PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of canmedLink to Publisher's site
 
Cancer Med. 2017 August; 6(8): 1847–1860.
Published online 2017 July 3. doi:  10.1002/cam4.1101
PMCID: PMC5548880

Meta‐analysis of first‐line therapies with maintenance regimens for advanced non‐small‐cell lung cancer (NSCLC) in molecularly and clinically selected populations

Abstract

Evidence has suggested survival benefits of maintenance for advanced NSCLC patients not progressing after first‐line chemotherapy. Additionally, particular first‐line targeted therapies have shown survival improvements in selected populations. Optimal first‐line and maintenance therapies remain unclear. Here, currently available evidence was synthesized to elucidate optimal first‐line and maintenance therapy within patient groups. Literature was searched for randomized trials evaluating first‐line and maintenance regimens in advanced NSCLC patients. Bayesian network meta‐analysis was performed within molecularly and clinically selected groups. The primary outcome was combined clinically meaningful OS and PFS benefits. A total of 87 records on 56 trials evaluating first‐line treatments with maintenance were included. Results showed combined clinically meaningful OS and PFS benefits with particular first‐line with maintenance treatments, (1) first‐line intercalated chemotherapy+erlotinib, maintenance erlotinib in patients with EGFR mutations, (2) first‐line afatinib, maintenance afatinib in patients with EGFR deletion 19, (3) first‐line chemotherapy + bevacizumab, maintenance bevacizumab in EGFR wild‐type patients, (4) chemotherapy+conatumumab, maintenance conatumumab in patients with squamous histology, (5) chemotherapy+cetuximab, maintenance cetuximab or chemotherapy + necitumumab, maintenance necitumumab in EGFR FISH‐positive patients with squamous histology, and (6) first‐line chemotherapy+bevacizumab, maintenance bevacizumab or first‐line sequential chemotherapy+gefitinib, maintenance gefitinib in patients clinically enriched for EGFR mutations with nonsquamous histology. No treatment showed combined clinically meaningful OS and PFS benefits in patients with EGFR L858R or nonsquamous histology. Particular first‐line with maintenance treatments show meaningful OS and PFS benefits in patients selected by EGFR mutation or histology. Further research is needed to achieve effective therapy for patients with EGFR mutation L858R or nonsquamous histology.

Keywords: advanced non‐small‐cell lung cancer (NSCLC), Bayesian network meta‐analysis, first‐line with maintenance therapy, molecularly and clinically selected patients

Introduction

Lung cancer remains the most common cause of cancer deaths in men 1, with only 30% surviving beyond 1 year and 8% beyond 5 years 2. Conventional first‐line therapy for patients with advanced NSCLC has been four to six cycles of chemotherapy doublets 3. On the other hand, emerging evidence has shown potential benefits of administering maintenance therapies to nonprogressing patients beyond four to six cycles of chemotherapy until disease progression 4. However, survival benefits have only been observed for particular maintenance treatments in targeted patient populations, for example, switch to or continue pemetrexed in patients with nonsquamous histology or continue gemcitabine in patients with squamous histology 4. Basic research has corroborated these clinical findings by showing that administration of effective therapies before disease progression could enhance kill of tumor cells before onset of treatment resistance 5, 6.

Optimal combined first‐line with maintenance treatment remains unknown. In fact, a sizeable portion of evidence on maintenance benefits has been derived from trials that randomized treatments only to patients not progressing after first‐line chemotherapy, and subsequently measured survival and progression times from start of maintenance therapy 7, 8, 9, 10. Many other trials have, in contrast, randomized patients at onset of first‐line therapy, and subsequently measured survival and progression times from start of first‐line therapy, then through maintenance 11, 12, 13. Moreover, many maintenance trials have evaluated maintenance after first‐line chemotherapy, while emerging evidence has shown benefits of first‐line targeted therapies compared to standard first‐line chemotherapy in particular patient populations 14. With that, there remains a paucity of evidence for optimal combination or sequence of first‐line and maintenance regimens.

Additionally, key trials in the first‐line setting have shown little or no survival benefit of first‐line therapies with maintenance regimens in unselected populations 7, 8, 9, 10. Nevertheless, survival benefits have been observed when patients are selected by particular biomarkers 14. For example afatinib has demonstrated survival benefits in patients harboring EGFR mutation subtype deletion 19 14. Survival benefits have been suggested within other clinically or molecularly selected populations 15. For example, patients with squamous histology and EGFR FISH positive showed survival gains with first‐line chemotherapy and necitumumab maintenance 15.

