Non-small cell lung cancer (NSCLC) remains the commonest

Original Article Clinical Utility of the Pretreatment Glasgow Prognostic Score in Patients with Advanced Inoperable Non-small Cell Lung Cancer Elaine Y.L. Leung, MB ChB,* Hazel R. Scott, MD, and Donald C. McMillan, PhD* Introduction: Traditional tumor-based staging systems provide limited information on the best treatment option for individual patients with advanced inoperable non-small cell lung cancer (NSCLC). The Glasgow prognostic score (GPS) reflects the host systemic inflammatory response and is a validated independent prognostic factor in these patients. The aim of this study was to examine the clinical application of the pretreatment GPS in a mature cohort of patients with inoperable NSCLC. Methods: The data of 261 patients with inoperable NSCLC were collected prospectively and before treatment. Information on patient demographics, body mass index, performance status (PS), the modified Glasgow prognostic score (mgps), the prognostic index, and treatment received were included. Results: The majority of patients were aged 65 years or older (68%), were men (59%), had a body mass index more than 20 (89%), and an Eastern Cooperative Oncology Group performance status (ECOG-PS) 0 or 1 (54%). Most patients had a pretreatment mgps = 1 (62%) and pretreatment prognostic index = 1 (56%). During the follow-up period, 248 (95%) patients died, 246 from their disease. The median survival was 8 months. On multivariate analysis, age (p = 0.001), ECOG-PS (p < 0.05), mgps (p < 0.0001), and tumor stage (p < 0.0001) were independently associated with cancer-specific survival. Using 5-year cancer-specific mortality as an end point, the area under the receiver operator curve was 0.735 (95% confidence interval [CI], 0.566 0.903; p = 0.024) for the mgps, 0.669 (95% CI, 0.489 0.848; p = 0.106) for ECOG-PS, and 0.622 (95% CI, 0.437 0.807; p = 0.240) for tumor, node, metastasis stage. Patients with an increased mgps were more likely to have a poorer ECOG-PS (p < 0.05), an increased white cell count (p < 0.05), and received palliative treatment (p < 0.05). Conclusion: The pretreatment mgps is a useful and important predictor of cancer-specific survival in patients with inoperable NSCLC. Basing clinical assessment on the mgps has implications for the routine monitoring and treatment of the patients. *University Department of Surgery, Faculty of Medicine University of Glasgow, Royal Infirmary, Glasgow, United Kingdom; and Department of Respiratory Medicine, Wishaw General Hospital, Lanarkshire, United Kingdom. Disclosure: The authors declare no conflicts of interest. Address for correspondence: Elaine Y. L. Leung, University Department of Surgery (University of Glasgow), 4th Floor, Walton Building, Glasgow Royal Infirmary, Glasgow, G31 2ER, United Kingdom. E-mail: e.leung@ clinmed.gla.ac.uk Copyright 2012 by the International Association for the Study of Lung Cancer ISSN: 1556-0864/12/0704-0655 Journal of Thoracic Oncology Volume 7, Number 4, April 2012 Key Words: Non-small cell lung cancer, Clinical utility, Glasgow prognostic score. (J Thorac Oncol. 2012;7:655 662) Non-small cell lung cancer (NSCLC) remains the commonest cause of cancer deaths in Western Europe and North America. Approximately 39,000 new cases are diagnosed each year in the United Kingdom. 1 Most patients present with advanced inoperable disease, and the prognosis despite radiotherapy and chemotherapy in this group of patients is poor. A Scottish analysis reported that half of such patients with lung cancer die within 4 months of diagnosis. 2 Traditional staging and grading systems provide limited information on determining the best treatment option for individual patients with advanced inoperable NSCLC. More recently, host factors, including nutritional status, comorbidity, performance status (PS), and the systemic inflammatory response, have been recognized to provide information on outcome in patients with advanced inoperable NSCLC. 