Breast cancer is the most commonly diagnosed cancer, and the fourth leading cause of cancer-related death in women worldwide (1). Although many advancements have been made in treatment, some breast cancer patients continue to experience recurrence and metastasis. Therefore, it is of great significance to find new biomarkers for early prediction of metastasis.

Body composition including muscle mass and adipose tissue has been increasingly recognized as a key factor to predict long-term prognosis in various types of cancers such as breast cancer (2), colon cancer (3), and hepatocellular carcinoma (4). In the past few decades, body composition has mostly focused on evaluation of adipose tissue, because obesity and its clinical implications have been extensively investigated (5). Currently, muscle mass has become a focal point for clinical research (6). Sarcopenia refers to a generalized and progressive loss of skeletal muscle mass and function (7). Previous studies have proven that sarcopenia is related to adverse clinical outcomes of cancer, including disability, poor response to chemotherapy, post-operative complications, and reduced overall survival (OS) (2). Recent research found that sarcopenia was associated with survival in breast cancer (6). Cancer cachexia has long been recognized as a consequence of malignancy and occurs in approximately 40% of breast cancer patients, and is recognized as a direct cause of reduced quality of life (8). Differential diagnosis of cachexia and sarcopenia might be difficult in clinical practice because they share certain characteristics and overlap in some of the criteria (9). Cachexia is defined by international consensus as weight loss greater than 5%, or weight loss greater than 2% in individuals already showing depletion (10). Thus, accurately measuring muscle loss associated with sarcopenia may dramatically improve early detection of cachexia.

Computed tomography (CT) images, the preferred examination method, could evaluate skeletal muscle mass, adipose tissue amount and distribution, and tissue-specific radiodensity values (11). The cross-sectional area of skeletal muscle at the level of the third lumbar vertebra (L3) was directly correlated with whole-body skeletal muscle and fat mass in cancer patients (12). Reduced skeletal muscle radiodensity (SMD), referred to as myosteatosis, reflects intramuscular fat infiltration, and directly affects survival (13). Skeletal muscle index (SMI), a marker for muscle mass, and mean muscle attenuation (MA), a marker for muscle quality, are the independent negative predictors for OS in cancer patients. Some postmenopausal women with higher body fat levels have a higher risk of invasive breast cancer (5). However, recent studies revealed body composition plays different roles at different clinical stages of breast cancer (14). In early breast cancer, higher amount of visceral adipose tissue (VAT) and the lower quality of VAT represented by the Hounsfield unit (HU) were associated with shorter distant disease-free survival (15).

Conventionally, CT-measured skeletal muscle mass is usually evaluated at the level of the L3 (16). However, in patients with respiratory or breast disease, abdominal CT scans are not routinely performed for evaluation and follow-up. A chest CT scan is a common examination to evaluate breast cancer. Recent studies have observed that pectoralis muscle area (PMA), measured from CT scans of the chest, is associated with prognosis in lung cancer (17) and diffuse large B-cell lymphoma (18). Currently, the measurable indicators of muscle mass at the T4 level include skeletal muscle area (SMA) and PMA (19), but SMA in the upper thorax varies depending on upper limb positions (i.e., arms above the head versus arms by the side) (20). Therefore, PMA on chest CT imaging was used to assess the muscle area in our study.

The purpose of this research was to determine whether preoperative PMA on chest CT imaging could predict distant metastasis and OS in breast cancer patients.

This study involved 493 breast cancer patients. Among them, the mean age was 50.4 ± 8.7 years (range 26-79). There were 238 (48.3%) pre-menopausal and 255 (51.7%) post-menopausal patients. Most patients were diagnosed with tumor grade I/II disease, and the most common histological type was ductal cancer. The numbers of patients with Luminal-A, Luminal-B, HER2+, and TNBC were 151, 139, 103, and 100, respectively.

The clinical and pathological characteristics of breast cancer patients according to metastasis status are presented in Table 1 . The patients with metastases were older and had lower PMA/T4, PMI/T4, SMA/T11, and SMI/T11 levels compared with the patients without metastases. There were significant differences in menopausal status, tumor size, Ki67 expression, ER status, PR status, molecular subtype, and adjuvant hormonal therapy between the two groups. However, BMI, histopathological type, multifocal disease, lymphovascular invasion, HER2 status, clinical stage, adjuvant radiotherapy, and adjuvant chemotherapy had no difference between the two groups.

Next, we identified the optimal cut-off value of PMI/T4 for distant metastasis by ROC under the curve analysis. According to the ROC curve analysis, the optimal cut-off value for PMI/T4 was 19.5 (cm²/m²), with sensitivity of 95.7%, and specificity of 42.3% (AUC = 0.715, 95% CI: 0.673-0.755, p < 0.0001) ( Figure 1 ). The patients were divided into two risk stratification groups. If patients had low PMI/T4, they were classified into the high-risk group (n = 302). Conversely, if patients had high PMI/T4, they were classified into the low-risk group (n = 191). The relationships between PMI/T4 levels and clinical characteristics are summarized in Table 2 . Our results showed that PMI/T4 was associated with BMI, lymphovascular invasion, Ki67 expression, multifocal disease, molecular subtype, and adjuvant chemotherapy. However, the associations between age, menopausal status, histopathological type, tumor size, lymph node status, clinical stage, ER status, PR status, HER2 status, adjuvant radiotherapy, adjuvant hormonal therapy, and PMI/T4 were not statistically significant.

