Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Bone metastases affect 50% to 70% of breast cancer (BC) patients and have a high mortality rate. Adipose tissue loss plays a pivotal role in the progression of cancer.

OBJECTIVE:

This study aims to evaluate the prognostic value of adipose tissue for bone metastasis in BC patients.

METHODS:

517 BC patients were studied retrospectively. Patients’ characteristics before the surgery were collected. Quantitative measurements of the subcutaneous fat index (SFI) were performed at the level of the eleventh thoracic vertebra. In order to adjust for the heterogeneity between the low SFI and high SFI groups, propensity score matching (PSM) was used. The Kaplan-Meier method was used to estimate the 5-year bone metastatic incidence. The prognostic analysis was performed with the Cox regression models.

RESULTS:

Compared with the patients without bone metastasis, the patients with bone metastasis had reduced SFI levels. In addition, Kaplan-Meier analysis revealed that patients with low SFI were more likely to develop bone metastases. The independent predictive value of SFI for bone metastases was confirmed by Cox regression analysis. The survival analysis was repeated after PSM with a 1:1 ratio, yielding similar results ( $P<$ 0.05).

CONCLUSIONS:

SFI is an independent predictor of bone metastasis in BC patients.

Keywords

Breast cancer bone metastasis subcutaneous fat index visceral fat index propensity score analysis

1. Introduction

Breast cancer (BC) is the most frequently diagnosed cancer and is the main reason for cancer-related mortality in women [1]. Most of the deaths from BC are due to distant metastases [2]. The bone is the most common site of metastasis, which has been shown to occur in between 50% and 70% of BC patients [3]. Patients with bone metastasis are accompanied by excessive, osteoclast-mediated bone destruction and have a 5-year overall survival (OS) rate of 22.8% [4, 5]. Early detection of bone metastasis is crucial for improving BC survival. However, the predictive factors for bone metastasis have not been sufficiently elucidated. Therefore, there is an urgent need for identifying novel markers that might provide new clues for prevention.

Recently, researchers have shown an increased interest in adipose tissue. Loss of adipose tissue is crucial to the development and progression of cancer [6, 7]. Adipose tissue is anatomically dispersed throughout the body in various compartments, and its function varies according to the region concerned. Subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) are physically distinct types of white adipose tissue [8]. The VAT is currently thought of as a metabolic and endocrine organ that can affect body weight balance and systemic immune conditions. SAT, a metabolic storehouse, is linked to the accumulation of visceral fat [9]. In a number of tumor types, including BC, colorectal cancer, hepatocellular carcinoma, and gastric cancer, elevated VAT has been associated with a lower OS rate [10, 11, 12]. Similarly, the SAT index can also be used to predict the outcomes of cancers, such as head and neck cancer [13], BC [14], and prostate cancer [15]. Visceral to subcutaneous fat area ratio (VSR) is an independent prognostic factor for poor prognosis in type 1 endometrial cancer and gastric cancer [16, 17]. Recent research has recognized the crucial role that interactions between adipocytes and BC cells play in the development of BC and associated bone metastases [18].

Conventionally, the third lumbar vertebra (L3) is the standard level at which computed tomography (CT)-measured body composition is assessed [19]. However, abdominal CT scans are not routinely performed on BC patients for assessment and follow-up. A recent study in solid tumors suggests strong correlations between body tissue components (including VAT and SAT) at both the intervertebral space from the 12th thoracic to first lumbar vertebrae (T12-L1) and L3 imaging levels [20]. In addition, in previous studies where L3 was not available, such as lung cancer, the level of the eleventh thoracic vertebra (T11) was used to assess body tissue components [21]. Therefore, we used measurements from the T11 anatomical level in this study.

The present study aimed to determine whether subcutaneous fat could be used as a prognostic parameter for bone metastasis in BC patients.

2. Materials and methods

2.1 Patient characteristics and specimens

The medical records of 517 consecutive female BC patients at Harbin Medical University Cancer Hospital from January 1 ${}^{\text{st}}$ , 2014, to December 31 ${}^{\text{st}}$ , 2015, were retrospectively reviewed. The following were the eligibility requirements for enrollment: (1) age $>$ 18 years; (2) histology-diagnosed BC after complete surgical resection; and (3) without treatment before diagnosis. The following was listed as an exclusion criterion: (1) received preoperative treatment; (2) had a previous diagnosis of cancer or metastatic lesions at diagnosis; (3) had incomplete clinical information and a loss of follow-up.

The pathologic stage was classified based on the 8th edition of the American Joint Committee on Cancer (AJCC). All BC patients were screened for metastases before operation. No patients had metastases at the time of confirmation of BC. Bone metastases are first defined by emission computed tomography (ECT) scans and then confirmed by X-ray, CT, and magnetic resonance imaging (MRI) scans. Bone metastasis-free survival was calculated from the date of surgery to the date of bone metastasis, or the last follow-up. Patients were followed up every three months for the first two years following surgery and every six months for the following three to five years. The last follow-up was on December 31, 2020. The median follow-up time was 65 months.

