International Journal of Colorectal Disease

, 24:275

Serum leptin, adiponectin, and resistin concentration in colorectal adenoma and carcinoma (CC) patients

Authors

  • Anna Kumor
    • Department of Laboratory DiagnosticsMedical University
    • Department of Digestive Tract DiseasesMedical University
  • Mirosława Pietruczuk
    • Department of Laboratory DiagnosticsMedical University
  • Ewa Małecka-Panas
    • Department of Digestive Tract DiseasesMedical University
Original Article

DOI: 10.1007/s00384-008-0605-y

Cite this article as:
Kumor, A., Daniel, P., Pietruczuk, M. et al. Int J Colorectal Dis (2009) 24: 275. doi:10.1007/s00384-008-0605-y

Abstract

Introduction

Leptin, adiponectin, and resistin are the proteins secreted by adipocytes, which affects the metabolism. While the role of leptin in colon carcinogenesis is documented, the effect of adiponectin and resistin remains unclear. It has been indicated that while leptin may potentiate the cancer cells growth, adiponectin and resistin may act oppositely.

Aim

The aim of this study is to determine the concentration of leptin, adiponectin, and resistin in patients with adenomatous polyps and colorectal cancer.

Methods

The serum concentration investigated adipohormones had been measured with ELISA in 37 patients with colorectal adenomas, 36 with colorectal cancer (CC) and in 25 controls with no colorectal pathology. Endoscopically removed polyps and CC biopsies had been evaluated with histopathology. Mean BMI value was calculated for all patients.

Results

Among 37 adenomas, 25 revealed high-grade dysplasia (HGD) and 12 low-grade dysplasia (LGD). All cases of CC were adenocarcinomas. No difference in the level of investigated adipohormones in serum between patients with HGD and LGD polyps was observed. The serum concentration of leptin and adiponectin in CC patients was lower than in patients with adenomas (p < 0.05; p < 0.05, respectively) as well as in controls (p < 0.01; p < 0.05, respectively). The concentration of resistin in CC was not significantly different in the adenoma group (p > 0.05) but higher than in controls (p < 0.05). There was a correlation between adiponectin and leptin serum concentration (r = 0.61).

Conclusion

We conclude that serum concentration of adiponectin and resistin may play an important role in colon carcinogenesis. We also assume that leptin may possibly have the prognostic value useful in clinical practice and its concentration is independent of BMI value.

Keywords

LeptinAdiponectinResistinColorectal adenomaColorectal cancer

Introduction

During the last two decades, fat tissue is increasingly viewed not only as an energy storage depot but also as an active endocrine organ that produces and secrets proteins that act as hormones [13]. To this day, 20 adipohormones have been described. According to their physiological role, they have been divided into two groups: insulin resistance-inducting factors; resistin, TNF alpha, and IL-6, and insulin-sensitizing factors; leptin, adiponectin, and visfatin [45]. Leptin, adiponectin, and resistin are the proteins produced and secreted mostly by white adipose tissue (WAT) [56]. Initially, leptin was thought to be expressed and secreted only by adipocytes; however, their production in placenta, gastric, colorectal, and mammary epithelial tissue has been documented [2].

Leptin, adiponectin, and resistin play an important role in energy homeostasis, glucose and lipid metabolism, and regulation of body weight [79]. Recent data have also revealed the functions of adipocytokines in immunity, cancer, and bone formation [10].

Based on many epidemiological data, it is suspected that increased body weight and insulin resistance can be associated with higher risk of severe human diseases, such as cardiovascular disorders, diabetes and development of cancers of various organs [11]. Many studies have shown the positive correlation between adiposity and increased risk of endometrium, prostate, and breast cancer in postmenopausal women [1216]. However, the mechanism of hypoleptinemia in hormonally independent tumors still remains unknown. Some authors suggested that low leptin serum concentration can be connected with weight loss observed in cachectic cancer patients [17]. On the other hand, others did not confirm these observations suggesting other, unknown mechanisms of hypoleptinemia in those patients [18]. It has been indicated, that serum concentration of leptin is much higher in hormonally dependent cancers than in hormonally independent ones [18]. In the latter, leptin may potentiate cancer cells growth and proliferation. The role of leptin in hormonally independent tumors still remains unknown. Adiponectin is abundant in serum of concentration approximately 1–15 μg/ml [19]. It was observed that adiponectin induced antiangiogenesis and antitumor activity [20, 21]. On the other hand, resistin effect on cell proliferation and differentiation needs further investigations. In addition, the role of these adipohormones in precancerous states especially adenomatous colorectal polyps has not been fully elucidated.