The median follow-up was 65 months (interquartile range, 62-69 months), with 5-year DMFS and OS being 90.7% and 94.5%, respectively. Overall, 46 (9.3%) cases had distant metastasis, and 27 (5.5%) patients died during follow-up. The Kaplan-Meier analysis was applied by log-rank test to assess the prediction capacity of PMI/T4. Patients with high PMI/T4 had better DMFS and OS than those with low PMI/T4 (log-rank P < 0.001) ( Figures 2 , 3 ).

Univariate and multivariate Cox regression analyses were performed in Tables 3 , 4 . Univariate analyses revealed that age of diagnosis, menopausal status, Ki67 expression, tumor size, molecular subtype, PMI/T4, and adjuvant hormonal therapy had a significant correlation with DMFS. Ki67 expression, lymph node status, molecular subtype, PMI/T4, SMI/T11, and adjuvant hormonal therapy were significantly associated with OS. All the parameters in the univariate analysis with a p-value less than 0.10 were then incorporated into a multivariate Cox proportional hazards regression model. Multivariate Cox analysis confirmed that Ki67 expression and PMI/T4 were the independent prognostic variables for DMFS. Lymph node status and PMI/T4 were the independent prognostic factors for OS.

This study showed that PMI/T4 was significantly associated with lymphovascular invasion, Ki67 expression, multifocal disease, and molecular subtype. In addition, PMI/T4 was an independent prognostic factor for DMFS and OS.

Sarcopenia, a useful parameter for reflecting body composition, was associated with the survival of breast cancer patients (6). Our results are consistent with those of previous studies. A previous report demonstrated that sarcopenia (defined as SMI/L3<41 cm²/m²) was an independent prognostic factor for disease-free survival (DFS) and OS in early breast cancer (25). Furthermore, another large study found that non-metastatic breast cancer patients with sarcopenia (defined as SMI/L3<40 cm²/m²) had higher overall mortality compared with those without (26). However, those studies just highlighted that sarcopenia is significantly associated with DFS and OS in patients with breast cancer (25, 26). To our knowledge, our study is the first to explore whether PMI can effectively predict DMFS in BC patients. DMFS is of vital clinical significance since distant metastasis is associated with a high rate of mortality and a difference in prognoses and responses to therapy in breast cancer patients (27). In addition, a close relationship between sarcopenia and metastasis has been verified in various types of cancers, including lung cancer (28), nasopharyngeal carcinoma (29), and colorectal cancer (30).

The mechanisms by which sarcopenia confers an increased risk of distant metastasis and mortality are still unclear, but the following reasons can be hypothesized. First, cancer cells utilize glucose and glutamine as a carbon skeleton and produce energy through lactate fermentation (31). It has been reported that high mobility group box-1 protein (HMGB1) secreted during tumorigenesis induces the degradation of host muscle tissues to supply glutamine to cancer cells as an energy source (32). Thus, expenditure of glutamine would lead to loss of muscle mass (33). Second, some studies found that myokines secreted from muscle cells could inhibit cancer cell growth and migration, and induce cancer cell death (34, 35). Therefore, we speculate that cancer cells utilize skeletal muscle as an energy repository, causing sarcopenia. Moreover, loss of muscle mass can lead to an impaired myokine response and an increased risk of distant metastasis and mortality.

Interestingly, our study revealed that PMI/T4, not SMI/T11, was an independent prognostic factor for DMFS and OS in breast cancer patients. Only two studies have been published on the association between PMA and breast cancer. One study demonstrated a close correlation between PMA on MRI and the psoas muscle area on CT scan in breast cancer patients (36). Another study showed that PMA on MRI is reduced after neoadjuvant chemotherapy (37). However, there is no study investigating the clinical implications of the PMI/T4 in CT images on distant metastasis and death outcome for breast cancer patients. A previous study found that low PMI/T4 was a risk factor for worse OS in lung cancer patients (17). Another study also detected that low PMI/T4 was strongly associated with worse PFS and OS in patients with diffuse large B-cell lymphoma (18). Our results confirmed that PMI/T4 may serve as a marker of adverse prognosis in breast cancer. However, the results of SMI/T11 in different cancer types were not consistent. A previous report detected that visceral fat density at T11 (not SMI/T11) could objectively predict the risk of hepatic decompensation and survival in patients with hepatocellular carcinoma undergoing transarterial chemoembolization (38). In contrast, one study found that patients with progressive disease had significantly lower SMI/T11 levels compared with stable disease in lung cancer (39). Similarly, another study reported that muscle mass at T8 level (not T10 level) was the best predictor of survival, and muscle mass at T12 showed no association with survival in lung cancer (40). In line with prior studies, our results revealed that PMI/T4 may be more helpful in survival prediction than SMI/T11 in patients with breast cancer. The muscle index on the different thoracic vertebral levels exhibited different predictive capacities for survival, due in part to the fact that pectoralis muscles at T4 support upper extremity and respiratory system function (40).

However, there are some limitations in this study. First, the patients in our study came from a single center, which limits the generalizability of the results. Second, similar to other retrospective findings, we were unable to determine the causal link between parameters. Third, the cut-off value of the PMI at the level of T4 was not a confirmed optimal cut-off value. In addition, measurement of muscle function was not included in our study. Finally, a few patients developed metastatic disease, thus partly limiting the analysis and conclusion. Further studies are needed.

In conclusion, PMI at the T4 level may serve as a marker of adverse prognosis in breast cancer.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

W-jH, WW, and M-lZ worked together to conceive and design the experiment. Q-Cj, XZ, SF, Y-xL, and J-rY completed the data collection together. S-dM and M-lZ completed the Data analysis together. W-jH wrote and edited the manuscript alone. R-tW and W-jH responsible for the quality control of the study and the review of the manuscript. All authors read and approved the final manuscript.

This work was supported by grants from the National Cancer Center climbing Foundation (NCC201908B09).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.