Abstract
The objectives for the treatment of advanced nonsmall cell lung cancer are palliative and include improvement of survival, symptom control, quality of life and cost. The level of evidence of these benefits is based on multiple randomised trials and meta-analyses. Cisplatin-based chemotherapy with one of the regimens shown to be effective should be preferred. Carboplatin may be substituted for cisplatin if medical contraindications exist. Nonplatinum-based regimens are indicated as first-line treatment for advanced nonsmall cell lung cancer in patients for whom platinum-based chemotherapy is contraindicated. Single drug chemotherapy may be considered in patients with poor performance status. The choice of the active drugs depends on the patient's medical condition. There is no conclusive evidence that high doses of cisplatin (100–120 mg·m−2) provide better results than standard lower doses (50–60 mg·m−2) in terms of survival. The optimal duration of chemotherapy is poorly documented in advanced nonsmall cell lung cancer. A minimum of four to six cycles is advised in responding patients. Second-line chemotherapy is now accepted as a standard and should be offered to patients with good performance status and failing platinum-based first-line chemotherapy. Evidence is in favour of docetaxel and in the case of adenocarcinoma and adequate renal function, pemetrexed is recommended.
SERIES “LUNG CANCER”
Edited by C. Brambilla
Number 4 in this Series
Medical treatment of advanced nonsmall cell lung cancer (NSCLC) has been improved over the last two decades, with the main increase in the number of active drugs, the development of effective regimens and the introduction of salvage therapy after failure of first-line treatment. The present review will focus on chemotherapy; targeted therapies will be covered in another article in the series.
FIRST-LINE CHEMOTHERAPY
The present section is based on the recently published guidelines of the European Lung Cancer Working Party (ELCWP) for advanced NSCLC 1, 2. The current review will focus on several main questions concerning the first-line medical management of the disease.
What benefits can patients expect from chemotherapy and what are the treatment objectives?
Randomised trials have shown benefits in terms of palliation, improvement of survival, symptom control, quality of life and cost.
In total, 13 trials have assessed the effect of combination chemotherapy (cisplatin-based in all but one) versus supportive care alone 3–15. One study included only patients with stage III NSCLC 10. Two thirds of the trials showed statistically significant improvements in survival. In addition, four trials have compared single agent chemotherapy, using one of the newer drugs with best supportive care alone 16–19. All but one 18 showed a significant survival improvement with chemotherapy.
Five meta-analyses 20–24, published in the 1990s and including one performed with individual patients data 23, have confirmed a modest but significant effect with chemotherapy in terms of survival (table 1⇓). Symptom control has also been demonstrated as summarised in the American College of Chest Physicians (ACCP) guidelines 25, showing a high rate of improvement for cough, haemoptysis, pain, dyspnoea, weight loss, anorexia and malaise. Quality of life has been assessed in eight trials, with significant improvements in all but one (table 2⇓). Finally, in terms of cost, Rapp et al. 4 have shown a reduced cost when chemotherapy is prescribed compared with supportive care alone 26.
What are the active chemotherapeutic drugs for which efficacy has been shown?
The drugs used in the published trials can be divided into three groups: inactive (also called first-generation); old (second-generation); and new (or modern or third-generation) drugs. The second-generation group of drugs has been the topic of a meta-analysis 27. They include cisplatin, ifosfamide, mitomycin C, vindesine and vinblastine. Each of these drugs is able to significantly improve the response rate of the disease. The third-generation of active drugs has also been the subject of a systematic review 28. They include gemcitabine, paclitaxel, docetaxel and vinorelbine, which are all available in Europe. In randomised trials, all of these drugs 16, 17, 19, except gemcitabine 18, have been shown to improve survival in comparison to supportive care alone.
What are the recommended regimens for first-line chemotherapy?
The recommendations of various scientific and academic associations are summarised in table 3⇓. In their guidelines, the Ontario Program and the Fédération Nationale des Centres de Lutte Contre le Cancer (Paris, France) 29 recommend cisplatin-containing chemotherapy, without further precision of the drug(s) to be combined. For the ACCP, chemotherapy should be platinum-based with a new single agent 25, 30. According to the American Society of Clinical Oncology (ASCO), it should be a two-drug combination regimen 31; nonplatinum containing chemotherapy may be used as an alternative to a platinum-based regimen. In patients with poor performance status, the ASCO recommends single-agent chemotherapy. For the ELCWP 2, cisplatin-based chemotherapy is proposed with one of the regimens shown to be effective. Single agent chemotherapy with a drug that is shown to be effective, may be considered in patients with poor performance status.
