Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Surg Res. Author manuscript; available in PMC 2012 September 1.
Published in final edited form as:
PMCID: PMC3154461

Obesity and Weight Loss at Presentation of Lung Cancer are Associated with Opposite Effects on Survival



Lung cancer is the second most common neoplasm and the leading cause of cancer deaths in the United States. In cancer, weight loss and obesity are associated with reduced survival. However, the effect of obesity or weight loss at presentation on lung cancer survival has not been well studied.

Materials and Methods

Using an extensive cancer dataset, we identified 76,086 patients diagnosed with lung cancer during the period of 1998–2002, of which 14,751 patients presented with obesity and/or weight loss. We examined the relationship between survival and weight loss or obesity at diagnosis using univariate and multivariate analysis.


Median survival time (MST) for all lung cancer patients was 8.7 months. Patients presenting with weight loss (15.8%) had shorter MST versus those who did not (6.4 vs. 9.2 months, p<0.001) and patients with weight loss had significantly shortened MST for all stages and histological subtypes. In contrast, obese patients at presentation (5.4%) had longer MST relative to non-obese patients (13.0 vs. 8.6 months, p<0.001), which was significant across all stages and histological subtypes. Multivariate analysis revealed that the absence of weight loss was an independent, positive predictor of improved survival (HR=0.087, p<0.001), while the absence of obesity was an independent predictor of worsened survival in lung cancer (HR=1.16, p<0.001).


Our results demonstrate an inverse relationship between survival and weight loss at presentation and a potentially protective effect of obesity in lung cancer survival, which could be due to greater physiologic reserves, thereby prolonging life by slowing the progress of cancer cachexia.

Keywords: lung cancer, obesity, weight loss, cachexia, outcomes, survival


Among all malignancies, lung cancer is the second most common neoplasm and the leading cause of cancer deaths in the United States (1). Despite recent advances in treatments, survival remains dismal among patients diagnosed with lung cancer with overall mortality rates of approximately 72% for men and 41% for women for the years 2001–2005 (2). Numerous studies have examined prognostic factors in lung cancer outcomes, and several studies have included co-morbid conditions in their analysis. However, data specifically examining the impact of obesity and/or weight loss at presentation on survival are scant.

Despite that obesity has become an epidemic globally, the influence obesity has on lung cancer survival has not been well established (3).There is a great deal of literature displaying the positive relationship obesity has on the incidence of many cancers, but literature on the impact obesity has on the survival of cancer patients is limited (4). This is especially true when reviewing the literature on the impact obesity has on lung cancer survival. Recently, there has arisen debate in the literature with the role obesity plays on survival in lung cancer patients. Some studies have identified an increased risk of mortality in obese lung cancer patients, while others have shown a decreased risk of mortality in this population (510). Also, there is discussion regarding the causality of smoking status (smokers, ex-smokers, lifelong non-smokers) on the relationship between obesity and lung cancer survival (4, 10, 11).

In an attempt to understand outcomes for lung cancer patients and potentially to improve survival, we examined a population-based registry to identify prognostic factors important in the treatment outcomes for patients with lung cancer. Due to the unresolved debate in the literature, our main focus was on the role obesity and the absence of weight loss at presentation has on lung cancer survival.

Materials and Methods

The University of Miami Institutional Review Board and the State of Florida Department of Health Institutional Review Board both approved this study. We studied data from Florida’s large, population based cancer registry, the Florida Cancer Data System (FCDS) to identify all incident cases of lung carcinoma diagnosed between the years of 1998–2002. This dataset was further enhanced with data linked to Florida's Agency for Health Care Administration (AHCA) dataset. The AHCA maintains two databases (Hospital Patient Discharge Data and Ambulatory Outpatient Data) on all patient encounters within hospitals and freestanding ambulatory surgical and radiation therapy centers in Florida. All hospitals have been required to report all discharges and outpatient encounters to AHCA since 1987. The AHCA data sets used in this study contain diagnoses and procedures performed during every hospitalization or outpatient encounter in the state of Florida, for the period 1998 – 2002. Cases in the FCDS and AHCA datasets were linked on the basis of unique identifiers and confirmed with the patient’s date of birth and gender. Postal codes listed in the FCDS-AHCA database were used to determine the community poverty levels where patients resided according to the 2007 US Census Bureau report (12). Non-Florida residents (seasonal residents) were not included in the analysis as follow-up for such patients, particularly survival information, may be inaccurate in up to 10% of such patients (FCDS personal communication).