Furthermore, the majority of first‐line trials with maintenance regimens have been compared to standard chemotherapy with no maintenance 11, 12, rendering a lack of reliable evidence on head‐to‐head comparisons of treatments. With emerging evidence of survival benefits in targeted populations 14, elucidating treatment strategies for clinically and molecularly selected patients is essential.

In order to elucidate first‐line and maintenance treatments that would have the most benefit for patients, one must consider outcomes that are clinically meaningful for patients. In lung cancer, an improvement of 3–4 months of survival or an HR around 0.80 might be considered clinically meaningful 16.

In this study, first‐line treatments with maintenance regimens are compared head‐to‐head via network meta‐analysis in terms of combined clinically meaningful overall survival (OS) and progression‐free survival (PFS) benefits. Treatments are compared from a precision medicine perspective in terms of treatment benefits within (1) molecularly selected patients in terms of EGFR mutation positive versus wild type, EGFR mutation subtype deletion 19 versus L858R, EGFR FISH, and (2) clinically selected patients in terms of histology and clinically enriched EGFR populations.

Methods

Systematic review and study selection

Search strategy

PubMed was searched for relevant studies published from 1 December 2003 to 19 March 2015. Phase II/III randomized controlled trials evaluating first‐line treatments with maintenance regimens in advanced NSCLC patients reporting OS or PFS relative efficacy estimates were included. Conference proceedings (ASCO 2014‐2015, ESMO 2014) were searched for additional relevant studies 17, 18. Studies which (1) randomized patients after first‐line treatment, (2) continued chemotherapy doublets beyond six cycles or until progression, (3) included more than 20% patients with performance (PS) 2–3, or (4) included surgery, radiation, or chemoradiation as treatment arms were excluded. Detailed accounting of studies is provided in Figure 1. Study screening was performed by two independent reviewers and disagreements were discussed with the team until consensus. Individual trial characteristics and relative efficacy estimates for OS and PFS were extracted from the included studies.

Figure 1
Search diagram for trials evaluating first‐line therapies followed by maintenance regimens in advanced NSCLC patients according to PRISMA 37 guidelines.

Outcomes evaluation

Combined clinically meaningful OS and PFS benefits was defined as hazard ratios (HRs) ≤0.80 16 with ≥95% posterior probability of the treatment being better than standard chemotherapy with no maintenance. Treatment efficacies were evaluated in terms of (1) surface under the cumulative ranking curve (SUCRA) 19, (2) posterior HR with corresponding 95% credible interval (CrI), (3) posterior probability better than standard chemotherapy with no maintenance, and (4) posterior probability the treatment is best. SUCRAs were computed as the average cumulative probabilities for a particular treatment to be ranked best, top two, top three, and so on 19.

Treatment efficacies were meta‐analyzed within the following molecularly and clinically selected subgroups, (1) EGFR mutation positive, (2) EGFR mutation subtype deletion 19, (3) EGFR mutation subtype L858R, (4) EGFR wild type, (5) EGFR FISH positive, (6) nonsquamous histology, (7) squamous histology, and (8) clinically enriched for EGFR mutation. The clinically enriched for EGFR mutation population was defined as patients of Asian/East‐Asian origins or who were light or never smokers 20. Additional head‐to‐head comparisons of treatments for EGFR mutation/subtypes and wild type were performed.

Statistical analysis

Bayesian network meta‐analysis (NMA) was performed by separately pooling individual studies’ reported OS and PFS hazard ratios on the logarithmic scale. Log hazard ratios were modeled as normally distributed centered on a treatment contrast‐specific mean subject to within‐ and between‐study heterogeneities. For studies which did not fully report HRs and confidence intervals (CIs), efficacies were computed using procedures outlined in Tierney et al. 21.

Prior distributions for within‐study heterogeneities were inverse gamma with mean matching the corresponding study's reported variance and variance proportional to the number of events reported for each endpoint in each study. Prior distributions for average treatment efficacies (log hazard ratios) were modeled uninformatively as normal centered at zero with large variance. Between‐study heterogeneity priors were weakly informative uniforms placing 95% of the prior mass on relative treatment efficacies varying up to twofold between studies.

Draws from the posterior distribution were generated using 10 Markov chain Monte Carlo chains, each with 100,000 burn‐in simulations followed by posterior sampling of 100,000 observations each to generate posterior efficacies in terms of treatment HRs and respective 95% CrIs, SUCRA rankings, probability best, and probability better than standard chemotherapy with no maintenance.