3 7 There is now good evidence that C-reactive protein (CRP), 8 10 albumin, 11 and their combination, 12 14 known as the Glasgow prognostic score (GPS), are validated and tumor stage-independent prognostic factors in patients with inoperable NSCLC. Therefore, the aim of this study was to examine the clinical application of the pretreatment modified GPS (mgps) in a large mature cohort of patients with inoperable NSCLC. METHODS Patients and Staging Patients presenting with inoperable NSCLC to a multidisciplinary clinic at Wishaw General Hospital, Lanarkshire, United Kingdom, between May 2001 and November 2004 were studied prospectively. All patients had cytologically or histologically confirmed disease and were staged on the basis of clinical findings, chest radiographs and, where appropriate, bronchoscopy, liver ultrasound, isotope bone scan, and computerized tomography of the thorax, according to the American Thoracic Society tumor, node, metastasis (TNM) classifications. 15 Patients with conditions known to evoke a systemic inflammatory response, either acute or chronic, were excluded. 655

Leung et al. Journal of Thoracic Oncology Volume 7, Number 4, April 2012 These were namely (i) previous chemoradiotherapy for other clinical indications before this study, (ii) clinical evidence of active pretreatment infection, or (iii) chronic active inflammatory diseases such as rheumatoid arthritis. The study was approved by the Research Ethics Committee at Wishaw General Hospital, Lanarkshire, United Kingdom. Body Mass Index and PS Body mass index (BMI) and PS (Eastern Cooperative Oncology Group [ECOG-PS]) were recorded at the time of diagnosis. PS was assessed by one of the two consultant respiratory physicians. mgps and Prognostic Index A venous blood sample was obtained, at diagnosis, for measurement of white cell count, albumin, and CRP concentrations. The coefficient of variation for these methods, over the range of measurement, was less than 5% as established by routine quality control procedures. The GPS was constructed as previously described. 16,17 In brief, CRP more than 10 mg/l and albumin less than 35 g/dl were each given a score of 1. The GPS was calculated as 0, 1, or 2. Importantly, hypoalbuminaemia alone in the absence of an increased CRP level did not confer a poorer cancer-specific survival in all patients with cancer. 7,17 As a result, the GPS was modified, and patients with hypoalbuminaemia alone were assigned a score of 0. 18 The prognostic index (PI) was constructed as previously described. 19 In brief, CRP more than 10 mg/l and white blood cells more than 11 10 9 /L are each given a score of 1. The PI is calculated as 0, 1, or 2. Treatment Received Patients were considered to have had palliative treatment or active treatment as previously described. 16 After diagnosis, treatment recommendations for each case were made after discussion at a multidisciplinary meeting according to various factors, including disease status, PS, comorbidities, and patient wishes. Patients were considered to have undergone active treatment if they received chemotherapy (mainly platinum based) and/or radical radiotherapy. Continuous hyperfractionated accelerated radiotherapy was not available at the time of recruitment of this cohort. Patients receiving palliative radiotherapy for symptom control and/or palliative care (symptom control without chemotherapy or radiation) were considered to have had palliative treatment. Statistics Grouping of the variables was carried out using standard thresholds for laboratory parameters. Data are presented as median and range. The relationships between the groups of patients were carried out using Mantel-Haenszel (χ 2 ) test for trend and the Kruskal-Wallis test as appropriate. Univariate survival analysis and multivariate survival analysis