The data included characteristics such as age, menstrual status, tumor size, histopathological type, lymph node metastasis, lymphovascular invasion, proliferation index (Ki-67) expression, estrogen receptor (ER) status, progesterone receptor (PR) status, clinical stage, molecular classification, and postoperative treatment. Venous blood samples were collected one week before the operation. Immunohistochemistry (IHC) and/or fluorescence in situ hybridization (FISH) were used to detect the ER, PR, human epidermal growth factor receptor-2 (HER-2), and Ki-67 status. HER-2 positivity was defined as IHC 3+ or FISH positive.

This study was approved by the Ethics Committee of the Harbin Medical University Cancer Hospital (KY2022-10). Informed consent was obtained from each participant.

2.2 Adipose tissue imaging analysis

The fat tissue was evaluated using preoperative chest CT scans. The areas of visceral and subcutaneous fat tissues were quantified in a single slice using Image J software version 1.53a (Wayne Rasband, National Institutes of Health, USA). Hounsfield Units (HU) thresholds between -190 and -30 were used to determine subcutaneous fat, while radiodensities between -150 and -50 HU were used to calculate visceral fat [22, 23]. Visceral fat index (VFI) and subcutaneous fat index (SFI) were calculated by dividing the corresponding fat area by the square of the patient’s height [12, 19]. The VSR was calculated by dividing the visceral fat tissue area (VFA) by the subcutaneous fat tissue area (SFA). Two independent radiologists who were blind to the patients’ information performed the CT scans within two weeks before surgery. The two measures’ mean was applied. The intra-observer coefficient of variation was less than 1.2%.

2.3 Statistical analysis

All statistical analyses were carried out using SPSS 25.0 software. The data were presented as percentages, or mean $\pm$ standard deviation. The comparison of continuous variables with normally distributed data was evaluated with the two-tailed Student $t$ -test, whereas categorical variables were compared with the Chi-square test. The incidence of bone metastases was estimated by Kaplan-Meier curves, and the analysis was carried out using the log-rank test. The predictive factors for bone metastases were analyzed using Cox univariate and multivariate proportional hazards regression models. All variables with a $P$ -value of $<$ 0.10 were included in the multivariate analysis. $P$ -values $<$ 0.05 were regarded as statistically significant.

3. Results

Table 1
Baseline characteristics of BC patients according to bone metastasis status

Variables	Without bone metastasis	Bone metastasis	$P$ value
N	442	75
Age (years)	50.2 $\pm$ 8.5	52.3 $\pm$ 9.5	0.048
BMI (kg/m ${}^{2})$	24.1 $\pm$ 3.1	24.3 $\pm$ 3.2	0.659
SFA (cm ${}^{2})$	162.1 $\pm$ 60.4	144.5 $\pm$ 75.3	0.025
SFI (cm ${}^{2}$ /m ${}^{2})$	49.2 $\pm$ 26.0	21.3 $\pm$ 5.7	$<$ 0.001
VFA (cm ${}^{2})$	72.0 $\pm$ 46.8	48.5 $\pm$ 37.4	$<$ 0.001
VFI (cm ${}^{2}$ /m ${}^{2})$	27.9 $\pm$ 18.3	19.2 $\pm$ 15.4	$<$ 0.001
VSR	0.5 $\pm$ 0.3	0.4 $\pm$ 0.2	0.020
Menopausal status			0.101
Pre	222 (50.2)	30 (40.0)
Post	220 (49.8)	45 (60.0)
Histologic type			$<$ 0.001
Ductal	388 (87.8)	31 (41.3)
Lobular	39 (8.8)	42 (56.0)
Others	15 (3.4)	2 (2.7)
Multifocal disease			$<$ 0.001
Yes	61 (13.8)	24 (32.0)
No	381 (86.2)	51 (68.0)
Tumor size (cm)			$<$ 0.001
$\geqslant$ 2.5	152 (34.4)	45 (60.0)
$<$ 2.5	290 (65.6)	30 (40.0)
Lymph node status			0.025
Negative	215 (48.6)	26 (34.7)
Positive	227 (51.4)	49 (65.3)
Clinical stage			0.018
I–II	378 (85.5)	56 (74.7)
III	64 (14.5)	19 (25.3)
Ki-67 (%)			0.196
$<$ 20%	132 (29.9)	28 (37.3)
$\geqslant$ 20%	310 (70.1)	47 (62.7)
ER			0.219
Positive	314 (71.0)	48 (64.0)
Negative	128 (29.0)	27 (36.0)
PR			0.013
Positive	290 (65.6)	38 (50.7)
Negative	152 (34.4)	37 (49.3)
HER2 status			0.101
Positive	129 (29.2)	15 (20.0)
Negative	313 (70.8)	60 (80.0)
Molecular subtype			0.764
Luminal-A	149 (33.7)	26 (34.7)
Luminal-B	123 (27.8)	17 (22.7)
HER2-enriched	87 (19.7)	15 (20.0)
TNBC	83 (19.3)	17 (22.7)
Adjuvant radiotherapy			0.411
Yes	114 (25.8)	16 (21.3)
No	328 (74.2)	59 (78.7)
Adjuvant hormonal therapy			$<$ 0.001
Yes	222 (50.2)	21 (28.0)
No	220 (49.8)	54 (72.0)
Adjuvant chemotherapy			0.323
Yes	414 (93.7)	73 (97.3)
No	28 (6.3)	2 (2.7)

BC, breast cancer; BMI, body mass index; SFA, subcutaneous fat tissue area; SFI, subcutaneous fat index; VFA, visceral fat tissue area; VFI, visceral fat index; VSR, visceral fat tissue to subcutaneous fat tissue ratio; Ki-67, proliferation index; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer.

