Meta-analysis comparing doublet and single cytotoxic agent therapy as first-line treatment in elderly patients with advanced nonsmall-cell lung cancer

Abstract

Objective

To perform a systematic review and meta-analysis comparing doublet versus single agent therapy in elderly patients with advanced nonsmall-cell lung cancer (NSCLC).

Methods

PubMed® and Cochrane databases, and American Society of Clinical Oncology, World Congress of Lung Cancer, and European Society of Medical Oncology abstracts were searched. Endpoints were overall survival (OS), 1-year survival rate (1-year SR), overall response rate (ORR), and grade 3/4 adverse events. Subgroup analyses were based on chemotherapy regimens and race.

Results

Out of 11 studies (13 randomized trials; n = 2782), doublet therapy was associated with significantly increased OS (hazard ratio [HR] 0.89, 95% confidence interval [CI] 0.83, 0.95), 1-year SR (risk ratio [RR] 1.15, 95% CI 1.04, 1.28), and ORR (RR 1.39, 95% CI 1.39, 1.86) versus single-agents. Chemotherapy regimen-based subgroup analyses favoured platinum-based doublet therapy for OS (RR 0.71, 95% CI 0.60, 0.84), 1-year SR (RR 1.28, 95% CI 1.11, 1.47), and ORR (RR 1.88, 95% CI 1.49, 2.38). Race-based subgroup analyses revealed increased benefit from doublet therapy in Asian populations for ORR (RR 1.70, 95% CI 1.29, 2.23) but not increased survival benefit. Higher incidences of grade 3/4 anaemia (RR 2.23, 95% CI 1.61, 3.09), thrombocytopenia (RR 2.47, 95% CI 1.17, 5.20), and fatigue (RR 1.36, 95% CI 1.06, 1.74) were observed with doublet versus single-agent therapy.

Conclusions

Doublet therapy was associated with significantly increased OS, 1-year SR and ORR compared with single agent therapy. Race may be considered when choosing doublet versus single-agent therapy as first-line treatment of NSCLC in elderly patients.

Keywords

Advanced nonsmall-cell lung cancer elderly chemotherapy

Introduction

Lung cancer is the leading cause of cancer-related death worldwide. Increasing life expectancy, with its concomitant increased cancer risk, has led to a significant increase in incidence of lung cancer in older people and represents a major societal disease burden.^1–4

Clinical trials of doublet therapies in elderly patients with advanced nonsmall-cell lung cancer (NSCLC) have reported survival benefits compared with monotherapy. The Southern Italy Cooperative Oncology Group trial, which compared gemcitabine plus vinorelbine versus vinorelbine alone in 152 patients aged ≥70 years, found that combination therapy improved overall median survival from 18 to 29 weeks.⁵ Another published trial comparing gemcitabine and carboplatin with gemcitabine monotherapy in NSCLC patients (median age, 79 years) found improved survival benefits in the doublet arm (overall survival [OS] increased from 12.6 to 15.4 months; progression-free survival increased from 2.4 to 3.9 months).⁶ Conversely, the Multicenter Italian Lung Cancer in the Elderly Study, which included 698 patients (median age 74 years),⁷ found that single agent therapies were as effective as nonplatinum-based combination therapy, but with lower toxicity. Another trial, which compared a weekly docetaxel–cisplatin combination with tri-weekly docetaxel alone, did not show a survival benefit for doublet therapy.⁸ Thus, the standard first-line treatment for elderly patients with advanced NSCLC remains unclear.

Despite the possible survival advantages of combination therapy in treating elderly patients with NSCLC, toxicity remains a major issue.⁹ Due to a lack of subgroup analyses and insufficient clinical research data, the type of elderly patients who would benefit most from doublet chemotherapy are yet to be identified. Thus, the present meta-analysis was performed in an attempt to provide information regarding the optimum choice of chemotherapy for elderly patients with advanced NSCLC.