What are the most effective platinum-based regimens for first-line chemotherapy?
Many cisplatin-based regimens are commonly used, combining cisplatin with old drugs such as vindesine, mitomycin C and/or ifosfamide or new drugs such as gemcitabine, docetaxel, paclitaxel or vinorelbine.
Two types of meta-analyses are available to help choose the most effective regimen 32. In the first type of meta-analyses (table 4⇓), the trials are compared according to the number of drugs in the regimen. Polychemotherapy is associated with better results than single agent treatment 33, 34. Two-drug regimens are superior to one-drug regimen, both in terms of response and survival; three-drug combinations are better than two-drugs only in terms of response 35. In the second type of meta-analyses (table 5⇓), the role of specific drugs is analysed. Addition of a drug to a platinum derivative is beneficial in terms of survival 36 but not the addition of mitomycin-C to a basic chemotherapy regimen 37. Gemcitabine appears to be associated with better outcomes in a meta-analysis of the literature but with much heterogeneity among the aggregated trials 38. The combination of cisplatin with docetaxel does not appear to result in better survival in comparison to other cisplatin-based regimens 39 but seems superior to regimens based on platinum and vinca alkaloids 40. In 2005, Canadian practice guidelines about the use of taxanes recommended paclitaxel or docetaxel plus cisplatin as one of a number of chemotherapy options in patients with good performance status 41. Finally, a study group tried to compare regimens with second-generation or third-generation platinum-based regimens but decided to not perform survival aggregation because of too high a heterogeneity 42. A recent systematic review of the literature on quality of life associated with standard chemotherapy in advanced NSCLC failed to show major differences between the various regimens 43.
In conclusion, chemotherapeutic regimens should include cisplatin with at least one other active drug. If the other drug is a new one, there is no evidence for the addition of a third agent outside the context of a clinical trial. There is also no evidence that combinations with new drugs are superior to those with old drugs in terms of survival. Cost of the treatment including supportive care and complications management should be taken into consideration in the choice of the regimen.
What is the indicated dosage of cisplatin?
There are five randomised trials that have investigated this question (table 6⇓), all of which were performed with old drugs 44–48. None of the studies were able to report a significant advantage in favour of high doses of cisplatin (100–120 mg·m−2) in comparison to lower doses (50–60 mg·m−2). In fact, the use of a high dose cisplatin is based on the observation of Gralla et al. 44 that responders to cisplatin plus vindesine survived longer when 120 mg·m−2 of cisplatin was administered instead of 60 mg·m−2. This difference was observed in a very small group of patients (n = 35). The ELCWP was unable to replicate the results in a much higher number of patients 45. High doses of cisplatin have the disadvantage of significantly higher renal, auditory and neurological toxicities 49.
Thus, there is no demonstration that high doses of cisplatin (100–120 mg·m−2) provide better results in terms of survival than standard lower doses (50–60 mg·m−2). Standard doses are associated with reduced toxicity and are thus recommended.
Can carboplatin be substituted for cisplatin?
The level of evidence is based on 10 published randomised trials 50–59 and three meta-analyses summarised in table 7⇓ 60–62. In randomised trials, the trend is in favour of cisplatin, both in terms of response and survival. The meta-analyses confirm this impression; the results are statistically significant in favour of cisplatin if the analysis is restricted to the regimens using new drugs combined with platinum derivatives.
Cisplatin should be preferred to carboplatin because of its better effect on survival. Carboplatin or a nonplatinum-based regimen may be prescribed if the patient is unable or unwilling to take cisplatin.
What is the optimal number of cycles?
The level of evidence is poor and based on a limited number of randomised trials, shown in table 8⇓. Two studies compared three cycles with six cycles 63, 64 or four cycles with six cycles 65 and another four cycles with treatment until disease progression 66. The last trials compared maintenance treatment using paclitaxel 67, vinorelbine 68, gemcitabine 69, 70 versus observation after induction chemotherapy. In none of the studies was prolongation of chemotherapy demonstrated as an advantage.