3.1 Baseline characteristics according to bone metastasis status

This study enrolled 517 female patients with BC. The average age was 50.5 $\pm$ 8.7 years (range: 26 to 76 years). 241 participants had no lymph node metastasis, whereas 276 people presented. 434 patients were found to be at stages I and II, while 83 were at stage III. 75 patients exhibited bone metastasis in 5 years. The baseline clinicopathological characteristics according to bone metastasis status are displayed in Table 1. Age, SFA, SFI, VFA, VFI, VSR, histologic type, multifocal disease, tumor size, lymph node status, clinical stage, PR, and adjuvant hormonal therapy were significantly associated with bone metastasis. There were no significant associations between other clinical features and bone metastases.

Table 2
Baseline clinico-pathological parameters of BC patients according to SFI levels before propensity matching

Variables	$>$ 30.87 cm ${}^{2}$ /m ${}^{2}$	$\leqslant$ 30.87 cm ${}^{2}$ /m ${}^{2}$	$P$ value
N	311	206
Age (years)	50.0 $\pm$ 8.1	51.2 $\pm$ 9.4	0.137
BMI (kg/m ${}^{2})$	24.2 $\pm$ 3.1	24.2 $\pm$ 3.5	0.058
Menopausal status			0.061
Pre	162 (52.1)	90 (43.7)
Post	149 (47.9)	116 (56.3)
Histologic type			$<$ 0.001
Ductal	273 (87.8)	146 (70.9)
Lobular	30 (9.6)	51 (24.8)
Others	8 (2.6)	9 (4.4)
Multifocal disease			$<$ 0.001
Yes	14 (4.5)	71 (34.5)
No	297 (95.5)	135 (65.5)
Tumor size (cm)			0.926
$\geqslant$ 2.5	118 (37.9)	79 (38.3)
$<$ 2.5	193 (62.1)	127 (61.7)
Lymph node status			0.019
Negative	158 (50.8)	83 (40.3)
Positive	153 (49.2)	123 (59.7)
Clinical stage			0.637
I–II	263 (84.6)	171 (83.0)
III	48 (15.4)	35 (17.0)
Ki-67 (%)			$<$ 0.001
$<$ 20%	73 (23.5)	87 (42.2)
$\geqslant$ 20%	238 (76.5)	119 (57.8)
ER			0.106
Positive	226 (72.7)	136 (66.0)
Negative	85 (27.3)	70 (34.0)
PR			0.011
Positive	211 (67.8)	117 (56.8)
Negative	100 (32.2)	89 (43.2)
HER2 status			0.038
Positive	97 (31.2)	47 (22.8)
Negative	214 (68.8)	159 (77.2)
Molecular subtype			$<$ 0.001
Luminal-A	124 (39.9)	51 (24.8)
Luminal-B	67 (21.5)	73 (35.4)
HER2-enriched	64 (20.6)	38 (18.4)
TNBC	56 (18.0)	44 (21.4)
Adjuvant radiotherapy			0.159
Yes	85 (27.3)	45 (21.8)
No	226 (72.7)	161 (78.2)
Adjuvant hormonal therapy			0.008
Yes	161 (51.8)	82 (39.8)
No	150 (48.2)	124 (60.2)
Adjuvant chemotherapy			0.714
Yes	292 (93.9)	195 (94.7)
No	19 (6.1)	11 (5.3)

BC, breast cancer; BMI, body mass index; SFI, subcutaneous fat index; Ki-67, proliferation index; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TNBC, triple-negative breast cancer.

3.2 Predictive ability of SFI for bone metastasis

We performed a receiver operating characteristic (ROC) analysis to evaluate the predictive ability of SFI for bone metastasis. The optimal cutoff value of SFI was 30.87 [AUC $=$ 0.844, 95% CI 0.810–0.874, $P<$ 0.001], with a corresponding sensitivity of 0.960 and specificity of 0.696 (Fig. 1). The patients were divided into two groups based on the optimal cut-off value: high SFI ( $>$ 30.87 cm ${}^{2}$ /m ${}^{2})$ ( $n=$ 311) and low SFI ( $\leqslant$ 30.87 cm ${}^{2}$ /m ${}^{2})$ ( $n=$ 206). We investigated the relationships between these clinicopathological variables and SFI. A statistically significant difference was observed in the status of histologic type, multifocal disease, lymph node status, Ki-67, PR status, HER-2 status, molecular subtype, and adjuvant hormonal therapy between the two groups divided by SFI threshold. However, no significant difference was observed in other clinicopathological variables (Table 2).