Patients and methods

Search strategy

An online literature search was performed using PubMed®, EMBASE and Cochrane databases for relevant articles updated to September 2012, using the following key words and various combinations: (non small cell OR adenocarcinoma OR squamous cell carcinoma) AND lung cancer; advanced; older OR elderly; first line; alone OR combination OR combined OR plus; versus OR vs OR comparing OR against. Papers and abstracts presented at the American Society of Clinical Oncology (updated to 2012), World Congress of Lung Cancer (updated to 2012), and European Society of Medical Oncology (updated to 2012) meetings were also searched. The language was restricted to English.

Eligibility criteria

Inclusion criteria for the manual selection of clinical trials comprised the following: (1) Patients diagnosed with advanced NSCLC that was previously untreated; (2) randomized controlled trials, involving patients aged >65 years, or including subgroup analysis of patients aged >65 years; (3) comparisons between doublets and single agents as first-line treatment, using a third-generation cytotoxic drug; (4) at least one of the endpoints detailed in the data extraction section. No study authors were contacted for additional data, if relevant data were missing from the publication. All review articles, including systematic reviews and meta-analyses were excluded.

Data extraction and validity assessment

The quantitative 5-point Jadad scale¹⁰ was used to assess the quality of included trials: trails with a score ≥2 were included for analyses. For analyses, the following data were extracted from eligible articles: median OS, overall response rate (ORR), 1-year survival rate (1-year SR), and percentage of patients experiencing grade 3–4 adverse events according to the National Cancer Institute Common Toxicity Criteria (NCICTC) version 2 or 3.¹¹ Data extraction was performed by two independent reviewers (S.F.W., Q.W.) who searched the literature and extracted the data independently. Any mismatches between reviewers were resolved by consensus after a re-check and discussion between the two reviewers.

Statistical analyses

The overall hazard ratio (HR) estimate was calculated for OS, and the risk ratio (RR) was calculated for ORR, 1-year SR, and grade 3–4 adverse events using Stata® software, version 11.0 (Stata Corp., College Station, TX, USA) for Windows®.¹² All statistical calculations (including graphical presentations and analyses) were performed as previously described,^13–15 using Stata® software. As for OS, all data including HR with 95% confidence intervals (CI) (lower limit [ll], upper limit [ul]) were used to evaluate the survival benefit for doublet compared with single-agent chemotherapy; this was extracted directly from papers included in the analysis, using the following code: 1) gen lnhr = ln(hr) gen lnll = ln(ll) gen lnul = ln(ul); 2) metan lnhr lnll lnul, eform label(namevar = study) xlabel(0.5,1,1.5) fixed/random effect(hr). For ORR, 1-year SR, and grade 3–4 adverse events, RR was calculated using the code: metan ncalive ncdeath ntalive ntdeath, label (namevar = study, yearvar = year) fixed/random rr counts.

Between-trial heterogeneity was estimated using the χ²-test-based Q-statistic and a P-value < 0.05 or I²> 50% was considered statistically significant.¹⁶ If heterogeneity existed, data were analysed in a random-effects model; otherwise, a fixed-effects model was used. Sources of heterogeneity were assessed by subgroup stratification analysis based on several study characteristics including chemotherapy regimen and race. A statistical test with a P-value > 0.05 was considered significant. A HR > 1 reflected more deaths or a higher rate of disease progression in the doublet combination group, while a RR > 1 indicated a higher incidence of toxicity-related adverse events or higher ORR in the doublet combination group. Publication bias was evaluated using the Begg’s Funnel Plot and Egger’s test.^17,18 All P-values were two-sided. All confidence intervals (CIs) had a two-sided probability coverage of 95%.

Results

Study selection

The initial retrieval performed in September 2012 found 215 articles. After screening titles and abstracts, 175 articles did not meet the inclusion criteria; 38 full-text articles and two abstracts were reviewed for further inclusion. Subsequently, three articles were excluded due to lack of sufficient data for extraction, and 26 articles were excluded due to lack of elderly (>65 years) subgroup analyses. Eleven publications were finally included for analyses (Figure 1; 13 trials and 2 782 patients.^{5–8,19–25} All included trials, together with their related Jadad scale scores, are shown in Table 1.

Figure 1.