with calculation of hazard ratios (HRs) were performed using Cox s proportional hazards model. A stepwise backward procedure was used to derive a final model of the variables that had a significant independent relationship with survival. Deaths until October 2010 were included in the analysis. Analysis was performed using SPSS software (SPSS, Inc., Chicago, IL). RESULTS Two hundred sixty-one patients with inoperable NSCLC were included in this study. The majority of patients were aged 65 years or older (68%), were men (59%), had BMI 20 (89%), and an ECOG-PS 0 or 1 (54%). Most patients had a pretreatment CRP above the reference range (reference range: ø10 mg/l, 57%) and therefore a pretreatment mgps = 1 (62%) and pretreatment PI = 1 (56%). Only two patients with hypoalbuminaemia had normal CRP concentrations (i.e., GPS = 1 and mgps = 0). The majority of patients had TNM stage III disease (51%) and received palliative treatment (63%). Of all patients, 104 (40%) had their symptoms controlled without chemotherapy or radiation. Sixty-one (23%) subsequently received palliative radiotherapy only, 59 (23%) received palliative chemotherapy only, 12 (5%) received radical radiotherapy, and 25 (10%) received chemoradiation. In patients with stage III disease (n = 134), 23 (17%) received chemotherapy only, 38 (28%) received radiation alone, and 18 (13%) received chemoradiation. All patients with stage IV disease (n = 127) received platinum doublets (e.g., cisplatin plus vinorelbine or carboplatin plus gemcitabine) unless contraindicated, and 35 (28%) received palliative radiotherapy alone. The number of treatment cycles was not available for the analysis. The median follow-up of the survivors was 83.1 months (range, 65.4 144.2 months). No patients were lost to follow-up. The median survival of our patients from the date of diagnosis was 8.0 months (0.4 63.1 months). During the follow-up period, 248 (95%) patients died, 246 from their disease (noncancer causes of death included 1 patient who died of myocardial infarction and another congestive heart failure). The relationship between clinicopathological characteristics and cancer-specific survival is shown in Table 1. On univariate survival analysis, age (p < 0.001), ECOG-PS (p < 0.0001), mgps (p < 0.0001), PI (p < 0.05), tumor stage (p ø t 0.001), and treatment received (p < 0.001) were significantly associated with cancer-specific survival. On multivariate analysis of these significant variables, age (HR: 1.40, 95% confidence interval [CI]: 1.14 1.72, p = 0.001), ECOG-PS (HR: 1.26, 95% CI: 1.05 1.50, p < 0.05), mgps (HR: 1.67, 95% CI: 1.28 2.19, p < 0.0001), and tumor stage (HR: 1.69, 95% CI: 1.26 2.26, p < 0.0001) were independently associated with cancer-specific survival (Table 1). The cancer-specific survival curves for the mgps, ECOG-PS, TNM stage, and treatment received are shown in Figures 1A D. Using 2-year cancer-specific mortality as an end point, the area under the receiver operator curve (ROC) was 0.618 (95% CI, 0.532 0.704; p = 0.014) for the mgps, 0.635 (95% CI, 0.548 0.722; p = 0.007) for ECOG-PS, and 0.574 (95% CI, 0.482 0.665; p = 0.122) for TNM stage. Using 5-year cancer-specific mortality as an end point, the area under the ROC was 0.735 (95% CI, 0.566 0.903; p = 0.024) for the mgps, 0.669 (95% CI, 0.489 0.848; p = 0.106) for ECOG-PS, and 0.622 (95% CI, 0.437 0.807; p = 0.240) for TNM stage. Baseline clinicopathological characteristics grouped according to the mgps are shown in Table 2. Patients with an increased mgps were more likely to have a poorer ECOG-PS (p < 0.05), an elevated white cell count (p < 0.05), and received 656 Copyright 2012 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 7, Number 4, April 2012 Clinical Utility of the GPS in NSCLC TABLE 1. Clinicopathological Characteristics in Patients with Inoperable Non-small Cell Lung Cancer (n=261). Univariate and Multivariate