The authors have declared that no competing interests exist.
Ghrelin and obestatin are gastrointestinal peptides, encoded by the same preproghrelin gene. Both are expressed in breast cancer tissue and ghrelin has been implicated in breast cancer tumorigenesis. Despite recent advances in breast cancer management the need for new prognostic markers and potential therapeutic targets in breast cancer remains high. We studied the prognostic impact of ghrelin and obestatin in women with node negative breast cancer.
Within a cohort of women with breast cancer with tumor size ≤ 50 mm, no lymph node metastases and no initiation of adjuvant chemotherapy, 190 women were identified who died from breast cancer and randomly selected 190 women alive at the corresponding time as controls. Tumor tissues were immunostained with antibodies versus the peptides.
Ghrelin expression was associated with better breast cancer specific survival in univariate analyses (OR 0.55, 95% CI 0.36–0.84) and in multivariate models, adjusted for endocrine treatment and age (OR 0.57, 95% CI 0.36–0.89). Obestatin expression was non-informative (OR 1.2, 95% CI 0.60–2.46). Ghrelin expression is independent prognostic factor for breast cancer death in node negative patients—halving the risk for dying of breast cancer. Our data implies that ghrelin could be a potential therapeutic target in breast cancer treatment.
Ghrelin is a 28 amino acid peptide which was first identified as a ligand of growth hormone (GH) secretagogue receptors (GHSRs) that regulates GH secretion [
Many of the ghrelin system constituents (ghrelin, obestatin, ghrelin splicing variants, and GHSRs) are present in normal breast tissue, breast tumors, and breast cancer cell lines [
Furthermore, specific receptors which bind natural ghrelin and synthetic GH secretagogues (GHSs) are present in human breast carcinomas. This binding is independent of tumor histological type, stage, ER status, Ki67, pre- or postmenopausal status, but correlates to the grade of tumor differentiation [
Studies on the functional role of the ghrelin system in the regulation of relevant processes in breast cancer development and progression are still limited and data are conflicting. Ghrelin causes significant inhibition of cell proliferation in human breast carcinoma cell lines and thus, may have a clinical application in breast cancer therapy [
The ghrelin system may be attractive target for new treatment possibilities. Potential therapeutic approaches using ghrelin mimetic compounds and antagonists in clinical disease are currently being developed. The orexigenic and GH releasing effects of ghrelin make it a good agent to be used in catabolic states/situations. In recent years, synthetic agonists and antagonists of the ghrelin receptor have been developed for the possible treatment of metabolic or nutritional disorders [
In the present study, we examined the roles of ghrelin and obestatin as prognostic factors in a clinically well-characterized set of cases and controls nested within a population based cohort of lymph node negative, chemotherapy-naive patients.
The source population of the study was a defined cohort of women diagnosed with breast cancer in the Uppsala-Örebro region, Sweden, from 1993–2004. Information about the patients was derived from the Uppsala-Örebro Breast Cancer Register, which is a population-based clinical database with coverage of over 98%. Inclusion criteria for this study were tumor size ≤50 mm, no lymph node metastases, and no adjuvant chemotherapy. Nine hundred women met the inclusion criteria. Within this cohort, cases were defined as women who died from breast cancer as reported to the national causes of death register. All eligible cases were selected. Eligible as controls were women in the cohort alive at the time of the corresponding case's death. Two-hundred and forty cases were identified. For each identified case, one control was randomly selected. Fifty cases and corresponding controls were excluded from the study for not meeting inclusion criteria after reviewing data from patient records and pathology reports, or because of missing tumor blocks: 26 patients had new/contralateral or locally advanced breast cancer, in 12 patients no paraffin blocks were found, six patients had non-breast cancer death, four patients had distant metastases at diagnosis, one patient received adjuvant chemotherapy, and one patient had no breast surgery performed. This resulted in a total of 190 cases and 190 controls. The average age was 66 years for cases and 61 years for controls. The average tumor size was 20 mm for cases and 16 mm for controls. All patients underwent surgery consisting of either modified radical mastectomy with axillary dissection, or sector resection with axillary dissection and post-operative irradiation of the breast. Fifty-three (28%) cases and 48 (25%) controls received antihormonal therapy. Patients’ characteristics including grade, hormone receptors and HER2 are shown in