Florida constitutes approximately 6% of the total United States population, making these datasets the most complete American cohort to date, providing multi-institutional data and the ability to evaluate and correct for the impact of patient demographic, co-morbidities and treatment variables in the examination of lung carcinoma.

The staging criteria used by the FCDS are consistent with the Surveillance, Epidemiology, and End Result (SEER, National Cancer Institute) summary staging and differ from the TNM (Tumor, Node, and Metastasis) staging guidelines. In this study, local staging represents disease that does not extend beyond the lung parenchyma, while those having positive lymph nodes at the time of resection were classified as having regional disease. Documentation of distant metastases during the peri-operative period led to classification of affected patients as having distant disease.

Statistical analysis was performed with SPSS Statistical Package version 15.0 (SPSS Inc., Chicago, IL). Correlations between categorical variables were made using the Chi-square test. Five-year survivals were calculated by the Kaplan-Meier method. Since the FCDS collects only primary cause of death, only overall survival was analyzed and not disease-specific survival. Survival was calculated from the time of the initial diagnosis to date of last contact (or date of death). The univariate analysis effects of demographic, clinical, and treatment variables on survival were tested by the Log-rank test for categorical values. Multivariate analysis was performed using a Cox regression model which included age, sex, race, ethnicity, smoking status, tumor stage (local, regional, distant), tumor grade (well-, moderately-, poorly- or un-differentiated), nodal status, surgery, radiation, chemotherapy, weight loss and obesity.

All co-morbidities, including weight loss and obesity were identified in the database by ICD-9 coding from The International Classification of Diseases, 9th Revision, Clinical Modification. Co-morbidities in the analysis were: congestive heart failure, cardiac arrhythmias, valvular disease, pulmonary circulation disorders, peripheral vascular disorders, hypertension, paralysis, other neurological disorders, chronic pulmonary disease, uncomplicated diabetes, complicated diabetes, hypothyroidism, renal failure, liver disease, peptic ulcer disease (excluding bleeding), AIDS/HIV, lymphoma, rheumatoid arthristis/collagen vascular disease, coagulopathy, obesity, weight loss, fluid and electrolyte disorders, blood loss anemia, deficiency anemia, alcohol abuse, drug abuse, psychoses, and depression. Obesity and weight loss were analyzed independently of each other, such that patients exhibiting both weight loss and obesity at presentation were not included in our analysis.


Patient demographics and clinical variables are summarized in Table 1. The cohort was divided near evenly among men (55.6%) and women (44.4%). The majority of the patients were Caucasian (93.2%), non-Hispanic (94.3%), and over the age of 70 years (51.5%). Over half of the patients (57.1%) were identified according to the US Census data as living in a community where 10% or less of the residents were at or below Florida’s poverty level.

Table 1
Patient Demographics

Among the reported co-morbid conditions, 15.4% of patients were reported to have weight loss at time of diagnosis while 4.8% of patients were reported to be obese. There were few obese patients who were also diagnosed with weight loss (335 patients, 9.58%) However, significantly fewer obese patients (9.58% or 335 patients) were reported to have weight loss than in the entire cohort overall (15.71% or 10,919 patients, p<0.001). In comparison to local disease (12.9%), weight loss was marginally seen more frequently in regional (15.4%, p<0.001) and distant disease (15.3%, p<0.001) (Table 2). Obesity was more commonly reported among local disease (5.5%) in contrast to regional (4.3%, p<0.001) and distant disease (3.1%, p<0.001). Also, there was a significant difference in obesity between regional disease and distant disease (4.31% vs. 3.05%, p<0.001).

Table 2
Distribution of Patients by Stage.