Flow chart showing identification of studies comparing doublet with single-agent therapy in elderly patients with advanced non-small cell lung cancer.

Table 1.

Characteristics of 13 clinical trials (from 11 publications) included in the current meta-analysis involving patients with advanced nonsmall-cell lung cancer aged >65 years, who were treated with doublet versus single-agent therapy. Data extracted from the Comella et al²¹ and Hainsworth et al²² studies were the outcomes of elderly patient subgroup analyses.

Reference	Publication Year	Patient age, years	PS	Chemotherapy regimen	n patients	Median OS (months)	1-year SR (%)	ORR (%)	Jadad score
Frasci et al.⁵	2003	≥70	0–2	Gemcitabine 1 200 mg/m² iv, d1, 8 + Vinorelbine 30 mg/m² iv, d1, 8 q.3.w.	60	6.8	30	22	3
Frasci et al.⁵	2003	≥70	0–2	Vinorelbine 30 mg/m² iv, d1, 8 q.3.w.	60	4.2	13	15	3
Kusagaya et al.⁶	2012	≥70	0–1	Gemcitabine 1 000 mg/m²+ Carboplatin (AUC) of 3 d, 15, q.4.w	31	12.6	NA	22.6	3
Kusagaya et al.⁶	2012	≥70	0–1	Gemcitabine 1000 mg/m² d1, 8, 15 q.4.w	30	15.4	NA	10	3
Gridelli et al.⁷	2003	≥70	0–2	Gemcitabine 1 000 mg/m² iv, d1, 8 + Vinorelbine 25 mg/m² iv, d1, 8 q.3.w.	232	7	30	21	3
				Gemcitabine 1 200 mg/m² iv, d1, 8 q.3.w.	233	6.5	28	16
				Gemcitabine 1 000 mg/m² iv, d1, 8 + Vinorelbine 25 mg/m² iv, d1, 8 q.3.w.	232	7	30	21
				Vinorelbine 30 mg/m² iv, d1, 8 q.3.w.	233	8.4	38	18
Abe et al.⁸	2011	≥70	0–1	Docetaxel 20 mg/m² iv, d1, 8, 15 + Cisplatin 25 mg/m² iv, d1, 8, 15 q.4.w.	131	13.3	54.5	34.4	2
Abe et al.⁸	2011	≥70	0–1	Docetaxel 60 mg/m² iv, d1 q.3.w.	126	14.8	58.2	24.6	2
Quoix et al.¹⁹	2012	≥70	0–2	Carboplatin (AUC) of 6 d1 + Paclitaxel 90 mg/m², d1, 8, 15 iv q.4.w.	225	10.3	44.5	27.1	3
Quoix et al.¹⁹	2012	≥70	0–2	Vinorelbine 25 mg/m², d1, 8 iv q.3.w. or Gemcitabine 1 150 mg/m², d1, 8 iv q.3.w.	226	6.2	25.4	10.2	3
Chen et al.²⁰	2008	≥70	1–2	Vinorelbine 22.5 mg/m² iv d1, 8 q.3.w + Cisplatin 50 mg/m² iv d1 q.3.w	34	11.3	47.2	32.4	3
Chen et al.²⁰	2008	≥70	1–2	Vinorelbine 25 mg/m² iv, d1, 8 q.3.w	31	12	50.9	16.1	3
Comella et al.²¹	2004	≥70	0–2	Gemcitabine 1 000 mg/m² iv, d1, 8 + Vinorelbine 25 mg/m², d1, 8 iv q.3.w.	68	9.7	32	23	3
				Gemcitabine 1 000 mg/m² iv, d1, 8 + Paclitaxel 80 mg/m² iv, d1, 8 q.3.w.	65	9.4	44	32
				Gemcitabine 1 200 mg/m² iv, d1, 8, 15 q.4.w.	68	5.1	29	18
				Paclitaxel 100 mg/m² iv, d1, 8, 15 q.4.w.	63	6.4	25	13
Hainsworth et al.²²	2007	≥65	0–2	Gemcitabine 800 mg/m² iv, d1, 8, 15 + Docetaxel 30 mg/m² iv, d1, 8, 15 q.4.w.	174	5.5	26	25	3
Hainsworth et al.²²	2007	≥65	0–2	Docetaxel 36 mg/m² iv, d1, 8, 15 q.4.w.	171	5.1	24	17	3
Tsukada et al.²³	2007	≥70	0–1	Docetaxel 20 mg/m² iv, d1, 8, 15 + Cisplatin 25 mg/m² iv, d1, 8, 15 q.4.w.	63	17	66.6	55	2
Tsukada et al.²³	2007	≥70	0–1	Docetaxel 25 mg/m² iv, d1, 8, 15 q.4.w.	63	10.7	45.2	26.2	2
Karampeazis et al.²⁴	2010	≥70	0–2	Docetaxel 30 mg/m² iv, d1, 8 + Gemcitabine 900 mg/m² iv, d1, 8 q.3.w.	49	15.9	NA	28.3	2
Karampeazis et al.²⁴	2010	≥70	0–2	Gemcitabine 1 200 mg/m² iv, d1, 8 q.3.w.	45	12.2	NA	11.6	2
Rijavec et al.²⁵	2010	≥70	0–2	Docetaxel 35 mg/m² iv, d1, 8, 15 + Gemcitabine 800 mg/m² iv, d1, 8, 15 q.4.w.	36	7.9	NA	30.6	2
Rijavec et al.²⁵	2010	≥70	0–2	Docetaxel 35 mg/m² iv, d1, 8, 15 q.4.w.	33	7.2	NA	15.2	2

