Effect of Lipid Levels on Tumor-Infiltrating Lymphocytes and Prognosis in Patients with Triple-Negative Breast Cancer Open Access

Breast cancer (BC) is the most common cancer and has the second highest mortality rate among female cancer patients [1]. Triple-negative breast cancer (TNBC) is a histological subtype of BC with poor prognosis, accounting for about 10–20% of all BC cases [1]. Due to the lack of common antigens, including estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 (HER2), TNBC is resistant to endocrine and HER2 targeted therapy. Additionally, TNBC has a high recurrence rate and a poor survival outcome, and about 40% of patients cannot survive for 5 years. TNBC is also pathologically characterized by an abundance of tumor-infiltrating lymphocytes (TILs) [2, 3].

It has been reported that cholesterols and their metabolites have a role in immunoregulation inside the tumor microenvironment (TME), which can influence tumor proliferation, malignancy, recurrence, and metastasis via numerous mechanisms [4]. Lipid levels contribute to the body’s immunity through influencing the immunological infiltration in TME [5]. It is considered that the clearance of small residual lesions after operation depends heavily on the immune status of the body. In this context, finding the relationship between lipid levels and tumor immune infiltration should help for a more accurate assessment of recurrence and prognosis for TNBC patients. Analysis of multiple significant lipid indicators, including serum total cholesterol (TC), triglycerides (TGs), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and TILs, can reveal the potential effects of lipid levels on tumor immune infiltration and prognosis in TNBC patients [6‒8]. Therefore, we aimed to explore the effects of lipid levels on TILs and prognosis in TNBC patients.

Two hundred and twenty-two patients pathologically diagnosed with TNBC at the First Affiliated Hospital of Sun Yat-sen University, Sun Yat-sen University cancer center and The Seventh Affiliated Hospital, Sun Yat-sen University from July 2007 and December 2019 were included in this retrospective study. Inclusion criteria: (1) female TNBC patients confirmed by pathological examination, as shown in Figure 1; (2) had complete clinical information; (3) received normal care; (4) aged more than 18 years old. Exclusion criteria: (1) patients with the history of other primary malignant tumors; (2) with preoperative radiotherapy, chemotherapy, and neoadjuvant endocrine therapy; (3) complicated with other serious organ dysfunctions, such as severe liver and kidney function damage and autoimmune diseases; (4) with serious complications such as hypertension and diabetes; (5) incomplete clinical data; (6) pathological tissues were lost or poorly preserved; (7) lost follow-up. All patients were aged between 27 and 78 years, with an average follow-up age of 50 years. All participants provided their written informed permission. This study was conducted in compliance with medical ethics rules and with the Ethics Committee’s approval of the Seventh Affiliated Hospital of Sun Yat-sen University [No.KYAF/SC-08/01.0].

The age of diagnosis, menstrual status, histological grade, tumor size (T grade), lymph node metastasis (N grade), LN positive count, postoperative stage, overall survival (OS), disease-free survival (DFS) rate, and median duration of follow-up were collected.

Paraffin sections of TNBC breast tissues were used for hematoxylin and eosin (H&E) staining, and one slice with a thickness of 4 μm was taken from each specimen for routine dewaxing. After staining with xylene I solution for 15 min, it was transferred to xylene II solution for 10 min. After hydration with alcohol in concentration gradient for 1–2 min, the slices were washed with running water. The slices were stained with hematoxylin solution for 10 min. After washing, it is differentiated by hydrochloric acid and alcohol and placed in hot water to return to blue. After washing, it was dyed with 0.5% eosin aqueous solution for 5 min. After gradient alcohol dehydration, dimethylbenzene was transparent, and the sheet was sealed with neutral gum. The staining results were judged under an optical microscope.

The fasting serum samples were collected before surgery and immediately sent to perform biochemical analysis of the levels of serum TC, TG, HDL-C, LDL-C, and apolipoprotein A (B) (Apo A(B)). The results from the electronic medical record system were evaluated. According to the 2016 ESC/EAS Guidelines for the Management of Dyslipidemias, dyslipidemia was diagnosed as follows: LDL-C, 3.1 mmol/L higher limit of normal (HLN); HDL-C, 0.9 mmol/L HLN; TC, 5.2 mmol/L HLN; TG, 1.7 mmol/L HLN. Apo A and Apo B were utilized in accordance with standard clinical diagnostic criteria. TNBC patients were classified based on their HLN values.

