Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.

com

ORIGINAL ARTICLE

Adipokines in Patients With Cancer

Anorexia and Cachexia
Joanna Smiechowska, MD,*Þ Anne Utech, MS, RD, LD,Þþ George Taffet, MD,*Þ
Teresa Hayes, MD, PhD,Þ§ Marco Marcelli, MD,Þþ and José M. Garcia, MD*Þþ

uals, a decrease in fat mass induced by caloric restriction (diet)

Background: Anorexia, cachexia, and insulin resistance are com- or by an increase in energy expenditure (ie, exercise) improves
monly seen in patients with cancer. Adipocyte-derived hormones or insulin sensitivity. However, this is also accompanied by a com-
adipokines play a role in the regulation of appetite, body weight, and pensatory increase in appetite that often leads to weight regain.
insulin sensitivity. However, their role in cancer-induced cachexia has These changes in insulin sensitivity and appetite are thought to be
not been well-established. The objective of this study was to determine mediated at least partially by changes in inflammatory markers
the levels of adipokines and their relation to appetite, weight loss, insulin and the adipokines leptin, adiponectin, and resistin.1,2
resistance, and other hormones in cancer cachexia. Leptin is a known satiety factor secreted proportionally to the
Methods: We measured adiponectin, resistin, and leptin plasma levels amount of fat mass and related to insulin resistance.3 Moreover,
in 21 men with cancer cachexia, 24 noncachectic cancer subjects, and 25 leptin administration induces anorexia and weight loss. In contrast,
noncancer controls matched by age, sex, and pre-illness body weight. adiponectin has been shown to stimulate appetite in rodents, and
Body weight change, appetite scores, insulin resistance assessed by its circulating levels are inversely related to body weight.1 Resistin
homeostasis model assessment, and other cytokines and hormones were is known to stimulate insulin resistance in mice. The role of resistin
also measured. Differences between groups were measured by analysis is less clear in humans, but it has been suggested that it also plays a
of covariance. Relations between variables were examined by linear role in energy homeostasis.4
regression analyses. Cachexia (a complex metabolic syndrome associated with
Results: Adiponectin levels were similarly elevated in cachectic and underlying illness and characterized by loss of muscle with or
noncachectic cancer patients compared with noncancer controls. Leptin without loss of fat mass5) is common in patients with cancer,6 and
levels were significantly decreased in cancer cachexia and were directly it is usually the result of lean and fat mass losses. Unlike what is
associated with appetite and insulin resistance, explaining 37% and 19% seen in diet-induced weight loss, cancer cachexia patients have
of the variance seen in cancer patients, respectively. Resistin levels were diminished appetite, food intake, and insulin sensitivity.7 Despite
not different between groups. the significant impact that cachexia and anorexia have on these
Conclusions: Leptin may play a role in the increased insulin resistance patients’ quality of life, the mechanisms underlying these symp-
seen in cancer patients. However, these patients are resistant to the toms are not fully understood.
orexigenic effects of hypoleptinemia. Other mechanisms besides weight In this study, we sought to determine the levels of adipokines
loss are responsible for the increased adiponectin level seen in cancer and the relation between adipokines, insulin resistance, and other
patients. It is unlikely that resistin plays a major metabolic role in this inflammatory markers in subjects with cancer anorexia and ca-
setting. chexia compared with noncachectic cancer patients and non-
Key Words: weight loss, cytokines, IL-6, TNF->, ghrelin, leptin, cancer controls.
adiponectin, resistin
(J Investig Med 2010;58: 554Y559) MATERIALS AND METHODS
Protocol and Experimental Subjects
The protocol was approved by the Baylor College of
BACKGROUND Medicine Institutional Review Board. All clinical investigation
Adipose tissue is increasingly being recognized as an im- described was in accordance with the Declaration of Helsinki.
portant endocrine organ that plays a role in the regulation of Informed consent was obtained from subjects before enrollment.
body weight, food intake, and insulin resistance by producing The study included 3 groups:
hormones known collectively as adipokines. In healthy individ- 1. Patients with cancer and cachexia, defined as an uninten-
tional weight loss of at least 5% of their pre-illness body
From the *The Huffington Center on Aging, Baylor College of Medicine; weight over the previous 6 months (n = 21).
†Michael E. DeBakey Veterans Affairs Medical Center; and ‡Division of 2. Noncachectic cancer patients, with stable body weight,
Diabetes, Endocrinology and Metabolism, Department of Medicine, and matched by age, sex, race, pre-illness body mass index
§Department of Oncology, Baylor College of Medicine, Houston, TX.
Received August 18, 2009, and in revised form November 2, 2009.
(BMI), and cancer staging to the previous group (n = 24).
Accepted for publication December 12, 2009. 3. Subjects without cancer with a stable body weight, matched
Reprints: José M. Garcia, MD, MEDVA Medical Center, 2002 Holcombe by age, sex, race, and BMI to the other 2 groups (n = 25).
Blvd, Houston, TX 77030. E-mail: [email protected].
This work is partly supported by The Huffington Center on Aging, Baylor
College of Medicine. Dr Garcia receives support from a MERIT Review
The inclusion criteria were male sex, 18 years or older,
Entry Program Grant from the Department of Veterans Affairs and a weight loss of 5% or more over the previous 6 months for the
South Central VA Healthcare Network Career Development Award from cachectic subjects and 2% or less for the other groups, and
the Department of Veterans Affairs. diagnosis of cancer other than nonmelanoma skin cancer for the
The authors have no financial disclosures related to this work.
Copyright * 2010 by The American Federation for Medical Research
cancer groups.
ISSN: 1081-5589 The exclusion criteria were blood malignancies or can-
DOI: 10.231/JIM.0b013e3181cf91ca cer involving the upper airway or upper gastrointestinal tract,