The aim of the current study was to determine the concentration of leptin, adiponectin, and resistin in patients with colorectal adenomatous polyps and colorectal cancer.

Material and methods

The study group comprised of 37 patients with colorectal adenomatous polyps (median age 62.4 ± 7.1; 14 men and 23 women) and 36 patients with CC (17 men and 19 women; median age 58.6 ± 9.1 years). In 37 patients with colorectal polyps, 45 polyps have been removed (in six cases, two and in two cases, three coexisting polyps have been detected). Among multiple polyps, only one with the highest grade of dysplasia has been further considered. Blood samples were obtained from 25 patients (ten men and 15 women; median age 60.1 ± 8.9 years) as controls. All the patients included into the control group underwent colonoscopy for numerous indications: screening colonoscopy, abdominal pain, change of bowel movement habits, and minor bleeding. In none of them was endoscopic colorectal pathology found. In addition, seven patients (28%) without clinical symptoms of colorectal pathology who took part in screening colonoscopy program were included in the control group (Table 1). The groups studied were demographically comparable to the control group. Pathology of endoscopically removed colorectal polyps and CC biopsy specimens has been further evaluated. All CC tumors have been classified according to Dukes’ and TNM staging systems, (Table 2) [22, 23]. The tumor was regarded as proximal when localized in ascending and transverse colon and distal—in descending colon, sigmoid, and rectum.
Table 1

Indications for the colonoscopy in examined groups and controls

Indications for the colonoscopy in control group

N (%)

 Rectal bleeding

6 (24)

 Abdominal pain

4 (16)

 Diarrhea

2 (8)

 Constipation

6 (24)

 Age between 50–65 years (colorectal cancer screening program)

7 (28)

Indications for the colonoscopy in examined groups (colorectal polyp and cancer)

 Rectal bleeding

14 (19)

 Diarrhea/constipation

22 (30)

 Pain in the abdomen

18 (25)

 Palpable pathological mass in abdomen

3 (4)

 Age between 50–65 years (colorectal cancer screening program)

16 (22)

Table 2

CC stage according to Dukes and TNM classifications In examined patients

No. of patients

TNM classification

 

Dukes’ classification

1

T1 N0 M0

I

A

3

T2 N0 M0

I

A

5

T3 N0 M0

IIA

B

7

T4 N0 M0

IIB

B

6

T1–2 N1 M0

IIIA

C

7

T3–4 N1 M0

IIIB

C

2

Any T N2 M0

IIIC

C

5

Any T any N M1

IV

D

Blood samples have been obtained in the morning after 12-h fasting from patients of all examined groups and controls before colonoscopy. In all patients, body mass index (BMI) value was calculated. Obese patients (BMI > 30) with higher risk of colorectal cancer have been excluded from the study. Blood serum was obtained after 30 min clotting and centrifugation at 2,000 rpm for 15 min at 4°C. Serum was removed and stored frozen at −80°C. Leptin, resistin, and adiponectin concentrations were measured with ELISA (R&D Systems, USA). The reference values specified by the R&D Systems company for healthy subjects for investigated adipohormones equal: leptin 7.36 ± 3.73 ng/ml for women and 3.84 ± 1.79 ng/ml for men; resistin 13.8 ± 4.64 ng/ml and for adiponectin range 0.86–21 μg/ml. The ELISA test for all investigated adipohormones had been performed at the same time in all patients.

