Fax: (011) 972-3-5352214
Adiponectin, ghrelin, and leptin in cancer cachexia in breast and colon cancer patients
Article first published online: 12 JAN 2006
Copyright © 2006 American Cancer Society
Volume 106, Issue 4, pages 966–973, 15 February 2006
How to Cite
Wolf, I., Sadetzki, S., Kanety, H., Kundel, Y., Pariente, C., Epstein, N., Oberman, B., Catane, R., Kaufman, B. and Shimon, I. (2006), Adiponectin, ghrelin, and leptin in cancer cachexia in breast and colon cancer patients. Cancer, 106: 966–973. doi: 10.1002/cncr.21690
- Issue published online: 3 FEB 2006
- Article first published online: 12 JAN 2006
- Manuscript Accepted: 20 SEP 2005
- Manuscript Revised: 16 SEP 2005
- Manuscript Received: 4 JUN 2005
- cancer cachexia;
- breast cancer;
- colon cancer
The hormone ghrelin and the adipocytokines leptin and adiponectin participate in body weight regulation. In response to weight loss, ghrelin and adiponectin levels increase and leptin decreases. Cancer cachexia is a complex metabolic state, characterized by loss of muscle mass and adipose tissue together with anorexia. The authors hypothesized that responses of these hormones may be attenuated in cancer cachexia.
Fasting plasma ghrelin, adiponectin, and leptin levels, as well as weight loss, were determined in 40 cancer patients: 18 of them suffered from cancer-induced cachexia, and 22 served as a comparison group. Hormone levels were measured before administration of cancer therapy.
A similar distribution of age, gender, and diagnosis was observed in both study groups, but the cachectic patients had higher rates of metastatic disease and lower albumin levels. No significant correlation was observed between plasma adiponectin levels and weight loss. Mean plasma ghrelin levels were higher among cachectic compared with noncachectic patients. Notably, the association between ghrelin levels and weight loss was only modest, and in a third of the cachectic patients, ghrelin levels were equal to or lower than those in the noncachectic group. Plasma leptin levels showed gender-dependent associations, and significantly lower levels were found among cachectic women but not among cachectic men.
Results suggested a gender-dependent attenuation of expected physiologic responses to weight loss among cancer cachexia patients. Thus, impaired response of adiponectin, ghrelin, and leptin may play a role in the pathogenesis of cancer cachexia syndrome. Cancer 2006. © 2006 American Cancer Society.
Cancer cachexia is one of the most frequent effects of malignancy, is often associated with poor prognosis, and may account for up to 20% of cancer deaths.1 Cancer cachexia is a complex metabolic disorder, involving loss of adipose tissue due to lipolysis, loss of skeletal muscle mass, elevation of resting energy expenditure, anorexia, and reduction in food intake.1–3 Although tissue catabolism is mainly mediated by the activity of cytokines, such as tumor necrosis factor α (TNF-α), interleukin-1β, interleukin-6, and interferon-γ,3, 4 the mechanisms associated with cancer anorexia have yet to be fully elucidated.5 Food intake and energy homeostasis are tightly regulated by a complex network of peripheral mediators, such as hormones, neuropeptides, and cytokines, which affect various central mechanisms.6 Among these mediators are the hormones adiponectin, ghrelin, and leptin.6, 7
Adiponectin (also known as ACRP30) is a member of a group of adipocyte-secreted proteins, collectively known as adipocytokines.8, 9 Adiponectin is secreted exclusively from adipose tissue, and its serum levels are high (5–30 μg/mL) and are gender dependent, i.e., higher in women compared with men.7–11 Adiponectin serum levels inversely correlate to body weight. Thus, low adiponectin levels are found in obesity,10, 11 and high levels are found in anorexia nervosa12 and during weight loss.13 Low adiponectin levels are also associated with insulin resistance and development of diabetes mellitus,14 and administration of adiponectin is associated with increased insulin sensitivity and fatty acid catabolism.15 Several reports have indicated association between low adiponectin levels and elevated risk of breast, endometrial, and gastric cancers.16–19 Recently, low adiponectin levels were reported among weight-losing patients with advanced lung cancer.20 The mechanisms responsible for regulation of adiponectin levels have not been fully elucidated. Yet, recent data suggest down-regulation of adiponectin by TNF-α, as well as by insulin.9, 21