554 Journal of Investigative Medicine & Volume 58, Number 3, March 2010

Copyright @ 2010 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
 Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.com

Journal of Investigative Medicine & Volume 58, Number 3, March 2010 Adipokines in Cancer Cachexia

dysphagia, substance abuse, heart failure, severe liver disease, groups were measured by analysis of covariance. Relations
chronic obstructive pulmonary disease, diabetes with hemo- between continuous variables were examined by linear regres-
globin level A1c higher than 7%, fasting glucose level higher than sion analyses. P values were 2-sided, and a P e 0.05 was
140 mg/dL or random glucose level higher than 200 mg/dL, thy- considered statistically significant. All the calculations were
roid disease, renal failure, active infection, use of glucocorticoids, performed using SPSS version 12.00 (SPSS Inc, Chicago, IL).
use of progesterone, testosterone or other orexigenic agents, his- Linear regression was used to test weight change pro-
tory of eating disorders, and history of cancer other than non- spectively (expressed as percentage of change from baseline
melanomatous skin cancer for the control group. to 12 months) as a secondary analysis. Weight change percent
Appetite was measured fasting in the morning at the time of was conceptualized as the outcome, and the predictors leptin,
blood draw by a visual analogue scale taken from the Edmonton adioponectin, and resistin levels were individually entered into the
Symptom Assessment System: BHow would you describe your regression analyses. Other predictors (TNF->, IL-6, HOMA-IR,
appetite?[ that ranged from 0 to 100 mm with words anchored at ghrelin, and IGF-1) were added in a separate stepwise fashion to
each end, expressing the most positive and negative ratings test for significance. Inclusion was set at probability F G 0.05, and
(where 0 was Bno appetite[ and 100 was Bvery good appetite[). exclusion was set at F 9 0.10. One-tailed correlations were used
This visual analogue scale for appetite is used extensively in the during these tests, with > = 0.05.
setting of cancer-induced cachexia and correlates well with 1-
and 7-day scores in test-retest evaluations.8Y10 RESULTS
Fasting blood samples were obtained by 0900 hours. Insu- All the groups were matched for age, BMI 6 months before
lin sensitivity was assessed using the homeostasis model assess- recruitment, race, and proportion of subjects with diabetes.
ment (HOMA-IR = fasting glucose (mmol/L)  fasting insulin Results for acylated ghrelin, peptide YY, IGF-1, IL-6, TNF->,
[KU/mL] / 22.5) as previously described. These estimates correlate insulin, HOMA-IR, appetite, and weight loss have been
well with the criterion standard hyperinsulinemic euglycemic published previously.12 The groups differed in their BMI at the
clamp method (r = 0.88, P G 0.0001).11 time of recruitment because weight loss was an inclusion
criterion for the cachectic group and an exclusion criterion for
Hormone Assays the other groups (Table 1). The cancer diagnoses for the
Blood was collected in EDTA-containing tubes and kept at cachectic group were small cell carcinoma of the lung (SCLC,
4-C during processing. Plasma was stored at j80-C until n = 8), colon cancer (n = 4), adenocarcinoma of the prostate