for Cancer Specific Survival. Univariate Multivariate Hazard Ratio (95% CI) P Hazard Ratio (95% CI) P Patient related Age (,65/65274/.75yr) 1.39 (1.1721.66),0.001 1.40 (1.1421.72) 0.001 Gender (male/female) 1.09 (0.8521.41) 0.500 BMI a (,20/<20) 0.86 (0.5821.28) 0.455 ECOG-PS b (0/1/2/3) 1.46 (1.2421.72),0.001 1.26 (1.0521.50) 0.011 White cell count (10 9 /L) (#11/11) 1.23 (0.9321.61) 0.143 mgps (0/1/2) 1.83 (1.4422.34),0.0001 1.67 (1.2822.19),0.0001 PI (0/1/2) 1.28 (1.0621.54) 0.011 Tumor related Pathology (Squamous/adenocarcinoma/ 0.96 (0.8321.12) 0.610 other/no pathology) Tumor stage (III/IV) 1.56 (1.2122.01) 0.001 1.69 (1.2622.26),0.0001 Treatment (palliative/active) 0.60 (0.4620.78),0.0001 a BMI, n 5 190. b ECOG-PS, n 5 206. mgps, modified prognostic score; CI, confidence interval; BMI, body mass index; ECOG-PS, Eastern Cooperative Oncology Group performance status; PI, prognostic index. palliative treatment (p < 0.05). The relationships between the mgps and cancer-specific survival in patients with better ECOG-PS (i.e., PS = 0 or 1) and poorer ECOG-PS (i.e., PS = 2 or 3) are shown in Figures 2A, B. The mgps was associated with poorer cancer-specific survival in these groups of patients (HR: 1.49, 95% CI: 1.01 2.20, p = 0.042 and HR: 1.70, 95% CI: 1.14 2.53, p = 0.009, respectively). Because the mgps was the most powerful prognostic factor, according to ROC analysis, the relationship between clinicopathological characteristics and cancer-specific survival in those patients with an mgps of 0, 1, and 2 was examined in Table 3. In those patients with an mgps of 0, 52 patients died of their cancer. On univariate survival analysis, BMI (p < 0.05) and ECOG-PS (p < 0.05) were significantly associated with cancer-specific survival. On multivariate analysis of these significant variables, only ECOG-PS (HR: 1.47, 95% CI: 1.06 2.05, p < 0.05) was independently associated with cancer-specific survival. In those patients with an mgps of 1, 158 patients died of their cancer. On univariate survival analysis, age (p [ltequ] 0.001), ECOG-PS (p [ltequ] 0.01), TNM stage, (p [ltequ] 0.001), and treatment received (p < 0.01) were significantly associated with cancer-specific survival. On multivariate analysis of these significant variables, only age (HR: 1.66, 95% CI: 1.27 2.16, p < 0.0001) and TNM stage (HR: 1.96, 95% CI: 1.37 2.82, p < 0.001) were independently associated with cancer-specific survival. In those patients with an mgps of 2, 38 patients died of their cancer. On univariate survival analysis, only treatment given (HR: 0.61, 95% CI: 0.45 0.84, p < 0.01) was significantly associated with cancer-specific survival. The relationships between the mgps and cancerspecific survival in patients who received palliative treatments and active treatments are shown in Figures 3A, B. The mgps was inversely associated with cancer-specific survival in patients who received palliative treatments (n = 165, HR: 2.19, 95% CI: 1.63 2.93, p < 0.0001), with a similar trend in patients who received active treatments (n = 96, HR: 1.39, 95% CI: 0.89 2.17, p = 0.142). CONCLUSION The results of this study demonstrate the clinical utility of the pretreatment mgps, the most powerful prognostic factor, because when the patients were stratified for the mgps, no other factor whether it be ECOG-PS and clinical TNM stage was consistently associated with cancer survival, independent of treatment received. For example, when those patients who had a good PS (ECOG-PS = 0 and 1; Figure 2A) and most likely to receive active treatment in current treatment regimens were analyzed, it was clear that the mgps further stratified those patients most likely to benefit from treatment. Also, when those patients with poorer PS (ECOG-PS 2; Figure 2B) and least likely to receive active treatment in current treatment regimens were analyzed, it was clear that the mgps further stratified those patients least