Key points

First-line treatment

In patients without targetable genetic alterations, the standard first-line therapy for advanced (stage IIIB or IV) non–small cell lung cancer (NSCLC) is chemotherapy with a platinum doublet for four to six cycles with or without bevacizumab. Historically, several drugs including paclitaxel, docetaxel, gemcitabine, and vinorelbine were considered acceptable platinum partners in the first-line metastatic setting with essentially no differences in progression-free survival (PFS) or overall survival (OS). More recently, additional agents have been approved in combination with platinum in this setting including pemetrexed and nab-paclitaxel. One particular advance in the last decade has been the recognition that histology should be considered in the choice of initial chemotherapy. This was discovered after an additional analysis of two studies showed pemetrexed to be more effective in nonsquamous histologies and less active in squamous tumors. Based on these findings, the choice of first-line agents in metastatic NSCLC is now strongly based on presenting histology, and this initial choice affects available second-line options.

First-line treatment

In patients without targetable genetic alterations, the standard first-line therapy for advanced (stage IIIB or IV) non–small cell lung cancer (NSCLC) is chemotherapy with a platinum doublet for four to six cycles with or without bevacizumab. Historically, several drugs including paclitaxel, docetaxel, gemcitabine, and vinorelbine were considered acceptable platinum partners in the first-line metastatic setting with essentially no differences in progression-free survival (PFS) or overall survival (OS). More recently, additional agents have been approved in combination with platinum in this setting including pemetrexed and nab-paclitaxel. One particular advance in the last decade has been the recognition that histology should be considered in the choice of initial chemotherapy. This was discovered after an additional analysis of two studies showed pemetrexed to be more effective in nonsquamous histologies and less active in squamous tumors. Based on these findings, the choice of first-line agents in metastatic NSCLC is now strongly based on presenting histology, and this initial choice affects available second-line options.

Maintenance therapy

Historically, patients treated with first-line platinum doublet chemotherapy who had objective responses or stable disease were placed on surveillance following completion of four to six cycles. However, over the last decade, new data suggests that there is benefit to the addition of maintenance therapy following initial chemotherapy. There are two maintenance strategies including continuation of an agent used in the first-line setting or switching to a previously unused agent (switch maintenance). There are data supporting the use of bevacizumab, pemetrexed, and erlotinib in the maintenance setting either as single agents or in combination. Prior maintenance therapy is of particular importance when discussing second-line chemotherapy options because the use of maintenance therapy, particularly when switch maintenance is used, influences the availability of agents in the second-line setting and beyond.

Chemotherapy as second-line treatment

Historically, nearly all patients received platinum doublet chemotherapy followed by single-agent chemotherapy in the second-line setting. However, with the discovery of targetable mutations, the development of tyrosine kinase inhibitors (TKI), the increasing use of maintenance strategies, and the introduction of immunotherapies to the current list of approved medications, choosing an appropriate second-line therapy has become more complicated. In general, those patients treated with targeted therapies receive a platinum doublet at the time of progression. For those who are wild-type for exploitable mutations, immunotherapy has become an increasingly popular second-line option because of its tolerability and potential for durable responses. However, there remains a role for additional treatments following second-line platinum doublets or immunotherapy, or alternatively, a role for second-line chemotherapy in those with contraindications to immunotherapy. This article discusses the available data for chemotherapy in the second-line treatment of NSCLC and beyond.

Chemotherapy in the second-line setting and beyond

With the exception of immunotherapy, there are four Food and Drug Administration (FDA)–approved agents for the second- and third-line treatment of advanced or metastatic NSCLC: (1) docetaxel, (2) pemetrexed, (3) erlotinib, and (4) ramucirumab. Docetaxel was the first agent approved for second-line treatment in 1999, and pemetrexed was approved in the second-line setting in 2004. Erlotinib was also approved in 2004 for second- and third-line treatment of advanced NSCLC and is currently the only FDA-approved third-line therapy. Ramucirumab was approved in 2014 in combination with docetaxel following progression on platinum-based chemotherapy.

Single agent versus combination regimens

In the first-line setting, doublet chemotherapy is clearly superior to single-agent treatment in advanced NSCLC. However, it is unclear whether combination therapy in the second-line setting improves outcomes over single-agent chemotherapy. A meta-analysis of six trials compared combination regimens with single-agent therapy in the second-line setting. The combination arm showed a statistically significant improvement in response rate (RR) (15.1% vs 7.3%; P = .0004) and median PFS (14 weeks vs 11.7 weeks; P = .0009; hazard ratio [HR], 0.79) compared with single-agent therapy. However, there was no difference in median OS between the two groups (37.3 weeks vs 34.7 weeks; HR, 0.92; P = .32). The doublet arms had significantly higher rates of grade 3/4 hematologic (41% vs 25%; P <.0001) and nonhematologic (28% vs 22%; P = .034) toxicity. Two additional meta-analyses assessing docetaxel alone versus docetaxel-based doublet chemotherapy and pemetrexed alone versus pemetrexed-based doublet chemotherapy demonstrated similar results. Both analyses showed improvements in RR and PFS with doublet chemotherapy but no improvement in OS. The doublet arms showed significant increases in hematologic toxicity for docetaxel and pemetrexed doublets and in nonhematologic toxicity for docetaxel doublets. Based on these findings, current second-line treatment approaches use mainly single-agent chemotherapy.