Table 3
Baseline clinico-pathological parameters of BC patients according to SFI levels after propensity matching

Variables	$>$ 30.87 cm ${}^{2}$ /m ${}^{2}$	$\leqslant$ 30.87 cm ${}^{2}$ /m ${}^{2}$	$P$ value
N	138	138
Age (years)	51.1 $\pm$ 8.8	51.6 $\pm$ 9.2	0.603
BMI (kg/m ${}^{2})$	24.1 $\pm$ 3.5	24.2 $\pm$ 3.2	0.911
Menopausal status			0.274
Pre	64 (46.4)	55 (39.9)
Post	74 (53.6)	83 (60.1)
Histologic type			0.401
Ductal	107 (77.5)	106 (76.8)
Lobular	23 (16.7)	28 (20.3)
Others	8 (5.8)	4 (2.9)
Multifocal disease			0.680
Yes	14 (10.1)	12 (8.7)
No	124 (89.9)	126 (91.3)
Tumor size (cm)			0.804
$\geqslant$ 2.5	53 (38.4)	51 (37.0)
$<$ 2.5	85 (61.6)	87 (63.0)
Lymph node status			0.904
Negative	64 (46.4)	63 (45.7)
Positive	74 (53.6)	75 (54.3)
Clinical stage			0.415
I–II	113 (81.9)	118 (85.5)
III	25 (18.1)	20 (14.5)
Ki-67 (%)			0.373
$<$ 20%	43 (31.2)	50 (36.2)
$\geqslant$ 20%	95 (68.8)	88 (63.8)
ER			0.797
Positive	94 (68.1)	92 (66.7)
Negative	44 (31.9)	46 (33.3)
PR			1.000
Positive	84 (60.9)	84 (60.9)
Negative	54 (39.1)	54 (39.1)
HER2 status			0.219
Positive	41 (29.7)	32 (23.2)
Negative	97 (70.3)	106 (76.8)
Molecular subtype			0.264
Luminal-A	47 (34.1)	36 (26.1)
Luminal-B	34 (24.6)	46 (33.3)
HER2-enriched	22 (15.9)	26 (18.8)
TNBC	35 (25.4)	30 (21.7)
Adjuvant radiotherapy			0.775
Yes	31 (22.5)	33 (23.9)
No	107 (77.5)	105 (76.1)
Adjuvant hormonal therapy			0.903
Yes	57 (41.3)	58 (42.0)
No	81 (58.7)	80 (58.0)
Adjuvant chemotherapy			0.812
Yes	128 (92.8)	129 (93.5)
No	10 (7.2)	9 (6.5)

3.3 Predictive value of SFI for bone metastasis

In order to adjust for the heterogeneity between the low SFI and high SFI groups, propensity score matching (PSM) was used. After 1:1 PSM, 67% of the low SFI group ( $n=$ 138) and 44% of the high SFI group ( $n=$ 138) were analyzed. The baseline characteristics were balanced in the matched groups (Table 3). After a follow-up of 65 months (range, 62–69 months), 75 (14.5%) patients had bone metastasis. Patients with lower SFI were more likely to develop bone metastasis than those with higher SFI (35.0% vs. 1.0%, respectively; $p<$ 0.001). Kaplan-Meier curves of SFI identified a significant difference between SFI $>$ 30.87 and $\leqslant$ 30.87 in BC (Fig. 2). After PSM, the lower SFI group still presented with a higher incidence of bone metastasis (11.4% vs. 2.9%, respectively; $p<$ 0.001) (Fig. 3).

Figure 1.

An optimized cut-off value was determined for SFI using ROC curve analysis. The ROC curve identified the optimal cutoff value of SFI as 30.87 cm ${}^{2}$ /m ${}^{2}$ , with a sensitivity of 0.960 and a specificity of 0.696. SFI, subcutaneous fat index; ROC curve, receiver operating characteristic curve.