Parameter | Case, n (%) | Control, n (%) | |
---|---|---|---|
Age, average (years) | 66 | 61 | |
Tumor size, average (mm) | 20 | 16 | |
Tumor histology | |||
Ductal | 163 (86) | 145 (76) | |
Lobular | 20 (10) | 23 (12) | |
Others | 7 (4) | 22 (12) | |
Histologic grade | |||
I | 19 (10) | 48 (25) | |
II | 94 (50) | 105 (55) | |
III | 76 (40) | 34 (18) | |
Not known | 1 (0) | 3 (2) | |
ER | |||
Positive | 103 (54) | 147 (77) | |
Negative | 79 (42) | 41 (22) | |
Not known | 8 (4) | 2 (1) | |
PgR | |||
Positive | 73 (38) | 127 (66) | |
Negative | 108 (57) | 60 (32) | |
Not known | 9 (5) | 3 (1) | |
Ki67 | |||
High (>22%) | 80 (42) | 56 (29) | |
Low | 100 (53) | 127 (67) | |
Not known | 10 (5) | 7 (4) | |
HER2 | |||
Overexpression (IHC 3+ or FISH pos) | 18 (10) | 13 (7) | |
Normal | 158 (83) | 161 (85) | |
Not known | 14 (7) | 16 (8) | |
Adjuvant radiotherapy | |||
Yes | 101 (53) | 116 (61) | |
No | 89 (47) | 74 (39) | |
Adjuvant endocrine therapy | |||
Yes | 53 (28) | 47 (25) | |
No | 137 (72) | 143 (75) | |
Total | 190 | 190 |
ER, estrogen receptor; FISH, fluorescent
The study was approved and the need for consent was waived by the local ethics committee, Regionala etikprövningsnämnden (EPN), in Uppsala, Sweden.
Tissue microarrays (TMA) were produced at the TMA facility, the Department of Laboratory Medicine, Center for Molecular Pathology, Lund University. Paraffin blocks from the patients' primary tumor were collected. Hematoxylin and eosin sections were reviewed and areas with invasive tumor were selected. Each tumor was reevaluated and reclassified according to the Elston and Ellis grading system (R-MA) [
The primary antibodies used for immunohistochemical staining were anti-obestatin, an in-house developed antibody (rabbit polyclonal), for which the production and characterization has been described previously [
Immunohistochemical staining was performed using the Dako EnVision Plus-HRP Detection Kit (K401111-2, Dako, Glostrup, Denmark) according to the manufacturer’s instructions. For antigen retrieval the sections were subjected to pre-treatment (microwave heating for 10 min at 750 W followed by 15 min at 380 W using Tris-HCl buffered saline, pH 8.0). The sections were incubated with the primary antibodies in PBS with 1% BSA over night at 4°C. Bound antibodies were visualized by incubation with liquid 3, 3′-Diaminobenzidine substrate chromogen for 5 min.
The analyses of the ghrelin and obestatin immunostainings were manually performed by two observers (MG and YN). None of the investigators had access to clinical data or outcomes. The intensity of the staining in the tumor cells was examined and scored on a scale as non-immunoreactive (non-IR, 0), weak (1), moderate (2) and strong (3). Each of the two cores from every tumor on the array was examined and scored separately. In case of conflicting results, a third evaluation was performed by the observers and consensus was reached. If one of the two tumor cores was lost, the remaining one was used for scoring. The entire core(s) from each tumor was examined and at least 200 tumor cells had to be evaluable to be designated an intensity score. Positive staining was defined as complete and/or partial (>50% IR tumor cells) staining at any intensity that could be differentiated from truly negative staining, background and diffuse non-specific staining. Cytoplasmic staining in high-power fields (40X objective) was accepted as positive reaction.
Photographs were taken using a Zeiss Observer Z1 microscope and the Axiovision software (Carl Zeiss, Gottingen, Germany).
This material has been used previously [
The specificity of the ghrelin and obestatin antibodies has been evaluated and presented previously [
Conditional logistic regression analysis was performed to estimate odds ratios (ORs) and confidence intervals (CIs) as a measure of the relative risk of being at decreased or increased risk of dying from breast cancer due to having a breast cancer with ghrelin and/or obestatin expression. Established prognostic factors such as age, tumor size and histological grade as well as ghrelin and obestatin were analyzed in univariate models.
Adjustment for adjuvant endocrine therapy and age was also performed. Correlations between ghrelin/obestatin and other clinicopathological parameters were assessed with Spearman's correlation test.