Patients with accompanying diseases as: hypertension, renal, and hepatic insufficiency, oral intake disturbances, other cancers and with family history of CC were excluded from the study. To exclude patients with diabetes mellitus, in all examined groups, oral glucose tolerance test (OGTT) was performed. Informed consent was obtained from all the subjects studied and the project was approved by the local Ethics Committee.

Statistical analysis

To perform statistical analysis STATISTICA version 5.5 PL program was used. The data are presented as mean ± SD. To test the distribution of results, the Kolmogorov–Smirnov test with Lilliefors correction was used. Comparison between the groups was made with Student t test or Mann–Whitney rank-sum test. We calculated Pearson correlation between investigated adipocytokines in examined groups. For all tests, a p value <0.05 was considered as statistically significant.

Results

In 37 patients with polyps, 37 adenomas qualified for further investigation, of which 17 (46%) were found in the distal part of the colon and 20 (54%) in the proximal one. Among 37 colorectal adenomas, 21 (57%) were tubulous and 16 (43%) tubulovillous. No villous polyps were found within the examined group.

Among 37 colorectal polyps, 25 (67%) adenomas revealed high-grade dysplasia (HGD) and 12 (33%), low-grade dysplasia (LGD). The grade of cellular dysplasia was assessed according to the WHO classification [22]. Among HGD polyps, 12 (48%) were tubulous and 13 (52%) tubulovillous. Among 12 LGD polyps, six (50%) were tubulous and six (50%) tubulovillous.

According to Dukes’ classification, four adenocarcinomas were stage A (TNM stage I), 12 stage B (TNM stage IIA + IIB), 15 stage C (TNM stage IIIA + IIIB + IIIC), and five stage D (TNM stage IV). All patients with CC of Dukes’ D had liver metastasis. Because of a small number of adenocarcinomas in A and D Dukes’ stage, patients from groups A and B as well as C and D were considered together.

In all cases of CC, tumor grade was established as G2 based on the degree of anaplasia and nuclear size according to the WHO criteria [23].

In 20 (56%) CC cases, tumors were located proximally and in 16 (44%), distally.

No difference in mean BMI between patients with polyps (26.8 ± 4.1), CC (25.1 ± 4.8), and controls (25.8 ± 3.8) has been observed. The leptin serum concentration in control group in women was slightly higher (p > 0.05) than in men.

Serum leptin concentration in men with colorectal polyps was 5.24 ± 2.56 ng/ml, which was only slightly lower without statistical significance than in women, 6.56 ± 2.31 ng/ml (p > 0.05). Similarly, there were no statistical differences in serum leptin concentration between men and women with colorectal cancer (3.62 ± 1.41 ng/ml vs. 4.51 ± 2.07 ng/ml; p > 0.05). Because of a lack of differences in serum leptin concentration between both genders, in colorectal polyps and colorectal cancer groups, we made further statistical calculation for both genders together.

The leptin serum concentration was statistically higher in control group (10.21 ± 3.1 ng/ml) than in patients with adenomatous polyps (7.59 ± 3.94 ng/ml; p < 0.05). There were no differences in leptin serum concentration between patients with LGD and HGD polyps (6.72 ± 2.04 ng/ml and 6.39 ± 2.52 ng/ml; p > 0.05) as well as between patients with tubulous (6.07 ± 2.84 ng/ml) and tubulovillous polyps (6.82 ± 1.97 ng/ml, p > 0.05; Table 3). There were also no significant differences in serum concentration of leptin depending on polyp localization (p > 0.05). On the other hand, the serum leptin concentration in CC was 4.12 ± 1.98 ng/ml which was significantly lower than in colorectal polyps (7.59 ± 3.94 ng/ml) and in control group (10.21 ± 3.1 ng/ml; p < 0.05; p < 0.01, respectively). There were also no significant differences in serum concentration of leptin between A + B vs. C + D CC stage (3.88 ± 2.01 ng/ml vs. 4.02 ± 2.29 ng/ml; p > 0.05) as well as its localization (p > 0.05).
Table 3