Co-morbidities and smoking status were analyzed, with odds ratios calculated for patients with and without weight loss as well as for patients with and without obesity (Table 3). Patients with weight loss showed significantly greater rates of every co-morbid condition analyzed, with the exceptions of simple diabetes, for which weight losing patients showed a lower rate (16.6% vs. 17.6% respectively, p<0.01), hypothyroidism and paralysis, of which the rates were statistically indistinguishable for weight losing and non-weight losing patients. Weight losing patients also exhibited a higher propensity (odds ratio > 1) for each co-morbid condition except for simple diabetes and rheumatoid arthritis. As might be expected, obese patients also showed increased rates for every co-morbid condition analyzed, with the exceptions of HIV/AIDS and lymphoma, the rates of which were indistinguishable between obese patients and non-obese patients. Obese patients also exhibited a higher propensity for each co-morbidity, except for HIV/AIDS. Smoking status in this study is a self-reported categorical variable and weight losing patients were slightly more likely to report smoking (91.9% vs. 90.6%, p<0.001, OR:1.17), while obese patients were slightly less likely to report smoking (89.2% vs. 90.8%, p<0.01, OR:0.84) (Table 3).

Table 3
Frequency of Co-morbidities


The median survival time (MST) for the entire cohort was 8.7 months. Overall, patients who were reported to have weight loss had a significantly shorter MST compared to those who did not have weight loss (6.4 versus 9.2 months respectively, p<0.001) (Figure 1 and Table 4). At every stage (Table 4) and histological subtype (Table 4), patients with weight loss showed reduced MST versus those without. Consistent with severity of the disease, patients with distant disease had the shortest MST, with longer survival for patients with regional disease and longest for those with local disease (p < 0.001) (Table 4). Patients with weight loss showed statistically reduced MST at every stage relative to non-weight losing patients (Table 4). The differences were fairly large. MST for weight losing patients with distant disease was 0.7 months or about 3 weeks less than that for patients without weight loss (p<0.001), while MST was reduced by 3.3 months for those with regional disease (p < 0.001), and reduced by 15.6 months for patients with local disease (p < 0.001). When analyzed within histological diagnoses, weight loss was also associated with reduced MST relative to patients without weight loss for small cell lung cancer (−1.6 months, p<0.001), as well as for non-small cell lung cancer (−3.3 months, p<0.001), and for other histological subtypes (−1.2 months, p<0.001) (Table 4). When restricting to non-smokers, the presence of weight loss had a significantly worse median survival as compared to non weight-losing patients (7.4 months vs. 12 months respectively) (P<0.0001) (Figure 2).

Figure 1
Kaplan-Meier survival curves for all patients with diagnosis of (A) lung cancer and obesity or (B) lung cancer and weight loss.
Figure 2
Kaplan-Meier survival curves for non-smoking patients with diagnosis of (A) lung cancer and obesity or (B) lung cancer and weight loss.
Table 4
Median Survival in Months by Weight Loss or Obesity and Stage, Histological Subtype.

Contrary to weight loss, diagnosis of obesity was associated with longer MST in lung cancer than for non-obese patients. Overall, MST was increased by 4.4 months in obese patients (13.0 months versus 8.6 months, p<0.001) (Figure 1 and Table 4). Consistent with severity of disease, MST declined with advancing stage, both in obese patients and non-obese patients (Table 4). But, when comparing obese patient to non-obese patients at every stage obese patients had significantly increased MST. Obese patients with distant disease, showed 0.7 months or 3 weeks increased MST as compared to non-obese patients (p < 0.001). MST in obese patients was increased by 6.7 months (p < 0.001) with regional disease and was increased by 25 months (p < 0.001) for patients with local disease. Across histological subtypes (Table 4), a diagnosis of obesity was associated with increased MST relative to patients without obesity. MST in obese patients was increased by 1 month (p < 0.007) with small cell lung cancer, increased by 6.4 months (p < 0.001) with non-small cell lung cancer, and increased by 6.9 months (p < 0.001) for those with other histological types. Also, when restricting to never smokers, the presence of obesity had a significantly improved median survival as compared to non obese patients (10.6 months vs. 17.5 months respectively) (P<0.0001) (Figure 2).