The hormone ghrelin is a 28 amino-acid peptide, unique for the esterification of its third serine residue by n-octanoic acid.22, 23 The major source of ghrelin is the stomach, where it is synthesized in distinct endocrine cell type, known as the X/A-like cells.23 The peptide is a powerful inducer of growth hormone (GH) release, acting at the pituitary and hypothalamic levels.23, 24 Yet, in addition to its GH-releasing action, ghrelin has been found to be a powerful orexigenic factor, stimulating food intake and inducing adiposity.25 Ghrelin participates directly in hypothalamic regulation of feeding, and its administration causes weight gain by reducing food utilization, increasing food intake, and inhibiting leptin-induced feeding reduction.25, 26 Similarly, intravenous administration of ghrelin to humans increased energy consumed from a free-choice buffet.27 Fasting plasma ghrelin levels correlate negatively with obesity, fall after meals, and increase before meals or during intentional weight loss.28–30 Ghrelin levels correlate positively with various cachectic states, such as anorexia nervosa and severe congestive heart failure.31, 32 Elevated ghrelin levels were recently reported in lung cancer-induced cachexia33 and in a cohort of male patients with mainly lung and prostate cancer.34 Interestingly, short-term ghrelin infusion has been reported recently to increase energy intake in cancer patients.35
Leptin is another member of the adipocytokines family. It is produced mainly by differentiated adipocytes and acts in the central nervous system to suppress food intake and stimulate energy expenditure.6, 7 Leptin plasma levels are higher among women6, 7 and decrease in response to diet36 and in anorexia nervosa patients37 but not cardiac cachexia patients.38 Data on the association between leptin levels and cancer are contradicting. Leptin levels were reported to be low in gastrointestinal39–41 and pancreatic cancer patients42 but high in breast and gynecologic cancer patients.43 Likewise, a correlation between lower leptin levels and cancer-induced weight loss was reported by some groups39, 40 but not by others.41
The aim of the present study was to evaluate associations among adiponectin, ghrelin, leptin, and cancer cachexia in a population of newly diagnosed breast and colon cancer patients.
MATERIALS AND METHODS
The study population included newly diagnosed, histologically confirmed, adult breast and colon cancer patients who were treated at the Oncology Institute of the Sheba Medical Center between September 2002 and August 2003. The protocol was approved by the Ethics Committee of the Sheba Medical Center, and informed consent was obtained from all participants before their inclusion in the study.
Exclusion criteria included: treatment by chemotherapy, radiotherapy, or a major operation during the 6 months before recruitment; brain metastasis; second malignancy; acute or chronic infection; dysphagia; other primary cachectic states (i.e., congestive heart failure, abnormal chronic obstructive pulmonary disease, cirrhosis); elevated bilirubin or liver enzymes (> 2 of the normal reference value); renal failure (creatinine > 2.0 mg/dL); history of eating disorders; or gastrectomy.
Demographic clinical and anthropometric data were collected upon recruitment. All pathology reports were reviewed, and data on tumor histology was recorded. Stage was defined according to the 1997 American Joint Committee on Cancer Staging System.42 Body mass index (BMI) was calculated as weight (kg) divided by height (m),2 and cachexia was defined as ≥ 5% reduction in BMI in the 6 months before recruitment, as calculated from reported weight differences given by these patients. Performance status was evaluated using the Eastern Cooperative Oncology Group (ECOG) performance status.
Blood samples were obtained from the antecubital vein, between 7:00 and 9:00 a.m., after overnight fast, for the measurement of adiponectin, ghrelin, and leptin, as well as for complete blood count and chemistry. For adiponectin, ghrelin, and leptin studies, the blood was immediately transferred into a chilled tube containing EDTA and centrifuged. The serum samples, obtained from the centrifugation, were then immediately frozen at −70 °C until further analysis.
Adiponectin, total ghrelin, and total leptin were determined using radioimmunoassay kits (Linco Research, St Charles, MO). The sensitivity of the adiponectin assay was 1 ng/mL, and the interassay coefficient of variation (CV) ranged from 6.9–9.3%. The sensitivity of the ghrelin assay was 100 pg/mL, and the interassay CV ranged from 15–20%. The sensitivity of the leptin assay was 0.5 ng/mL, and the interassay CV ranged from 3– 6%.
The study variables were compared between the study groups using T-tests for continuous variables and Fisher exact test for categorical variables. Pearson correlation coefficient was used to determine the relation between continuous variables. Multiple linear regression analysis was performed to ascertain independent effects of BMI, after adjustment for gender and type of cancer, on adiponectin, ghrelin, and leptin levels. All significance tests were two-tailed. All calculations and statistical tests were performed using the software package S-PLUS 2000 (MathSoft, Seattle, WA).