Multiple efficacy estimates from the same study were modeled as multivariate normal with a study correlation. A uniform prior on 0 to 0.95 was specified for the within‐study correlation. Bayesian meta‐analysis was implemented using JAGS 22 run via the R interface rjags 23. All other statistical analyses were performed in R 24. WebPlotDigitizer was used to recover HRs and respective CIs reported graphically in individual studies 25.

Results

A total of 87 records and 56 trials evaluating first‐line with maintenance treatments in advanced NSCLC were included for meta‐analysis (Fig. 1). Studies and treatment comparisons within‐patient groups are shown in Figure 2 and Appendix Table 1.

Figure 2
Network of studies in (A) EGFR mutation positive (top left), (B) EGFR wild type (top right), (C) nonsquamous (bottom left), and (D) squamous (bottom right).

Molecularly selected populations

EGFR mutation

EGFR mutation positive

In EGFR mutation‐positive patients, first‐line intercalated chemotherapy + erlotinib with erlotinib maintenance was the only treatment which showed combined clinically meaningful OS and PFS benefits. First‐line intercalated chemotherapy + erlotinib with erlotinib maintenance showed the best survival SUCRA, along with posterior HR 0.48 (0.26–0.88) and 99% posterior probability of outperforming chemotherapy with no maintenance (Table 1 and Fig. 3). Clinically meaningful PFS benefits were demonstrated with first‐line intercalated chemotherapy + erlotinib, erlotinib + bevacizumab, afatinib, chemotherapy + bevacizumab, erlotinib, gefitinib, chemotherapy (gefitinib maintenance), and pemetrexed + gefitinib, each with maintenance regimens as illustrated in Table 1 and Appendix Figure 4.

Figure 3
Overall survival hazard ratio (95% CrI), surface under the cumulative ranking curve (SUCRA) probability, and probabilitya better than standard chemotherapy with no maintenance by EGFR mutation status and subtypes for first‐line therapies with ...
Table 1
Overall survival and progression‐free survival by EGFR mutation status and subtypes

Head‐to‐head comparisons for OS showed that first‐line intercalated chemotherapy + erlotinib with erlotinib maintenance outperformed first‐line chemotherapy + erlotinib, erlotinib, gefitinib, chemotherapy + cetuximab, and chemotherapy (gefitinib maintenance), each with maintenance regimens as illustrated in Appendix Table 17. Head‐to‐head comparisons for PFS benefits are shown in Appendix Table 18.

EGFR mutation subtype Del 19

In patients with EGFR mutation subtype Del 19, first‐line afatinib with maintenance afatinib was the only treatment which showed combined clinically meaningful OS and PFS benefits. First‐line afatinib with maintenance afatinib showed the best survival SUCRA, along with posterior HR 0.59 (0.43–0.80) and >99% posterior probability of outperforming standard chemotherapy with no maintenance (Table 1 and Fig. 3). Clinically meaningful PFS benefits were demonstrated with first‐line afatinib, gefitinib, erlotinib, and erlotinib+bevacizumab, each with maintenance regimens as illustrated in Table 1 and Appendix Figure 4.

Head‐to‐head comparisons for OS showed that first‐line afatinib with afatinib maintenance outperformed first‐line erlotinib with erlotinib maintenance as illustrated in Appendix Table 19. Head‐to‐head comparisons for PFS benefits are shown in Appendix Table 20.

EGFR mutation subtype L858R

In patients with EGFR mutation subtype L858R, no treatment demonstrated clinically meaningful OS benefit compared to standard chemotherapy with no maintenance (Table 1 and Fig. 3). Clinically meaningful PFS benefits were demonstrated with first‐line erlotinib, gefitinib, afatinib, and erlotinib + bevacizumab, each with maintenance regimens as illustrated in Table 1 and Appendix Figure 21.

Head‐to‐head comparisons for OS and PFS showed no strong evidence of differences among first‐line afatinib, gefitinib, erlotinib, and chemotherapy (gefitinib maintenance) as shown in Appendix Table 22.

EGFR mutation wild type

In EGFR wild‐type patients, first‐line chemotherapy+bevacizumab with bevacizumab maintenance was the only treatment which showed combined clinically meaningful OS and PFS benefits. First‐line chemotherapy + bevacizumab with bevacizumab maintenance showed the best survival SUCRA, along with posterior HR 0.57 (0.35–0.94) and 99% posterior probability of outperforming standard chemotherapy with no maintenance (Table 1 and Fig. 3). Clinically meaningful PFS benefits were demonstrated with first‐line chemotherapy + bevacizumab and chemotherapy + gefitinib, each with maintenance regimens as illustrated in Table 1 and Appendix Figure 4.