assayed. Adiponectin level was measured by radioimmunoassay (n = 3), nonYSCLC (NSCLC, n = 2), renal cell carcinoma
(RIA) with a commercially available kit purchased from Linco (n = 1), fibrosarcoma (n = 1), adenocarcinoma of unknown
Research (St Charles, MO) in diluted plasma samples (1:450) as origin (n = 1), and squamous cell carcinoma of unknown origin
indicated by the manufacturer. The lower and upper detection
limits were 1.56 and 200 Kg/mL, respectively. Resistin level was
measured by the enzyme-linked immunosorbent assay accord-
ing to the protocol provided by the manufacturer (Biovendor, TABLE 1. Baseline Characteristics
Candler, NC). The lower and upper detection limits were 1 and
50 ng/mL, respectively. Leptin level was measured by RIA Cancer Cancer
(Linco Research) following the manufacturer’s instructions. Cachexia Noncachexia Controls
Acylated ghrelin levels were measured by a commercially (n = 21) (n = 24) (n = 25) P
available RIA kit from Linco Research in plasma treated with 1-N Age, mean (SD), yr 66 (8) 69 (11) 64 (12) 0.67
hydrochloric acid and phenylmethylsulfonyl fluoride added im- Race, n (%) 0.92
mediately after collection. This RIA kit uses an antibody raised
White 12 (57) 14 (58) 17 (68)
in guinea pig against octanoylated ghrelin and 125I-octanoylated
African 8 (38) 8 (33) 6 (24)
ghrelin as the tracer. This assay has been found to be highly
American
specific for acylated ghrelin with lower than 0.1% cross reactivity
Other 1 (4.8) 2 (8.4) 2 (8)
for des-octanoyl ghrelin and no cross reactivity with ghrelin
14Y28, motilin-related peptide, leptin, insulin, glucagon, or glu- Initial BMI, 27.5 (4) 27 (4) 28.5 (3) 0.4
mean (SD)
cagonlike peptide 7Y36. The lower and upper detection limits were
10 and 2000 pg/mL, respectively. The intra-assay coefficient Final BMI, 24 (4) 28 (4) 30 (4) 0.001
mean (SD)
of variation was 5.3%. Insulin levels were measured by a RIA
kit purchased from Linco Research. Insulinlike growth factor Staging, n(%) 0.12
1 (IGF-1) levels were extracted using acid-methanol and measured I 3 (13)
by a RIA kit from Nichols Laboratories (San Juan Capistrano, II 3 (15) 7 (29)
CA). The measurements of the circulating levels of interleukin III 3 (15) 5 (21)
6 (IL-6) and tumor necrosis factor > (TNF->) were performed at IV 14 (70) 9 (38)
the Michael E. DeBakey Veterans Affairs Medical Center as we Chemotherapy*, 14 (67) 18 (75) 0.74
have previously described.12 For all hormones, plasma from an n (%)*
equal number of subjects from each group was included in the first Current 9 (42) 9 (36) 0.59
assay, and the remaining samples were assayed in a second assay chemotherapy,
to minimize interassay variability in hormone levels between the n (%)†
groups. Time to diagnosis, 627 (1018) 1324 (1366) 0.06
mean (SD), d
Statistical Analyses *Patients who had received chemotherapy at any time point before
Continuous variables are expressed as mean (SEM). enrollment.
Nonnormally distributed data (leptin, resistin, and adiponectin) †Patients receiving chemotherapy at the time of enrollment.
were log transformed before analysis. Differences between the

* 2010 The American Federation for Medical Research 555

Copyright @ 2010 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
 Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.com

Smiechowska et al Journal of Investigative Medicine & Volume 58, Number 3, March 2010