The study population included 40 patients who met all inclusion criteria. Eighteen (45%) of the patients suffered from ≥ 5% reduction in BMI in the 6 months before recruitment and were, therefore, defined as cachectics. The rest of the patients (n = 22) were defined as noncachectics. The demographic and clinical characteristics of the 2 groups are shown in Table 1. No significant differences were noted between the 2 groups in age, gender, type of cancer, performance status, and previous or current BMI. Mean BMI loss among cachectic patients during the 6 months prior to diagnosis was 11.5%; BMI loss among noncachectic patients was only 1.1% (P < 0.0001). Higher rates of metastatic disease and lower albumin and hemoglobin were also noted among the cachectic patients, but no differences were noted in cholesterol levels or white blood cell counts. No differences were noted in liver or kidney function tests or in the prevalence of other comorbidity (data not shown).
|Characteristics||Noncachexia (N = 22)||Cachexia (N = 18)||P|
|Age, mean ± SD||62.2 ± 13.9||68.0 ± 12.9||0.18|
|Gender, no. (%)||0.52|
|Male||10 (45)||6 (33)|
|Female||12 (54)||12 (67)|
|Diagnosis, no. (%)||1.0|
|Breast cancer||8 (36)||6 (33)|
|Colorectal cancer||14 (64)||12 (67)|
|Metastatic disease, no. (%)||0.0007|
|No||20 (91)||7 (39)|
|Yes||2 (9)||11 (61)|
|ECOG performance status, no. (%)||0.1|
|0||10 (45)||4 (22)|
|1||10 (45)||7 (39)|
|2||2 (9)||4 (22)|
|Mean ± S.D.|
|Predisease BMI, kg/m2||25.9 ± 4.2||27.8 ± 5.0||0.22|
|Current BMI, kg/m2||25.6 ± 4.2||24.6 ± 4.8||0.47|
|Change in BMI, %||−1.1 ± 2.9||−11.5 ± 6.0||< 0.0001|
|Albumin, g/dL||4.3 ± 0.3||4.0 ± 0.5||0.024|
|Cholesterol, mg/dL||202 ± 34||197 ± 36||0.71|
|Hb, g/dL||14.2 ± 1.4||12.7 ± 1.2||0.001|
|WBC, ×109/dL||7.3 ± 2.1||8.0 ± 1.9||0.29|
|Lymphocytes, %||22.9 ± 5.9||20.4 ± 8.1||0.27|
Adiponectin Levels and Cancer Cachexia
No correlation was found between adiponectin levels and BMI loss over the previous 6 months (Fig. 1A; r = 0.15, P = 0.4). Mean adiponectin levels were higher in cachectic patients compared with noncachectic patients but the differences reached only borderline statistical significance (Fig. 1B; 12.3 ± 5.2 vs. 9.6 ± 3.6 μg/mL, respectively, P = 0.06). Analysis of mean adiponectin levels by cachexia and gender also revealed higher levels among cachectic compared with noncachectic men and women, but the differences did not reach statistical significance (Fig. 1B).
Multiple regression analysis was used to evaluate the role of BMI loss as a continuous variable, along with age, gender, and type of cancer, to predict adiponectin levels (Table 2). Results of the model indicate that only age and gender, but not BMI loss or type of cancer, were significant independent predictors of adiponectin levels. Increasing age and being female were associated with higher mean adiponectin levels.
|Type of cancer||−0.27||1.82||0.88|
Ghrelin Levels and Cancer Cachexia
A significant positive correlation was found between ghrelin levels and BMI loss over the previous 6 months (Fig. 2A; r = 0.5, P = 0.001). Ghrelin levels were significantly higher in cachectic patients compared with noncachectic patients (1969 ± 673 vs. 1447 ± 462 pg/mL, respectively, P = 0.01), but analysis by gender revealed statistically significant differences only among women (Fig. 2B). It is important to mention that despite the significant differences noted, 27% of noncachectic patients showed high ghrelin levels, and 39% of cachectic patients showed normal or low ghrelin levels.
Multiple regression analysis was used to evaluate the role of BMI loss as a continuous variable, along with age, gender, and type of cancer, to predict ghrelin levels (Table 3). The results of the model indicate that only BMI loss was a significant independent predictor of ghrelin levels. Analysis of the association of ghrelin and other laboratory data revealed only a negative correlation between ghrelin and hemoglobin levels (data not shown).
|Type of cancer||343.9||225.8||0.13|
Leptin Levels and Cancer Cachexia
As leptin levels are considerably higher among women compared with men,8 the correlation between leptin levels and BMI loss over the previous 6 months was calculated separately for men (Fig. 3A) and women (Fig. 3B). Among men, a positive correlation between leptin levels and BMI loss, with borderline statistical significance, was observed (r = 0.46, P = 0.07). However, among women, a statistically significant negative correlation was observed (r = −0.45, P = 0.03). Similarly, although nonsignificant, leptin levels were higher among cachectic compared with noncachectic men (8.3 ± 5.5 vs. 6.1 ± 2.6 ng/mL, respectively, P = 0.29) but significantly lower among cachectic compared with noncachectic women (Fig. 3C; 11.4 ± 4.2 vs. 19.3 ± 10.0 ng/mL, respectively, P = 0.02).