Head‐to‐head comparisons for OS showed that first‐line chemotherapy + bevacizumab with bevacizumab maintenance outperformed first‐line erlotinib + bevacizumab, gefitinib, and erlotinib, each with maintenance regimens as illustrated in Appendix Table 23. Head‐to‐head comparisons for PFS benefits are shown in Appendix Table 24.

EGFR FISH positive

In EGFR FISH‐positive patients with squamous histology, both first‐line chemotherapy + cetuximab with cetuximab maintenance and chemotherapy + necitumumab with necitumumab maintenance showed combined clinically meaningful OS and PFS benefits. First‐line chemotherapy + cetuximab with cetuximab maintenance and chemotherapy + necitumumab with necitumumab maintenance showed respective posterior OS HRs 0.56 (0.35–0.89) and 0.70 (0.48–1.01) with 99% and 97% posterior probabilities of outperforming standard chemotherapy with no maintenance (Appendix Table 12). On the contrary, for EGFR FISH‐positive and unselected histology, first‐line chemotherapy+cetuximab with cetuximab maintenance did not show clinically meaningful OS or PFS benefits (Appendix Table 12).

Clinically selected populations

Histology

Nonsquamous

In nonsquamous histology, no treatment demonstrated clinically meaningful OS benefit compared to standard chemotherapy with no maintenance (Table 2 and Fig. 4). Clinically meaningful PFS benefits were demonstrated with first‐line chemotherapy + bevacizumab and chemotherapy + bevacizumab+dulanermin, each with maintenance regimens as illustrated in Table 2 and Appendix Figure 5.

Figure 4
Overall survival hazard ratio (95% CrI), surface under the cumulative ranking curve (SUCRA) probability, and probabilitya better than standard chemotherapy with no maintenance by histology (EGFR mutation unselected) for first‐line therapies with ...
Table 2
Overall survival and progression‐free survival by histology

Squamous

In squamous histology, first‐line chemotherapy+conatumumab with conatumumab maintenance was the only treatment which showed combined clinically meaningful OS and PFS benefits. First‐line chemotherapy+conatumumab with conatumumab maintenance showed best survival SUCRA, along with posterior HR 0.51 (0.23–1.14) and 95% posterior probability of outperforming standard chemotherapy with no maintenance (Table 2 and Fig. 4). Clinically meaningful PFS benefits were demonstrated with first‐line chemotherapy + conatumumab, chemotherapy + celecoxib, chemotherapy + cetuximab + cilengitide, and chemotherapy + ipilimumab, each with maintenance regimens as illustrated in Table 2 and Appendix Figure 5.

Clinically enriched for EGFR mutations

In patients clinically enriched for EGFR mutations with nonsquamous histology, both first‐line chemotherapy+bevacizumab with bevacizumab maintenance and sequential chemotherapy + gefitinib with gefitinib maintenance showed combined clinically meaningful OS and PFS benefits. First‐line chemotherapy + bevacizumab with bevacizumab maintenance and sequential chemotherapy + gefitinib with gefitinib maintenance showed respective posterior HRs 0.78 (0.59–1.05) and 0.79 (0.60–1.05), both with 95% posterior probabilities of outperforming standard chemotherapy with no maintenance (Appendix Table 13). Clinically meaningful PFS benefits were demonstrated with first‐line chemotherapy + bevacizumab with bevacizumab maintenance, sequential chemotherapy + gefitinib with gefitinib maintenance, and intercalated chemotherapy + erlotinib with erlotinib maintenance, each with maintenance regimens as illustrated in Appendix Table 13.

Discussion

This meta‐analysis showed combined clinically meaningful OS and PFS benefits of particular first‐line with maintenance treatments in advanced NSCLC patients selected by molecular and/or clinical biomarkers. Results suggest the following treatment and patient selection strategies; (a) for molecularly selected patients, the following showed combined clinically meaningful OS and PFS benefits; (i) first‐line intercalated chemotherapy+erlotinib, maintenance erlotinib in patients with EGFR mutations, (ii) first‐line afatinib, maintenance afatinib in patients with EGFR deletion 19, (iii) first‐line chemotherapy+bevacizumab, maintenance bevacizumab in EGFR wild‐type patients, and (iv) first‐line chemotherapy+cetuximab, maintenance cetuximab or first‐line chemotherapy+necitumumab, maintenance necitumumab in EGFR FISH‐positive patients with squamous histology, whereas (b) for clinically selected patients, the following showed combined clinically meaningful OS and PFS benefits; (i) first‐line chemotherapy+conatumumab, maintenance conatumumab in patients with squamous histology and (ii) first‐line chemotherapy + bevacizumab, maintenance bevacizumab or first‐line sequential chemotherapy + gefitinib, maintenance gefitinib in patients clinically enriched for EGFR mutations with nonsquamous histology. No treatment showed combined clinically meaningful OS and PFS benefits in patients with EGFR L858R or nonsquamous histology.