Thus, it is reasonable to recommend a minimum of four to six cycles in responding patients. Prolongation with a single drug appears ineffective in terms of survival. The attitude to continue treatment until best response merits further assessment.
Can nonplatinum-based regimens be substituted for platinum-based chemotherapy as first-line treatment?
ASCO is the only scientific society recommending nonplatinum regimens as an alternative to platinum-based chemotherapy as first-line treatment of patients with advanced NSCLC 31. All other societies recommend first-line platinum-based chemotherapy for advanced NSCLC patients. There are multiple randomised published trials on this topic 71–90. In terms of survival, there is no statistically significant difference between the two types of treatment in all but one trial. In 2005, Barlesi and Pujol 91 performed a systematic review of phase III trials available in the literature (table 9⇓). They concluded that the approach is still debatable when doublet regimens with new drugs are considered and did not report a meta-analysis. D'Addario et al. 92 have performed a meta-analysis of the published literature. When all trials were considered (irrespectively of using old or new drugs), there was a significant advantage both for response rate and 1-yr survival in favour of platinum-based treatment. The increase in 1-yr survival was 5%. When the analysis was restricted to combination regimens with new drugs, there was no significant difference in survival but response rate was significantly improved with platinum-based treatment.
In conclusion, it is reasonable to recommend that nonplatinum-based regimens may be used as a first-line treatment for advanced NSCLC in cases where platinum-based chemotherapy is contraindicated. For all other patients, they should be used only in the context of clinical trials.
Sequential chemotherapy
Two phase II trials have been published on sequential chemotherapy (table 10⇓) 94, 95. The ELCWP completed a large phase III trial 95 where patients without disease progression after three courses of cisplatin-based chemotherapy were randomised between further platinum-based chemotherapy or paclitaxel with crossover at the time of progression. There was no difference in survival between the two approaches; however, there was a trend in favour of the nonsequential approach. There is thus no indication for sequential chemotherapy with taxanes (or other drugs) in the management of advanced NSCLC.
Customised chemotherapy
Customised or tailored chemotherapy results from a treatment decision based on an analysis of biomarkers of response and resistance to cytotoxic drugs. Evidence in favour of that approach comes from post hoc analysis of adjuvant chemotherapy trials in resected NSCLC 97–99 and from chemotherapy trials where the enzymes regulatory subunit of the ribonucleotide reductase and excision repair cross-complementation group (ERCC) 1 were assessed. If their increased expression is associated with a better prognosis after surgery 100, a lower expression in the tumour results in a better survival for platinum- and gemcitabine-based chemotherapy 101, 102. The first randomised trial performed on the topic showed an improved response rate when treatment was guided by the tumour ERCC1 expression 103.
Targeted therapy
Epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs), gefitinib and erlotinib, have been used as first-line therapy in combination with chemotherapy and results were disappointing. Three large randomised trials were conducted, two with gefitinib, by Giaccone et al. 104 and Herbst et al. 105, and one with erlotinib by Herbst et al. 106. In a recent trial carried out in patients with poor performance status (performance status 2), erlotinib single-agent therapy was associated with significantly shorter survival in comparison with platinum-based chemotherapy 107.
Bevacizumab, a monoclonal antibody against vascular endothelial growth factor, has been associated with better survival when added to chemotherapy compared with paclitaxel and carboplatin 108. The final results of a European confirmatory trial are awaiting publication.
In conclusion, to date, targeted therapy has no routine application in the first-line treatment of advanced NSCLC.
Conclusion
In advanced NSCLC, chemotherapy is recommended as first-line treatment in patients with good performance status. Treatment objectives are survival, quality of life and symptom control improvement. Cisplatin-based chemotherapy with one of the effective regimens should be used. If the second drug is new, there is no evidence for the addition of a third agent outside the context of a clinical trial. There is also no evidence that combinations with new drugs are superior to those with old drugs in terms of survival. Carboplatin may be prescribed if the patient is unable or unwilling to receive cisplatin. Nonplatinum-based regimens are indicated in patients for whom platinum-based chemotherapy is contraindicated. Cost of treatment including supportive care and complication management has to be taken into consideration when choosing a regimen. Clinical research is ongoing in order to develop customised chemotherapy.