Mean serum leptin concentration and p value in examined groups

Group of patients

Serum leptin concentration (ng/ml)

Control group

10.21 ± 3.1

LGD colorectal adenoma

6.72 ± 2.04a

HGD colorectal adenoma

6.07 ± 2.84b

Adenocarcinoma, Dukes’ A + B

3.88 ± 2.01c

Adenocarcinoma, Dukes’ C + D

4.02 ± 2.29d

ap < 0.05 compared to control group

bp > 0.05 compared to LGD adenoma

cp < 0.05 compared to HGD adenoma

dp > 0.05 compared to adenocarcinoma, Dukes’ A + B

The concentration of adiponectin in control group (17.8 ± 7.8 μg/ml) was only slightly higher than in colorectal adenomas (15.4 ± 7.6 μg/ml; p > 0.05) and significantly higher than in CC (12.0 ± 4.9 μg/ml; p < 0.05; Table 4). There was no significant difference in adiponectin serum concentration between patients with LGD (14.7 ± 6.2 μg/ml) and HGD polyps (15.9 ± 8.0 μg/ml). No significant differences in adiponectin serum concentration depending on pathology: tubulous (15.7 ± 7.4 μg/ml) vs. tubulovillous adenoma structure (15.0 ± 8.1 μg/ml) as well as its localization have been observed (p > 0.05; p > 0.05, respectively). There were also no significant differences in serum concentration of adiponectin between A + B vs. C + D CC stage (12.3 ± 3.9 μg/ml vs. 11.4 ± 3.2 μg/ml; p > 0.05) as well as depending on tumor localization (p > 0.05).
Table 4

Mean serum adiponectin concentration and p value in examined groups

Group of patients

Serum adiponectin concentration (μg/ml)

Control group

17.8 ± 7.8

LGD adenoma

14.7 ± 6.2a

HGD adenoma

15.9 ± 8.0b

Adenocarcinoma, Dukes’ A + B

12.3 ± 3.9c

Adenocarcinoma, Dukes’ C + D

11.4 ± 3.2d

ap > 0.05 compared to control group

bp > 0.05 compared to LGD adenoma

cp < 0.05 compared to HGD adenoma

dp > 0.05 compared to adenocarcinoma, Dukes’ A + B

The serum concentration of resistin in CC patients (6.79 ± 2.41 ng/ml) as well as in colorectal adenomas (5.29 ± 2.44 ng/ml) was significantly higher than in the control group (3.6 ± 1.08 ng/ml; p < 0.05; p < 0.05, respectively). On the other hand, the concentration of resistin did not differ in patients with colorectal adenomas and those with CC (p > 0.05; Table 5). There were no differences in resistin concentration depending on the grade of dysplasia and yielded: 4.89 ± 2.15 ng/ml in patients with LGD vs. 5.7 ± 2.89 ng/ml in patients with HGD adenomas, p > 0.05. No differences depending on polyp localization in resistin values have been found (p > 0.05). Similarly, no differences considering resistin level between patients with tubulous (5.46 ± 3.03 ng/ml) and tubulovillous polyps (5.11 ± 1.79 ng/ml) have been detected (p > 0.05).
Table 5

Mean serum resistin concentration and p value in examined groups

Group of patients

Serum resistin concentration (ng/ml)

Control group

3.6 ± 1.08

LGD adenoma

4.89 ± 2.15a

HGD adenoma

5.7 ± 2.89b

Adenocarcinoma, Dukes’ A + B

5.86 ± 3.1c

Adenocarcinoma, Dukes’ C + D

6.79 ± 2.41d

ap < 0.05 compared to control group

bp > 0.05 compared to LGD adenoma

cp > 0.05 compared to HGD adenoma

dp > 0.05 compared to adenocarcinoma, Dukes’ A + B

There were also no significant differences in serum concentration of resistin between A + B vs. C + D CC stage (5.86 ± 3.1 ng/ml vs. 6.76 ± 4.2 ng/ml; p > 0.05) as well as its localization (p > 0.05). Among all examined patients with CC and adenomas, the correlation between adiponectin and leptin serum concentration (r = 0,61) has been observed (Fig. 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00384-008-0605-y/MediaObjects/384_2008_605_Fig1_HTML.gif
Fig. 1

