Is Overweight or Obese Status an Independent Prognostic Factor in Various Solid Organ Cancers?

Abstract

While obesity has been found to be a risk factor for several types of cancer, its use as a prognostic tool is unclear. The authors have reviewed literature interrogating the link between overweight status and obesity and cancer prognosis for five distinct solid organ malignancies: colon, breast, ovarian, lung, and renal. There appears to be a positive correlation between overnutrition and poor prognosis in breast and colon cancer. Similarly, there may well be a survival advantage for overweight or obese patients with newly diagnosed renal cell carcinoma and smoking-associated lung cancer. The relationship between ovarian cancer and obesity is less clear, as is the relationship between lung cancer in never/scant-smokers and obesity.

Introduction

Obesity is increasing in the United States, with an average of 64% of adults that are either overweight or obese, according to the National Health and Nutrition Survey performed between 1999 and 2000.¹ A more recent report from the CDC highlights the fact that 25% of Americans are now considered obese, a statistic that may grossly underestimate the actual prevalence of the condition because of the limits of self-report.² This large-scale health problem is linked to increased morbidity from heart disease, stroke, hypertension, diabetes, osteoarthritis and breathing problems, and is a likely link to many common cancers such as breast, ovarian and colon cancer.³ In fact, obesity has been found to be a causative factor in several cancer types and may also adversely affect prognosis.⁴

This is a problem with potentially severe consequences. Similar to the detrimental health effects of smoking, there may be long-term implications of obesity on health. In an illuminating study of over 10,000 Glasgow students attending school between 1948 and 1968, with a mean follow-up of 41 years, researchers sought to determine the long-term effects of obesity as measured by physicians during annual health check-ups.⁵ There was an association between adolescent Body Mass Index (BMI) and subsequent cancer mortality, controlling for tobacco use. In cancer not related to smoking, there was a positive association between overweight status during adolescence and cancer risk later in life. This study was limited by the lack of data during the interceding years, but still offers compelling evidence regarding the health implications of obesity.

The role of using weight status in helping to determine treatment-related decisions and in offering a long-term outlook for patients is unclear. The authors have reviewed available literature to evaluate the relationship between overweight and/or obese status and cancer prognosis. Namely, the authors are interested in determining whether overweight status or obesity is an independent prognostic factor in colon, breast, ovarian, lung, and renal cancers.

Materials and Methods

A literature review was performed using the databases PubMed and CINAHL, employing the following search terms: colon cancer, lung cancer, breast cancer, ovarian cancer, renal cancer, obesity, and prognosis. Research that specifically discussed the use of overweight and/or obese status as an independent prognostic indicator was included in the literature review. Thirty-three studies ranging in dates from 1996 through 2010, including Web-based information, were reviewed. The majority of articles described studies using retrospective chart reviews of overweight and/or obesity as a prognostic factor in the above-mentioned cancer types with few exceptions. Research addressing overweight status and obesity only as a risk factor for cancer as opposed to having prognostic implications was excluded, as were review articles without direct clinical application. The literature on the effect of obesity on rectal cancer prognosis was scarce, so for the purposes of this review we have limited articles on rectal cancer. For the sake of brevity, in-depth information related to the specific tumor types has been limited in favor of a discussion of the relevant literature.

Colorectal Cancer

Colorectal cancer is the third most common cancer and second leading cause of cancer death among men and women, with approximately 140,000 people diagnosed each year and over 53,000 deaths affecting men and women equally.⁶ Even with advanced screening technologies, the majority of patients are diagnosed with regional (lymph node involvement) or distant (metastatic to other organs) disease where 5-year survival rates are 68% and 11%, respectively.⁷

The data on the relationship between obesity and colon cancer prognosis have been inconsistent because of the confounding influences of dietary intake, sedentary lifestyle, and adiposity, according to Siegal et al.⁷ A possible explanation for this is that the majority of the studies producing these data have been retrospective analyses; thus, these variables were not addressed at the time of data collection. The majority of the data examining relationships between body size and recurrence of colon cancer after surgical resection and overall survival have been extrapolated from clinical trials or retrospective data.⁷ (Please see Table 1 for concise information regarding the studies reviewed.) Most study participants have been diagnosed with high-risk stage II or III colon cancer and have undergone surgical resection for the disease, and it is not until after resection has taken place that obesity as a prognostic factor has been evaluated.^7,8,9 Adjuvant chemotherapy is recommended for stage III disease to decrease the chance of disease recurrence; however, adjuvant chemotherapy in stage II disease is still controversial and may be a confounding factor when analyzing this literature.¹⁰ BMI is the most commonly utilized measurement of overweight or obese status in the studies reviewed, where a BMI ≥30–35kg/m² is considered obese and a BMI ≥35 kg/m² is considered very obese. Obesity decreases overall survival and disease-free survival as well as increases the risk of death by 19% versus the risk experienced by normal-weight patients.⁹ In addition, obese patients tended to have other poorer prognostic variables, such as a higher tumor stage, increased number of lymph nodes involved by cancer, and distal tumors, versus their normal-weight counterparts.⁹ In a prospective observational study by Meyerhardt et al. where height and weight were self-reported by patients, the analysis showed no relationship between BMI and risk of recurrence or death for stage III colon cancer.¹¹ However, the study did show an increase in the risk of colon cancer recurrence after resection of disease for those patients with a BMI ≥35 kg/m², even though statistical significance was not met. In a 2003 cohort study, Meyerhardt et al. showed that there were gender differences in the risk of recurrent colon cancer in obese patients.¹² Women who were obese (BMI ≥30 kg/m²) had a higher likelihood of recurrent disease than those of a normal weight, but this relationship did not reach statistical significance. In contrast, in men BMI did not influence mortality or recurrent disease in normal, overweight, or obese patients.¹¹ These data conflict with those of Sinicrope et al., who found that the relationship between BMI and men was more predictive of outcome than in women, where men who had a BMI ≥35 kg/m² had a statistically significant 35% increased risk of death, in comparison with very obese women, for whom the relative risk of death was 11% and was not considered significant. However, those women with a BMI of 30–34 kg/m² still had worse overall survival in comparison to their normal-weight counterparts, and this relationship was not observed in men.⁹ These gender differences can possibly be explained by central adiposity, which is more characteristic in men than in women, as well as a possible protective effect of hormonal replacement therapy in postmenopausal women.⁹ One can see from the above-mentioned information that the data speak to colon cancer directly, versus the combined entity of colorectal cancer. Interestingly, when assessing the risk of rectal cancer with waist circumference, waist:hip ratio, and BMI, Pischon et al. found no statistically significant association; however, they were strongly associated with colon cancer risk in both men and women.¹³ In another study by Meyerhardt et al., prognosis of stage II and III rectal cancer was not influenced by BMI; however, gender differences were seen, and men with a BMI ≥30 had an increased risk of both local and overall cancer recurrence.¹⁴

Table 1.

