintenance therapy, issued in 2011, and increased awareness about the benefits of this approach. Clinical investigations in this area have helped demonstrate that second-line chemotherapy is subsequently given to about two-thirds of patients who have a treatment holiday after disease stabilization with four to six cycles of chemotherapy. Maintenance therapy, either as continuation or switch, leads to improved survival for patients with NSCLC.

Historical Maintenance Trials

In 1989, a study to evaluate the effect of prolonging chemotherapy for patients with stable disease beyond two or three cycles of a four-drug regimen (methotrexate, doxorubicin, cyclophosphamide, and lomustine) found no benefit for a longer duration. The study was small, with 74 patients randomly assigned to maintenance chemotherapy or to discontinuation of chemotherapy, and the results showed a nonsignificant trend toward longer overall survival of nearly 4 months for maintenance chemotherapy, but cast doubt on the use of prolonged first-line chemotherapy or maintenance therapy. Despite this uncertainty, patients and physicians seemed to favor continued therapy, as evidenced by difficulty in recruiting to a randomized trial of six courses of mitomycin-C, vinblastine, and cisplatin compared with observation after three cycles of the same regimen. The authors noted that most patients who declined to enroll in the study said they refused because they preferred to continue treatment. However, despite this challenge, the trial was completed and it showed no improvement in survival for prolonged chemotherapy and demonstrated an increase in fatigue and other types of toxicities, further supporting the idea that less is better when considering the continuation of combination cytotoxic regimens. At that time, the number of treatment cycles varied, as did the regimens used for first-line therapy.

Socinski et al. published the results of a practice-changing trial in 2002. The backbone regimen for this trial was carboplatin and paclitaxel, based on the findings of multiple phase III studies performed in the United States and Europe that had demonstrated the tolerability and efficacy of this regimen. All patients received four cycles of carboplatin (area under the curve [AUC] of 6) and paclitaxel (200 mg/m ² ) on a 21-day regimen, with disease assessment after every two cycles. Patients in one arm of the trial had a break from treatment after four cycles, with assessments for progression done every 6 weeks, and patients in the other arm received chemotherapy every 3 weeks until disease progression or until the decision was made to end treatment. It was planned that all patients were to receive second-line therapy with weekly paclitaxel (80 mg/m ² ) given at the time of progression. A total of 230 patients were enrolled, predominantly between 1998 and 1999. Response rates were 22% and 24%, respectively, in the two arms, with no additional responses after four cycles for patients who had received continued chemotherapy. The median survival times were 6.6 and 8.5 months, respectively, but the difference was not significant ( p = 0.63). Of note, 45% of patients received second-line chemotherapy, and more patients received continued chemotherapy than were given a treatment holiday. Toxicities, particularly neuropathy, were higher among patients who received continued chemotherapy, but there was no clear difference in quality of life between the two treatment arms. The conclusion drawn from this trial was that treatment beyond four cycles of a platinum-based doublet did not lead to improved survival and could lead to increased toxicity. Thus, the standard practice shifted toward this approach, with an additional two cycles offered to patients who had a response, such that four to six cycles of a platinum-based doublet, followed by a treatment holiday, was the standard approach. Two subsequent studies addressed the question of four versus six cycles of first-line chemotherapy, and both failed to show a clear benefit with the additional two cycles, thus supporting four cycles of first-line doublet therapy as the standard of care.

Another contemporary phase III study compared three cycles with six cycles of carboplatin (equivalent to AUC of 5) given on day 1 and vinorelbine (25 mg/m ² ) on days 1 and 8 every 3 weeks. A total of 297 patients were enrolled, and the median survival was 28 weeks for three cycles and 32 weeks for six cycles (hazard ratio [HR], 1.04; 95% confidence interval [CI], 0.82–1.31; p = 0.75), casting further doubt on the additional benefit of prolonged cycles of first-line chemotherapy.

All of these studies evaluated continuation of the initial regimen beyond four to six cycles compared with a true maintenance approach assessing prolongation of chemotherapy only in patients benefiting from platinum-based chemotherapy. An early maintenance study comparing a continuation and a switch approach enrolled 493 patients in 2000–2004 to receive three cycles of a triplet regimen (gemcitabine, ifosfamide, and cisplatin) on an every-3-week schedule. After three cycles, the 281 patients who did not have disease progression were randomly assigned to receive continued chemotherapy until disease progression or intolerability, or to receive switch maintenance therapy with paclitaxel (225 mg/m ² ) every 3 weeks, also continued until disease progression or intolerability. The progression-free survival was similar in both arms (4.4 vs. 4.0 months; p = 0.56). Although numerically the median overall survival favored continued chemotherapy (11.9 vs. 9.7 months), the difference was not significant ( p = 0.17). The 1-year survival rate was 49% for continued chemotherapy and 42% for switch therapy. Putting this trial into context with other maintenance studies is challenging, as patients in both arms continued with treatment until disease progressed, the randomization occurred after only three cycles of a platinum-based combination, and a substantial number of patients in each arm received the opposite regimen at the time of disease progression (69 of the 140 patients in the continued chemotherapy arm subsequently received paclitaxel).