Multivariate analysis

The results of a multivariate analysis using the Cox regression method, including patient demographics, tumor characteristics, co-morbid conditions, and treatment variables are summarized in Table 5. Independent predictors of worse survival in lung carcinoma were the absence of obesity (HR=1.12, p <0.001), increasing pathological grade (moderate, poor, undifferentiated) (HR:1.38-1.9, P<0.0001), the absence of radiation (HR:1.2, P<0.0001), chemotherapy (HR:1.51, P<0.0001) or surgery (HR:4.6, P<0.0001), and a positive smoking history (HR:1.073, P=0.007). Conversely, the absence of weight loss was a positive predictor of improved overall outcomes (HR=0.087, p<0.001).

Table 5
Multivariate Analysis


It has been extensively documented in the literature that obesity has a positive relationship with the incidence of numerous cancers and negatively impacts survival among cancer patients (4, 13, 14). However, details regarding individual sites of cancers and their relation to mortality are less available. Associations of obesity with an increased risk of mortality have been noted for various cancers including endometrial, kidney, gallbladder, esophageal, and post-menopausal breast cancer (4, 8, 14, 15). However, the data are scant and inconsistent regarding obesity and the risk of mortality among lung cancer patients. Our study shows the beneficial role obesity or the absence of weight loss at presentation has on lung cancer survival, which has been previously suggested in the literature (46, 810).

In our cohort, approximately 15% and 5% of patients presented with weight loss and obesity, respectively at time of diagnosis. In the evaluation of co-morbid conditions, weight losing patients displayed a higher propensity (odds ratio > 1) for almost every co-morbid condition as compared to non-weight losing patients, suggesting that weight loss might be a marker of /contributor towards general overall lower health. Obese patients, however, also showed increased rates and propensities (odd ratios > 1) of nearly all co-morbidities analyzed, indicating that obese patients also suffered reduced overall health relative to non-obese patients.

Interestingly, despite their overall increased morbidity, patients who were reportedly obese had a 4.4 month overall survival advantage over patients who were not obese (14 vs. 8.6 months respectively). This inverse relationship between obesity and lung cancer survival has been previously documented in the literature (5, 6, 810). In a prospective study of close to 900,000 U.S. Adults, Calle et al. demonstrated a significant inverse relationship between obesity and lung cancer mortality. This inverse relationship was present when analyzing data for both smokers and non-smokers. But, interestingly, when excluding smokers from the analysis, the apparent inverse relationship between BMI and lung cancer mortality was not present. This later data was not statistically significant but there appeared to be a trend of decreasing relative risk of mortality with increasing BMI in lung cancer patients (4). Also, in Great Britain’s prospective Million Women Study with 45,037 incident cancers and17,203 cancer deaths, they showed an inverse association between increasing BMI and cancer mortality in premenopausal breast, squamous cell carcinoma of the esophagus and lung cancer (16). Also, Leung et al. performed a prospective cohort analysis in China (2000–2003) following 60,000 patients for 3 years ultimately showing that obesity was associated with lower lung cancer mortality (6). Parr et al. studied the relationship between BMI and cancer mortality in a large (424,519) Asia-Pacific (Asia, Australia, New Zeland) cohort and showed increased mortality along all types of cancers except for lung and upper aerodigestive tract cancers, with both displaying an inverse relationship (8).

This inverse association supports the phenomenon of paradoxically improved survival with obesity in the face of a chronic condition, as seen in congestive heart failure with preserved systolic function (1719). This population appears to display an “obesity paradox”, where being overweight increases one’s likelihood of suffering from congestive heart failure, but being obese with heart failure decreases their mortality (17). Therefore, it appears an “obesity paradox” may also be present in lung cancer since multiple studies, as well as ours have shown an increased risk of lung cancer with increasing BMI as well as waist circumference (5, 11, 20, 21) and a decreasing mortality when overweight (610, 22).

Weight loss among cancer patients has been reported in up to 50% of all cancer patients (23) and lung cancer is one of the more prevalent cancers where weight loss has been observed (24). Among patients with small cell cancer, weight loss has been noted in nearly 57% of patients; similarly patients with non-small cell cancer have had reported weight loss rates of approximately 54% (25). The observed rate of weight loss among our cohort was 15.4%; however, this rate was static, observed at time of diagnosis, and could not be measured as a function of disease progression. Despite the static nature of this variable, weight loss noted at diagnosis remained an independent negative predictor of overall outcomes with shorter MST in patients with weight loss (6.4 months) versus those without weight loss (9.2 months). Our observations support current literature demonstrating that weight loss is a poor prognostic factor in cancer survival (26, 27).