Obesity and Colon Cancer

Author	Year	Study design and sample composition	Measurement (definition of obesity and overweight)	Findings
Meyerhardt et al.	2003	Cohort study within large randomized trial of 3,759 men and women with high-risk stage II–III colon carcinoma	BMI ≥30 kg/m²	In women with stage II–III colon cancer, obesity increased overall mortality, and significantly increased disease recurrence. Men were not affected.
Statin et al.	2004	Case-controlled study of serum from men diagnosed with cancer of the colon or rectum. 378 colorectal cancer cases and 378 controls were analyzed.	N/A	Increased leptin levels were associated with increase risk of colon cancer, and serum leptin levels were correlated to obesity, insulin resistance, and serum insulin levels.
Meyerhardt et al.	2004	Cohort study within large randomized trial of 1,688 men and women with stage II–III rectal cancer	BMI ≥30 kg/m²	Obesity had no influence on overall mortality or recurrence among women. Obese men were more likely to have local and overall recurrence than normal-weight men.
Pischon et al.	2006	Multicenter prospective cohort study of 368,277 healthy men and women monitored for 6.1 yrs. 1570 were diagnosed with colorectal cancer during that time	BMI ≥30kg/m² WC ≥102cm (men), ≥88cm (women), WHR ≥0.95 (men), ≥0.80 (women)	WC and WHR strongly associated with colon cancer risk in men and women. None of the measures of obesity were statistically significant in rectal cancer.
Meyerhardt et al.	2008	Prospective observational study of 1,053 patients with stage III colon cancer as part of a large randomized trial of adjuvant chemotherapy	BMI ≥30kg/m²	BMI was not significantly associated with increased risk of colon cancer recurrence or death.
Gonullu et al.	2010	Experimental trial looking at relation between serum adiponectin and resistin levels in 36 CRC patients	BMI ≥27.2 ± 4.5 WHR ≥0.92 ± 0.08	Adiponectin and resistin may be associated with colon cancer carcinogenesis.
Sincrope et al.	2010	Meta-analysis of 868 stage II–III colon cancer patients from 7 randomized controlled trials	BMI ≥30–34.9kg/m² = class 1 obesity BMI ≥35kg/m² = class 2,3 obesity	Obesity is an independent prognostic variable in colon cancer, and there are apparent gender-related differences.

WHR = waist:hip ratio, WC = waist circumference, BMI = Body Mass Index, CRC = colorectal cancer, and 8-OHdG = 8-hydroxydeoxyguanosine.

Breast Cancer

According to the CDC, breast cancer is the second most common type of cancer diagnosed and the third leading cause of cancer death.¹⁵ There will be an estimated 207,000 new cases of breast cancer diagnosed in 2010, but less than 40,000 deaths from breast cancer.¹⁶ Obesity and overweight status has been associated with increasing the risk for developing breast cancer, but studies linking breast cancer prognosis with weight status have been inconsistent.¹⁷ (Please see Table 2 for concise information regarding the studies reviewed.)

Table 2.

Obesity and Breast Cancer

Author	Year	Study design and sample composition	Measurement (definition of obesity and overweight)	Findings
Bastarrachea et al.	1994	Retrospective review of clinical course and characteristics of 735 lymph-node-positive patients with stages II and III primary breast cancer treated using three consecutive post-operative adjuvant chemotherapy protocols (10 year follow-up)	Obesity defined as 20% over ideal body weight for each patient	Obesity was associated with postmenopausal status, as well as larger, later-stage tumor at diagnosis. Obesity was found to be an independent prognostic factor associated with shorter disease-free survival and overall survival; differences remained after correcting for pre-treatment characteristics.
Kroenke et al.	2005	Retrospective analysis of 5,204 participants in the Nurses' Health Study diagnosed with breast cancer between 1976 and 2000, excluding metastatic or in situ diagnoses	BMI cut points of <21, 21–22, 23–24, 25–29, and >30	Body weight before diagnosis positively correlated with breast cancer death and all-cause mortality but not recurrence, especially in never-smokers. Women who gained between 0.5 and 2.0 kg after diagnosis had an elevated risk of breast cancer death in follow-up, especially premenopausal women.
Tao et al.	2005	Cohort study from a larger population-based case control study using interviews of patients or family members in the Shanghai Cancer Registry; sample of 1,455 breast cancer patients diagnosed between 8/96 and 3/98 (5-year follow-up)	BMI of 25 or greater highest cut point; also used waist:hip ratio and waist circumference	Increased BMI found to be an independent prognostic factor correlated with shorter overall survival and disease-free survival. Waist:hip ratio and waist circumference were not independently associated with OS or DFS.
Pierce et al.	2007	Prospective analysis of 1,490 women diagnosed with early-stage breast cancer between 1991 and 2005 (mean follow-up of 6.7 years)	BMI between 25 and 29 considered overweight, and >30 considered obese	Women who adhered to a healthy lifestyle were found to have a 50% reduction in mortality, even in the obese population.
Litton et al.	2008	Retrospective analysis of 1,169 early-stage breast cancer patients registered in the Breast Cancer Management System database of MD Anderson Cancer Center (median follow-up of 4.1 years)	BMI cut points of <25 (normal or underweight), 25–30 (overweight), and >30 (obese)	Obese patients tended to present with hormone-negative tumors and higher grade tumors. Higher BMI was also associated with a worse response to neoadjuvant chemotherapy.
Núñez et al.	2008	Prospective RCT using ovariectomized and nonovariectomized mice to determine impact of obesity on tumor growth	Cut points used were lean, normal, overweight, very overweight, and obese based on mouse biometrics	Obesity accelerated tumor growth in ovariectomized but not in nonovariectomized mice.
Imkampe et al.	2009	Retrospective analysis of 2,298 breast cancer patients diagnosed between 1983 and 2007 (mean follow-up of 7 years); data initially collected prospectively on an actively managed database	BMI cut points were >35, 30–35, 22–29, and <22; obesity defined as >30	BMI independently associated with recurrence-free survival, most prominently in patients with node-positive disease and <60 years old. Breast cancer outcomes worse for patients with increased BMI, especially in younger patients (<60).
Healy et al.	2009	Case control study comparing 200 postmenopausal breast cancer cases to 519 age-matched healthy female controls	BMI 20–25 (normal weight), >30 (obese)	Obesity is associated with larger tumors and more advanced staging at diagnosis but not with involvement of the axillary nodes. Median survival and overall survival were equivalent in obese and nonobese patients, negating the survival impact of later staging at diagnosis.