SECOND-LINE CHEMOTHERAPY
Despite demonstrated improvements in first-line treatment, most stage IIIB/IV patients experience disease progression and ∼50–60% of them are fit enough to receive a second-line treatment. The availability of new active drugs allows significant improvement in survival and symptom control without a major detrimental effect on quality of life. Thus, many thoracic oncologists prescribe not only first-line chemotherapy but also usually second-line chemotherapy or EGFR TKIs followed by a third-line of treatment in a large number of patients. The current article is only concerned with cytotoxic chemotherapy and the present authors have focused on chemotherapy drugs registered or studied in the setting of second-line treatment. Other important emerging issues in second-line treatment will also be considered. Hence, as chemotherapy gains wider acceptance for use in earlier stages of NSCLC, particularly in the adjuvant and neoadjuvant setting, physicians face a growing population of high performance status patients who have relapsed after their first-line chemotherapy. The type of second-line chemotherapy after initial adjuvant or neoadjuvant treatment with a platinum-based regimen remains largely undefined. Some might consider rechallenging patients with a platinum-based combination whereas others might treat these patients according to second-line clinical guidelines. Most relapses occurring after perioperative chemotherapy and surgery are either locally advanced relapses or metastatic diseases. Some differences exist between these post-surgical relapses and the progressions occurring after the first-line nonsurgical treatment of a stage IIIB/IV patient. Therefore, patients are more likely to have a low performance status (0–1), progression is often asymptomatic and diagnosed in the post surgical follow-up, the time between the first-line of treatment and the treatment of the relapse is generally longer than in stage IIIB/IV and in most cases the dose of chemotherapy previously administered is lower than that administered in first-line treatment of a stage IIIB/IV patient. These differences might be associated with a more chemosensitive disease and might justify the treatment of these patients according to first-line guidelines. The level of evidence of such a therapeutic recommendation is scarce and studies are ongoing to answer this question 109.
Which patients should receive second-line treatment?
In stage IIIB-IV patients, response to first-line therapy is generally short lived and progression occurs an average 4–6 months after treatment is discontinued. Many of these patients continue to have a good performance status and are candidates for second-line therapy, although not all receive it. Recent studies indicate that <50% of patients receive second-line treatment; furthermore, the characteristics of patients who receive a second-line treatment have not been well described in the literature. Hensing et al. 110 studied 230 patients with stage IIIB or IV NSCLC who received first-line therapy with carboplatin and paclitaxel. Of these patients, only 101 (44%) received second-line therapy. Factors increasing the likelihood of second-line therapy included high performance status, female sex and nonsquamous histology, while early termination of first-line therapy decreased the likelihood of further therapy. In another study 111, sex, stage at diagnosis, performance status at the start of second-line therapy and best response to initial therapy were associated with improved survival outcome in multivariate analyses. Thus, these factors should be used to select the patients who will benefit most from second-line chemotherapy.
The impact of second-line chemotherapy has been studied in a large cohort of 4,318 patients in 19 phase III trials 112. In the current review, the objective response rates to chemotherapeutic agents are lower than those in the first-line setting in cases of advanced NSCLC. The median intention-to-treat objective response rate was only 6.8%, whereas the median disease control rate was 42.4%. A median survival time of 6.6 months showed no correlation with the objective response rate (p = 0.6992) but, in contrast, was better associated with the disease control rate (p = 0.0129). This indicates that not only tumour shrinkage, but also disease stabilisation, contributes to survival benefit in the second-line setting.
The second-line treatment of NSCLC has been widely studied in the last decade and as a result, clinical practice guidelines are available and are based on randomised clinical trials. However, the setting of second-line treatment should not be limited to this well known situation. The recent introduction of targeted therapies as a potential player in the second-line has complicated the therapeutic algorithm. Clinical trials are still ongoing and the identification of predictive factors of response and survival will be a critical point in the selection of the best treatment for a given patient.
Which are the recommended second-line regimens?
The vast majority of chemotherapy guidelines recommend docetaxel or pemetrexed for stage III–IV NSCLC patients who fail first-line chemotherapy 1, 113, 114. Promising results of phase II trials of docetaxel in previously treated patients prompted two phase III trials, which have established docetaxel as the first chemotherapeutic agent with proven benefit for patients with recurrent or refractory disease following initial chemotherapy 115, 116. The registration of docetaxel was based on data from these phase III trials.