MST was also clearly related to stage of disease, as advanced disease demonstrated shorter MST which is consistent with reports documenting weight loss and cachexia with increased lung cancer staging (28). In our study a greater percentage of patients reported to have weight loss had regional and distant disease. Similarly, patients with advanced disease were less likely to be obese.

Even under multivariate analysis, when accounting for possible confounding variables (demographics, SES, smoking status, tumor characteristics, and treatment strategies), we were able to show that the absence of obesity was an independent negative predictor of survival (HR: 1.12, P<0.0001)) and the absence of weight loss was an independent positive predictor of improved overall survival (HR:0.792, P<0.0001).

When evaluating lung cancer survival one must take into account smoking status because it may be a confounding variable. It is well known that smoking is a major cause of lung cancer and lung cancer patients that are current smokers present with lower BMI’s as compared to former and never smokers(20, 29). Therefore, heavy smokers will be more likely to be in the lower BMI categories and at greater risk of lung cancer. However, the relationship of obesity and lung cancer mortality when taking into account smoking status has not been well documented. Some studies have displayed an inverse relationship between obesity and lung cancer mortality when not restricting for smoking status. But, when restricting the analysis to never smokers the inverse relationship between obesity and lung cancer mortality disappears (4, 9, 16, 20). The Cancer Prevention Study II demonstrated through complex statistical control models, the inverse relationship between obesity and lung cancer mortality disappeared entirely when restricted to never smokers (20). They suggested the inverse relationship between lower BMI and increased lung cancer mortality could be explained by possible pre-clinical disease upon disease presentation. On the other hand Leung et al. found an inverse relationship between BMI and lung cancer mortality in both smokers and non-smokers (6). Therefore, it remains to be clarified whether in fact smoking status is a confounding variable in the overall survival of obese lung cancer patients.

Our dataset was limited to self reported current smoking status which by and of itself is a statistic notoriously under representing the actual prevalence of the condition (30). However, weight losing patients reported smoking slightly more often than non-weight losing patients (OR:1.17), while obese patients reported smoking slightly less often than non-obese patients (OR:0.84) (Table 3). Also, when limiting survival analysis to non smokers we found obesity and the absence of weight loss to significantly improve survival (Figure 2). Under multivariate analysis smoking was found to be an independent negative predictor of survival (HR:1.073, P<0.0001) (Table 5).

The power of our study is improved by the linkage of the FCDS to the AHCA database giving us access to a very large cohort which represents approximately 6% of the total United States population. This provides us with additional data modifiers including co-morbid conditions, improved follow-up, socioeconomic details, and improved treatment information. Data from such a large population-based cancer registry provides insight into tumor behavior and it allows examination of outcomes based on current treatment strategies (3134). A database of this size, however, is not without its limitations. When examining obesity and weight loss, it would be ideal to quantify obesity in terms of BMI. Without access to BMI we were unable to discern whether the degree of obesity influenced mortality. We can simply state that among our cohort a diagnosis of obesity at presentation was associated with improved overall outcomes even when accounting for demographics, co-morbidities, tumor characteristics, and treatment variables. Also, both obesity and weight loss in this dataset are provided as a categorical variable as opposed to a continuous variable. Thus, we cannot abstract either co-morbid condition as a response to disease progression, nor can we examine the degree of weight loss or obesity in each patient. Moreover, we cannot discern whether obesity is a risk factor for developing lung cancer, but this has already been well established in the literature (11, 20, 21). Therefore, with a decreased mortality associated with obesity, it is imperative in future studies to delineate whether the improved mortality is a result of a positive prognostic factor and whether or how obesity affects disease progression in the course of lung cancer.

Our observations support the literature documenting weight loss and cachexia being poor prognostic factors in lung cancer survival. Currently, data in the literature are inconsistent regarding the prognostic strength of obesity in outcomes for lung cancer. However, our observations reinforce the idea that patients diagnosed with obesity appear to have a survival advantage over those patients who are not obese. This is similar to the improved survival of obese patients in diseases displaying high catabolic states, including CHF, chronic kidney disease, chronic heart failure, chronic obstructive lung disease, AIDS, rheumatoid arthritis, and the elderly. These data may reflect an overall protective effect of greater physiological reserve in the form of excess fat and muscle, both of which are lost in the final phases of lung cancer. The knowledge of this protective effect could ultimately play a significant role in prolonging the lives of lung cancer patients. Therefore, more studies are needed to evaluate the mechanisms explaining the inverse association between obesity and lung cancer mortality.