BMI = Body Mass Index, OS = overall survival, and DFS = disease-free survival.

One issue in determining the independent prognostic value of overweight status and obesity may be the confounding variable of stage and grade of tumor at diagnosis. Namely, researchers have questioned whether overweight or obese patients present with more dangerous disease states. Healy et al. found that obesity is associated with larger tumors and more advanced staging at diagnosis but not with involvement of the axillary nodes.¹⁸ Even so, median survival and overall survival were equivalent in obese and nonobese patients, negating the survival impact of later staging at diagnosis. Bastarrachea et al. also found that obese patients tended to be diagnosed at later stages and have larger tumors.¹⁹ Unlike the Healy study, however, this study concluded that obesity was an independent prognostic factor associated with shorter disease-free survival and overall survival, and these differences remained after correcting for other pre-treatment characteristics. Another study utilizing a cohort of 1,455 breast cancer patients found that increased BMI at diagnosis or soon after was associated with poorer overall survival and disease-free survival.²⁰ Litton et al. found that obese patients were more likely to have hormone-negative tumors, higher grade tumors, and worse overall survival than their normal-weight or overweight counterparts. Higher BMI was also associated with a worse response to neoadjuvant chemotherapy.²¹

Menopausal status and presence or absence of hormonal therapy at the time of breast cancer diagnosis may also play a role in whether overweight status or obesity is a significant prognostic factor, but even these data are contradictory. Obesity has been found to be associated with postmenopausal status, which may also correlate with increased age at diagnosis.¹⁹ Timing of weight gain after diagnosis may also be a factor in overall survival and recurrence. Another study found that increased weight before diagnosis was positively associated with recurrence and death in never-smokers in multivariate analysis.²² Also, in this same study, women who gained between 0.5 and 2.0 kg/m² after diagnosis had an elevated risk of breast cancer death in follow-up. Both of these associations were stronger in pre-menopausal than postmenopausal women.²² Pre-clinical studies have also yielded some interesting results. A study at NIH concluded that obesity accelerated tumor growth in ovariectomized mice, but did not have the same effect on mice with intact ovaries, supporting the assertion that overweight status and obesity have some kind of relationship with hormonal status, which impacts breast cancer incidence and mortality.²³ A more recent study found that increased BMI was independently associated with poorer recurrence-free survival, and these outcomes were worst for younger patients with increased BMI and node-positive disease.²⁴

The link between healthy lifestyle choices and longevity may in the end prove to be equally as important as that between overnutrition and cancer. An excellent prospective study of 1,490 women with early-stage breast cancer found that women who adhered to a healthy lifestyle, including increased fruit and vegetable intake, physical exercise, and maintenance of a normal BMI, had a significant survival advantage over those who did not.²⁵ “Among those women who adhered to this healthy lifestyle, there was no apparent effect of obesity on survival.”²⁵ Although the link between normal-weight status and overall survival could not be excluded, neither could obesity be implicated when healthy lifestyle choices were controlled for.

Ovarian Cancer

Ovarian cancer is the sixth leading cause of cancer death in the United States and will account for about 21,000 new cases of cancer this year.²⁶ Ovarian cancer is most often treated using multimodality therapy, including surgical intervention and chemotherapy. Relatively few studies have evaluated a link between overweight status and/or obesity and ovarian cancer prognosis is rare but available. Similar to colon and breast cancer, studies to date consist of retrospective chart reviews, and the data are underwhelming. Also similar to other studies discussed thus far, BMI is overwhelmingly used as the method to evaluate overweight, obese, and normal-weight patients, with little to no mention of waist:hip ratio or waist circumference. (Please see Table 3 for concise information regarding the studies reviewed.)

Table 3.

Obesity and Ovarian Cancer

Author	Year	Study design and sample composition	Measurement (definition of obesity and overweight)	Findings
Li et al.	2007	Genotype analysis of 81 patients with epithelial ovarian cancer	BMI <18.5 (underweight), 18.5–24.9 (ideal weight), 25–29.9 (overweight), >30 (obese)	Obesity combined with short AR allelotypes may promote more aggressive ovarian cancer phenotypes.
Skírnisdóttir and Sorbe	2007	Retrospective chart review of 635 ovarian cancer patients who underwent primary surgery between 1974 and 2004	BMI <18.5 (underweight), 18.5–25 (ideal weight), 25–30 (overweight), >30 (obese)	Overweight and obesity significantly associated with older age. No significant difference in FIGO staging, recurrence, recurrence-free survival, overall survival, or cancer-specific survival among various BMI categories.
Barrett et al.	2008	Prospective study of 1,077 patients receiving chemo for ovarian cancer	BMI <18.5 (underweight), 18.5–24.9 (ideal weight), 25–29.9 (overweight), >30 (obese)	No significant differences in overall survival or progression-free survival in all groups. No association found between BMI and tumor staging and grading or debulking surgery.
Münstedt et al.	2008	Retrospective chart review of 824 patients with ovarian cancer	BMI stratified by the following categories: <20, 20–25, 25–30, 30–40, >40	No significant influence of obesity on survival, but there was a trend toward improved survival in obese patients.
Matthews et al.	2008	Retrospective chart review of 304 patients with epithelial ovarian cancer	BMI >30 considered obese	No survival distinctions between obese and nonobese patients when optimal debulking took place.

BMI = Body Mass Index. FIGO = Federation International of Gynecologic Oncology.