In the first trial 115, docetaxel (75 mg·m−2 every 3 weeks) significantly prolonged median and 1-yr survival duration compared with best supportive care (median survival 7.5 versus 4.6 months; p = 0.010; 1-yr survival 37 versus 12%), although the response rate was low (5.5%). In the second study 116 the 6-months and median survival rates were similar for docetaxel and vinorelbine or ifosfamide. However, the 1-yr survival rate was significantly greater with docetaxel (75 mg·m−2) than ifosfamide or vinorelbine (32 versus 19%; p = 0.025). In both studies docetaxel significantly improved some parameters of quality of life. Since these two pivotal studies, new potential second-line drugs were compared with the docetaxel standard of care 117–123. With regards to the therapeutic results of the docetaxel arm of these studies, it must be emphasised that response rates and survival data were highly and significantly reproducible (table 11⇓).
The standard 3-weekly dosing regimen has been challenged by a weekly schedule, and trials have shown that while weekly docetaxel does not result in better survival rates when compared with a 3-week docetaxel regimen, it may produce better compliance, better response rates and a lower rate of neutropenia 124–127. Docetaxel-based combination regimens have not been found to be superior to docetaxel alone as second-line therapy and monochemotherapy remains the standard of care in this setting 128.
Pemetrexed is a multi-targeted antifolate drug. The targets of pemetrexed are the enzyme thymidylate synthase, glycinamide ribonucleotide formyl transferase and dihydrofolate reductase 129. These enzymes are critical for the synthesis of purine nucleotides and thymidine. The initial trials with pemetrexed, mostly in patients with mesothelioma, revealed high rates of myelosuppression, mucositis and diarrhoea 130. Later studies demonstrated that the incidence of grade three or four mucositis and diarrhoea was correlated with elevated levels of both homocysteine and methylmalonic acid 131. These results led to the hypothesis that folic acid and vitamin B12 supplementation could lower the incidence of toxicity. This last point has been demonstrated in the pivotal mesothelioma trial 132 and folic acid and vitamin B12 supplementation is now recommended when pemetrexed is prescribed. The Food and Drug Administration approved pemetrexed for second-line NSCLC 4 yrs ago based on data from a single randomised, phase III trial comparing this new drug to docetaxel 117. In that trial, median survival with pemetrexed (500 mg·m−2 every 3 weeks) was 8.3 versus 7.9 months with docetaxel (75 mg·m−2 every 3 weeks; not significantly different). Response rates and time to disease progression for both agents were comparable. The incidence of side-effects (grade three or four neutropenia, febrile neutropenia and neutropenia with infections) with pemetrexed was significantly lower than with docetaxel (p≤0.004), and hospitalisations for neutropenic fever (p<0.001) and other toxicities (p = 0.092) were also lower with pemetrexed. Furthermore an analysis of survival without grade three/four toxicity 133 suggested a benefit-to-risk profile that favours pemetrexed over docetaxel. The analysis of the impact of NSCLC histology on overall survival demonstrated clinically relevant differences in survival according to histology. Recent evidence suggests that some subtypes of NSCLC such as adenocarcinoma or large cell carcinoma benefit more from the use of pemetrexed 134 and very recently the European Medicines Agency recommended the use of pemetrexed for patients with nonsquamous histologies. These data obtained in a first-line randomised trial comparing pemetrexed and cisplatin with gemcitabine and cisplatin, were retrospectively shown in the study comparing pemetrexed with docetaxel 135. An analysis of the impact of NSCLC histology on the treatment effect on overall survival was in favour of pemetrexed versus docetaxel for other than predominantly squamous histologies (n = 399; 9.3 versus 8 months; p = 0.047) and was in favour of docetaxel for squamous cell carcinoma histology (n = 172; 6.2 versus 7.4 months; p = 0.018).