Table 4

The predictors of bone metastases in patients with breast cancer

Variables	Univariate analysis		Multivariate analysis
	HR (95% CI)	$P$ -value	HR (95% CI)	$P$ -value
Before propensity matching
Age (years)	1.027 (1.001–1.054)	0.041	1.038 (1.003–1.074)	0.033
Histologic type
Ductal	1		1
Lobular	6.109 (1.478–25.254)	0.012	5.126 (1.205–21.812)	0.027
Others	0.632 (0.151–2.639)	0.529	0.951 (0.119–7.623)	0.962
Multifocal disease
(Positive vs Negative)	2.552 (1.570–4.146)	$<$ 0.001	0.645 (0.352–1.182)	0.156
Tumor size (cm)
( $\geqslant$ 2.5 vs $<$ 2.5)	2.584 (1.628–4.102)	$<$ 0.001	1.739 (0.880–3.438)	0.111
Lymph node status
(Positive vs Negative)	1.725 (1.072–2.775)	0.025	0.961 (0.519–1.781)	0.900
Clinical stage
(III vs I–II)	1.884 (1.120–3.171)	0.017	1.051 (0.521–2.120)	0.889
PR
(Positive vs Negative)	0.566 (0.360–0.891)	0.014	2.822 (1.370–5.814)	0.005
HER2
(Positive vs Negative)	0.508 (0.288–0.894)	0.019	0.575 (0.270–1.227)	0.152
Adjuvant hormonal therapy
(Yes vs No)	0.407 (0.246–0.674)	$<$ 0.001	0.285 (0.134–0.605)	0.001
SFI (cm ${}^{2}$ /m ${}^{2})$	0.908 (0.885–0.932)	$<$ 0.001	0.904 (0.868–0.941)	$<$ 0.001
VFI (cm ${}^{2}$ /m ${}^{2})$	0.968 (0.952–0.985)	$<$ 0.001	0.986 (0.943–1.032)	0.551
VSR	0.177 (0.062–0.506)	0.001	0.161 (0.015–1.711)	0.130
After propensity matching
Age (years)	1.012 (0.980–1.044)	0.473
Histologic type
Ductal	1		1
Lobular	5.388 (3.018–9.621)	$<$ 0.001	6.424 (3.302–12.498)	$<$ 0.001
Others	1.523 (0.359–6.458)	0.569	1.843 (0.421–8.059)	0.417
Multifocal disease
(Positive vs Negative)	0.817 (0.293–2.273)	0.698
Tumor size (cm)
( $\geqslant$ 2.5 vs $<$ 2.5)	2.499 (1.407–4.437)	0.002	1.668 (0.881–3.157)	0.116
Lymph node status
(Positive vs Negative)	1.365 (0.765–2.435)	0.292
Clinical stage
(III vs I–II)	1.624 (0.829–3.185)	0.158
PR
(Positive vs Negative)	0.870 (0.490–1.544)	0.634
HER2
(Positive vs Negative)	0.949 (0.494–1.824)	0.875
Adjuvant hormonal therapy
(Yes vs No)	0.664 (0.364–1.210)	0.181
SFI (cm ${}^{2}$ /m ${}^{2})$	0.923 (0.894–0.953)	$<$ 0.001	0.916 (0.885–0.948)	$<$ 0.001
VFI (cm ${}^{2}$ /m ${}^{2})$	0.965 (0.944–0.986)	0.001	0.988 (0.961–1.016)	0.398
VSR	0.160 (0.043–0.603)	0.007	0.238 (0.036–1.553)	0.134

BC, breast cancer; BMI, body mass index; SFI, subcutaneous fat index; VFI, visceral fat index; VSR, visceral fat tissue to subcutaneous fat tissue ratio; Ki-67, proliferation index; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; HR, hazard ratio; CI, confidence interval.

The possible predictive variables for bone metastases are examined using Cox univariate and multivariate proportional regression models. Before propensity matching, age, SFI, VFI, VSR, histologic type, multifocal disease, tumor size, lymph node status, clinical stage, PR, HER-2 status, and adjuvant hormonal therapy were significantly associated with bone metastasis in univariate analysis. SFI, histologic type, and adjuvant hormonal therapy were independently predictive factors of bone metastasis in multivariate analysis (Table 4). After PSM, the univariate analysis indicated that SFI, VFI, VSR, histologic type, and tumor size were related to bone metastasis. The multivariate analysis revealed that SFI and histologic type were independently predictive factors of bone metastasis (Table 4). The predictive power of SFI in the molecular subtype of BC is provided in supplementary figures (Figs S1–4).

4. Discussion

In this paper, we calculated the impact of preoperative SFI on bone metastasis in BC patients. By accurately evaluating SFI in chest CT scans at the T11 level, the results revealed that low SFI presented as a potential predictor for bone metastasis in BC patients. These results remind clinicians to pay close attention to adipose tissue in BC patients.

To our knowledge, no researcher has considered the relationship between SFI and postoperative bone metastases in BC patients. Consistent with our results, previous reports have shown that low SFI is independently associated with increased mortality and shorter survival in cancer patients [24]. Similarly, Dong et al. found that low subcutaneous fat was a risk factor for the prognosis of gastric cancer, including OS and disease-free survival [25]. Black et al. reported that low levels of subcutaneous fat are associated with reduced survival for patients with primary operable colorectal cancer [7]. Moreover, a recent study confirmed that lower SFI was an independent predictor of poor OS in hepatocellular carcinoma [6]. Another study suggests that high SFI and VFI were independently associated with better OS in patients with bone metastases [26]. However, Bradshaw et al. observed that elevated SAT was associated with worse survival among women with non-metastatic BC [14]. The conflicting data may be the result of different study populations, different inclusion criteria, and different indicators of SAT. Thus, further study is needed to clarify the role of SAT in BC.

To date, potential explanations for the predictive effect of SFI on bone metastases have not been clearly identified. SAT is the major source of adiponectin and leptin, which play crucial roles in regulating bone health and bone metastasis in BC [27]. Adiponectin has pro-apoptotic and anti-proliferative properties in human BC cells [28, 29]. It has been reported that adiponectin inhibits the metastatic process via suppression of the adhesion, invasion, and migration of BC cells through activation of the AMPK/S6K axis and upregulation of live kinase B1 (LKB1) [28]. Overall, the prevailing evidence indicates that low levels of adiponectin in the circulation portend a poorer BC risk and prognosis [30]. Adipose tissue mass directly correlates with leptin synthesis and plasma levels, improving lipid metabolism and insulin sensitivity [31]. Leptin helps regulate bone health by modulating bone density, growth, and adiposity. Some studies found lower serum leptin levels in premenopausal BC patients compared with healthy controls [32]. In addition, leptin was also associated with a better prognosis in patients with colorectal cancer [33]. Therefore, adipose tissue-derived adipokines and leptin affect different stages of the bone metastatic process. However, more research is needed to confirm whether and how SAT protects against bone metastasis in BC.