Correlation between leptin and adiponectin

Discussion

Leptin, adiponectin, and resistin play the physiologic role in the reduction of food intake, energy homeostasis, and regulation of body weight. Simons JP et al. and Wallace AM et al. suggested that low-serum leptin concentration in their studies could be related to decreased body fat mass which is in opposition to our results [2426]. In the present study, we did not observe a relationship between serum leptin concentration and weight loss in investigated groups. These data are with agreement with the study of Arpaci et al. who also observed decreased serum level of leptin in CC patients without weight loss and anorexia [1]. Based on our observations, we suspect that weight loss is not a causative factor of hypoleptinemia and some unknown mechanisms may be responsible for low-serum leptin level in colorectal adenoma and CC patients.

We observed, to our knowledge, for the first time, lower leptin concentration in patients who presented with colorectal adenomas. We also found out that serum concentration of leptin did not differ in patients with colorectal adenomas depending on the grade of dysplasia, histopatology, and localization, what also have not been studied. Hardwick JC et al. observed increased proliferation of human colon cancer cells lines in studies in vitro and in vivo after stimulation with leptin. Based on this, the authors suspect that leptin represents the growth factor for colonic epithelial cells and can stimulate proliferation of the colon mucosa cells [27]. Similarly, Garofalo et al. suggested that leptin not only stimulates cell growth via ERK1/2 pathway in colorectal cancer but also reduces cell apoptosis [28]. However, Aparicio et al. observed that leptin stimulates proliferation only in vitro but did not promote tumor growth in in vivo study. We assume that these observations may be caused due to a much higher leptin concentration used in in vitro study. Tessitore et al. observed increased serum leptin concentration in breast cancer patients in comparison to control group [18]. The stimulatory effects of leptin on hormone-dependent tumors growth, such as breast and prostate cancer growth were considered to occur primarily via activation of the estrogen receptor [29, 30]. In our study, we observed significantly lower serum leptin concentration in CC patients than in patients with colorectal adenomas as well as in the control group. We did not observe differences between CC depending on stage (Dukes’ and TNM scale), which is in agreement with the study of Levy et al. The authors did not observe a statistically significant difference in leptin and adiponectin serum concentration between early and metastatic stages of colorectal cancer [31]. Our observations are also in agreement with the study of Bolukbas et al. Those authors observed significantly lower serum leptin concentration in CC patients than in control group. On the other hand, Tessitore L et al. showed that plasma leptin levels in CC were similar to controls [18].

Fukumoto et al. observed in patients with adenoma, slightly lower adiponectin serum concentration than in healthy controls [32]. As far as we are aware, this is the first analysis of the differences in adiponectin serum concentration in colorectal adenoma patients depending on grade of dysplasia, histopatology, and polyp localization. In our study, we also observed higher serum adiponectin concentration in control group than in patients with colorectal adenomas and CC. Our data correspond with the observations of Ishikawa M et al. who found lower adiponectin serum level in patients with gastric cancer in comparison to the control group [33]. Moreover, they observed that adiponectin serum concentration inversely correlated with tumor size and TNM stage. The latter may give support to our data concerning patients with different CC stages. In addition, a prospective study by Wei E et al. has detected an increased risk of CC in patients with low adiponectin concentration [34]. We hypothesized that probably, a low concentration of adiponectin in CC patients may be caused by overexpression of proinflammatory cytokines often observed in carcinogenesis. Wang et al. suspect that low concentration of adiponectin in CC patients serum can be caused by inhibition of this adipohormone not only by TNF alpha but also other cytokines [35]. On the other hand, a relatively low concentration of adiponectin may be caused by elevated expression of adipohormone receptors—AdipoR1 and Adipo R2—which was recently observed in vivo in colorectal cancer tissue and in vitro study [36]. Ogunwobi et al. have shown that adiponectin inhibits leptin-induced cell proliferation via adiponectin receptor 1 (Adipo R1) in esophageal cancer tissue. The authors suspect that relative adiponectin deficiency may cause increased proliferation that predispose to malignant transformation [37].