This work was supported by grants to T.A.Z. from the National Cancer Institute (R01CA122596), the American Cancer Society (RSG TBE-111831), by individual allocation from an ACS Institutional Grant to the University of Miami, and by the Papanicoulau Corps for Cancer Research, the Women’s Cancer Association of the University of Miami, the Wendy Will Case Cancer Fund, and the Department of Defense Breast Cancer Research Program (BCRP) BC045536, and the Amit family. The contents are solely the responsibility of the authors and do not necessarily represent the official views of the supporting agencies.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Competing interests

The authors have no competing interests to declare.

Authors’ contributions

T.A.Z. conceived of the project, M.C., M.M.B, and L.G.K. designed the experiment, R.Y., M.M.B, and M.C. performed analysis of the data, and R.Y., M.C., F.E.P, L.G.K. and T.A.Z. interpreted the data. R.Y. F.E.P and T.A.Z. wrote and edited the paper.

None of the authors declares a conflict of interest.


1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, et al. Cancer statistics, 2008. CA Cancer J Clin. 2008;58:71–96. [PubMed]
2. Jemal A, Thun MJ, Ries LA, Howe HL, Weir HK, et al. Annual report to the nation on the status of cancer, 1975–2005, featuring trends in lung cancer, tobacco use, and tobacco control. J Natl Cancer Inst. 2008;100:1672–1694. [PMC free article] [PubMed]
3. Stewart ST, Cutler DM, Rosen AB. Forecasting the effects of obesity and smoking on U.S. life expectancy. N Engl J Med. 2009;361:2252–2260. [PMC free article] [PubMed]
4. Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med. 2003;348:1625–1638. [PubMed]
5. Kollarova H, Machova L, Horakova D, Cizek L, Janoutova G, et al. Is obesity a preventive factor for lung cancer? Neoplasma. 2008;55:71–73. [PubMed]
6. Leung CC, Lam TH, Yew WW, Chan WM, Law WS, et al. Lower lung cancer mortality in obesity. Int J Epidemiol. 2010 [PubMed]
7. Nonemaker JM, Garrett-Mayer E, Carpenter MJ, Ford ME, Silvestri G, et al. The risk of dying from lung cancer by race: a prospective cohort study in a biracial cohort in Charleston, South Carolina. Ann Epidemiol. 2009;19:304–310. [PubMed]
8. Parr CL, Batty GD, Lam TH, Barzi F, Fang X, et al. Body-mass index and cancer mortality in the Asia-Pacific Cohort Studies Collaboration: pooled analyses of 424,519 participants. Lancet Oncol. 2010;11:741–752. [PMC free article] [PubMed]
9. Whitlock G, Lewington S, Sherliker P, Clarke R, Emberson J, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet. 2009;373:1083–1096. [PMC free article] [PubMed]
10. Yang L, Yang G, Zhou M, Smith M, Ge H, et al. Body mass index and mortality from lung cancer in smokers and nonsmokers: a nationally representative prospective study of 220,000 men in China. Int J Cancer. 2009;125:2136–2143. [PubMed]
11. Kabat GC, Kim M, Hunt JR, Chlebowski RT, Rohan TE. Body mass index and waist circumference in relation to lung cancer risk in the Women's Health Initiative. Am J Epidemiol. 2008;168:158–169. [PMC free article] [PubMed]
12. DeNavas-Walt CPBRM. Current Population Reports. US Census Bureau 2007. 2007:60–226.
13. Bergstrom A, Pisani P, Tenet V, Wolk A, Adami HO. Overweight as an avoidable cause of cancer in Europe. Int J Cancer. 2001;91:421–430. [PubMed]
14. Carroll KK. Obesity as a risk factor for certain types of cancer. Lipids. 1998;33:1055–1059. [PubMed]