In one study of the relationship between obesity and ovarian cancer using fairly high BMI values, Matthews et al. found no significant differences between two BMI groups (BMI <30 versus BMI >30) in terms of progression-free survival or overall survival when optimal surgical debulking took place, although acute wound complications were more prevalent in obese patients.²⁷ As noted in a similar study of breast cancer, overweight and obese status in ovarian cancer patients is significantly associated with the older population.²⁸ Skirnisdottir et al. reviewed cases of surgery, surgery and radiation, and chemotherapy at a single institution over 29 years. Patients were stratified based on BMI as well as treatment modalities, and the authors found no significant difference in FIGO staging, recurrence, recurrence-free survival, overall survival, or cancer-specific survival.²⁸ Similarly, Barrett et al. stratified patients based on BMI and found no significant differences in progression-free survival or overall survival between groups.²⁹ This study also found no difference in chemo dose intensity, which has been posited as a possible explanation for long-term poor outcomes in metastatic ovarian cancer. No association was found between BMI and tumor staging/grading or debulking surgery.²⁹ Munstedt et al. actually found that, in ovarian cancer, there is a trend toward improved survival in obese patients, but this was not clinically significant.³⁰ This study also concluded that BMI has no influence on histological tumor type and found no significant results linking BMI and ovarian cancer prognosis.³⁰

Others have evaluated for links between tumor histology and obesity. Li et al. found that obesity combined with short androgen receptor (AR) allelotypes may promote more aggressive ovarian cancer phenotypes, possibly by supplying androgenic hormones.³¹ Because adipose tissue is capable of producing these androgenic hormones, it is thought that obese patients may have a greater likelihood of presenting with more aggressive tumors.³¹ Whether this would have a survival impact has yet to be shown.

Lung Cancer

Lung cancer is the leading cause of death in the United States each year. There are over 222,520 new cases and 157,300 deaths from this disease, which exceeds the number of deaths caused by breast, colorectal, and ovarian cancer combined.³² Lung cancer is not usually associated with obesity, as smoking suppresses appetite, and there is an increase of caloric expenditure (work of breathing). (Please see Table 4 for concise information regarding the studies reviewed.) In a prospective cohort study of over 200,000 Chinese men, there was an inverse relationship between BMI and mortality from lung cancer (p < 0.0001 for trend) in current smokers, with each 5kg/m² lower BMI associated with a 35% increased mortality.³³ In the same study, there was no association between BMI and mortality from lung cancer in nonsmoking patients.³³ Singh and Lindsted reported on the relationship between BMI and the risk of mortality of women who never smoked, and again there was an inverse relationship between BMI and respiratory mortality.³⁴ That study consisted of 16,946 Seventh-Day Adventist females, whose lifestyle is characterized by minimal smoking and a lacto-ovo-vegetarian diet, which was felt by the authors to decrease the confounding variables that are normally associated with a Western lifestyle (smoking and a high-fat, high-meat diet). Over a 15–26-year time period, females who had a BMI >27kg/m² had decreased mortality from respiratory etiology but an increase in overall mortality from other disease categories such as cardiovascular events, cerebrovascular events, and hormone-related cancers. The decrease seen in respiratory mortality of overweight females was overshadowed by the significant weight loss associated with disease onset.³⁴ Although this study does not discuss the relationship between BMI and lung cancer mortality specifically, it may highlight an important link between overnutrition and decreased overall survival due to multiple comorbidities.

Table 4.

Obesity and Lung Cancer

Author	Year	Study design and sample composition	Measurement (definition of obesity and overweight)	Findings
Singh et al.	1998	26-year cohort study of 12,576 non-Caucasian women who never smoked and BMI	BMI ≥27kg/m²	Inverse association between BMI and respiratory mortality.
Kabat et al.	2007	Cohort study within a randomized control trial of screening for breast cancer of 89,835 women, 750 of whom were incident lung cancer cases	Mean BMI of study population ≥24.6 ± 4.3; definition of overweight not stated	An inverse association was found between BMI and lung cancer in current female smokers, and a positive association found with BMI and never-smokers.
Kondo et al.	2007	Cohort study of 29,350 Japanese men over 11 years, evaluating cancer risk	Mean BMI of study population ≥22.3 ± 3.0; definition of overweight status was not stated	Never-smokers with high BMI at an earlier age (∼20 years old) predicts lung cancer mortality, but BMI loss in adulthood elevates the risk for lung cancer in current smokers in this population.
Kabat et al.	2008	Cohort study that used data from the Women's Health Initiative to study anthropometric factors associated with lung cancer risk; 1,353 cases of lung cancer found	Definition of overweight status not stated; however, mean BMI <23.1 = 23% of study population, BMI ≥23.1–28.3 = 40% of study population, BMI ≥28.3–32.2 = 20% population, and BMI ≥32.2 = 17% of study population	BMI was inversely associated with lung cancer risk in current smokers; however, waist circumference is positively associated with risk in current smokers.
Yang et al.	2009	Prospective cohort study of 217,180 Chinese men with no prior history of cancer	Mean baseline BMI 21.1kg/m²	Strong inverse association between lung cancer mortality and BMI. Lifelong nonsmokers showed no apparent association between BMI and mortality

BMI = Body Mass Index.

The inverse relationship of BMI and lung cancer mortality in Chinese men was described by Yang et al.³³ However, in a review of the data from the Canadian National Breast Screening Study by Kabat et al., there was a positive relationship between BMI and lung cancer risk in never-smokers, yet in the smokers the relationship was still inversely associated.³⁵ In a study of over 29,000 Japanese men, high BMI positively correlated with mortality from lung cancer in never-smokers and the possible explanation was that obese individuals tend to have higher levels of circulating insulin, thus promoting carcinogenesis, a relationship that is also seen in colon cancer.³⁶ The inverse relationship between smoking-associated lung cancer and BMI could be due to mortality from other comorbid conditions, significant weight loss prior to lung cancer diagnosis, and the fact that heavy smokers have a lower BMI, and thus the smoking is the driving factor of mortality.³² Nonabdominal and visceral fat can be assessed by BMI and waist circumference, but the loss of lean muscle mass that is associated with smoking may not be fully captured with these measures.³⁷ Kabat et al. showed that once BMI and waist circumference were controlled for, there was a positive association with waist circumference and lung cancer risk in current and former female smokers.³⁵ The positive association seen between BMI and never-smokers, both male and female, warrants further research as this population of lung cancer patients is becoming more prevalent.