likely to benefit from treatment. In contrast, ECOG-PS did not consistently stratify patients within the mgps groups confirming the superior prognostic value of the mgps in patients with advanced NSCLC. Therefore, although mgps and ECOG-PS are significantly associated, the mgps has considerable clinical utility because it provides a superior objective pretreatment predictor of survival compared with the factors included in a routine clinical assessment, in particular, ECOG-PS. Given the rapid progression of the disease and short survival of patients with inoperable NSCLC, these results would suggest that the simple, objective, and well-standardized mgps should form a regular part of a clinical pretreatment assessment. The results of this study are consistent with the work of Crumley et al. 20 who have proposed a new clinical staging Copyright 2012 by the International Association for the Study of Lung Cancer 657

Leung et al. Journal of Thoracic Oncology Volume 7, Number 4, April 2012 FIGURE 1. A, The relationship between the modified Glasgow prognostic score (mgps; 0, 1, 2 from top to bottom) and cancer-specific survival in patients with inoperable non-small cell lung cancer (p < 0.0001). B, The relationship between tumor, node, metastasis (TNM) stage (III, IV from top to bottom) and cancer-specific survival in patients with inoperable non-small cell lung cancer (p = 0.001). C, The relationship between Eastern Cooperative Oncology Group performance status (ECOG-PS) (PS = 0 to 3 from top to bottom) and cancer-specific survival in patients with inoperable non-small cell lung cancer (p < 0.001). D, The relationship between treatment received (palliation only to chemoradiation from top to bottom) and cancer-specific survival in patients with inoperable non-small cell lung cancer (p < 0.0001). system based on mgps in patients presenting with gastroesophageal cancer, also an aggressive cancer. It is of interest that Iimura et al. 21 recently reported the development and validation of the TNM-CRP score in patients undergoing nephrectomy for renal clear cell cancer. A similar approach has also been carried out in bladder cancer. 22 Recently, in almost 9000 patients, the mgps was shown to be superior to other markers of the systemic inflammatory response across a wide variety of cancers. 23 Therefore, we believe that the mgps should be incorporated into the routine clinical pretreatment assessment of patients with cancer. As demonstrated in this study, the pretreatment mgps identifies a subgroup of patients with advanced NSCLC who are at higher risk of dying from their disease, independent of 658 Copyright 2012 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 7, Number 4, April 2012 Clinical Utility of the GPS in NSCLC TABLE 2. Clinicopathological Characteristics and Treatment Received in Patients in Patients with Inoperable Non-small Cell Lung Cancer (n= 261). mgps 5 0 n 5 59 (%) mgps 5 1 n 5 163 (%) mgps 5 2 n 5 39 (%) Patient related Age,65 22 (37) 49 (30) 13 (33) 0.571 65274 22 (37) 76 (47) 11 (28) 75 15 (25) 38 (23) 15 (38) Gender Female 28 (47) 102 (63) 24 (62) 0.122 Male 31 (53) 61 (37) 15 (38) BMI a,20 10 (19) 16 (13) 4 (14) 0.459 20 42 (81) 104 (87) 14 (86) ECOG-PS b 0 13 (25) 19 (15) 2 (8) 0.015 1 18 (35) 53 (41) 7 (29) 2 13 (25) 46 (35) 4 (17) 3 8 (15) 12 (9) 11 (46) White cell count (10 9 /L) 11 48 (81) 114 (70) 22 (56) 0.029.11 11 (19) 49 (30) 17 (44) PI 0 48 (81) 0 (0) 0 (0) <0.0001 1 11 (19) 114 (70) 22 (56) 2 0 (0) 49 (30) 17 (44) Tumor related Pathology Squamous 17 (29) 66 (40) 12 (31) 0.968 Adenocarcinoma 16 (27) 35 (21) 8 (21) Other 4 (7) 10 (6) 5 (13) No pathology 22 (37) 52 (32) 14 (44) TNM Stage III 27 (46) 92 (56) 15 (38) 0.082 IV 32 (54) 71 (44) 24 (62) Treatment Palliative 40 (68) 93 (57) 32 (82) 0.011 Active 19 (32) 70 (43) 7 (18) a BMI, n 5 190. b ECOG-PS, n 5 206. mgps, modified prognostic score; BMI, body mass index; ECOG-PS, Eastern