(n = 1). For the noncachectic group, the diagnoses were

adenocarcinoma of the prostate (n = 10), NSCLC (n = 6), SCLC
(n = 3), adenocarcinoma of unknown origin (n = 1), bladder
cancer (n = 1), chondrosarcoma (n = 1), breast cancer (n = 1),
and melanoma (n = 1). However, the difference in cancer
diagnoses did not reach significance (P = 0.09). There was also a
trend toward a more advanced stage (P = 0.12) and shorter
duration of disease (P = 0.06) in the cachectic group, although it
did not reach significance. The 2 groups were comparable
regarding the proportion of subjects receiving chemotherapy
(Table 1).
The appetite scores were also lower for cancer cachexia
subjects compared with the cancer noncachexia subjects and the
controls (21.1 [3], 48 [3], and 49 [3], respectively; analysis of
variance, P G 0.001), and insulin resistance as measured by FIGURE 2. Leptin levels in the cancer cachexia, cancer
HOMA-IR was elevated in cancer noncachexia subjects noncachexia, and noncancer control subjects. *P G 0.05 when
compared with the other 2 groups (8.5 [0.94%], 4.67 [0.94%], compared with the control group. §P G 0.05 compared with
and 5.22 [0.92%], respectively; analysis of variance, P G 0.02). the cancer cachexia group.

Adipokine Levels
The adiponectin levels were significantly higher in both levels remained significantly associated with anorexia, acylated
groups of subjects with cancer when compared with the noncancer ghrelin and IL-6 were associated with weight loss, and only
controls. However, there was no difference between the 2 cancer leptin was associated with body weight and insulin resis-
groups (Fig. 1). The adiponectin levels were not correlated with tance. Adiponectin and resistin were not associated with any of
IL-6 or TNF-> levels (r = 0.18, P = NS or r = j0.12, P = NS, the 4 dependent variables (body weight, body weight change,
respectively). The leptin levels were significantly decreased in appetite, and insulin resistance) on multiple regression analysis
cancer cachexia compared with the other 2 groups (Fig. 2), and (Table 3).
they correlated negatively with IL-6 (r = j0.46, P G 0.005) but not Prospectively, the weight change percent between the base-
with TNF-> (r = 0.21, P = NS). The resistin levels were similar line and more than 12 months in the cancer cachectic versus
between the groups (Fig. 3, P = 0.14) and were not correlated with noncachectic groups was not significantly predicted by baseline
the IL-6 or TNF-> levels (r = 0.25, P = 011 or r = 0.26, P = 0.09, levels of leptin (F1,26 = 0.204, P = 0.655), adiponectin
respectively). These results persisted after adjusting for BMI and (F1,21 = 3.28, P = 0.084), or resistin (F1,26 = 0.12, P = 0.915).
cancer staging. Baseline adiponectin level was negatively correlated (R = j0.368,
P = 0.42) with weight change percent, however. The stepwise
Adipokines, Body Weight Changes, Appetite, regression analysis did not find any significant predictive value nor
and Insulin Resistance correlations of TNF->, IL-6, HOMA-IR, ghrelin, nor IGF-1 with
We used simple linear regression analyses to establish the weight change percent (data not shown). Similarly, nonsignificant
relationship between adipokines and other hormones regarding results were found when all the 3 subject groups (adding controls)
body weight, weight loss, appetite, and insulin resistance in the were entered into the equation (data not shown).
2 cancer groups (Table 2). Leptin, adiponectin, IGF-1, and
insulin were significantly associated with body weight. Leptin, DISCUSSION
IL-6, ghrelin, IGF-1, and insulin were associated with appetite Cachexia affects 50% of cancer patients, worsening their
and weight change, whereas only leptin and IGF-1 were asso- quality of life and survival. Paradoxically, the weight loss seen in
ciated with insulin resistance. When multiple linear regres- this setting does not cause an increase in appetite and insulin
sion analysis models were used including parameters with sensitivity usually seen in noncancer subjects losing weight
significant correlations in univariate analyses, leptin and IL-6 voluntarily. It is thought that adipokines play a role in regulating

FIGURE 1. Adiponectin levels in the cancer cachexia, cancer

noncachexia, and noncancer control subjects. *P G 0.05 when FIGURE 3. Resistin levels in the cancer cachexia, cancer
compared with the control group. noncachexia, and noncancer control subjects. P = 0.14.