PS, performance status; iv, intravenous; OS, overall survival; 1-year SR, 1-year survival rate; ORR, overall response rate; AUC, area under the curve; NA, not applicable.

Publication bias

There was no obvious asymmetry revealed using Begg’s Funnel Plot (P = 0.246 for ORR) or Egger’s Test (P = 0.175 for OS; P = 0.371 for 1-year SR; P = 0.067 for ORR).

Overall survival

Overall survival was defined as time from randomization to death from any cause, censoring patients who had not died at the date last known alive. Pooled OS data available from 10 trials favoured doublet therapy, with a HR of 0.89 (95% CI 0.83, 0.95, P = 0.001), using a random-effects model. Subgroup analysis based on chemotherapy regimens and race found that OS was significantly improved by platinum-based doublet therapy compared with single-agent therapy (HR 0.71, 95% CI 0.60, 0.84, P < 0.001), but there was no significant difference between nonplatinum based doublet therapy and single-agent therapy (HR 0.93, 95% CI 0.86, 1.01, P = 0.078). In the East Asian subgroup, OS was not significantly improved by doublet therapy (HR 0.87, 95% CI 0.66, 1.16, P = 0.336) but OS was improved by doublet therapy in the non-East Asian subgroup (HR 0.89, 95% CI 0.83, 0.95, P = 0.001) (Figure 2).

Figure 2.

Meta-analysis of overall survival hazard ratios (HRs) between double and single-agent therapy in elderly patients with advanced nonsmall-cell lung cancer. HR < 1 corresponds to a higher survival for doublet compared with single-agent chemotherapy. Subgroup analysis based on race and chemotherapy regimens were also performed. G, gemcitabine; V, vinorelbine.

Survival at 1 year

Pooled RR based on data from nine trials for 1-year SR favoured doublet over single-agent therapy (RR 1.15, 95% CI 1.04, 1.28, P = 0.007) using a random-effects model. Subgroup analysis showed that 1-year SR was significantly improved by platinum-based doublets (HR 1.28, 95% CI 1.11, 1.47, P = 0.001), but not by nonplatinum based doublets (HR 1.06, 95% CI 0.91, 1.23, P = 0.446) compared with single-agent therapy. 1-year SR was significantly improved by doublet therapy in the non-East Asian population (pooled RR, 1.19; 95% CI 1.05, 1.36, P = 0.007). In the East Asian population, however, there was no significant difference in pooled RR between doublet and single therapy (RR 1.06, 95% CI 0.90, 1.28, P = 0.494).