Renal Cancer

Renal cancer, or renal cell carcinoma (RCC), is the seventh most common type of cancer diagnosed in the United States.³⁸ In this type of cancer, similar to smoking-associated lung cancer, overweight or obese status at diagnosis, as measured by BMI, may confer a survival advantage, although these studies are also limited and contradictory. (Please see Table 5 for concise information regarding the studies reviewed.)

Table 5.

Obesity and Renal Cancer

Author	Year	Study design and sample composition	Measurement (definition of obesity and overweight)	Findings
Kamat et al.	2004	Retrospective chart review of 400 patients who underwent nephrectomy for localized renal cell carcinoma between 1985 and 1998	BMI <25 (normal), 25–30 (overweight), >30 (obese)	Improved disease-specific survival in overweight and obese patients. Even in patients with organ-confined disease, patients with normal BMI did worse than overweight or obese patients. BMI was found to be an independent prognostic factor in multivariate analysis for progression-free survival, overall survival, and disease-specific survival.
Schips et al.	2004	Prospective analysis of 683 renal cell carcinoma patients undergoing surgery between 1994 and 2000	BMI <25, 25–29.9, 30–34.9, >35	Patients with a BMI >25 had better outcomes compared with normal-weight patients, but multivariate analysis showed no statistical difference. There was a survival advantage for overweight patients, but this was not significant when tumor-specific disease was controlled for, and overweight status was not an independent prognostic factor.
Parker et al.	2006	Retrospective record review of 1,345 patients at the Mayo Clinic who underwent nephrectomy for renal cell carcinoma between 1988 and 2002	BMI 25–29.9 (overweight), BMI >30 (obese)	Patients with higher BMI tended to present with less aggressive tumor types, but when aggressiveness of tumor was factored into multivariate analysis, higher BMI did not independently confer a survival advantage.
Donat et al.	2006	Review of database information from 1,137 renal cell carcinoma patients undergoing radical or partial nephrectomy between 1995 and 2003	BMI <25 (normal weight), 25–29.9 (overweight), >30 (obese)	Age (>65), staging, systemic symptoms, and type of surgery all impacted progression-free survival, and there was a trend toward improved overall and progression-free survival in overweight but not obese patients.
Awakura et al.	2007	Prospective analysis of 264 Japanese patients undergoing surgery for renal cell carcinoma at a single institution between 1991 and 2002	BMI <23 (normal weight), 23–25 (overweight), >25 (obese), following the cutoff points for Asians recommended by the World Health Organization	Nodal or metastatic disease state was the best predictor of survival, followed by BMI. This was true for overall survival and cancer-specific survival, but BMI showed no predictive value for recurrence-free survival.
Haferkamp et al.	2008	Prospective analysis of 780 patients undergoing nephrectomy for renal cell carcinoma between 1990 and 2005	BMI <18.5 (underweight), 18.5–<25 (normal weight), 25–<30 (overweight), >30 (obese)	Underweight status conferred significantly worse prognosis in RCC. Overweight and obese patients showed a tendency toward less aggressive disease, but this difference was not statistically significant.
Schrader et al.	2009	Retrospective analysis of 771 patients who underwent renal surgery for renal cell carcinoma between 1990 and 2005 (mean follow-up of 5.48 years)	BMI <18.5 (underweight), 18.5–24.99 (normal weight), 25–29.99 (pre-obese), >30 (obese)	Overweight status was an independent prognostic factor in organ-confined, tumor-specific renal cell carcinoma, although they found no significant relationship between survival and BMI in metastatic disease.
Jeon et al.	2010	Retrospective chart review of 1,017 Korean renal cell carcinoma patients who underwent curative surgery between 1988 and 2006 (mean follow-up of 76.9 months)	BMI <23 (normal), 23–27.5 (overweight), 27.5 (obese)	There was a significant positive correlation between increased BMI and decreased pathological T stage and metastatic stage at presentation.

BMI = Body Mass Index.

Higher BMI has been shown to be a risk factor for the development of renal cell carcinoma, but may contribute to the development of less aggressive disease in affected patients.^39,40,41 Jeon et al. found a significant positive correlation between increased BMI and decreased pathological T stage and metastatic state at presentation.³⁹ They also found that BMI was an independent prognostic factor in both univariate and multivariate analysis, conferring a survival advantage over normal-weight patients.³⁹ Parker et al. also found that patients with higher BMI tended to present with less aggressive tumor type.⁴⁰ However, when the aggressiveness of tumors was factored into multivariate analysis, higher BMI did not independently confer a survival advantage.⁴⁰ Kamat et al. also reported improved disease-specific survival in overweight and obese patients.⁴¹ Even in patients with organ-confined disease, patients with normal BMI did worse than overweight or obese patients. In fact, BMI was found to be an independent prognostic factor in multivariate analysis for progression-free survival, overall survival, and disease-specific survival.⁴¹ Schrader et al. added to the body of literature in 2009, reporting that overweight status was an independent prognostic factor in organ-confined, tumor-specific renal cell carcinoma, although they found no significant relationship between survival and BMI in metastatic disease.⁴² Haferkamp et al. were just as concerned with the impact of underweight status on prognosis and found that underweight status conferred significantly worse prognosis in RCC. Overweight and obese patients showed a tendency toward less aggressive disease, but this difference was not statistically significant.⁴³

Donat et al. found that several factors other than BMI were far more prognostically important in renal cell carcinoma.⁴⁴ Age (>65), staging, systemic symptoms, and type of surgery all impacted progression-free survival, and there was a trend toward improved overall and progression-free survival in overweight but not obese patients.⁴⁴ Awakura et al. also found that factors beyond BMI proved be more important, although BMI still played a role.⁴⁵ In their study, multivariate analysis showed that nodal or metastatic disease state was the best predictor of survival, followed by BMI. This was true for overall survival and cancer-specific survival, but BMI showed no predictive value for recurrence-free survival.⁴⁵ Schips et al. found that in univariate analysis, patients with a BMI >25 had better outcomes compared with normal-weight patients, but multivariate analysis showed no statistical difference.⁴⁶ There was a survival advantage for overweight patients, but this was not significant when tumor-specific disease was controlled for, and overweight status was not an independent prognostic factor. ⁴⁶

Discussion

There appears to be a positive correlation between overnutrition and poor prognosis in breast and colon cancer. Similarly, there may well be a survival advantage for overweight or obese patients with newly diagnosed renal cell carcinoma and in smoking-associated lung cancer. The relationship between ovarian cancer and obesity is less clear, as is the relationship between lung cancer and obesity in nonsmokers.