556 * 2010 The American Federation for Medical Research

Copyright @ 2010 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
 Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.com

Journal of Investigative Medicine & Volume 58, Number 3, March 2010 Adipokines in Cancer Cachexia

TABLE 2. Simple Linear Regression Analysis Between Adipokines and Body Weight, Weight Loss, Anorexia, and Insulin Resistance

Body Weight Weight Change Appetite Score HOMA-IR

Parameter r P r P r P r P
Leptin 0.87 G0.001 0.55 G0.001 0.60 G0.001 0.44 0.003
Adiponectin j0.36 0.03 j0.19 0.27 j0.23 0.18 j0.18 0.29
Resistin 0.19 0.23 0.16 0.31 j0.21 0.18 0.001 0.99
Acylated ghrelin j0.25 0.1 j0.54 G0.001 j0.37 0.01 j.13 0.39
IL-6 j0.26 0.08 j0.5 G0.001 j0.51 G0.001 j0.16 0.29
TNF-> 0.2 0.18 0.01 0.97 j0.11 0.47 j0.2 0.18
Peptide YY 0.15 0.34 0.18 0.25 0.10 0.53 j0.02 0.88
IGF-1 0.44 0.003 0.38 0.01 0.40 0.008 0.34 0.02
Insulin 0.51 G0.001 0.32 0.03 0.44 0.003
P G 0.05 appear in bold.

appetite and insulin resistance in other settings, but their role in weight loss.22,23 Recently, resistin level was found to be higher in
cancer cachexia is not well defined. cachectic compared with noncachectic lung and stomach cancer
The role of adiponectin in the setting of cancer is contro- patients24 and in esophageal cancer patients compared with non-
versial, and its relationship with anorexia has not been well cancer controls.13,25 In our study, resistin levels were comparable
established. Although subjects with obesity-related cancers have between cancer cachexia, cancer without cachexia, and noncancer
lower adiponectin levels at the time of diagnosis in some but not controls. This discrepancy could be explained by our smaller
all reports,13,14 this is not the case in lung or prostate cancer sample size, but other contributing factors such as the more severe
patients.15,16 In this study, we found increased adiponectin levels anorexia usually seen in esophageal or gastric cancer patients may
in cancer cachexia subjects compared with noncancer controls. have also played a role. Nevertheless, the fact that weight loss,
We expected these changes in cachectic subjects because weight appetite, and insulin resistance were not associated with resistin
loss is associated with increased adiponectin levels in non- levels suggests that it is unlikely that resistin plays a major meta-
cancer subjects.17 However, adiponectin levels were also ele- bolic role in this setting.
vated in cancer subjects without cachexia, suggesting that other
mechanisms besides weight loss are responsible for the in-
creased adiponectin levels in this setting. Although adiponectin
was inversely related to body weight, it was not a significant TABLE 3. Multivariate Analyses for Appetite, Body Weight,
predictor of weight change, appetite, or insulin resistance in this HOMA-IR, and Weight Change in Cancer Patients
population.
Given adiponectin’s antiproliferative properties,18 it is pos- Dependent Regression
sible that this increase in adiponectin levels may represent a Variable Variable Coefficient, A SE P
compensatory response present in subjects with advanced cancer Appetite Constant 2.66 0.77 0.001
aimed at controlling the disease. On the other hand, several Leptin level 0.77 0.31 0.01
publications recently have indicated that adiponectin may have
IL-6 level j0.03 0.01 0.01
proangiogenic properties in animal models of cancer19 or under
caloric restriction.20 Hence, the increase in adiponectin in this IGF-1 level 0.003 0.004 0.52
setting could be a marker of increased proangiogenic environ- Insulin level 0.009 0.014 0.5
ment in the presence of cancer. Further studies are needed to test Body weight Constant 20 2.46 0.001
these hypotheses and to clarify the role of adiponectin in this Leptin level 3.43 0.61 0.001
setting. Adiponectin level j0.78 0.79 0.33
Adiponectin’s orexigenic properties in rodents raise the IGF-1 level 0.01 0.01 0.32
possibility that this pathway may be a therapeutic target for Insulin level 0.01 0.04 0.66
anorexia and cachexia in the setting of cancer.1 Although our HOMA-IR Constant j5.45 45.56 0.9
data does not support the hypothesis that low adiponectin levels Leptin level 55.26 24.47 0.02
contribute to anorexia in this setting given the lack of correlation
IGF-1 level 0.35 0.37 0.34
with anorexia and that endogenous adiponectin levels are
already elevated, investigation of this pathway is warranted Weight change Constant j1.78 3.46 0.61
because further activation of the adiponectin receptor may still Acyl-ghrelin level j0.05 0.01 0.002
prove useful in the same way that exogenous ghrelin adminis- IL-6 level j0.11 0.04 0.01
tration increases appetite and food intake in cancer subjects Leptin level 2.57 1.39 0.07
despite their high endogenous ghrelin levels.12,21 Insulin level j0.02 0.06 0.73
Resistin, a product of white adipose tissue in mice and of IGF-1 level 0.02 0.01 0.21
mononuclear cells in humans, is associated with insulin resistance
Multiple linear regression analysis models include those predictors
and obesity, but its relationship with weight change is not clear. with significant correlations in univariate analyses. Significance appears
Some authors have reported a decrease in circulating resistin level, in bold.
whereas others have found an increase in its levels in response to