This meta‐analysis highlights the importance of testing for specific subtypes of EGFR mutation (Del19/L858R) as results suggest that deletion 19 and L858R could be clinically distinct and exhibit different treatment outcomes. Feasibility of wide‐spread EGFR testing has been greatly extended by recent advances in ‘liquid biopsy’ or plasma‐based genotyping 26, 27. In particular, plasma‐based genotyping has been shown to detect both EGFR deletion 19 and L858R rapidly and accurately, reducing the need for traditional invasive biopsies 26, 27. Further trials in the EGFR mutation setting should study treatments distinctly by subtype. Urgent research is needed to identify treatments that will benefit patients with L858R subtype as they make up approximately 40% of identified EGFR mutations 14. This study has found little evidence of effective treatments for patients with this subtype.

When interpreting OS benefit of first‐line intercalated chemotherapy+erlotinib in patients with EGFR mutations, it is worth noting that result was derived from exploratory analysis of a single trial (FASTACT‐2) 28. Additionally, biomarker analyses revealed that the majority of this study population had 23% EGFR mutation Del 19 and 14% L858R, which might corroborate evidence from this meta‐analysis on the preferential OS benefit in Del 19 compared to L858R, although confirmatory studies are needed 29. Furthermore, only 57% of the trial population was tested and testing was not mandatory. Hence, results may not be representative of the full trial population 29.

Previous studies in the maintenance setting have suggested survival benefits with maintenance pemetrexed in nonsquamous patients who did not progress after first‐line cisplatin/pemetrexed 4, 30. However, in this meta‐analysis of first‐line trials followed by maintenance regimens, no clinically meaningful survival benefit was observed for patients with nonsquamous histology. It is important to note that in the earlier maintenance trial 30, only patients who had disease control were randomized to a maintenance therapy. In this study all patients randomized to first‐line therapy contributed to comparative estimates of first‐line with maintenance regimens, whereas only patients with disease control were given maintenance therapy. The PRONOUNCE trial, for example, contained 42% disease progressors 31. This raises an important question for future research, namely, is first‐line cisplatin/pemetrexed with pemetrexed maintenance beneficial for all nonsquamous patients, or are its benefits limited to patients with disease control after first‐line therapy.

In this study, aggregated data meta‐analysis, as opposed to individual patient data meta‐analysis, was performed. However, we believe that findings are robust due in particular to the systematic selection of well‐designed trials and objective outcomes. An additional limitation of this study is that trials which had no common comparator arms could not be included within the network of comparisons, for example, oral versus intravenous vinorelbine as single first‐line agent 32 or first‐line chemotherapy with maintenance vinorelbine versus gemcitabine 33.

Notably, the level of uncertainty (and corresponding power to detect differences, if present) varied across comparisons, and was driven by several factors, including the number of studies informing the comparison, the level of uncertainty/sample sizes of the relevant studies, the estimate of study‐to‐study heterogeneity in the efficacy of treatments, and how indirect the evidence pertaining to the specific comparison was. In particular, some comparisons were relatively uncertain, as reflected by wide credible and predictive intervals, and this uncertainty was a function of the data used for analysis. The analysis data were constrained in two ways, by availability and by inclusion criteria. Data availability reflected both population sizes of disease subgroups and current clinical thinking. Here, the inclusion criteria balance inclusiveness with transparency and robustness of modeling and results.

In particular, several treatments within subgroups failed to show a combined clinically meaningful benefit, but this does not necessary mean that there is evidence that the treatment is not beneficial. Importantly, if the comparison of interest was uncertain, then there may not have been enough evidence to make strong conclusions of any kind, such as concluding that there was a combined clinically meaningful benefit. Consider, for example, results for the nonsquamous subgroup. Several treatments showed some degree of promise, but none crossed the combined clinically meaningful benefit thresholds, which measure both the size of effect (hazard ratio) and uncertainty (probability better than standard chemotherapy with no maintenance). More broadly, several of the treatments within subgroups which showed promise, but failed to achieve combined clinically meaningful benefit, in particular those for which results were quite uncertain, may warrant further investigation.