Two experienced physicians studied stromal TILs (STILs) and total TILs (TTILs) under double-blind settings. Low abundance of TILs was characterized as 0–10% positive lymphocytes (LN), moderate abundance as 11–40% positive LN, and large abundance as 41–100% positive LN. In reference to the N9831, GeparQuattro (G4), and GeparQuinto (G5) studies, BC with TILs was characterized as 60% positive LN.

SPSS 24.0 was used to conduct statistical analysis. The data from the census were presented in percentages. Normally distributed measurement data were represented as mean ± standard deviation. OS and DFS were analyzed using the Kaplan-Meier technique. Univariate and multivariate logistic regression analyses were used to analyze the potential risk factors for TNBC. p < 0.05 was considered statistically different.

TG was found to be significantly correlated with patient’s age (p = 0.005), tumor size (p = 0.023), positive LN count (p = 0.033), postoperative stage (p = 0.005), and menstrual history (p = 0.012) (Table 1). Furthermore, Apo B was significantly correlated with patients’ age (p = 0.004) and menstrual history (p = 0.01) (Table 2). Moreover, LDL-C was clearly correlated with STILs (p = 0.003), lymphocyte-predominant breast cancer (LPBC) (p = 0.013), and TT (Table 3).

The median duration of follow-up was 67.2 months (3–159 months). Patients with elevated LDL-C level had significantly lower OS (p = 0.0001) and DFS (p = 0.048) than those with low LDL-C levels (Fig. 2). Patients with low STILs had shorter DFS (p = 0.023) than those with moderate and high STILs but not OS (p = 0.172) (Fig. 3). Moreover, there was a significant correlation between LDL-C and STILs (LPBC) in TNBC patients (r = −0.136, p = 0.043) (Fig. 4).

In TNBC patients, univariate analysis demonstrated that tumor size, positive LN count, postoperative stage, TTILs, LDL-C, LPBC, and Apo B had the predictive values for survival. LDL-C (HR = 2.644, 95% CI: 1.88–3.70, p 0.01), LPBC (HR = 0.47, 95% CI: 0.27–0.80, p = 0.006), and Apo B (HR = 2.15, 95% CI: 1.547–3.001, p 0.01) were independently predictive factors for OS in TNBC patients. Additionally, positive LN count (HR = 1.04, 95% CI: 1.01–1.08, p = 0.005), postoperative stage (HR = 2.52, 95% CI: 1.33–4.74, p = 0.004), and TTILs (HR = 0.47, 95% CI: 0.25–0.87, p = 0.021) were independently predictive factors for DFS in TNBC patients (Table 4).

According to epidemiology, exogenous or endogenous cholesterol excess can be a risk factor for BC [9]. It has been noted that BC patients exhibit variable degrees of dyslipidemia, especially postmenopausal patients. A prospective study involving 244 BC patients has revealed that high LDL-C is predictive of high tumor grade, advanced tumor stage, and high tumor cell proliferation potential [10]. It is also reported that a higher LDL-C is correlated with a shorter DFS [10].

With the characteristics of high invasiveness and poor prognosis, the treatment of TNBC has received considerable attention in recent years. Prior study has shown that elevated levels of TC and LDL-C are the risk factors for TNBC [11]. It has been determined that TNBC patients have significantly greater levels of TC and LDL-C than HR-positive BC patients. Moreover, TNBC is frequently associated by high LDL-C blood syndrome, resulting in a poor DFS [12]. Lofterød et al. [13] found that the 5-year OS in TNBC patients with TG of 1.23 mmol/L and TG of 0.82 mmol/L was 65% and 84%, respectively. In the same study, TG levels were shown to be negatively related to OS in HER2-positive BC patients. We observed that higher LDL-C levels were related with poorer OS and DFS rates in TNBC patients. Additionally, there were relationships between TG level and patient age (p = 0.005), tumor size (p = 0.023), positive LN count (p = 0.033), surgical stage (p = 0.005), and menstruation history (p = 0.012). Moreover, Apo B level was linked with patients’ age (p = 0.004) and menstruation history (p = 0.01). These findings are consistent with previously published research.