* 2010 The American Federation for Medical Research 557

Copyright @ 2010 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
 Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.com

Smiechowska et al Journal of Investigative Medicine & Volume 58, Number 3, March 2010

Leptin suppresses appetite in other settings, and lack of new therapies that could prolong survival and increase quality of
leptin or defects in leptin signaling are associated with hyper- life in these patients.
phagia and massive obesity.26,27 In agreement with previous
reports, the leptin level was significantly lower in our cachectic
patients compared with noncachectic cancer subjects and healthy REFERENCES
controls.28 This is expected because leptin is secreted by adipo- 1. Kubota N, Yano W, Kubota T, et al. Adiponectin stimulates
cytes in proportion to fat mass and excludes leptin excess as AMP-activated protein kinase in the hypothalamus and increases
a cause for anorexia in this setting. We also found leptin level food intake. Cell Metab. 2007;6:55Y68.
to be directly associated with appetite, explaining 37% of the 2. Schwartz MW, Porte D Jr. Diabetes, obesity, and the brain. Science.
variance seen in cancer patients. That these patients are resistant 2005;307:375Y379.
to the orexigenic effects of hypoleptinemia has been proposed 3. Morton GJ. Hypothalamic leptin regulation of energy homeostasis
before.29 Our findings that leptin is inversely associated with and glucose metabolism. J Physiol. 2007;583:437Y443.
appetite in this setting is novel and further supports this hy-
4. Kos K, Harte AL, da Silva NF, et al. Adiponectin and resistin in
pothesis. We also found leptin level to be directly associated
human cerebrospinal fluid and expression of adiponectin receptors
with insulin resistance, explaining 19% of the variance seen in
in the human hypothalamus. J Clin Endocrinol Metab. 2007;92:
cancer patients. These findings suggest that leptin also may play 1129Y1136.
a role in the increased insulin resistance seen in noncachectic
cancer subjects. 5. Evans WJ, Morley JE, Argiles J, et al. Cachexia: a new definition.
Several in vivo studies in rodent models suggest that the Clin Nutr. 2008;27:793Y799.
increase in proinflammatory cytokines may be responsible for the 6. Levine JA, Morgan MY. Preservation of macronutrient preferences
decrease in appetite and food intake in this setting by acting in cancer anorexia. Br J Cancer. 1998;78:579Y581.
centrally.30,31 Our findings support this hypothesis because IL-6 7. McCall JL, Tuckey JA, Parry BR. Serum tumour necrosis factor
remained a significant predictor of anorexia in multivariate alpha and insulin resistance in gastrointestinal cancer. Br J Surg.
analysis. The leptin receptor is part of the cytokine receptor 1992;79:1361Y1363.
superfamily and activates several pathways that are normally 8. Bruera E, Kuehn N, Miller MJ, et al. The Edmonton Symptom
activated by IL-6 such as the c-Jun N-terminal kinase pathway.32 Assessment System (ESAS): a simple method for the assessment
It is possible that this increase in cytokines may also be of palliative care patients. J Palliat Care. 1991;7:6Y9.
responsible for the failure of low leptin levels to increase appetite 9. Chang VT, Hwang SS, Feuerman M. Validation of the Edmonton
in this setting.29,33 Symptom Assessment Scale. Cancer. 2000;88:2164Y2171.
Baseline laboratory values (leptin, adiponectin, resistin,
10. Heedman PA, Strang P. Symptom assessment in advanced palliative
TNF->, IL-6, HOMA-IR, ghrelin, and IGF-1) did not predict
home care for cancer patients using the ESAS: clinical aspects.
weight change percent in the following 12 months after the initial