A directed acyclic graph (DAG) was used to determine factors to include a multivariate model, indicating that none of the factors should be included in the model [
Comparison of the agreement of the two observers’ results was performed to evaluate the reproducibility of the scoring of the immunohistochemical results. The material was manually scored using a light microscope. The degree of concordance between the two investigators was quantified as the chance-corrected measure of agreement, known as kappa [
Results from routine stainings for hormone receptors and HER2 are described in the patients’ characteristics (
Case, n (%) | Control, n (%) | Total, n (%) | |
---|---|---|---|
0 | 113 (64) | 88 (49) | 201 (57) |
1 | 38 (21) | 41 (23) | 79 (22) |
2 | 18 (10) | 34 (19) | 52 (15) |
3 | 8 (5) | 15 (9) | 23 (6) |
Total, n | 177 | 178 | 355 |
0 | 17 (9) | 19 (11) | 36 (10) |
1 | 21 (12) | 14 (8) | 35 (10) |
2 | 66 (37) | 55 (32) | 121 (35) |
3 | 74 (42) | 84 (49) | 158 (45) |
Total, n | 178 | 172 | 350 |
The intensity of the staining in the tumor cells scored on a scale as non-immunoreactive (0), weak (1), moderate (2) and strong (3).
Various patterns of immunostaining intensity were observed. The majority of the cases/controls had a low to moderate intensity in the ghrelin stainings. A tendency towards a higher intensity in the obestatin stainings was observed. Representative photos from the ghrelin immunostainings are shown in
Representative images of ghrelin with 0 (non-immunoreactive), 1 (weak), 2 (moderate) and 3 (strong) immunostaining. Scale bar = 100 μm.
Ghrelin and obestatin were positively correlated to each other (Spearman correlation). No strong correlations could be identified between ghrelin/obestatin to any of the investigated clinicopathological parameters. Obestatin had statistically significant correlations with ER, NHG and Ki67, but still at negligible levels with Spearman correlation coefficients <0.20. Spearman’s correlations are presented in
Ghrelin | ρ | p-value | n |
---|---|---|---|
Tumor size | -0.01 | 0.87 | 178 |
Age | -0.14 | 0.06 | 178 |
ER | 0.10 | 0.20 | 176 |
PgR | 0.12 | 0.10 | 176 |
HER2 | -0.05 | 0.50 | 164 |
NHG | -0.12 | 0.12 | 176 |
Ki67 | -0.12 | 0.11 | 174 |
Tumor size | -0.09 | 0.26 | 172 |
Age | 0.02 | 0.83 | 172 |
ER | 0.18 | 0.02 | 170 |
PgR | 0.11 | 0.16 | 170 |
HER2 | 0.04 | 0.58 | 158 |
NHG | -0.16 | 0.03 | 169 |
Ki67 | -0.17 | 0.03 | 167 |
Ghrelin | 0.22 | 0.004 | 165 |
ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; NHG, Nottingham Histological Grade; PgR, progesterone receptor; ρ, Spearman’s correlation test coefficient.
A statistically significant association was observed between breast cancer death and expression of ghrelin (p-value 0.006), histological grade (p-value 0.00), age (p-value 0.00) and tumor size (p-value 0.001) using conditional logistic regression in a univariate model. Overexpression of HER2 was not statistically associated with breast cancer death (p-value 0.83), neither was obestatin (p-value 0.59). The numerically highest ORs for breast cancer death were predicted by grade (OR ≥ 3). OR for breast cancer death using Ki67 was numerically lower (OR < 2). Data is summarized in
OR (95% CI) | p-value | ||
---|---|---|---|
Tumor size | ≤20 mm vs. >20 mm | 2.2 (1.3–3.5) | 0.001 |
NHG | I+II vs. III | 3.0 (1.8–4.9) | 0.00 |
Ki67 | ≤22% vs. >22% [ |
1.7 (1.1–2.6) | 0.02 |
HER2 | Negative vs. positive | 0.9 (0.4–2.1) | 0.83 |
Age | <70 yrs vs. ≥70 yrs | 2.4 (1.5–3.9) | 0.00 |
Ghrelin intensity | 0 vs. 1+2+3 [ |
0.6 (0.4–0.8) | 0.006 |
Obestatin intensity | 0 vs. 1+2+3 [ |
1.2 (0.6–2.5) | 0.59 |
Ghrelin intensity |
0 vs. 1+2+3 [ |
0.6 (0.4–0.9) | 0.007 |
Endocrine treatment | Negative vs. positive | 1.0 (0.6–1.7) | 0.93 |
Ghrelin intensity |
0 vs. 1+2+3 [ |
0.6 (0.4–0.9) | 0.01 |
Endocrine treatment | Negative vs. positive | 1.0 (0.6–1.7) | 0.95 |
Age | <70 yrs vs. ≥70 yrs | 2.4 (1.4–3.9) | 0.001 |
Odds ratio (OR) and 95% confidence intervals (CI) obtained from univariate and multivariable Cox conditional logistic models. HER2, human epidermal growth factor receptor 2; NHG, Nottingham Histological Grade.