Figure 2.

Incidence of bone metastasis based on SFI levels before propensity score matching. The 5-year cumulative hazard of bone metastasis in patients with high SFI (blue, $n=$ 311) or low SFI (red, $n=$ 206) is plotted. Kaplan-Meier analysis demonstrated that low SFI was significantly correlated with the incidence of bone metastasis ( $P<$ 0.001, log-rank test). The median bone metastasis time was 53.4 months (95% confidence interval [CI] 51.9–54.8). SFI, subcutaneous fat index.

Figure 3.

Incidence of bone metastasis based on SFI levels after propensity score matching. After 1:1 propensity score matching, the lower SFI group presented with a higher incidence of bone metastasis (11.4% vs. 2.9%, respectively; $p<$ 0.001, log-rank test). The median bone metastasis time was 51.9 months (95% confidence interval [CI] 49.8–54.1). SFI, subcutaneous fat index.

Meanwhile, no association between body mass index (BMI) and bone metastasis was observed in our study. The relationship between BMI and postoperative outcomes in cancer patients has been a subject of controversy. A previous report revealed that the association between BMI and cancer survival is not consistent across all cancer types and stages [34]. The difference in the prognostic value of BMI may be partly explained by the fact that individuals with similar BMI may have different body compositions (e.g., more fat or muscle in one patient than in another) [35]. In other words, BMI cannot distinguish between fat and fat-free mass or different fat stores for proper assessment of body composition [36]. In this study, we used regional measures of obesity, such as SFI and VFI, which reduced the risk of this error estimate and enhanced the consistency of the findings.

Because our findings showed that patients with lower SFI are more likely to develop bone metastasis, SFI may be a novel marker for predicting bone metastasis in BC. This predictive marker can be easily acquired from CT scans collected in routine clinical care. However, this study has several limitations. First, this study was a single-institutional retrospective study and lacked a mechanistic explanation. Second, the cutoff value of SFI at the T11 slice has not been reported in the literature. Thus, multicentric studies are needed to verify our results. In addition, this was a single T11 slice analysis. SFIs in other CT slices were not assessed. Finally, there is a significant difference in BC incidence, subtype distribution, and cancer survival by ethnicity [37, 38]. The participants in our study were all Chinese. Thus, the results cannot be extrapolated to other ethnic groups.

5. Conclusions

In this paper, we found that SFI is an independent predictor of bone metastasis in BC patients. Our study indicated that SFI provides additional information on the prevention of skeletal-related events in patients with BC. Further studies are warranted to observe whether early intervention to maintain adipose tissue could improve prognosis in patients with low SFI levels.

Author contributions

Conception: WW, W-J H, and P-P L.

Data collection: SF, M-L Z, and XZ.

Interpretation or analysis of data: WW, W-J H, P-P L, SF, M-L Z, and XZ.

Preparation of the manuscript: WW, W-J H, and P-P L.

Revision for important intellectual content: R-T W and Y-X H.

Supervision: R-T W and Y-X H.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-230011.

Footnotes

Acknowledgments

The design of the study was supported by the Harbin Medical University Cancer Hospital [Grant Number JJQN2019-06].

References

DeSantis

C.E.

Miller

K.D.

Dale

Mohile

S.G.

Cohen

H.J.

Leach

C.R.

Goding Sauer

Jemal

and Siegel

R.L.

, Cancer statistics for adults aged 85 years and older, CA Cancer J Clin 69 (2019), 452–467.

Jin

and Mu

, Targeting Breast Cancer Metastasis, Breast Cancer (Auckl) 9 (2015), 23–34.

Yardley

D.A.

, Pharmacologic management of bone-related complications and bone metastases in postmenopausal women with hormone receptor-positive breast cancer, Breast Cancer (Dove Med Press) 8 (2016), 73–82.

Haider

M.T.

Ridlmaier

Smit

D.J.

and Taipaleenmaki

, Interleukins as Mediators of the Tumor Cell-Bone Cell Crosstalk during the Initiation of Breast Cancer Bone Metastasis, Int J Mol Sci 22 (2021).

Xiong

Deng

Huang

Xie

Wang

Shuang

and Wang

, Bone metastasis pattern in initial metastatic breast cancer: a population-based study, Cancer Manag Res 10 (2018), 287–295.

Kobayashi

Kawai

Nakano

Abe

Kamimura

Sakamaki

Kamimura

Tsuchiya

Takamura

Yamagiwa

and Terai

, Prognostic value of subcutaneous adipose tissue volume in hepatocellular carcinoma treated with transcatheter intra-arterial therapy, Cancer Manag Res 10 (2018), 2231–2239.

Black

Mackay

Ramsay

Hamoodi

Nanthakumaran

Park

K.G.M.

Loudon

M.A.

and Richards

C.H.

, Prognostic Value of Computed Tomography: Measured Parameters of Body Composition in Primary Operable Gastrointestinal Cancers, Ann Surg Oncol 24 (2017), 2241-2251.