One of the first true maintenance studies evaluated switch maintenance therapy with vinorelbine after four cycles of mitomycin, ifosfamide, and cisplatin. The study registered 573 patients with advanced stage NSCLC treated either with chemoradiation therapy for stage III disease or chemotherapy for stage IV disease, but only 181 were randomly assigned to the treatment arms of maintenance therapy with vinorelbine or no maintenance therapy. Of the 91 patients in the maintenance therapy arm, 7 died as a result of toxicity. No survival advantage was noted, which dampened enthusiasm for this approach.

A randomized phase II trial demonstrated more robust activity with continuation maintenance therapy with paclitaxel. The trial included 401 patients who were randomly assigned to one of three treatment groups: weekly paclitaxel with every-4-week carboplatin or weekly paclitaxel with weekly carboplatin according to two different schedules. Patients who did not have disease progression at week 16 were further randomly assigned to maintenance therapy with weekly paclitaxel (70 mg/m ² ) for 3 of 4 weeks or to observation (65 patients in each group). The every-4-week carboplatin regimen was superior in terms of response, and the maintenance therapy resulted in a 9-week longer time to progression, a 15-week improvement in median survival, and improvements in 1- and 2-year survival rates. This study was conducted to determine the optimal weekly regimen of paclitaxel and carboplatin, and the efficacy of maintenance therapy was not a key question. However, these results led to adoption of maintenance therapy in the subsequent phase III trial, in which 444 patients were randomly assigned to weekly paclitaxel (100 mg/m ² ) for 3 of 4 weeks plus carboplatin (AUC of 6) on day 1 of an every-4-week cycle or to standard every-3-week paclitaxel (225 mg/m ² ) on day 1 with carboplatin (AUC of 6). Patients in both treatment groups subsequently received maintenance therapy with paclitaxel (70 mg/m ² ) for 3 of 4 weeks until disease progressed. Although the toxicity profiles differed, the efficacy outcomes did not. Because maintenance therapy with paclitaxel was given to patients in both groups, its contribution is not clear.

Positive results with a continuation maintenance approach were found in a trial evaluating gemcitabine. Of the 352 patients enrolled who received cisplatin (80 mg/m ² ) on day 1 and gemcitabine (1250 mg/m ² ) on days 1 and 8 every 3 weeks, 206 patients had no disease progression and were eligible for random assignment to continuation with gemcitabine or to no further treatment (2:1 randomization). Maintenance therapy with gemcitabine was associated with a significantly improved time to progression and a trend in favor of overall survival that did not reach statistical significance (median, 13.0 vs. 11.0 months; p = 0.195). Because of the lack of a significant improvement in overall survival, the results did not have a major impact on clinical practice.

Although study findings suggested a benefit with maintenance therapy, particularly continuation maintenance with paclitaxel or gemcitabine, additional positive results were not reported until 2008. The tolerability of docetaxel, gemcitabine, and, more recently, pemetrexed, led to continued exploration of the concept of maintenance therapy.

Modern Maintenance Trials

Switch Maintenance With Chemotherapy ( Table 46.1 )

Fidias et al. pioneered the modern use of the switch maintenance approach in a clinical trial in which 309 (54.6%) of 566 patients with nonprogressive disease after four cycles of first-line gemcitabine and carboplatin were randomly assigned to second-line treatment with docetaxel (maximum of six cycles) either immediately or at the time of disease progression. The median progression-free survival was longer for patients treated with immediate docetaxel than for patients treated with delayed docetaxel (5.7 vs. 2.7 months; HR, 0.71; p = 0.0001). The difference in median overall survival between the two treatment approaches did not reach significance (12.3 vs. 9.7 months; HR, 0.84; p = 0.0853) in this undersized trial. Overall survival was the primary end point of the trial, which lessened the impact of the other results. Approximately 37% of the patients assigned to receive delayed docetaxel never received it because of substantial symptomatic deterioration, death, or the investigator’s decision. A subanalysis restricted to patients who did receive docetaxel in both arms showed that overall survival was identical in both arms (12.5 months), suggesting that the trend toward improved outcomes was associated with more patients in the immediate group receiving docetaxel. The toxicity profiles were similar for the two treatment approaches, and no differences in quality-of-life factors were found.