Limitations of the studies reviewed include the fact that most did not take into account any measurement of overnutrition beyond BMI, such as waist:hip ratio or waist circumference, which may be more indicative of central adiposity. Some of the studies relied on self-measurement and self-report of height and weight to calculate BMI, thus either over- or underestimating values. Furthermore, Siegal et al. believe that BMI may not be the best way to assess obesity because BMI does not take into account muscle mass, which can give someone a higher BMI but not necessarily mean that they are overweight or obese.⁷ In one of the only studies reviewed using alternative methods of measurement for obesity, waist:hip ratio and waist circumference were not independently associated with overall survival or disease-free survival.⁴⁷

Another point for consideration is the use of overall survival instead of progression-free and/or disease-specific survival as study end points. Follow-up using overall survival as an endpoint may not allow for enough differentiation between cancer mortality and mortality associated with other obesity-related comorbidities.

The exact mechanisms by which obesity influences cancer are not well understood, but recent studies have discussed possible etiologies such as insulin resistance and hyperinsulinemia, increased leptin levels, and low adiponectin levels. Insulin can affect the growth of both normal and neoplastic cells, and there are insulin growth factor receptors in normal colonic mucosa.⁴⁷ When hyperinsulinemia occurs, there is an inhibition of apoptosis and an increase in secretion of vascular endothelial growth factor (VEGF), which are two important factors in the evolution of carcinogenesis.⁴⁸

Hormonal effects of overnutrition have been long thought to play the biggest role in increasing cancer risk, and may also be related to differences in obesity-related prognosis, if indeed there are any. Leptin is an adipose-derived cytokine found to be elevated in obese patients and significantly associated with carcinogenesis.⁴⁹ It causes increased energy consumption and inhibition of appetite, which may lead to weight loss.⁵⁰ Obese patients tend to have a decreased sensitivity to leptin, and leptin production is directly related to percentage of body fat; thus, the higher the body fat, the more circulating leptin.⁵¹ Leptin has been shown to be a growth factor in certain cell types but most importantly was present in not only human colon cancer cells but also precancerous polyps and colonic mucosa.⁵² Leptin overexpression is seen in 59.2% of epithelial ovarian cancer and is significantly associated with poor progression-free survival. In vitro, leptin has been found to stimulate ovarian carcinoma proliferation and inhibit apoptosis.⁵³ The impact of leptin on tumor growth is not universal. On the contrary, in a study of nearly 500 breast cancer patients presenting at various stages and receiving a variety of treatment modalities, leptin was not significantly associated with overall survival or distant disease-free survival in multivariate analysis, even though there was an association found with overall survival in univariate analysis.⁵⁴

Other hormones that are absent or unavailable in sufficient supply in obese patients may also be important to study. Adiponectin is an adipose tissue–derived hormone that provides several benefits to our bodies such as inhibiting plaque formation in arteries, anti-inflammatory activity, and regulating glucose and fatty acid metabolism.⁴⁷ Gonullu et al. showed that the levels of adiponectin were significantly lower in obese patients with colon cancer than in the normal controls, thus contributing to insulin resistance and chronic hyperinsulinemia. In addition, there was an inverse relationship between adiponectin levels and stage of disease, where patients with advanced-stage disease had lower levels of adiponectin, thus conferring a poorer prognosis.⁴⁸

Studies have also looked at the relationship between oxidative stress and cancer prognosis. Oxidative stress may play a role in the development of breast cancer,⁵⁵ colorectal cancer,⁵⁶ prostate cancer,⁵⁷ lung cancer,⁵⁸ and ovarian cancer.⁵⁹ Alterations in metabolic pathways in tumor cells, a chaotic and tortuous tumor vasculature, and macrophage infiltration of the tumor tissue may be a cause of oxidative stress and contribute to increased metastasis, increased tumor mutation rate, increased tumor blood supply, and resistance to therapy, all conferring a poorer prognosis.⁶⁰ Oxidative stress may also be a potential mutagen that causes errors in repeated sequences of DNA, thus promoting tumorigenesis.⁶¹

Conclusion

While a clear relationship, regarding obesity as a prognostic factor in solid tumors is not evident, it does seem clear that there are a series of correlations between overnutrition and the pathogenesis, histology, and long-term outcomes in some cancers. There may be hormonal links between poor cancer prognosis and hormone-mediated cancer. If oxidative stress was a driving force, how do we account for the survival advantage conferred to obese patients with renal cancer? The inverse relationship between overweight or obese status and nonsmokers with lung cancer, in direct contradiction to smoking-related lung cancer, seems to offer important insights into the existence of an important, and not clearly understood, link between overnutrition and cancer.

One of the most important conclusions to draw from the studies reviewed is that information about the impact of overnutrition on cancer prognosis cannot be generalized from one tumor type to another. There appears to be a unique metabolic relationship between obesity and the development and maturation of specific tumor cells, and this is not yet well understood. Finding out why certain malignancies develop differently in the presence of increased adipose tissue may offer important clinical insights leading to improved outcomes.

Looking beyond BMI and using waist:hip ratio or waist circumference, which may be more indicative of central adiposity, may prove to be beneficial when looking at obesity and cancer prognosis. Furthermore, how much we eat may not be the only problem contributing to tumorigenesis, but what we eat as well.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

National Cancer Institute. Obesity and Cancer: Questions and Answers. U.S. National Institutes of Health. www.cancer.gov/cancertopics/factsheet/Risk/obesity. 2010 July 25.

Centers for Disease Control and Prevention. Adult Obesity: Obesity Rises among Adults. CDC Vital Signs. www.cdc.gov/VitalSigns/pdf/2010-08-vitalsigns.pdf. 2010 July 25.

Nazario

. Weight Loss: Health Risks Associated with Obesity. WebMD. www.webmd.com/cholesterol-management/obesity-health-risks. 2010 July 25.

Ceschi

, Gutzwiller

, Moch

, Eichholzer

, Probst-Hensch

. Epidemiology and pathophysiology of obesity as cause of cancer. Swiss Med Weekly, 2007; 137:50–56.

Okasha

, McCarron

, McEwen

, Smith

. Body mass index in young adulthood and cancer mortality: A retrospective cohort study. J Epidemiol Comm Health, 2002; 56:780–784.