Hwang

and Kim

J.B.

, Two Faces of White Adipose Tissue with Heterogeneous Adipogenic Progenitors, Diabetes Metab J 43 (2019), 752–762.

Mittal

, Subcutaneous adipose tissue & visceral adipose tissue, Indian J Med Res 149 (2019), 571-573.

10.

Iwase

Sangai

Fujimoto

Sawabe

Matsushita

Nagashima

Sato

Nakagawa

Masuda

Nagashima

and Ohtsuka

, Quality and quantity of visceral fat tissue are associated with insulin resistance and survival outcomes after chemotherapy in patients with breast cancer, Breast Cancer Res Treat 179 (2020), 435–443.

11.

Basile

Bartoletti

Polano

Bortot

Gerratana

Di Nardo

Borghi

Fanotto

Pelizzari

Lisanti

Garutti

Buriolla

Ongaro

Andreuzzi

Montico

Balestreri

Miolo

Toffoli

Aprile

Puglisi

and Buonadonna

, Prognostic role of visceral fat for overall survival in metastatic colorectal cancer: A pilot study, Clin Nutr 40 (2021), 286–294.

12.

Fujiwara

Nakagawa

Kudo

Tateishi

Taguri

Watadani

Nakagomi

Kondo

Nakatsuka

Minami

Sato

Uchino

Enooku

Kondo

Asaoka

Tanaka

Ohtomo

Shiina

and Koike

, Sarcopenia, intramuscular fat deposition, and visceral adiposity independently predict the outcomes of hepatocellular carcinoma, J Hepatol 63 (2015), 131–40.

13.

Pai

P.C.

Chuang

C.C.

Chuang

W.C.

Tsang

N.M.

Tseng

C.K.

Chen

K.H.

Yen

T.C.

Lin

C.Y.

Chang

K.P.

and Lei

K.F.

, Pretreatment subcutaneous adipose tissue predicts the outcomes of patients with head and neck cancer receiving definitive radiation and chemoradiation in Taiwan, Cancer Med 7 (2018), 1630–1641.

14.

Bradshaw

P.T.

Cespedes Feliciano

E.M.

Prado

C.M.

Alexeeff

Albers

K.B.

Chen

W.Y.

and Caan

B.J.

, Adipose Tissue Distribution and Survival Among Women with Nonmetastatic Breast Cancer, Obesity (Silver Spring) 27 (2019), 997–1004.

15.

Lopez

Newton

R.U.

Taaffe

D.R.

Singh

Buffart

L.M.

Spry

Tang

Saad

and Galvao

D.A.

, Associations of fat and muscle mass with overall survival in men with prostate cancer: a systematic review with meta-analysis, Prostate Cancer Prostatic Dis (2021).

16.

Wada

Yamaguchi

Yamakage

Inoue

Kusakabe

Abiko

Takakura

Konishi

and Satoh-Asahara

, Visceral-to-subcutaneous fat ratio is a possible prognostic factor for type 1 endometrial cancer, Int J Clin Oncol 27 (2022), 434–440.

17.

Lin

Y.C.

Lin

and Yeh

T.S.

, Visceral-to-subcutaneous fat ratio independently predicts the prognosis of locally advanced gastric cancer – highlighting the role of adiponectin receptors and PPARalpha, beta/delta, Eur J Surg Oncol 47 (2021), 3064–3073.

18.

Soni

Torvund

and Mandal

C.C.

, Molecular insights into the interplay between adiposity, breast cancer and bone metastasis, Clin Exp Metastasis 38 (2021), 119–138.

19.

Hamaguchi

Kaido

Okumura

Kobayashi

Hammad

Tamai

Inagaki

and Uemoto

, Proposal for new diagnostic criteria for low skeletal muscle mass based on computed tomography imaging in Asian adults, Nutrition 32 (2016), 1200–5.

20.

Joffe

Shen

Shadid

Jin

and Ladas

E.J.

, Skeletal muscle and adipose tissue changes in the first phase of treatment of pediatric solid tumors, Cancer Med 10 (2021), 15–22.

21.

Chen

Yang

Jiang

Zhu

Liu

Pang

Zeng

Chen

Gong

Yin

and Chen

B.T.

, CT-Based Sarcopenic Nomogram for Predicting Progressive Disease in Advanced Non-Small-Cell Lung Cancer, Front Oncol 11 (2021), 643941.

22.

Martin

Birdsell

Macdonald

Reiman

Clandinin

M.T.

McCargar

L.J.

Murphy

Ghosh

Sawyer

M.B.

and Baracos

V.E.

, Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index, J Clin Oncol 31 (2013), 1539–47.

23.

Prado

C.M.

Lieffers

J.R.

McCargar

L.J.

Reiman

Sawyer

M.B.

Martin

and Baracos

V.E.

, Prevalence and clinical implications of sarcopenic obesity in patients with solid tumours of the respiratory and gastrointestinal tracts: a population-based study, Lancet Oncol 9 (2008), 629–35.

24.

Ebadi

Martin

Ghosh

Field

C.J.

Lehner

Baracos

V.E.

and Mazurak

V.C.