Cooperative Oncology Group performance status; PI, prognostic index; TNM, tumor, node, metastasis. P tumor stage and regardless of treatment they subsequently received. It is reasonable to propose that the mgps is used routinely to inform the planning of supportive measures for these patients, and patients with a systemic inflammatory response may benefit from closer follow-up or earlier involvement of palliative care. The basis of the relationship between the systemic inflammatory response (mgps) and poorer cancer-specific survival is unclear and likely to be multifactorial. However, it is clear that systemic inflammatory markers, such as CRP, play an important role in the tumor-host relationship. Increased CRP level is associated with lymphocytopenia and impaired T lymphocytic response within the tumor, 24,25 as well as the upregulation of components of the innate immune system, including complement and macrophage function. 24,26 Moreover, proinflammatory cytokines and growth factors, which may promote and maintain tumor growth, are also increased as part of the systemic inflammatory response. 27,28 The mgps resulted from the realization that the process of cachexia in patients with cancer resulted mainly from an ongoing systemic inflammatory response. The rationale and development of the mgps, from the selective combination of CRP and albumin, have been previously described in detail. 18 A strength of this study is that it examines the clinical utility of the mgps in a mature cohort of patients with advanced NSCLC, i.e., 95% of patients had died on follow-up. Copyright 2012 by the International Association for the Study of Lung Cancer 659

Leung et al. Journal of Thoracic Oncology Volume 7, Number 4, April 2012 FIGURE 2. A, The relationships between modified Glasgow prognostic score (mgps) and cancer-specific survival in patients with inoperable non-small cell lung cancer and Eastern Cooperative Oncology Group performance status (ECOG-PS) = 0 or 1 (p = 0.042). B, The relationships between mgps and cancer-specific survival in patients with inoperable non-small cell lung cancer and ECOG-PS = 2 or 3 (p = 0.009). However, given that the patients studied were recruited between 2001 and 2004, a limitation of this study is that treatment regimes have changed over this period, and it is not clear whether these results can be extrapolated to current treatment regimes. Nevertheless, it is important to note the consistency TABLE 3. Clinicopathological Characteristics and Inoperable Non-small Cell Lung Cancer. Univariate and Multivariate Survival Analyses for Cancer Specific Survival According to mgps (n 5 261). mgps 5 0 (n 5 59) mgps 5 1 (n 5 163) mgps 5 2 (n 5 39) Multivariate Univariate Multivariate Univariate Multivariate Univariate HR (95% CI) p HR (95% CI) p HR (95% CI) p HR (95% CI) p HR (95% CI) p HR (95% CI) p Patient related 1.22 (0.8421.78) 0.293 1.51 (1.1921.91) 0.001 1.66 (1.2722.16),0.001 1.14 (0.7821.67) 0.488 Age (<65/ 65274/ >75years) 1.03 (0.5921.79) 0.923 1.12 (0.8121.55) 0.486 1.43 (0.6922.96) 0.336 Gender (male/female) 0.40 (0.1920.85) 0.018 1.19 (0.6922.05) 0.525 1.95 (0.6026.36) 0.268 BMI a (,20/ 20) 1.47 (1.0622.04) 0.020 1.47 (1.0622.05) 0.022 1.34 (1.0721.67) 0.010 1.42 (0.9122.22) 0.124 ECOG2PS b (0/1/2/3) 1.41 (0.6922.88) 0.349 1.09 (0.7721.53) 0.635 0.92 (0.4721.79) 0.797 White cell count (10 9 /L) 11/.11 1.41 (0.6922.88) 0.349 1.09 (0.7721.53) 0.621 0.92 (0.4721.79) 0.797 PI (0/ 1/ 2) Tumor related 0.900 (0.6421.26) 0.531 0.96 (0.7921.16) 0.649 1.10 (0.7821.56) 0.573 Pathology (squamous/adenocarcinoma/ 1.39 (0.7822.48) 0.271 1.72 (1.2522.37) 0.001 1.96 (1.3722.82),0.001 1.35 (0.6822.69) 0.390 other/no pathology) Tumour stage (III/IV) 0.77 (0.4221.41) 0.400 0.61 (0.4520.84) 0.003 0.61 (0.4520.84) 0.61 (0.4520.84) 0.003 Treatment (palliative/ active) a BMI, n 5 190. b ECOG-PS, n 5 206. mgps, modified prognostic score; HR, hazard ratio; BMI, body mass index; ECOG-PS, Eastern Cooperative Oncology Group performance status; PI, prognostic index. 660 Copyright 2012 by the International Association for the Study of Lung Cancer