In the present study, we observed that highest resistin serum concentration in CC patients and lowest in control group. Resistin is produced mostly by monocytes and macrophages of peripheral blood. Lehrke et al. observed that TNF alpha and IL-6 induced production of resistin mRNA as well as protein secretion by primary human macrophages [38]. These data can explain the highest concentration of resistin in colorectal cancer, which is strongly connected with inflammation [39]. They also found out that insulin sensitizers that have anti-inflammatory properties, including a synthetic PPARγ agonist (rosiglitazone), as well as aspirin, suppress macrophage resistin expression. To our best knowledge, there are no published data on resistin in colorectal adenomas and cancer; however, increased serum resistin concentration in patients with other cancers has been reported. Kan JH et al. observed significantly higher resistin level in breast cancer group 5.23 ± 6.9 ng/mL vs. control 1.46 ± 2.0 ng/mL. They also noticed that the risk of breast cancer was significantly increased in the highest tertile group for serum resistin level compared to the lowest tertile one [40].

Chronic inflammation role in cancer development is well known. Inflammation may be represented by similar biomarkers as early carcinogenesis in various organs. Thus, the correlation between plasma resistin levels and breast cancer risk in the study by Kan JH et al. has been explained by accompanying inflammation [41].

Increased serum resistin concentration has been evaluated in other precancerous colon states. Konrad A et al. observed in patients with IBD significantly higher resistin levels compared with controls. They also noticed that resistin concentrations were significantly correlated with elevated white blood cell count, C-reactive protein (CRP), and disease activity [40]. Oppositely, Karmiris K et al. did not find a correlation between resistin and C-reactive protein in IBD patients [42]. Moreover, Mu H et al. found out that resistin induces not only human endothelial cell proliferation and migration but also promotes capillary-like tube formation. Resistin upregulates the mRNA expression of vascular endothelial growth factor receptors (VEGFR-1 and VEGFR-2) and matrix metalloproteinases (MMP-1 and MMP-2) at both mRNA and protein levels which play a pivotal role in carcinogenesis and metastasis formation [43]. Calabro P et al. found in human aortic smooth muscle, the increased proliferative effect in cells after stimulation with increasing concentrations of resistin through both ERK 1/2 and Akt signaling pathways [44]. To date, the role of resistin in colorectal carcinogenesis is not fully elucidated yet. We can speculate that resistin may promote colonocytes proliferation, however, further research are necessary. To our best knowledge, this is the first study in which resistin serum concentration was investigated both in CC and colorectal adenomas. In our study, we did not observe any differences in resistin serum concentration in colon adenomas of different grade of cellular dysplasia, histopatology, and adenomas localization.

We have observed the correlation between adiponectin and leptin serum concentration in patients with colorectal adenomas and cancer. Therefore, we can assume that both these adipohormones may possibly play a significant and opposite role in both CC and even in CA formation, however, both adipohormones deserve further attention. On the basis of the present data, we can speculate that increased mucosal proliferation in colorectal carcinogenesis is stimulated with leptin and not inhibited with adiponectin due to the relative decreased concentration of the latter in colorectal cancer tissue. Based on our data, the decrease of leptin serum concentration in colorectal adenoma and CC patients is independent on BMI and weight loss. Due to the decrease in leptin and adiponectin serum concentration with the malignant transformation progression, we speculate that the estimation of those adipohormones might provide an additional tool for colorectal neoplasia prognosis. The possible role of all mentioned adipohormones in colorectal carcinogenesis prognosis needs further investigations.

Acknowledgement

Supported by grant no 503-1002-1 and 507-11-333.

Copyright information

© Springer-Verlag 2008