Overall response rates

Overall response rates were defined as the sum of partial and complete response rates according to Response Evaluation Criteria in Solid Tumors (RECIST).²⁶ Pooled ORR based on data from 12 trials were significantly in favour of doublet therapy (RR 1.61, 95% CI 1.39, 1.86, P < 0.001), using a fixed-effects model. In addition, subgroup analysis based on chemotherapy regimens showed that platinum-based and nonplatinum- based regimens favoured doublet therapy in terms of ORR (RR 1.88, 95% CI 1.49, 2.38, P < 0.001; RR 1.46, 95% CI 1.21, 1.76, P < 0.001, platinum-based and nonplatinum based, respectively) compared with single-agent therapy. In the East Asian and non-East Asian populations, pooled ORR favoured doublet over single-agent therapy (RR 1.70, 95% CI 1.29, 2.23, P < 0.001; RR 1.58, 95% CI 1.33, 1.88, P < 0.001, East Asian and non-East Asian, respectively).

Safety

Adverse events were calculated according to the NCICTC (version 2 or 3).¹¹ Grade 3–4 toxicity was reported in 11 trials, with the doublet group having a significantly higher number of patients experiencing toxicity (anaemia, thrombocytopenia, fatigue) than the single-agent group. The non-East Asian population had more grade 3–4 toxicities with doublet therapy compared with single agents, however, the number of patients experiencing grade 3–4 thrombocytopenia and fatigue were similar for doublet and single agent therapy in the East Asian population (Table 2).

Table 2.

Summary of grade 3 or 4 adverse events (according to the National Cancer Institute Common Toxicity Criteria, version 2 or 3), in the current meta-analysis of 13 clinical trials (from 11 publications) involving patients with advanced nonsmall-cell lung cancer aged >65 years, who were treated with doublet versus single-agent therapy.

Toxicity-related adverse event	Studies, n	Therapy type		Heterogeneity		RR (95% CI)	Statistical significance
Toxicity-related adverse event	Studies, n	Double-agent therapy	Single-agent therapy	Statistical significance	I ²	RR (95% CI)	Statistical significance
Grade 3–4 anaemia	10	101/1351	44/1337	NS	19%	2.24 (1.59, 3.16)	P < 0.0001
Grade 3–4 neutropenia	11	291/1582	259/1562	P < 0.0001	92%	1.35 (0.68, 2.70)	NS
Grade 3–4 thrombocytopenia	8	66/1252	28/1241	NS	44%	2.29 (1.50, 3.50)	P < 0.0001
Grade 3–4 diarrhoea	6	24/1340	20/1330	NS	0	1.13 (0.63, 2.04)	NS
Grade 3–4 fatigue	8	118/1204	82/1192	NS	0	1.40 (1.09, 1.79)	P = 0.007
Grade 3–4 neurotoxicity	6	15/1243	6/1235	NS	4.60%	2.32 (0.90, 6.02)	NS
Grade 3–4 nausea and vomiting	8	39/1434	31/1421	NS	40%	1.12 (0.69, 1.82)	NS

Data presented as n patient incidence/total n patients (χ²-test).

RR, risk ratio; CI, confidence interval.

NS, no statistically significant between-group differences (P > 0.05; χ2-test).

Discussion

In the present meta-analysis, double-agent chemotherapy was associated with significantly increased OS (HR 0.89, 95% CI 0.83, 0.95, P = 0.001), 1-year SR (RR 1.15, 95% CI 1.04, 1.28, P = 0.007), and ORR (RR 1.61, 95% CI 1.39, 1.86, P < 0.001) compared with single agents. Subgroup analysis showed OS and 1-year SR were significantly improved by platinum-based doublets but not by nonplatinum-based doublets compared with single-agent therapy. Unlike the non-Asian population, however, no significantly increased benefit regarding OS or 1-year SR was associated with doublet therapy in the Asian population. Notably, in the East Asian population, grade 3–4 thrombocytopenia and fatigue were similar with doublet and single-agent therapy. In the non-East Asian population, the incidence of toxicity-related fatigue, thrombocytopenia and anaemia were significantly more frequent with double- compared with single-agent therapy.