U.S. Cancer Statistics Working Group. United States Cancer Statistics: 1999–2006 Incidence and Mortality Web-Based Report. Atlanta, GA: Department of Health and Human Services, Centers for Disease Control and Prevention, and National Cancer Institute, 2010. www.cdc.gov/uscs. 2010 July 25.

Siegel

, Ulrich

, Poole

, Holmes

, Jacobsen

, Shibata

. The effects of obesity and obesity-related conditions on colorectal cancer prognosis. Cancer Control, 2010; 17:52–57.

Centers for Disease Control and Prevention. Defining Overweight and Obesity. April 29, 2005. www.cdc.gov/nccdphp/dnpa/obesity/defining.htm. 2010 July 25.

Sinicrope

, Foster

, Sargent

, O'Connell

, Rankin

. Obesity is an independent prognostic variable in colon cancer survivors. Clin Cancer Res, 2010; 16:1884–1893.

10.

Marshall

. Risk Assessment in Stage II colorectal cancer. Oncology, 2010; 24:9–13.

11.

Meyerhart

, Niedzwiecki

, Hollis

, Saltz

, Mayer

, Nelson

, Whittom

, Hantel

, Thomas

, Fuchs

. Impact of body mass index and weight change after treatment on cancer recurrence and survival in patients with Stage III colon cancer: Findings from cancer and leukemia Group B 89803. J Clin Oncol, 2008; 26:4109–4115.

12.

Meyerhardt

, Catalano

, Haller

, Mayer

, Benson AL

III

, Macdonald

, Fuchs

. Influence of body mass index on outcomes and treatment-related toxicity in patients with colon carcinoma. Cancer, 2003; 98:484–495.

13.

Pischon

, Lahmann

, Boeing

, Friedenreich

, Norat

, Tjønneland

et al. Body size and risk of colon and rectal cancer in the european prospective investigation into cancer and nutrition. J Natl Cancer Inst, 2006; 98:920–931.

14.

Meyerhardt

, Tepper

, Niedzwiecki

, Hollis

, McCollum

, Brady

, O'Connell

, Mayer

, Cummings

, Willett

, Macdonald

, Benson

, Fuchs

. Impact of body mass index on outcomes and treatment-related toxicity in patients with Stage II and III rectal cancer: Findings from intergroup trial 0114. J Clin Oncol, 2004; 22:648–657.

15.

Centers for Disease Control and Prevention. 2006. Top Ten Cancers. United States Cancer Statistics, National Program of Cancer Registries. http://apps.nccd.cdc.gov/uscs/toptencancers.aspx. 2010 July 25.

16.

National Cancer Institute. Breast Cancer. U.S. National Institutes of Health. www.cancer.gov/cancertopics/types/breast. 2010 July 25.

17.

Gunter

, Hoover

, Yu

, Wassertheil-Smoller

, Rohan

, Manson

et al. Insulin, insulin-like growth factor-I, and risk of breast cancer in postmenopausal women. JNCI, 2009; 101:48–60.

18.

Healy

, Ryan

, Rowley

, Boyle

, Connolly

, Kennedy

et al. Obesity increases the risk of postmenopausal breast cancer and is associated with more advanced stage at presentation but no impact on survival. Breast J, 2010; 16:95–97.

19.

Bastarrachea

, Hortobagyi

, Smith

, Kau

, Buzdar

. Obesity as an adverse prognostic factor for patients receiving adjuvant chemotherapy for breast cancer. Ann Intern Med, 1994; 120:18–25.

20.

Tao

, Shu

, Ruan

, Gao

, Zheng

. Association of overweight with breast cancer survival. Am J Epidemiol, 2006; 163:101–107.

21.

Litton

, Gonzalez-Angulo

, Warneke

, Buzdar

, Kau

, Bondy

et al. Relationship between obesity and pathologic response to neoadjuvant chemotherapy among women with operable breast cancer. J Clin Oncol, 2008; 26:4072–4077.

22.

Kroenke

, Chen

, Rosner

, Holmes

. Weight, weight gain, and survival after breast cancer diagnosis. J Clin Oncol, 2005; 23:1370–1378.

23.

Núñez

, Perkins

, Smith

, Berrigan

, Berendes

, Varticovski

et al. Obesity accelerates mouse mammary tumor growth in the absence of ovarian hormones. Nutr Cancer, 2008; 60:534–541.

24.

Imkampe

, Bates

. Impact of a raised body mass index on breast cancer survival in relation to age and disease extent at diagnosis. Breast J, 2010; 16:156–161.

25.

Pierce

, Stefanick

, Flatt

, Natarajan

, Sternfeld

, Madlensky

et al. Greater survival after breast cancer in physically active women with high vegetable-fruit intake regardless of obesityw. J Clin Oncol, 2007; 25:2345–2351.

26.

National Cancer Institute. Ovarian Cancer. U.S. National Institutes of Health. www.cancer.gov/cancertopics/types/ovarian. 2010 July 25.

27.

Matthews

, Straughn

Jr , Kemper

, Hoskins

, Wang

, Rocconi

. The effect of obesity on survival in patients with ovarian cancer. Gynecol Oncol, 2009; 112:389–393.

28.

Skírnisdóttir

, Sorbe

. Prognostic Impact of body mass index and effect of overweight and obesity on surgical and adjuvant treatment in early-stage epithelial ovarian cancer. Int J Gynecol Cancer, 2008; 18:345–351.

29.

Barrett

, Paul

, Hay

, Vasey

, Kaye

, Glasspool

. Scottish Gynaecological Cancer Trials Group. Does body mass index affect progression-free or overall survival in patients with ovarian cancer? Results from SCOTROC I trial. Ann Oncol, 2008; 19:898–902.

30.

Münstedt

, Wagner

, Kullmer

, Hackethal

, Franke

. Influence of body mass index on prognosis in gynecological malignancies. Cancer Causes Control, 2008; 19:909–916.

31.

, Elmore

, Pavelka

, Karlan

. Hyperandrogenism, mediated by obesity and receptor polymorphisms, promotes aggressive epithelial ovarian cancer biology. Gynecol Oncol, 2007; 107:420–423.

32.

National Cancer Institute. Lung Cancer. 2010. U.S. National Institutes of Health. www.cancer.gov/cancertopics/types/lung. 2010 July 25.

33.

Yang

, Yan

, Zhou

, Smith

, Ge

, Boreham

, Hu

, Peto

, Wang

, Chen

. 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.