In addition to tumor cells, immune effector cells and immunosuppressive cells are major components of the TME. They are referred to as TILs having distinct pro- and anti-tumorigenic activities. A recent meta-analysis [14] has focused on the predictive values of TILs in 22,964 patients from 25 publications. TILs are independent of DFS and OS in BC patients, but there is a positive correlation in TNBC patients. Adams et al. [15] demonstrated TILs as a possible prognostic predictor for TNBC in a cohort analysis with 481 TNBC patients from the E2197 and E1199 trials of the Eastern Oncology Group (ECOG). Moreover, a higher level of STILs is found to be a prognostic factor for a reduced risk of recurrence and mortality, as well as independently prognostic value for the OS and DFS in TNBC patients. The current study showed that patients with low STILs had considerably lower DFS than those with moderate or high STILs.

Recent research has shown that cholesterols and their metabolites are involved in immunoregulation inside TME and can promote tumor growth, malignancy, recurrence, and metastasis through numerous mechanisms [4]. 27-hydroxycholesterol (27-HC), one of the cholesterol metabolites, demonstrates a clear link between cholesterol and BC [16], as its raised level in tumor tissue is pathogenic and can stimulate the proliferation, invasion, and migration of tumor cells [17, 18]. Recent research by Baek et al. [19] demonstrated that 27-HC was the primary biochemical mediator of BC metastasis in hypercholesterolemic individuals, with polymorphonuclear neutrophils and T cells playing the most significant roles. It has been demonstrated that 27-HC-enriched polymorphonuclear neutrophils and T cells can inhibit the cytotoxic action of CD8+ T cells on the tumors, thereby promoting metastasis. Therefore, the ratio of CD8/CD4 changed, and the amount of IL-9, a cytokine marker, also changed. Through regulation of immunosuppressive pathways, the dyslipidemia-induced rise of cholesterol metabolite 27-HC appears to play a crucial role in the start and progression of BC. Nonetheless, the precise regulatory process needs further clarification.

It is known that high cholesterol has immunosuppressive properties and is associated with increased expressions of immunological checkpoints. Yi and colleagues [20] discovered in 2018 that a lower amount of cholesterol predicted a greater anticancer potential of T cells in liver cancer models and that the level of cholesterol in tumor tissues was the greatest among immune organs [21]. In addition, they discovered that the expressions of immune checkpoints, such as PD-1, LAG-3, TIM-3, and 2B4, were increased with the rise of cholesterol in immune cells. Yet the immune cells were more sensitive to apoptosis and had diminished cell-killing and proliferation capacities. Further GeneChip study revealed that excessive cholesterol can impair the lipid metabolism network, inducing the stress response of the endoplasmic reticulum and the upregulation of XBP1, one of the endoplasmic reticulum stress sensors [22, 23]. The XBP1 sensor boosts the expression of immunological checkpoints and restricts T cells, hence accelerating the progression of cancer. Overall, the lipid metabolism in TNBC patients may influence the immunological state of the body. LDL-C in TNBC was linked with STILs (p = 0.027) but was independent of ITILs (p = 0.235) and LPBC (>60% STILs, p = 0.217) in the present investigation.

A limitation of the present study is the small sample size, necessitating a larger study to determine the predictive importance of TG in TNBC patients. In addition, certain clinical indications, such as BMI and ki-67, are not included in this study. The role of lipid levels in immunity and the prognosis of BC may be further elucidated, and potential processes may be investigated through future study on a grander scale with more clinical indications.

We found a specific effect of lipid levels on immune infiltration in TME and demonstrated a substantial connection between dyslipidemia and postoperative disease progression in TNBC patients. Newly diagnosed TNBC patients should have their lipid levels measured in the clinic. If a patient’s LDL-C level is greater than or equal to 3.1, they must undergo extensive management. If necessary, a more tailored therapeutic regimen is required.

All participants provided their written informed permission. This study was conducted in compliance with medical ethics rules and with the Ethics Committee’s approval of the Seventh Affiliated Hospital, Sun Yat-sen University [No.KY AF/SC-08/01.0].

There are no potential conflicts of interest to disclose.

The authors have not received any funding support.

NiJiati^.AiErken is responsible for the literature research, clinical studies, statistical analysis, and manuscript preparation; Nan Shao is responsible for the study concepts and definition of intellectual content; Yuhong Liu is responsible for the data acquisition; Huijuan Shi is responsible for the experimental studies; Yawei Shi is responsible for the data analysis; Zhongyu Yuan is responsible for the manuscript review; and Ying Lin is responsible for the guarantor of integrity of the entire study, study design, and manuscript editing. All authors read and approved the final manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.