Journal of Thoracic Oncology Volume 7, Number 4, April 2012 Clinical Utility of the GPS in NSCLC to confirm whether targeting patients with advanced NSCLC and a systemic inflammatory response with active oncological treatment confers significant clinical benefit. In summary, in this study, we have shown the clinical utility of the pretreatment mgps as the basis of predicting likely cancer-specific survival in patients with inoperable NSCLC. The mgps is an objective measurement readily applied in a standard clinical setting, and consideration should be given to its routine clinical use. ACKNOWLEDGMENTS Supported by the University Department of Surgery, Royal Infirmary, Glasgow. The authors acknowledge the assistance of Lynn M. Forrest for her role in data collection for this study. FIGURE 3. A, The relationships between modified Glasgow prognostic score (mgps) and cancer-specific survival in patients with inoperable non-small cell lung cancer and received palliative treatment (n = 165; p < 0.0001). B, The relationships between mgps and cancer-specific survival in patients with inoperable non-small cell lung cancer and who received active treatment (n = 96; p = 0.142). and reliability of the prognostic value of the mgps across tumor types and treatment regimes, 6,23,29,30 which make it likely that the mgps will also have clinical utility in the most recent treatment regimens. Prospective studies, ideally within the context of randomized controlled trials, will be required REFERENCES 1. Cancer Research UK (CRUK). Cancer statistics: lung cancer; 2008. Available at: http://info.cancerresearchuk.org/cancerstats/types/lung/. 2. Scottish Executive Health Department. Cancer scenarios: an aid to planning cancer services in Scotland in the next decade; 2001. Available at: http://www.sehd.scot.nhs.uk/publications/csatp/csatp-00.htm. 3. Scott HR, McMillan DC, Forrest LM, et al. The systemic inflammatory response, weight loss, performance status and survival in patients with inoperable non-small cell lung cancer. Br J Cancer 2002;87:264 267. 4. Gioulbasanis I, Baracos VE, Giannousi Z, et al. Baseline nutritional evaluation in metastatic lung cancer patients: mini nutritional assessment versus weight loss history. Ann Oncol 2011;22:835 841. 5. MacDonald N. Cancer cachexia and targeting chronic inflammation: a unified approach to cancer treatment and palliative/supportive care. J Support Oncol 2007;5:157 162. 6. McMillan DC. Systemic inflammation, nutritional status and survival in patients with cancer. Curr Opin Clin Nutr Metab Care 2009;12:223 226. 7. Proctor MJ, Morrison DS, Talwar D, et al. An inflammation-based prognostic score (mgps) predicts cancer survival independent of tumour site: a Glasgow Inflammation Outcome Study. Br J Cancer 2011;104:726 734. 8. Wilop S, Crysandt M, Bendel M, et al. Correlation of c-reactive protein with survival and radiographic response to first-line platinum-based chemotherapy in advanced non-small cell lung cancer. Onkologie 2008;31:665 670. 9. Koch A, Fohlin H, Sörenson S. Prognostic significance of c-reactive protein and smoking in patients with advanced non-small cell lung cancer treated with first-line palliative chemotherapy. J Thorac Oncol 2009;4:326. 10. Gagnon B, Abrahamowicz M, Xiao Y, et al. Flexible modeling improves assessment of prognostic value of C-reactive protein in advanced nonsmall cell lung cancer. Br J Cancer 2010;102:1113 1122. 11. Arrieta O, Ortega M, Rosa M, et al. Association of nutritional status and serum albumin levels with development of toxicity in patients with advanced non-small cell lung cancer treated with paclitaxel-cisplatin chemotherapy: a prospective study. BMC Cancer 2010;10:50. 