To our knowledge, this study is the first multiple treatment comparison meta‐analysis to evaluate efficacies of first‐line therapies with maintenance regimens head‐to‐head and elucidate treatments with combined clinically meaningful OS and PFS benefits in patients selected by molecular and clinical biomarkers. An earlier meta‐analysis has shown clinically meaningful survival benefits of maintenance treatments in advanced NSCLC 4. However, those maintenance trials randomized only nonprogressing patients after first‐line chemotherapy rendering pooling of evidence with studies in the current meta‐analysis inappropriate due to differences in the distribution of disease trajectories of patients between the two distinct study designs.

Recently, first‐line immunotherapies as a monotherapy have been tested in PD‐L1‐positive advanced NSCLC 34, 35. Interestingly, pembrolizumab showed improved OS in patients with high PD‐L1 expression. (≥50%), whereas nivolumab showed no survival benefits in patients with low PD‐L1 expression (≥1%) 34, 35. Further research on strategies to identify patients who will have the greatest benefit from immunotherapies in the first‐line setting of advanced NSCLC is needed. In addition, a trial is currently underway to compare efficacy of the third‐generation EGFR TKI osimertinib with first‐generation erlotinib/gefitinib as monotherapy in the first‐line setting of activating EGFR mutant advanced NSCLC 36. When data from these promising studies become available in the future, it will add to the knowledge and evidence base of treatment options and patient selection strategies for improving treatment outcomes in advanced NSCLC patients.

In conclusion, this meta‐analysis of current evidence shows that particular first‐line with maintenance treatments show clinically meaningful OS and PFS benefits in molecularly and/or clinically selected populations. Further research is needed to achieve effective therapy for patients with EGFR mutation L858R or nonsquamous histology.

Conflict of Interest

GL reports grants and personal fees from Roche, grants and personal fees from Sanofi Aventis, grants and personal fees from Astra Zeneca, grants and personal fees from Boehringer Ingelheim, grants and personal fees from Lilly, grants and personal fees from Merck Serono, grants and personal fees from Merck, Sharp and Dhome, and grants and personal fees from BMS, outside the submitted work.

Supporting information

Appendix S1. Supplementary Tables and Figures.

Acknowledgments

This research did not receive any specific grant from funding agencies in the public, commercial, or not‐for‐profit sectors. This study constitutes part of the author's (PST) doctoral thesis.