*Model adjusted for endocrine treatment.
**Model adjusted for endocrine treatment and age.
The DAG analysis showed that none of the factors influenced ghrelin in such way that they should be included in a multivariate model. To further exclude confounding, two additional models were performed. An analysis on ghrelin adjusted for endocrine treatment showed no change in the prognostic power of ghrelin with an OR of 0.55 (95% CI, 0.36–0.85) in the adjusted analysis as compared with OR 0.55 (95% CI, 0.36–0.84) in the non-adjusted analysis. Thus, endocrine treatment did not appreciably affect the prognostic value of ghrelin. In a model adjusted for both endocrine treatment and age, ghrelin was still a prognostic factor with OR 0.57 (95% CI, 0.36–0.89).
The two investigators (MG and YN) examined the ghrelin and obestatin immunostainings “blindly” and separately. The kappa value for ghrelin was 0.81 and for obestatin 0.83 (almost perfect agreement), demonstrating very good reproducibility of the scorings.
In all experiments the positive control showed IR cells in the deeper third part of the gastric mucosa as expected (data not shown).
This study shows that expression of ghrelin is a prognostic factor for lymph node negative breast cancer patients, not treated with adjuvant chemotherapy, in both uni-and multivariate analyses.
The prognostic value of ghrelin and obestatin was assessed in a well-defined group of women with breast cancer as the study was designed with the intention to investigate potential prognostic factors in breast cancer. The number of cases (breast cancer deaths), i.e., the events that drive the statistical power, is high for a study in the field. Thus, our data should reflect results that would have been achieved if the whole cohort had been studied.
One limitation of this study is that women received adjuvant endocrine treatment according to standard guidelines spanning over a time period of more than 20 years. Endocrine treatment was inconsistently prescribed, where 73% of the women receiving adjuvant endocrine treatment had ER positive tumors and the rest (27%) ER negative tumors
To determine which factors to be included in the multivariate analysis, DAG depicting causal relations was used. This was done to assess whether adjustment for a factor minimizes or introduces bias and to select possible confounding factors where causal relations are depicted. Clinicopathologic factors included in this study, such as ER, proliferation index and age, are all well-known factors associated with breast cancer survival. To our knowledge, there is no indication that ghrelin influences these factors in such a way that it in turn would affect breast cancer prognosis; hence, they should not be included in the statistical model. This is in line with the results from the correlation analyses/DAG, where none of the factors is strongly correlated to ghrelin. Additional analyses to further exclude the risk of confounding showed that ghrelin was a prognostic factor independent from endocrine treatment and age.
The data from this study suggest that patients with tumors non-IR for ghrelin have approximately 2 times higher risk for breast cancer-specific death, similar to established prognostic factors, giving ghrelin a possible role as a new prognostic marker. The prognostic importance of ghrelin seems robust since we have obtained similar results in another study [
In this study, obestatin did not provide any prognostic information as opposed to ghrelin, despite being derived from the same precursor. Although there was a significant correlation between ghrelin and obestatin, it was weak (rho value 0.22). Thus, it was anticipated that the two proteins would not provide different similar prognostic information. Nonetheless, the result is in accordance with our previous studies [
The results from this study indicate that ghrelin could be used as a prognostic factor in breast cancer. Additionally, ghrelin represents an interesting and attractive target [
In conclusion, ghrelin is a prognostic factor for breast cancer related death in women with node-negative breast cancer. This finding suggests a potential role of ghrelin in breast cancer—both as a prognostic marker but possibly also as a potential therapeutic target. To validate the function of ghrelin in breast cancer, larger population studies are necessary, which also should include functional and genetic studies. With growing understanding in the functionality of ghrelin and the molecular pathways involved, it is a promising therapeutic target.
The authors wish to thank Åsa Forsberg for skillful technical assistance.