, Subcutaneous adiposity is an independent predictor of mortality in cancer patients, Br J Cancer 117 (2017), 148–155.

25.

Dong

Q.T.

Cai

H.Y.

Zhang

Zou

H.B.

Dong

W.X.

Wang

W.B.

Song

H.N.

Luo

Chen

X.L.

and Huang

D.D.

, Influence of body composition, muscle strength, and physical performance on the postoperative complications and survival after radical gastrectomy for gastric cancer: A comprehensive analysis from a large-scale prospective study, Clin Nutr 40 (2021), 3360–3369.

26.

Chuang

W.C.

Tsang

N.M.

Chuang

C.C.

Chang

K.P.

Pai

P.C.

Chen

K.H.

Chou

W.C.

Tai

S.F.

Liu

S.C.

and Lei

K.F.

, Association of subcutaneous and visceral adipose tissue with overall survival in Taiwanese patients with bone metastases – results from a retrospective analysis of consecutively collected data, PLoS One 15 (2020), e0228360.

27.

Maroni

, Leptin, Adiponectin, and Sam68 in Bone Metastasis from Breast Cancer, Int J Mol Sci 21 (2020).

28.

Saxena

N.K.

and Sharma

, Metastasis suppression by adiponectin: LKB1 rises up to the challenge, Cell Adh Migr 4 (2010), 358–62.

29.

Dalamaga

Diakopoulos

K.N.

and Mantzoros

C.S.

, The role of adiponectin in cancer: a review of current evidence, Endocr Rev 33 (2012), 547–94.

30.

Libby

E.F.

Frost

A.R.

Demark-Wahnefried

and Hurst

D.R.

, Linking adiponectin and autophagy in the regulation of breast cancer metastasis, J Mol Med (Berl) 92 (2014), 1015–23.

31.

Avram

A.S.

Avram

M.M.

and James

W.D.

, Subcutaneous fat in normal and diseased states: 2. Anatomy and physiology of white and brown adipose tissue, J Am Acad Dermatol 53 (2005), 671–83.

32.

Mantzoros

C.S.

Bolhke

Moschos

and Cramer

D.W.

, Leptin in relation to carcinoma in situ of the breast: a study of pre-menopausal cases and controls, Int J Cancer 80 (1999), 523–6.

33.

Paik

S.S.

Jang

S.M.

Jang

K.S.

Lee

K.H.

Choi

and Jang

S.J.

, Leptin expression correlates with favorable clinicopathologic phenotype and better prognosis in colorectal adenocarcinoma, Ann Surg Oncol 16 (2009), 297-303.

34.

Greenlee

Unger

J.M.

LeBlanc

Ramsey

and Hershman

D.L.

, Association between Body Mass Index and Cancer Survival in a Pooled Analysis of 22 Clinical Trials, Cancer Epidemiol Biomarkers Prev 26 (2017), 21–29.

35.

Garcia-Estevez

Cortes

Perez

Calvo

Gallegos

and Moreno-Bueno

, Obesity and Breast Cancer: A Paradoxical and Controversial Relationship Influenced by Menopausal Status, Front Oncol 11 (2021), 705911.

36.

Lennon

Sperrin

Badrick

and Renehan

A.G.

, The Obesity Paradox in Cancer: a Review, Curr Oncol Rep 18 (2016), 56.

37.

Iqbal

Ginsburg

Rochon

P.A.

Sun

and Narod

S.A.

, Differences in breast cancer stage at diagnosis and cancer-specific survival by race and ethnicity in the United States, JAMA 313 (2015), 165–73.

38.

Albain

K.S.

Gray

R.J.

Makower

D.F.

Faghih

Hayes

D.F.

Geyer

C.E.

Dees

E.C.

Goetz

M.P.

Olson

J.A.

Lively

Badve

S.S.

Saphner

T.J.

Wagner

L.I.

Whelan

T.J.

Ellis

M.J.

Wood

W.C.

Keane

M.M.

Gomez

H.L.

Reddy

P.S.

Goggins

T.F.

Mayer

I.A.

Brufsky

A.M.

Toppmeyer

D.L.

Kaklamani

V.G.

Berenberg

J.L.

Abrams

Sledge

G.W.

and Sparano

J.A.

, Race, Ethnicity, and Clinical Outcomes in Hormone Receptor-Positive, HER2-Negative, Node-Negative Breast Cancer in the Randomized TAILORx Trial, J Natl Cancer Inst 113 (2021), 390–399.

Lower subcutaneous fat index predicts bone metastasis in breast cancer

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Patient characteristics and specimens

2.2 Adipose tissue imaging analysis

2.3 Statistical analysis

3. Results

Table 1 Baseline characteristics of BC patients according to bone metastasis status

Table 2 Baseline clinico-pathological parameters of BC patients according to SFI levels before propensity matching

Table 3 Baseline clinico-pathological parameters of BC patients according to SFI levels after propensity matching

Author contributions

Supplementary data

Footnotes

Acknowledgments

References

Table 1
Baseline characteristics of BC patients according to bone metastasis status

Table 2
Baseline clinico-pathological parameters of BC patients according to SFI levels before propensity matching

Table 3
Baseline clinico-pathological parameters of BC patients according to SFI levels after propensity matching