15. Peto J. Cancer epidemiology in the last century and the next decade. Nature. 2001;411:390–395. [PubMed]
16. Reeves GK, Pirie K, Beral V, Green J, Spencer E, et al. Cancer incidence and mortality in relation to body mass index in the Million Women Study: cohort study. BMJ. 2007;335:1134. [PMC free article] [PubMed]
17. Kapoor JR, Heidenreich PA. Obesity and survival in patients with heart failure and preserved systolic function: a U-shaped relationship. Am Heart J. 2010;159:75–80. [PubMed]
18. Horwich TB, Fonarow GC, Hamilton MA, MacLellan WR, Woo MA, et al. The relationship between obesity and mortality in patients with heart failure. J Am Coll Cardiol. 2001;38:789–795. [PubMed]
19. Lavie CJ, Osman AF, Milani RV, Mehra MR. Body composition and prognosis in chronic systolic heart failure: the obesity paradox. Am J Cardiol. 2003;91:891–894. [PubMed]
20. Henley SJ, Flanders WD, Manatunga A, Thun MJ. Leanness and lung cancer risk: fact or artifact? Epidemiology. 2002;13:268–276. [PubMed]
21. Kabat GC, Miller AB, Rohan TE. Body mass index and lung cancer risk in women. Epidemiology. 2007;18:607–612. [PubMed]
22. Kalantar-Zadeh K, Horwich TB, Oreopoulos A, Kovesdy CP, Younessi H, et al. Risk factor paradox in wasting diseases. Curr Opin Clin Nutr Metab Care. 2007;10:433–442. [PubMed]
23. Skipworth RJ, Stewart GD, Dejong CH, Preston T, Fearon KC. Pathophysiology of cancer cachexia: much more than host-tumour interaction? Clin Nutr. 2007;26:667–676. [PubMed]
24. Dewys WD, Begg C, Lavin PT, Band PR, Bennett JM, et al. Prognostic effect of weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. Am J Med. 1980;69:491–497. [PubMed]
25. Tan BH, Fearon KC. Cachexia: prevalence and impact in medicine. Curr Opin Clin Nutr Metab Care. 2008;11:400–407. [PubMed]
26. Lew EA, Garfinkel L. Variations in mortality by weight among 750,000 men and women. J Chronic Dis. 1979;32:563–576. [PubMed]
27. McMillan DC. Systemic inflammation, nutritional status and survival in patients with cancer. Curr Opin Clin Nutr Metab Care. 2009;12:223–226. [PubMed]
28. Temel JS, Pirl WF, Lynch TJ. Comprehensive symptom management in patients with advanced-stage non-small-cell lung cancer. Clin Lung Cancer. 2006;7:241–249. [PubMed]
29. Flanders WD, Lally CA, Zhu BP, Henley SJ, Thun MJ. Lung cancer mortality in relation to age, duration of smoking, and daily cigarette consumption: results from Cancer Prevention Study II. Cancer Res. 2003;63:6556–6562. [PubMed]
30. Gorber SC, Schofield-Hurwitz S, Hardt J, Levasseur G, Tremblay M. The accuracy of self-reported smoking: a systematic review of the relationship between self-reported and cotinine-assessed smoking status. Nicotine Tob Res. 2009;11:12–24. [PubMed]
31. Cheung MC, Perez EA, Molina MA, Jin X, Gutierrez JC, et al. Defining the role of surgery for primary gastrointestinal tract melanoma. J Gastrointest Surg. 2008;12:731–738. [PubMed]
32. Hodgson N, Koniaris LG, Livingstone AS, Franceschi D. Gastric carcinoids: a temporal increase with proton pump introduction. Surg Endosc. 2005;19:1610–1612. [PubMed]
33. Perez EA, Koniaris LG, Snell SE, Gutierrez JC, Sumner WE, 3rd, et al. 7201 carcinoids: increasing incidence overall and disproportionate mortality in the elderly. World J Surg. 2007;31:1022–1030. [PubMed]
34. Perez EA, Livingstone AS, Franceschi D, Rocha-Lima C, Lee DJ, et al. Current incidence and outcomes of gastrointestinal mesenchymal tumors including gastrointestinal stromal tumors. J Am Coll Surg. 2006;202:623–629. [PubMed]