In the present study, platinum-based doublet therapy appeared to be more efficacious than single third-generation agents, and the non-Asian population was associated with a greater benefit from doublet-agent versus single-agent therapy compared with the Asian population. Even with single-agent therapy, elderly Asians experienced serious adverse events that were not significantly different compared with double-agent therapy. Severe adverse effects are directly associated with poor health-related quality-of-life and psychological burden.²⁷ In the present meta-analysis, the incidence of serious adverse events may be the reason why double-agent therapy was not associated with survival benefits in the Asian population.

The results of the present systematic review and meta-analysis should be interpreted with caution, since: (1) The definition of older patients was not clearly established. In previous studies, a cut off has been set at 65 years of age, particularly in epidemiological studies, but this is problematic as the median age of lung cancer diagnosis is 70–71 years.^19,28 The cut off for entry into therapeutic trials is usually 70 years of age, probably because treatment may not require age-related adaptation before this age cut off;^19,28 (2) The treatment schedules differed between the included trials. Because of these limitations it could not be concluded that platinum-based doublet therapy should be considered as first-line treatment in patients aged >70 years of age. (3) The results of the present systematic review may be influenced by four large trials,^7,8,19,22 which may be responsible for the lack of survival benefit seen in East Asian patients. It is possible that even after pooling the data, the power may have been inadequate to detect a difference. Thus, further high-quality, randomized controlled trials are warranted to clarify the efficacy of doublet versus single-agent therapy as first line treatment in clearly defined elderly patients.

In conclusion, doublet therapy was superior to single third-generation cytotoxic agents as first-line treatment for elderly patients with advanced NSCLC, when patients were observed to tolerate anaemia, thrombocytopenia and fatigue. There was evidence from the present meta-analysis to suggest that race may be a factor for consideration when choosing double-agent or single-agent therapy as first line treatment of NSCLC in the elderly.

Footnotes

Declaration of conflicting interest

The authors declare that there are no conflicts of interest.

Funding

The present study was supported by the Special Scientific Research for Traditional Chinese Medicine in 2013; (201307006); Longhua Medicial Project (LYTD-25); Trans–Century Training Programme Foundation for the Talents by the State Education Commission (to J. Z.); Young Investigator grant from Shanghai Municipal; Health Bureau and Zhuoshi grant from Fudan University; Ministry of Education Scientific Research Foundation; and Shanghai science and technology commission foundation key project.

References

Siegel

Naishadham

Jemal

. Cancer statistics, 2012. CA Cancer J Clin 2012; 62: 10–29.

Siegel

DeSantis

Virgo

. Cancer treatment and survivorship statistics, 2012. CA Cancer J Clin 2012; 62: 220–241.

Jemal

Bray

Center

. Global cancer statistics. CA Cancer J Clin 2011; 61: 69–90.

Vamvakas

Saloustros

Karampeazis

. Advanced non–small-cell lung cancer in the elderly. Clin Lung Cancer 2009; 10: 158–167.

Frasci

Lorusso

Panza

. Gemcitabine plus vinorelbine versus vinorelbine alone in elderly patients with advanced non-small-cell lung cancer. J Clin Oncol 2003; 18: 2529–2536.

Kusagaya

Inui

Karayama

. Biweekly combination therapy with gemcitabine and carboplatin compared with gemcitabine monotherapy in elderly patients with advanced non-small-cell lung cancer: a randomized, phase-II study. Lung cancer 2012; 77: 550–555.

Gridelli

Perrone

Gallo

. Chemotherapy for elderly patients with advanced non-small-cell lung cancer: the Multicenter Italian Lung Cancer in the Elderly Study (MILES) phase III randomized trial. J Natl Cancer Inst 2003; 95: 362–372.

Abe

Yokoyama

Takeda

. Randomized phase III trial comparing weekly docetaxel (D)-cisplatin (P) combination with triweekly D alone in elderly patients (pts) with advanced non-small cell lung cancer (NSCLC): an intergroup trial of JCOG0803/WJOG4307L. J Clin Oncol 2011; 29: 7509–7509.