Docetaxel

Docetaxel was the first agent approved for the second-line treatment of patients with advanced NSCLC. This approval came based on two phase III clinical trials. In the TAX 317 trial, 104 patients previously treated with at least one platinum-based regimen were randomized to receive either docetaxel (100 mg/m ² or 75 mg/m ² every 3 weeks) or best supportive care (BSC) ( Table 1 ). The primary end point of the study was OS, and this favored the docetaxel arms with a median OS of 7.0 versus 4.6 months ( P = .047). The group treated with 75 mg/m ² had a numerically better OS than those treated with 100 mg/m ² (7.5 months vs 5.9 months) and demonstrated a much better toxicity profile. A subsequently published quality of life analysis of this trial demonstrated a significant improvement in pain scores for the docetaxel arms and a trend toward improved overall quality of life.

A second study evaluating docetaxel in the second-line setting following progression on platinum-based chemotherapy was the TAX 320 trial. This trial compared docetaxel at 75 mg/m ² or 100 mg/m ² every 3 weeks with a control arm treated with either vinorelbine (30 mg/m ² on Days 1, 8, and 15) or ifosfamide (2 mg/m ² /d on Days 1–3) repeated every 3 weeks. RR ( P = .002) and PFS ( P = .005) were better in the docetaxel arms. The median OS was not different among the three groups, although the 1-year OS was 32% in the docetaxel 75 mg/m ² versus 19% in the control arm ( P = .025). The docetaxel 75 mg/m ² arm again demonstrated much less hematologic and nonhematologic toxicity compared with the 100 mg/m ² arm. These two trials firmly established docetaxel 75 mg/m ² as the standard second-line therapy in 1999.

Several studies have compared weekly dosing of docetaxel with treatment every 3 weeks. A meta-analysis, published in 2007, included 865 individual patients and five trials comparing these dosing strategies. Median OS was 27.4 weeks in the every 3 week treatment arm and 26.1 weeks in the weekly arm ( P = .24), suggesting no difference in efficacy. There were also no significant differences in the rates of anemia, thrombocytopenia, or nonhematologic toxicity, but there was less febrile neutropenia ( P <.00001 for both) in the weekly dosing arm. Weekly dosing with docetaxel is an acceptable treatment alternative for second-line advanced NSCLC, and both treatment schedules are frequently used in practice.

At least three phase III trials have compared alternative chemotherapy agents with docetaxel in the second-line setting. A randomized controlled trial comparing docetaxel with pemetrexed is discussed in more detail later. The first of the other two trials compared docetaxel with vinflunine, a fluorinated vinca alkaloid, in a 1:1 randomization. This study met its noninferiority end point with a median PFS of 2.3 months for both arms. RR stable disease rate and OS were numerically superior in favor of docetaxel but did not reach statistical significance, and the vinflunine arm had higher rates of several hematologic and nonhematologic adverse events. Another study compared docetaxel with oral topotecan following progression on platinum-based therapy. This study also met its primary noninferiority end point; however, PFS at 1-year (25.1% vs 28.7%) and median OS (27.9 weeks vs 30.7 weeks) were both higher in the docetaxel group, although not statistically significant. The docetaxel arm did show a significant improvement in time to progression at 11.3 versus 13.1 weeks ( P = .02), and grade 3/4 toxicity was similar in the two groups. Vinflunine and topotecan have activity in the second-line setting but neither has shown clear improvement over docetaxel in terms of either efficacy or toxicity. Neither agent is FDA-approved in the United States for the treatment of NSCLC.

Following the approval of docetaxel as a standard second-line option in NSCLC, at least three phase III trials have compared docetaxel with or without a targeted agent. In the ZODIAC trial, 1331 patients with advanced NSCLC were randomized to receive docetaxel plus placebo or docetaxel plus the oral multikinase inhibitor vandetanib, an inhibitor of vascular endothelial growth factor (VEGF) receptor, EGFR, and rearranged during transfection tyrosine kinases. The primary end point was met with a median PFS in the vandetanib group of 4.0 months versus 3.2 months in the placebo group (HR, 0.79; P <.0001), and RR was also improved in the vandetanib group (17% vs 10%; P = .0001). OS was a secondary end point in this study. There was no difference between the two groups with a median OS of 10.3 months in the treatment arm versus 9.9 months in the placebo (HR, 0.95; P = .371). The LUME-Lung 1 trial compared docetaxel plus placebo with docetaxel plus nintedanib, an oral angiokinase inhibitor that blocks VEGFR 1 to 3, fibroblast growth factor receptors 1 to 3, and platelet-derived growth factor receptors α and β. The combination arm met its primary end point of PFS at 3.4 months versus 2.7 months (HR, 0.79; P = .0019). However, it failed to show a difference in OS between the two groups, although in a prespecified subgroup analysis, there was an improvement in OS in patients with adenocarcinoma (12.6 months vs 10.3 months; HR, 0.83; P = .0359). Based on this trial, nintedanib in combination with docetaxel was approved in Europe, but not the United States, for the second-line treatment of patients with NSCLC with adenocarcinoma.

Most recently, the REVEL trial randomized 1253 patients who had progressed after first-line platinum-based therapy to receive docetaxel plus either ramucirumab, an IgG1 monoclonal antibody against VEGF receptor 2, or placebo. The primary end point was OS. This study met its primary end point with a median OS of 10.5 months in the ramucirumab arm versus 9.1 months in the placebo arm (HR, 0.86; P = .023). PFS was also improved in the treatment arm at 4.5 months versus 3 months (HR, 0.76; P <.0001). There were slightly higher rates of grade 3 neutropenia, febrile neutropenia, and leukopenia in the ramucirumab arm and higher grade 3 fatigue and hypertension. Based on this study, the FDA approved ramucirumab in combination with docetaxel for the second-line treatment of NSCLC following progression on platinum-based chemotherapy. Although the OS advantage is modest, this regimen remains a consideration for patients in the second-line setting.