12. Forrest LM, McMillan DC, McArdle CS, et al. Comparison of an inflammation-based prognostic score (GPS) with performance status (ECOG) in patients receiving platinum-based chemotherapy for inoperable non-small-cell lung cancer. Br J Cancer 2004;90:1704 1706. 13. Forrest LM, McMillan DC, McArdle CS, et al. A prospective longitudinal study of performance status, an inflammation-based score (GPS) and survival in patients with inoperable non-small-cell lung cancer. Br J Cancer 2005;92:1834 1836. 14. Brown DJF, Milroy R, Preston T, et al. The relationship between an inflammation-based prognostic score (Glasgow Prognostic Score) and changes in serum biochemical variables in patients with advanced lung and gastrointestinal cancer. J Clin Pathol 2007;60:705 708. 15. Mountain CF, Greenberg SD, Fraire AE. Tumor stage in non-small cell carcinoma of the lung. Chest 1991;99:1258 1260. 16. Forrest LM, McMillan DC, McArdle CS, et al. Evaluation of cumulative prognostic scores based on the systemic inflammatory response in Copyright 2012 by the International Association for the Study of Lung Cancer 661

Leung et al. Journal of Thoracic Oncology Volume 7, Number 4, April 2012 patients with inoperable non-small-cell lung cancer. Br J Cancer 2003;89:1028 1030. 17. McMillan DC, Crozier JE, Canna K, et al. Evaluation of an inflammationbased prognostic score (GPS) in patients undergoing resection for colon and rectal cancer. Int J Colorectal Dis 2007;22:881 886. 18. McMillan DC. An inflammation-based prognostic score and its role in the nutrition-based management of patients with cancer. Proc Nutr Soc 2008;67:257 262. 19. Kasymjanova G, Macdonald N, Agulnik JS, et al. The predictive value of pre-treatment inflammatory markers in advanced non-small-cell lung cancer. Curr Oncol 2010;17:52 58. 20. Crumley AB, Stuart RC, McKernan M, et al. Comparison of pre-treatment clinical prognostic factors in patients with gastro-oesophageal cancer and proposal of a new staging system. J Gastrointest Surg 2010;14:781 787. 21. Iimura Y, Saito K, Fujii Y, et al. Development and external validation of a new outcome prediction model for patients with clear cell renal cell carcinoma treated with nephrectomy based on preoperative serum c-reactive protein and TNM classification: the TNM-C score. J Urol 2009;181:1004 1012. 22. Gakis G, Todenhoefer T, Renninger M, et al. Development of a new outcome prediction model in carcinoma invading the bladder based on preoperative serum C-reactive protein and standard pathological risk factors: the TNR-C score. BJU Int 2011;108:1800 1805. 23. Proctor MJ, Morrison DS, Talwar D, et al. A comparison of inflammationbased prognostic scores in patients with cancer. A Glasgow Inflammation Outcome Study. Eur J Cancer 2011;47:2633 2641. 24. Du Clos TW, Mold C. C-Reactive protein: an activator of innate immunity and a modulator of adaptive immunity. Immunol Res 2004;30:261 277. 25. Nozoe T, Matsumata T, Sugimachi K. Preoperative elevation of serum c-reactive protein is related to impaired immunity in patients with colorectal cancer. Am J Clin Oncol 2000;23:263. 26. Coussens LM, Werb Z. Inflammation and cancer. Nature 2002;420:860 867. 27. Abramovitch R, Marikovsky M, Meir G, et al. Stimulation of tumour growth by wound-derived growth factors. Br J Cancer 1999;79:1392 1398. 28. Canna K, Hilmy M, McMillan DC, et al. The relationship between tumour proliferative activity, the systemic inflammatory response and survival in patients undergoing curative resection for colorectal cancer. Colorectal Dis 2008;10:663 667. 29. Roxburgh CS, McMillan DC. Role of systemic inflammatory response in predicting survival in patients with primary operable cancer. Future Oncol 2010;6:149 163. 30. Clarke SJ, Chua W, Moore M, et al. Use of inflammatory markers to guide cancer treatment. Clin Pharmacol Ther 2011;90:475 478. 662 Copyright 2012 by the International Association for the Study of Lung Cancer