Recent phase III trials have also evaluated the use of new drugs in this setting. Oral topotecan, polyglutamated paclitaxel and vinflunine were studied in randomised trials. None of these drugs are yet approved in second-line treatment. Topotecan is a topoisomerase-I inhibitor. An oral form of topotecan has been developed that may offer a treatment option for patients who prefer oral to i.v. therapy. A randomised, phase III study compared the efficacy and safety of oral topotecan with i.v. docetaxel, as a second-line treatment of advanced NSCLC 118. This study demonstrated that oral topotecan is active and tolerable in patients with previously treated advanced NSCLC. The lack of difference in the primary end-point (1-yr survival rates) indicated that oral topotecan was not inferior to docetaxel based on the prespecified 10% noninferiority margin. However, the docetaxel treatment group had a higher survival rate than the oral topotecan treatment group (p = 0.057). Median time to progression favoured the docetaxel group, with an absolute difference of 1.8 weeks (p = 0.02). With respect to adverse events, the two treatments offered similar risk profiles, although each produced a different set of toxicities. Grade three/four neutropenia occurred more frequently with docetaxel, whereas grade three/four anaemia and grade three thrombocytopenia were more frequent with topotecan. Nausea, diarrhoea and vomiting were more frequent in the topotecan group, whereas alopecia, neuropathy and fever were more frequent in the docetaxel group. Overall, both treatments showed a progressive worsening of the quality of life symptom scores. This large trial (829 patients), demonstrated the activity of topotecan in NSCLC; however, this drug appeared globally slightly inferior to docetaxel. Oral topotecan would have provided an option for patients who desire an oral treatment after relapse, this option is probably unrealistic given the fact that EGFR TKIs are also oral treatments but with a sharply lower toxicity profile.
Paclitaxel, the other available taxane, has also demonstrated potential activity in the second-line setting. Phase II studies 126–142 and a follow-up study of all the patients who received paclitaxel as second-line chemotherapy after a randomised phase III trial 143 have demonstrated the activity of the single agent paclitaxel. The equivalent of docetaxel and pemetrexed was evaluated in one small randomised phase II study which found no statistically significant difference in terms of response rate (14 versus 3%) and median survival (105 versus 184 days) 144.
Bonomi et al. 119 recently reported a second-line phase III trial that compared docetaxel with paclitaxel poliglumex in 849 patients. Paclitaxel poliglumex, a macromolecule drug conjugate linking paclitaxel to polyglutamic acid, reduces systemic exposure to peak concentrations of free paclitaxel. Patients received 175 or 210 mg·m−2 of paclitaxel poliglumex or 75 mg·m−2 of docetaxel. The study enrolled 849 previously treated NSCLC patients with advanced disease. Median survival was 6.9 months in both arms, (p = 0.257), 1-yr survival was 25% for paclitaxel polyglumex and 29% for docetaxel (p = 0.134), and time to progression (median: paclitaxel polyglumex 2 months and docetaxel 2.6 months; p = 0.075) were similar between treatment arms. Paclitaxel poliglumex was associated with significantly less grade three or four neutropenia (p<0.001) and febrile neutropenia (p = 0.006). Grade three or four neuropathy (p<0.001) was more common in the paclitaxel polyglumex arm. Patients receiving paclitaxel polyglumex had less alopecia and did not receive routine pre-medications. Paclitaxel poliglumex and docetaxel produced similar survival results but had different toxicity profiles. Compared with docetaxel, paclitaxel poliglumex patients had less alopecia and less febrile neutropenia, shorter infusion times and a higher rate of the elimination of routine use of medications to prevent hypersensitivity reactions.
Vinorelbine is a second-generation semi-synthetic vinca alkaloid agent. Phase II studies have studied vinorelbine in combination with other agents such as gemcitabine, cisplatin 145–150 and mitomycin C 151. These second-line studies have shown results consistent with those previously demonstrated with other drugs in phase II. Notably a standard reproducible activity and a manageable tolerance profile has been demonstrated. However, disappointing results shown in a phase III comparison of docetaxel versus vinorelbine or ifosfamide 116 have limited the use of this agent in routine clinical practice. In the same generation of drugs vinflunine is a novel tubulin-targeted agent obtained by semi-synthesis. The actions of vinflunine on microtubules produce effects on mitotic spindle functions leading to modifications of cell cycle progression and cell killing 152. Vinflunine prevents microtubule assembly during mitosis 153–155. The affinity profile of vinflunine shows features which suggest that it will have greater effects on mitotic rather than axonal tubulin and so will cause less neurotoxicity 156. Vinflunine showed antitumour activity in a multicentre, single-arm, phase II trial in patients with advanced NSCLC previously treated with a platinum-based regimen 157. In total, 63 patients were included, the response rate was 7.9%, median progression free survival was 2.6 months (95% confidence interval 1.4–3.8), and median survival was 7.0 months. Grades three to four neutropenia was reported in 50% of patients; febrile neutropenia was observed in two patients (3.2%); grades three to four myalgia and grade three constipation were experienced by 10 (15.9%) and six (9.5%) patients, respectively. Constipation was manageable, noncumulative and could be prevented with laxative prophylaxis. The encouraging results from this phase II study led to a phase III trial comparing vinflunine to docetaxel. In total, 547 stage IIIB/IV pre-treated patients were treated with vinflunine 320 mg·m−2 (272 patients) or docetaxel 75 mg·m−2 (275 patients) 120. Response rates were <5% in both arms and overall survival and progression free survival were similar. The toxicity profile seemed to disfavour vinflunine, with pretty comparable haematological toxicity but a little more fatigue, abdominal pain and constipation. Overall, vinflunine emerged as a reasonable alternative to taxotere in this setting, but there was no obvious reason to recommend it over the already existing drugs.