TABLE 46.1

Recent Trials of Switch Maintenance Treatment With Chemotherapy

	Induction Treatment ^a	Maintenance Treatment						Poststudy Treatment (%)		Progression-Free Survival			Overall Survival
Author/Trial	Regimen	Regimen	Median Age (Y)	PS 2 (%)	SCC Histology (%)	Never-Smokers (%)	Women (%)	Study Drug	Any	HR (95% CI)	Median (Mo)	p	HR (95% CI)	Median (Mo)	p	Toxicity	Quality of Life
Fidias et al.	Carboplatin AUC 5 on day 1; gemcitabine 1000 mg/m ² on days 1 & 8, every 3 wk × 4	Docetaxel 75 mg/m ² every 3 wk × 6 ( n = 153) Observation ( n = 156)	65.4 65.5	5.9 10.3	16.3 18.8	NR NR	37.9 37.8	NR 63	NR NR	0.71 (0.55–0.92)	5.7 2.7	0.0001	0.84 (0.65–1.08)	12.3 9.7	0.0853	Neutropenia: 27.6%; febrile neutropenia: 3.5%; fatigue: 9.7% Neutropenia: 28.6%; febrile neutropenia: 2%; fatigue: 4.1%	No differences (LCSS)
JMEN	Platinum-based doublet (without pemetrexed) every 3 wk × 4	Pemetrexed 500 mg/m ² every 3 wk + best supportive care ( n = 441) Placebo + best supportive care ( n = 226)	60.6 60.4	0 0	26 30	26 28	27 27	<1 18	51 67	0.50 (0.42–0.61)	4.3 2.6	0.0001	0.79 (0.65–0.95)	13.4 10.6	0.012	Total: 16% Fatigue: 5%; anemia: 3%; infection: 2% Total: 4% Fatigue: 1%; anemia: 1%; infection: 0%	No overall differences; better control of pain & hemoptysis

AUC , area under the curve; HR , hazard ratio; NR , not recorded; PS 2 , performance status of 2; SCC , squamous cell carcinoma.

a In the trial by Fidias et al. 566 patients received induction treatment, and 309 (54.6%) were randomly assigned to maintenance therapy; the number of patients in JMEN who received induction treatment was not recorded.

The JMEN trial evaluated pemetrexed as single-agent switch maintenance therapy. The trial design did not incorporate mandatory poststudy therapy, randomization was 2:1, and progression-free survival was the primary end point, which drew criticisms compared with the trial by Fidias et al., in which overall survival was the primary end point. However, the advantages of the JMEN trial were that its statistical assumptions were more realistic and the sample size allowed for more robust comparisons. In this trial, 663 patients with stage IIIB or IV disease who did not have disease progression during four cycles of platinum-based chemotherapy (without pemetrexed) were randomly assigned to receive best supportive care with or without pemetrexed until disease progression. Maintenance therapy with pemetrexed significantly improved the median progression-free survival (4.3 vs. 2.6 months; HR, 0.50; 95% CI, 0.42–0.61; p < 0.0001) and overall survival (13.4 vs. 10.6 months; HR, 0.79; 95% CI, 0.65–0.95; p = 0.012) ( Fig. 46.1A ). Of note, relatively fewer patients in the pemetrexed group received systemic postdiscontinuation therapy (51% vs. 67%; p = 0.0001), and 19% of patients in the control group received salvage treatment with pemetrexed. A prespecified analysis showed a significant interaction between treatment and histology, consistent with the findings in similar prior trials in different NSCLC settings. For patients who had tumors with nonsquamous cell histology, pemetrexed was associated with a greater benefit in terms of both progression-free survival (4.4 vs. 1.8 months; HR, 0.47; 95% CI, 0.37–0.60; p < 0.00001) and overall survival (median, 15.5 vs. 10.3 months; HR, 0.70; 95% CI, 0.56–0.88; p = 0.002), compared with patients who had tumors with squamous cell histology ( Fig. 46.1B ). In a subgroup analysis, the overall survival advantage with pemetrexed was greater for patients with stable disease at the end of induction chemotherapy (HR, 0.61) than for patients who had a partial or complete response (HR, 0.81). Treatment discontinuations due to drug-related toxic effects were more frequent with pemetrexed than with placebo (5% vs. 1%), as were drug-related grade 3 or 4 adverse events (16% vs. 4%; p < 0.0001), particularly fatigue (5% vs. 1%; p = 0.001) and neutropenia (3% vs. 0%, p = 0.006). No pemetrexed-related deaths occurred. Quality-of-life evaluations showed no global differences but a significant delay in worsening of pain and hemoptysis. As a result of this trial, both the Food and Drug Administration (FDA) and the European Medicines Association (EMA) approved pemetrexed as switch maintenance therapy for metastatic NSCLC, specifically for patients with nonsquamous cell tumors in whom disease has not progressed after first-line platinum-based chemotherapy.

Fig. 46.1, Overall survival in all patients (A) and in patients with nonsquamous cell tumors (B) in the JMEN trial of switch maintenance therapy with pemetrexed. CI, confidence interval; HR, hazard ratio.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here