Notes

Cancer Medicine 2017; 6(8):1847–1860

References

1. GLOBOCAN 2012: Estimated Cancer Incidence, Mortality and Prevalence Worldwide in 2012 . 2012. Int. Agency Res. Cancer. http://globocan.iarc.fr/Pages/fact_sheets_population.aspx. Cancer Research UK. Lung
3. Rossi A., Chiodini P., Sun J.‐M., O'Brien M. E. R., von Plessen C., Barata F., et al. 2014. Six versus fewer planned cycles of first‐line platinum‐based chemotherapy for non‐small‐cell lung cancer: a systematic review and meta‐analysis of individual patient data. Lancet Oncol. 15:1254–1262. [PubMed]
4. Tan P. S., Lopes G., Acharyya S., Bilger M., and Haaland B.. 2015. Bayesian network meta‐comparison of maintenance treatments for stage IIIb/IV non‐small‐cell lung cancer (NSCLC) patients with good performance status not progressing after first‐line induction chemotherapy: Results by performance status, EGFR mutation, hi. Eur. J. Cancer 51:2330–2344. [PubMed]
5. Gerber D. E., and Schiller J. H.. 2013. Maintenance chemotherapy for advanced non‐small‐cell lung cancer: new life for an old idea. J. Clin. Oncol. 31:1009–1020. [PubMed]
6. Goldie J. H., and Coldman A. J.. 1979. A mathematic model for relating the drug sensitivity of tumors to their spontaneous mutation rate. Cancer Treat. Rep. 63:1727–1733. [PubMed]
7. Lynch T. J., Patel T., Dreisbach L., et al. 2010. Cetuximab and first‐line taxane/carboplatin chemotherapy in advanced non‐small‐cell lung cancer: results of the randomized multicenter phase III trial BMS099. J. Clin. Oncol. 28:911–917. [PubMed]
8. Herbst R. S., Prager D., Hermann R., et al. 2005. TRIBUTE: a phase III trial of erlotinib hydrochloride (OSI‐774) combined with carboplatin and paclitaxel chemotherapy in advanced non‐small‐cell lung cancer. J. Clin. Oncol. 23:5892–5899. [PubMed]
9. Herbst R. S., Giaccone G., Schiller J. H., et al. 2004. Gefitinib in combination with paclitaxel and carboplatin in advanced non‐small‐cell lung cancer: a phase III trial–INTACT 2. J. Clin. Oncol. 22:785–794. [PubMed]
10. Giaccone G., Herbst R. S., Manegold C., et al. 2004. Gefitinib in combination with gemcitabine and cisplatin in advanced non‐small‐cell lung cancer: a phase III trial–INTACT 1. J. Clin. Oncol. 22:777–784. [PubMed]
11. Sandler A., Gray R., Perry M. C., et al. 2006. Paclitaxel‐carboplatin alone or with bevacizumab for non‐small‐cell lung cancer. N. Engl. J. Med. 355:2542–2550. [PubMed]
12. Thatcher N., Hirsch F. R., Luft A. V., et al. 2015. Necitumumab plus gemcitabine and cisplatin versus gemcitabine and cisplatin alone as first‐line therapy in patients with stage IV squamous non‐small‐cell lung cancer (SQUIRE): an open‐label, randomised, controlled phase 3 trial. Lancet Oncol. 16:763–774. [PubMed]
13. Rosell R., Carcereny E., Gervais R., et al. 2012. Erlotinib versus standard chemotherapy as first‐line treatment for European patients with advanced EGFR mutation‐positive non‐small‐cell lung cancer (EURTAC): a multicentre, open‐label, randomised phase 3 trial. Lancet Oncol. 13:239–246. [PubMed]
14. Yang J. C., Wu Y., Schuler M., et al. 2015. Afatinib versus cisplatin‐based chemotherapy for EGFR mutation‐positive lung adenocarcinoma (LUX‐Lung 3 and LUX‐Lung 6): analysis of overall survival data from two randomised phase 3 trials. Lancet Oncol. 2045:1–11. [PubMed]
15. Hirsch F. R., Boyle T. A., Thatcher N., Paz‐Ares L., Varella‐Garcia M., Kowalewski A. A., et al. 2015. ORAL32.05 EGFR IHC and FISH Correlative Analyses (SQUIRE Trial): Necitumumab + Gemcitabine‐Cisplatin vs Gemcitabine‐Cisplatin in 1st‐Line Squamous NSCLC. J Thorac Oncol. 10(Suppl 2):S797.
16. Ellis L. M., Bernstein D. S., Voest E. E., et al. 2014. American society of clinical oncology perspective: raising the bar for clinical trials by defining clinically meaningful outcomes. J. Clin. Oncol. 32:1277–1280. [PubMed]
17. ASCO Meeting Abstracts . 2014. ‐2015. http://meetinglibrary.asco.org/abstracts.
18. Table of Contents (ESMO 2014 abstracts) . 2014. http://annonc.oxfordjournals.org/content/25/suppl_4.toc.
19. Salanti G., Ades A., and Ioannidis J. P.. 2011. Graphical methods and numerical summaries for presenting results from multiple‐treatment meta‐analysis: an overview and tutorial. J. Clin. Epidemiol. 64:163–171. [PubMed]
20. Shigematsu H., Lin L., Takahashi T., et al. 2005. Clinical and biological features associated with epidermal growth factor receptor gene mutations in lung cancers. J. Natl Cancer Inst. 97:339–346. [PubMed]