How to select a second-line regimen
A short list of agents is now available in the second-line setting; new agents are eagerly awaited, however, no dramatic breakthroughs have appeared from the existing published or ongoing trials. The available agents appear to have similar efficacies in terms of response and overall survival, but also have significantly different toxicity profiles. Currently, the selection of a second-line agent depends on a number of factors, including patient preference, physician preference, performance status and patient comorbidities, smoking history, response to first-line chemotherapy, toxicities related to first-line chemotherapy and, according to recent studies 111, 134, the histological type of the tumour. Given the incurable nature of advanced NSCLC and the modest survival seen in the second-line setting, patient convenience and preference should be considered first when selecting a second-line agent 158. In addition to patient preference, performance status and comorbidities may also impact the selection of second-line therapies. Pemetrexed is contraindicated in patients with renal insufficiency (glomerular filtration rate <40 mL·min−1). Docetaxel may be given to patients with renal insufficiency but requires dose adjustments for those with hepatic impairment. Docetaxel has a higher rate of neurotoxicity than pemetrexed. Pemetrexed or an EGFR TKI may be preferable in patients with diabetic neuropathy or residual neuropathy from first-line therapy.
Nonsmokers with lung cancer are at high probability to respond to EGFR TKIs, this last point is critical in the selection of a second-line treatment. Response to first-line treatment appears to be a prognostic factor in patients receiving second-line treatment, furthermore nonresponding patients are often more likely to be treated with an EGFR TKI instead of chemotherapy. The emerging role of histology has been recently shown either for adenocarcinomas which are more likely to benefit to EGFR TKIs, and for large cell carcinomas and adenocarcinomas which are more likely to respond to pemetrexed whereas epidermoid carcinomas benefit more from docetaxel.
It is highly probable that in the future the use of molecular markers will assist the therapeutic decision-making. Research efforts continue to focus on identifying molecular markers and corresponding clinical features that will allow physicians to individualise patients' therapy. Furthermore a better understanding of prognostic factors in the second-line setting may allow clinicians to better select patients for second-line therapy and lead to better-designed second-line trials. Several new agents have shown activity in phase III trials; however, their efficacy and toxicity profile is not superior to that of existing agents and this raises doubts on their regulatory approval in the future. New drugs, such as enzastaurin, bortezomib, vorinostat and epidermal growth factor receptor or vascular epidermal growth factor targeted therapies administered alone or combined with either pemetrexed or docetaxel are currently being tested in clinical trials. These drugs may be integrated into second-line therapy as single agents or in combination with current agents in the future.
Statement of interest
A statement of interest for D. Moro-Sibilot can be found at www.erj.ersjournals.com/misc/statements.dtl
Footnotes
-
Previous articles in this series: No. 1: De Wever W, Stroobants S, Coolen J, Verschakelen JA. Integrated PET/CT in the staging of nonsmall cell lung cancer: technical aspects and clinical integration. Eur Respir J 2009; 33: 201–212. No. 2: Rami-Porta R, Tsuboi M. Sublobar resection for lung cancer. Eur Respir J 2009; 33: 426–435. No. 3: McWilliams A, Lam B, Sutedja T. Early proximal lung cancer diagnosis and treatment. Eur Respir J 2009; 33: 656–665.
- Received August 26, 2008.
- Accepted September 11, 2008.
- © ERS Journals Ltd