Gridelli

Cigolari

Gallo

. Activity and toxicity of gemcitabine and gemcitabine + vinorelbine in advanced non-small-cell lung cancer elderly patients Phase II data from the Multicenter Italian Lung Cancer in the Elderly Study (MILES) randomized trial. Lung Cancer 2001; 31: 277–284.

10.

Jadad

Moore

Carroll

. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 1–12.

11.

National Institutes of Health, National Cancer Institute. Common Terminology Criteria for Adverse Events, version 3.0, http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/ctcae_index.pdf (2003, accessed 07 July 2015).

12.

Therasse

Arbuck

Eisenhauer

. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000; 92: 205–216.

13.

Zhang

Chen

Zhang

. Adjuvant chemotherapy in oesophageal cancer: a meta-analysis and experience from the Shanghai Cancer Hospital. J Int Med Res 2008; 36: 875–882.

14.

Chen

Zhang

Wang

. The platinum-based treatments for advanced non-small cell lung cancer, is low/negative ERCC1 expression better than high/positive ERCC1 expression? A meta-analysis. Lung Cancer 2010; 70: 63–70.

15.

Liu

Song

. Meta-analysis added power to identify variants in FTO associated with type 2 diabetes and obesity in the Asian population. Obesity (Silver Spring) 2010; 18: 1619–1624.

16.

Zintzaras

Ioannidis

. Heterogeneity testing in meta-analysis of genome searches. Genet Epidemiol 2005; 28: 123–137.

17.

Begg

Mazumdar

. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50: 1088–1101.

18.

Egger

Davey Smith

Schneider

. Bias in metaanalysis detected by a simple, graphical test. BMJ 1997; 315: 629–634.

19.

Quoix

. Therapeutic options in older patients with metastatic non-small cell lung cancer. Ther Adv Med Oncol 2012; 4: 247–254.

20.

Chen

Perng

Shih

. A phase II randomized study of vinorelbine alone or with cisplatin against chemo-naïve inoperable non-small cell lung cancer in the elderly. Lung Cancer 2008; 61: 214–219.

21.

Comella

Frasci

Carnicelli

. Gemcitabine with either paclitaxel or vinorelbine vs paclitaxel or gemcitabine alone for elderly or unfit advanced non-small-cell lung cancer patients. Br J Cancer 2004; 91: 489–497.

22.

Hainsworth

Spigel

Farley

. Weekly docetaxel versus docetaxel/gemcitabine in the treatment of elderly or poor performance status patients with advanced nonsmall cell lung cancer: a randomized phase 3 trial of the Minnie Pearl Cancer Research Network. Cancer 2007; 110: 2027–2034.

23.

Tsukada

Yokoyama

Nishiwaki

. Randomized controlled trial comparing docetaxel (D)-cisplatin (P) combination with D alone in elderly patients (pts) with advanced non-small cell lung cancer (NSCLC):JCOG0207. J Clin Oncol 2007; 25(18S): 7629–7629.

24.

Karampeazis

Vamvakas

Pallis

. Docetaxel (D) plus gemcitabine (G) compared with G in elderly patients with advanced non-small cell lung cancer (NSCLC): preliminary results of a randomized phase III Hellenic Oncology Research Group trial. J Clin Oncol 2010; 28(15S): 7605–7605.

25.

Rijavec

Belvedere

Aita

. Docetaxel (D) versus docetaxel/gemcitabine (D&G) in the treatment of older patients with advanced non-small cell lung cancer (NSCLC): an Alpe Adria Thoracic Oncology Multidisciplinary Group randomized phase II trial (ATOM 017). J Clin Oncol 2010; 28(15S): e18028–e18028.

26.

Eisenhauer

Therasse

Bogaerts

. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009; 45: 228–247.

27.

Goffin

Lacchetti

Ellis

. First-line systemic chemotherapy in the treatment of advanced non-small cell lung cancer: a systematic review. J Thorac Oncol 2010; 5: 260–274.

28.

Lebovic

Mahmood

Chiappori

. Lung cancer after 70: is it a different disease? Oncology (Williston Park) 2010; 24: 1106–1111.