34.

Singh

, Lindsted

. Body mass and 26-year risk of mortality from specific diseases among women who never smoked. Epidemiology, 1998; 9:246–254.

35.

Kabat

, Miller

, Rohan

. Body mass index and lung cancer risk in women. Epidemiology, 2007; 18:607–612.

36.

Kondo

, Hori

, Yatsuya

, Tamakoshi

, Toyoshima

, Nishino

et al.

Lung cancer mortality and body mass index in a Japanese cohort: Findings from the Japan Collaborative Cohort Study (JACC Study)

Cancer Causes Control, 2007; 18:229–234.

37.

Kabat

, Kim

, Hunt

, Chlebowski

, Rohan

. 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.

38.

39.

Jeon

, Jeong

, Lee

, Kwak

, Kim

, Lee

. Prognostic value of body mass index in Korean patients with renal cell carcinoma. J Urol, 2010; 183:448–454.

40.

Parker

, Lohse

, Cheville

, Thiel

, Leibovich

, Blute

. Greater body mass index is associated with better pathologic features and improved outcome among patients treated surgically for clear cell renal cell carcinoma. Urology, 2006; 68:741–746.

41.

Kamat

, Shock

, Naya

, Rosser

, Slaton

, Pisters

. Prognostic value of body mass index in patients undergoing nephrectomy for localized renal tumors. Urology, 2004; 63:46–50.

42.

Schrader

, Rustemeier

, Timmesfeld

, Varga

, Hegele

, Olbert

, Hofmann

. Overweight is associated with improved cancer-specific survival in patients with organ-confined renal cell carcinoma. J Cancer Res Clin Epidemiol, 2009; 135:1693–1699.

43.

Haferkamp

, Pritsch

, Bedke

, Wagener

, Pfitzenmaier

, Buse

, Hohenfellner

. The influence of body mass index on the long-term survival of patients with renal cell carcinoma after tumour nephrectomy. Br J Urol Int, 2008; 101:1243–1246.

44.

Donat

, Salzhauer

, Mitra

, Yanke

, Snyder

, Russo

. Impact of body mass index on survival of patients with surgically treated renal cell carcinoma. J Urol, 2006; 175:46–52.

45.

Awakura

, Nakamura

, Ito

, Yamasaki

, Kamba

, Kamoto

, Ogawa

. Influence of body mass index on prognosis of japanese patients with renal cell carcinoma. Urology, 2007; 70:50–54.

46.

Schips

, Lipsky

, Zigeuner

, Gidaro

, Salfellner

, Rehak

, Pummer

, Hubmer

. Does overweight impact on the prognosis of patients with renal cell carcinoma? A Single Center Experience Of 683 Patients. J Surg Oncol, 2004; 88:57–61discussion, 61–62.

47.

Pais

, Silaghi

, Rusu

, Dumitrascu

. Metabolic syndrome and risk of subsequent colorectal cancer. World of Gasteroenterol, 2009; 15:5141–5148.

48.

Gonullu

, Kahraman

, Bedir

, Bektas

, Yucel

. Association between adiponectin, resistin, insulin resistance, and colorectal tumors. Int J Colorectal Dis, 2010; 25:205–212.

49.

Brennan

, Mantazoros

. Metabolism: Drug insight: The role of leptin in human physiology and pathophysiology: Emerging clinical applications. Nature Rev Endocrinol, 2006; 2:318–327.

50.

Werynska

, Kosacka

, Gotecki

, Jankowska

. Leptin serum levels in cachetic and non-cachetic lung cancer patients. Pneumonol Alergolog Pols, 2009; 77:500–506.

51.

Considine

, Sinha

, Heiman

, Kriauciunas

, Stephens

, Nyce

. et al. Serum immunoreactive-leptin concentrations in normal-weight and obese humans. NEJM, 1996; 334:292–295.

52.

Stattin

, Lukanova

, Biessy

, Soderberg

, Palmqvist

, Kaaks

, Olsson

, Jellum

. Obesity and colon cancer: Does leptin provide a link? Int J Cancer, 2004; 109:149–152.

53.

Uddin

, Bu

, Ahmed

, Abubaker

, Al-Dayel

, Bavi

, Al-Kuraya

. Overexpression of leptin receptor predicts unfavorable outcome in Middle Eastern ovarian cancer. Molecular Cancer, 2009; 8.

54.

Goodwin

, Ennis

, Fantus

, Pritchard

, Trudeau

, Koo

et al.

Is leptin a mediator of adverse prognostic effects of obesity in breast cancer?

J Clin Oncol, 2005; 23:6037–6042.

55.

Dai

, Gao

, Shu

, Yang

, Milne

, Cai

, Wen

, Rothman

, Cai

, Li

, Xiang

, Chow

, Zheng

. Oxidative stress, obesity, and breast cancer risk: Results from the Shanghai Women's Health Study. J Clin Oncol, 2009; 27:2482–2488.

56.

Sato

, Takeda

, Otake

, Yokozawa

, Nishise

, Fujishima

, Orii

et al. Increased plasma levels of 8-Hydroxydeoxyguanosine are associated with development of colorectal tumors. J Clin Biochem Nutr, 2010; 47:59–63.

57.

Kumar

, Koul

, Khandrika

, Meacham

, Koul

. Oxidative stress is inherent in prostate cancer cells and is required for aggressive phenotype. Cancer Research, 2008; 68:1777–1785.

58.

Gupta

, Srivastava

, Prasad

, Natu

, Mittal

, Negi

MPS

, Srivastava

. Smoking intensity, oxidative stress and chemotherapy in non-small cell lung cancer: A correlated prognostic study. BioSci Trends, 2009; 3:191–199.

59.

Pylväs

, Puistola

, Kauppila

, Soini

, Karihtala

. Oxidative stress-induced antioxidant enzyme expression is an early phenomenon in ovarian carcinogenesis. Eur J Cancer, 2010; 46:1661–1667.

60.

Brown

, Bicknell

. Hypoxia and oxidative stress in breast cancer, oxidative stress: Its effects on the growth, metastatic potential and response to therapy of breast cancer. Breast Cancer Res, 2001; 3:323–327.

61.

Gasche

, Chang

, Rhees

, Goel

, Boland

. Oxidative stress increases frameshift mutations in human colorectal cancer cells. Cancer Res, 2001; 61:7444–7448.