Multiple regression analysis was used to evaluate the role of BMI loss as a continuous variable, along with age, as a predictor of leptin levels among men (Table 4) and along with age and type of disease as a predictor of leptin levels among women (Table 5). The results of these models indicate for both men and women that BMI loss is probably associated with leptin levels. However, in men, this association was negative and only of borderline significance (P = 0.09), whereas among women, BMI loss was significantly and positively associated with leptin levels.
|Type of cancer||0.63||3.64||0.86|
In the present study, we investigated associations among adiponectin, ghrelin, and leptin levels and cancer cachexia in a population of newly diagnosed or newly recurrent, untreated, breast and colon cancer patients. Our results indicate complex interactions between these hormones and cancer cachexia.
Adiponectin levels are reported to be inversely correlated with body weight. Thus, voluntary weight loss, as well as anorexia nervosa, is associated with elevated adiponectin levels.9–12 However, in this study, we found no correlation between BMI loss and adiponectin levels, even after adjusting for gender and tumor type. Adiponectin levels are regulated mainly by changes in the adipose tissue.7–9 Unlike starvation and other cachectic states, cancer cachexia is often characterized by preferential loss of skeletal mass rather than adipose tissue.2–4 Thus, the lack of association between adiponectin levels and weight loss may simply reflect the preservation of adipose tissue. Another explanation for our observation may come from recent studies, which found inhibition of adiponectin secretion from adipocytes by various cytokines, including TNF-α.21, 43 Thus, the lack of elevation of adiponectin levels after cancer cachexia, may reflect altered regulation of adiponectin in this condition. Interestingly, lower adiponectin levels were also found in a cohort of cachectic patients with very advanced stage of lung cancer compared with healthy volunteers.20
Elevated levels of total or active ghrelin in cancer cachexia have been reported in cohorts of mainly male lung cancer patients.33, 34 In this study, we report on elevated ghrelin levels in a cohort of breast and colon cancer patients. However, ghrelin levels overlapped considerably in the cachectic and the noncachectic groups, and, in about a third of the cachectic patients, no elevation of ghrelin levels was noted. Moreover, we found a stronger correlation between ghrelin levels and cachexia in women compared with men. Notably, low levels of ghrelin were also found among a significant number of cachectic lung cancer patients.33 This subset of patients, identified by us to have low ghrelin levels despite cachexia, may benefit most from administration of exogenous ghrelin. Administration of ghrelin has recently been reported to increase energy intake in a group of 7 metastatic cancer patients, and it is currently being tested in a Phase II clinical trial as a novel treatment for cancer cachexia.35 Thus, our results suggest that measurement of ghrelin levels may have important clinical implications in treating cancer cachexia syndrome.
Although leptin levels are directly associated with weight loss after dieting,7, 36 associations between leptin levels and cancer cachexia are not yet fully elucidated. Thus, lower leptin levels were found in patients with gastrointestinal cancers, regardless of the degree of weight loss and in colon cancer patients who had no weight loss at all.41, 44–45 However, association between leptin levels and weight loss was noted in a cohort of lung46 and pancreatic cancer patients.39 We found gender-dependent association between leptin levels and cancer-induced weight loss. Leptin levels correlated, although with only borderline statistical significance, positively with weight loss in men but negatively in women. These findings were noted in all 3 analyses models used by us. Interestingly, analysis of adipose tissue revealed that leptin secretion from tissues obtained from women was much more sensitive to various stimulations compared with tissues obtained from men.47 This phenomenon may explain gender differences found in our study.
The number of patients included in this study was relatively small. However, we believe that the very strict inclusion and exclusion criteria, and specifically the exclusion of previously treated patients and the focus on only 2 tumor types, allowed us to eliminate several major confounding factors that could interfere with elucidating associations between cancer cachexia and adiponectin, ghrelin, or leptin.
An important issue that has not been addressed yet is the association between weight loss rate and levels of adiponectin, ghrelin, and leptin. Perhaps, slower rate of weight loss, as commonly observed in breast and colon cancers compared with lung and pancreatic cancers,1 may lead to adaptation and, therefore, be associated with less pronounced changes in levels of these hormones. More studies are needed to substantiate associations between cancer cachexia and adiponectin, ghrelin, or leptin in the population of breast and colon cancer, as well as in additional types of cancer.
- 42FlemingID, CooperJS, HensonDE, editors. American Joint Committee on Cancer Staging Manual. 5th edition. Philadelphia: Lippincott-Raven, 1997.