Anticancer Res. 2001;21:4077Y4082.
data collection. These results may be limited to interpretation
because only baseline laboratory values were available, whereas 11. Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis model
weights were followed prospectively. Adiponectin level was assessment: insulin resistance and beta-cell function from fasting
negatively correlated with weight change percent, however, when plasma glucose and insulin concentrations in man. Diabetologia.
1985;28:412Y419.
all the groups were analyzed and when just cancer cachectic
subjects were compared with cancer noncachectic subjects. In the 12. Garcia JM, Garcia-Touza M, Hijazi RA, et al. Active ghrelin levels
future, it would be useful to measure changes in adiponectin and active to total ghrelin ratio in cancer-induced cachexia. J Clin
levels over time to study this relationship further, for reasons Endocrinol Metab. 2005;90:2920Y2926.
previously explained. 13. Nakajima TE, Yamada Y, Hamano T, et al. Adipocytokines and
Limitations of our study include the small sample size squamous cell carcinoma of the esophagus. J Cancer Res Clin
and the heterogeneity in cancer diagnosis that could explain Oncol. 2009;136:261Y266.
the lack of difference in resistin levels. Given the biological 14. Wolf I, Sadetzki S, Kanety H, et al. Adiponectin, ghrelin, and leptin
differences between visceral and subcutaneous fat regard- in cancer cachexia in breast and colon cancer patients. Cancer.
ing production of adipokines, assessing body composition by 2006;106:966Y973.
computed tomography or magnetic resonance imaging would 15. Freedland SJ, Sokoll LJ, Platz EA, et al. Association between
have been useful in determining the association between dif- serum adiponectin, and pathological stage and grade in men
ferent fat deposits and the level of these hormones. Although undergoing radical prostatectomy. J Urol. 2005;174:1266Y1270.
HOMA-IR correlates well with hyperinsulinemic euglycemic 16. Petridou ET, Mitsiades N, Gialamas S, et al. Circulating adiponectin
clamp, most of the data comes from large epidemiological levels and expression of adiponectin receptors in relation to lung
studies. The use of the clamp technique would have provided cancer: two case-control studies. Oncology. 2007;73:261Y269.
stronger support to the clinical relevance of the HOMA-IR
17. Sheu WH, Chang TM, Lee WJ, et al. Effect of weight loss on
data. These issues should be considered in the design of future
proinflammatory state of mononuclear cells in obese women.
studies.
Obesity (Silver Spring). 2008;16:1033Y1038.
In summary, adiponectin level was not associated with
appetite or insulin resistance in this group of men with cancer, but 18. Yokota T, Oritani K, Takahashi I, et al. Adiponectin, a new member
it was higher in the cachectic and noncachectic cancer patients of the family of soluble defense collagens, negatively regulates the
compared with that in the noncancer controls. This suggests that growth of myelomonocytic progenitors and the functions of
macrophages. Blood. 2000;96:1723Y1732.
other mechanisms besides weight loss are responsible for these
differences. Leptin is a marker for anorexia and insulin resistance 19. Landskroner-Eiger S, Qian B, Muise ES, et al. Proangiogenic
in cancer patients. It is unlikely that resistin plays a major contribution of adiponectin toward mammary tumor growth in
metabolic role in this setting. vivo. Clin Cancer Res. 2009;15:3265Y3276.
A better understanding of the mechanisms regulating 20. Kondo M, Shibata R, Miura R, et al. Caloric restriction stimulates
appetite and body weight may help with the development of revascularization in response to ischemia via adiponectin-mediated