21. Tierney J. F., Stewart L. A., Ghersi D., Burdett S., and Sydes M. R.. 2007. Practical methods for incorporating summary time‐to‐event data into meta‐analysis. Trials 8:16. [PubMed]
22. JAGS ‐ Just Another Gibbs Sampler . http://mcmc-jags.sourceforge.net/.
23. Plummer M. RJAGS version 3‐11: Bayesian graphical models using MCMC . 2012.
24. The R Project for Statistical Computing . http://www.r-project.org.
25. Rohatgi A. WebPlotDigitizer. http://arohatgi.info/WebPlotDigitizer.
26. Weber B., Meldgaard P., Hager H., et al. 2014. Detection of EGFR mutations in plasma and biopsies from non‐small cell lung cancer patients by allele‐specific PCR assays. BMC Cancer 14:294. [PubMed]
27. Zill O. A., Mortimer S., Banks K. C., et al. 2016. Somatic genomic landscape of over 15,000 patients with advanced‐stage cancer from clinical next‐generation sequencing analysis of circulating tumor DNA. | 2016 ASCO Annual Meeting | Abstracts | Meeting Library. J. Clin. Oncol. 34. (suppl; abstr LBA11501).
28. Wu Y.‐L., Lee J. S., Thongprasert S., et al. 2013. Intercalated combination of chemotherapy and erlotinib for patients with advanced stage non‐small‐cell lung cancer (FASTACT‐2): a randomised, double‐blind trial. Lancet Oncol. 14:777–786. [PubMed]
29. Mok T., Ladrera G., Srimuninnimit V., et al. 2016. Tumor marker analyses from the phase III, placebo‐controlled, FASTACT‐2 study of intercalated erlotinib with gemcitabine/platinum in the first‐line treatment of advanced non‐small‐cell lung cancer. Lung Cancer 98:1–8. [PubMed]
30. Paz‐Ares L. G., de Marinis F., Dediu M., et al. 2013. PARAMOUNT: Final overall survival results of the phase III study of maintenance pemetrexed versus placebo immediately after induction treatment with pemetrexed plus cisplatin for advanced nonsquamous non‐small‐cell lung cancer. J. Clin. Oncol. 31:2895–2902. [PubMed]
31. Zinner R. G., Obasaju C. K., Spigel D. R., et al. 2015. PRONOUNCE: randomized, open‐label, phase iii study of first‐line pemetrexed + carboplatin followed by maintenance pemetrexed versus paclitaxel + carboplatin + bevacizumab followed by maintenance bevacizumab in patients ith advanced nonsquamous non‐small‐C. J. Thorac. Oncol. 10:134–142. [PubMed]
32. Hirsh V., Desjardins P., Needles B. M., et al. 2007. Oral versus intravenous administration of vinorelbine as a single agent for the first‐line treatment of metastatic nonsmall cell lung carcinoma (NSCLC). Am. J. Clin. Oncol. 30:245–251. [PubMed]
33. Martoni A., Marino A., Sperandi F., et al. 2005. Multicentre randomised phase III study comparing the same dose and schedule of cisplatin plus the same schedule of vinorelbine or gemcitabine in advanced non‐small cell lung cancer. Eur. J. Cancer 41:81–92. [PubMed]
34. Socinski M., Creelan B., Horn L., et al. 2016. CheckMate 026: a Phase 3 Trial of Nivolumab vs Investigator's Choice (IC) of Platinum‐Based Doublet Chemotherapy (PT‐DC) as First‐Line Therapy for Stage IV/Recurrent Programmed Death Ligand 1 (PD‐L1) − Positive NSCLC. Ann Oncol. p. 27 (suppl_6): LBA7_PR. Available from http://www.nejm.org/doi/10.1056/NEJMoa1606774.
35. Reck M., Rodríguez‐Abreu D., Robinson A. G., et al. 2016. Pembrolizumab versus Chemotherapy for PD‐L1–Positive Non–Small‐Cell Lung Cancer. N. Engl. J. Med. 375:1823–1833. Available from http://www.nejm.org/doi/10.1056/NEJMoa1606774 [PubMed]
36. AZD9291 Versus Gefitinib or Erlotinib in Patients With Locally Advanced or Metastatic Non‐small Cell Lung Cancer (FLAURA) . 2017. ClinicalTrials.gov. https://clinicaltrials.gov/ct2/show/record/NCT02296125.
37. Moher D., Liberati A., Tetzlaff J., and Altman D. G.. 2009. Preferred reporting items for systematic reviews and meta‐analyses: the PRISMA statement. PLoS Med. 6:e1000097. [PubMed]
38. Pirker R., Pereira J. R., Szczesna A., et al. 2009. Cetuximab plus chemotherapy in patients with advanced non‐small‐cell lung cancer (FLEX): an open‐label randomised phase III trial. Lancet 373:1525–1531. [PubMed]
39. Pirker R., Pereira J. R., von Pawel J., et al. 2012. EGFR expression as a predictor of survival for first‐line chemotherapy plus cetuximab in patients with advanced non‐small‐cell lung cancer: analysis of data from the phase 3 FLEX study. Lancet Oncol. 13:33–42. [PubMed]
40. Douillard J.‐Y., Pirker R., O'Byrne K. J., et al. 2014. Relationship between EGFR expression, EGFR mutation status, and the efficacy of chemotherapy plus cetuximab in FLEX study patients with advanced non–small‐cell lung cancer. J. Thorac. Oncol. 9:717–724. [PubMed]

Articles from Cancer Medicine are provided here courtesy of Wiley-Blackwell