558 * 2010 The American Federation for Medical Research

Copyright @ 2010 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
 Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.com

Journal of Investigative Medicine & Volume 58, Number 3, March 2010 Adipokines in Cancer Cachexia

activation of endothelial nitric-oxide synthase. J Biol Chem. 2009; leptin receptor gene causes obesity and pituitary dysfunction.
284:1718Y1724. Nature. 1998;392:398Y401.
21. Neary NM, Small CJ, Wren AM, et al. Ghrelin increases energy 28. Wallace AM, Kelly A, Sattar N, et al. Circulating concentrations
intake in cancer patients with impaired appetite: acute, randomized, of Bfree[ leptin in relation to fat mass and appetite in gastrointestinal
placebo-controlled trial. J Clin Endocrinol Metab. 2004;89:2832Y2836. cancer patients. Nutr Cancer. 2002;44:157Y160.
22. Bougoulia M, Triantos A, Koliakos G. Effect of weight loss with or 29. Mantovani G, Maccio A, Madeddu C, et al. Serum values of
without orlistat treatment on adipocytokines, inflammation, and proinflammatory cytokines are inversely correlated with serum
oxidative markers in obese women. Hormones (Athens). 2006;5: leptin levels in patients with advanced stage cancer at different sites.
259Y269. J Mol Med. 2001;79:406Y414.
23. Koebnick C, Wagner K, Garcia AL, et al. Increase in serum 30. Cahlin C, Korner A, Axelsson H, et al. Experimental cancer
resistin during weight loss in overweight subjects is related to lipid cachexia: the role of host-derived cytokines interleukin (IL)-6,
metabolism. Int J Obes (Lond). 2006;30:1097Y1103. IL-12, interferon-gamma, and tumor necrosis factor alpha evaluated
24. Karapanagiotou EM, Tsochatzis EA, Dilana KD, et al. The in gene knockout, tumor-bearing mice on C57 Bl background and
significance of leptin, adiponectin, and resistin serum levels in eicosanoid-dependent cachexia. Cancer Res. 2000;60:5488Y5493.
non-small cell lung cancer (NSCLC). Lung Cancer. 2008;61:391Y397. 31. Barton BE. IL-6-like cytokines and cancer cachexia: consequences
25. Kerem M, Ferahkose Z, Yilmaz UT, et al. Adipokines and ghrelin of chronic inflammation. Immunol Res. 2001;23:41Y58.
in gastric cancer cachexia. World J Gastroenterol. 2008;14:3633Y3641. 32. Wang Y, Kuropatwinski KK, White DW, et al. Leptin receptor
26. Montague CT, Farooqi IS, Whitehead JP, et al. Congenital action in hepatic cells. J Biol Chem. 1997;272:16216Y16223.
leptin deficiency is associated with severe early-onset obesity in 33. Dulger H, Alici S, Sekeroglu MR, et al. Serum levels of leptin
humans. Nature. 1997;387:903Y908. and proinflammatory cytokines in patients with gastrointestinal
27. Clement K, Vaisse C, Lahlou N, et al. A mutation in the human cancer. Int J Clin Pract. 2004;58:545Y549.

* 2010 The American Federation for Medical Research 559

Copyright @ 2010 American Federation for Medical Research. Unauthorized reproduction of this article is prohibited.
 Downloaded from http://jim.bmj.com/ on August 20, 2016 - Published by group.bmj.com

Adipokines in Patients With Cancer

Anorexia and Cachexia
Joanna Smiechowska, Anne Utech, George Taffet, Teresa Hayes,
Marco Marcelli and José M. Garcia

J Investig Med 2010 58: 554-559

doi: 10.2310/JIM.0b013e3181cf91ca

Updated information and services can be found at:

http://jim.bmj.com/content/58/3/554

These include:

References This article cites 31 articles, 6 of which you can access for free at:
http://jim.bmj.com/content/58/3/554#BIBL

Email alerting Receive free email alerts when new articles cite this article. Sign up
service in the box at the top right corner of the online article.

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to:

http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to:

http://group.bmj.com/subscribe/