Low Levels of High-Density Lipoprotein are Associated With Poor Prognosis in Patients With Early Triple-Negative Breast Cancer

Abstract

Background

Although dyslipidemia has been shown to be associated with breast cancer prognosis, the prognostic value of high-density lipoprotein (HDL) in triple-negative breast cancer (TNBC) remains uncertain. The purpose of this study was to evaluate the relationship between preoperative HDL levels and prognosis in TNBC patients.

Methods

Preoperative HDL levels and clinical data from 287 patients who underwent TNBC surgery were retrospectively collected. Overall survival (OS) and disease-free survival (DFS) were analyzed via the Kaplan-Meier method and the Cox proportional hazards regression model.

Results

Via the combination of receiver operating characteristics and Kaplan-Meier analysis, we found that low HDL levels were a prognostic factor for TNBC. Patients with lower HDL levels had a significantly lower T1 ratio than patients with higher HDL levels and exhibited a significantly higher neoadjuvant chemotherapy ratio than patients with higher HDL levels. In univariate analysis, the DFS and OS of patients with low HDL levels were worse than those of patients with high HDL levels in terms of DFS (HR: 0.452; 95% CI: 0.237-0.862; P = 0.016) and OS (HR: 0.438; 95% CI: 0.199-0.968; P = 0.041). However, in the multivariate analysis, there was no difference observed between patients with low HDL levels and those with high HDL levels in terms of DFS or OS. According to the subgroup analysis, lymph node-negative (N-) patients with low levels of HDL had worse DFS and OS than did patients with high HDL levels. However, in lymph node-positive (N+) patients, there was no difference observed in prognosis between the high- and low-HDL groups.

Discussion

Low HDL levels are correlated with the progression of TNBC. Moreover, low HDL levels (particularly in patients who are N-) have been identified as a significant prognostic factor impacting the prognosis of TNBC patients; however, these levels do not represent an independent prognostic factor.

Keywords

triple-negative breast cancer high-density lipoprotein prognosis

Introduction

With 2.3 million new cases predicted in the year 2020, the incidence of breast cancer is expected to surpass that of other malignancies, and breast cancer has emerged as the predominant form of cancer in women, whereby it constitutes approximately 30% of all cases. Moreover, breast cancer is the most prevalent cancer among females, with a mortality-to-incidence ratio of 15%.^1-3 Triple-negative breast cancer (TNBC), which is associated with a more unfavorable prognosis than other types of breast cancer, accounts for approximately 15%-20% of all diagnosed cases of breast cancer.^4-6

Several types of cancer, including TNBC, are associated with abnormal lipid and lipoprotein metabolism.^7-9 In addition to promoting cholesterol metabolism, High density lipoprotein (HDL) also participates in reverse cholesterol transport and can act as an antioxidant and anti-inflammatory agent.¹⁰ HDL levels are associated with poor patient outcomes in a wide range of cancer types, including digestive organ cancer, breast cancer, skin cancer, urinary tract cancer and lymphoid and hematopoietic cancer.^10-12 Moreover, several other studies have obtained inconclusive results regarding how HDL levels are related to breast cancer risk and survival rates.^13-15 Furthermore, there are few studies examining the correlation between TNBC and HDL levels.¹⁶ In summary, the prognostic significance of HDL in patients with TNBC remains uncertain.

Based on our hypothesis, we conducted a comprehensive retrospective study with a considerable sample size and duration to assess the potential of HDL as a reliable indicator. Specifically, we investigated the correlation between HDL levels and clinical outcomes in patients with TNBC who underwent radical surgery.

Materials & Methods

We reviewed the data of all TNBC patients from January 2013 to December 2017. The inclusion criteria for all of the patients were as follows: female sex, aged ≥18 years at the time of diagnosis, histologically confirmed invasive breast cancer, and performance of radical surgery. Patients who had previously (or during the follow-up period) been diagnosed with other tumors, patients with HER2 statuses that were ambiguous or unknown, patients with distant metastasis and patients with insufficient medical data (including the absence of HDL data) were excluded from the study. The clinicopathological parameters of the patients, including sex, age of onset, tumor size, lymph node status, Ki67 index, histological grade, adjuvant or neoadjuvant chemotherapy administration and operation, were collected. Estrogen receptor (ER) and progesterone receptor (PgR) staining were considered to be negative if immunostaining was observed in less than 1% of the tumor nuclei. HER2 negativity was defined as negative IHC0/1+ or IHC2+/FISH results.¹⁷ The main purpose of this study was to determine the effects of HDL levels on the DFS and OS of patients with TNBC. DFS was defined as the interval between surgery and any course of recurrence or death. OS was calculated as the time from diagnosis to death from any cause. This study was approved by the ethics committee of the Sir Run Run Shaw Hospital, Zhejiang University School of Medicine (ethics number: 2023-0649) on November 9, 2023, with an exemption for obtaining patient consent. The reporting of this study conforms to remark guidelines.¹⁸

Statistical Analysis

The cutoff value for the stratification of patients at high risk of death according to HDL cholesterol concentration was determined via receiver operating characteristic (ROC) curve analysis. The correlations between HDL levels and clinical features were tested via the chi-square test. Kaplan-Meier and log-rank tests were used to calculate survival curves and compare differences. The Cox proportional hazard model was used for multivariate analysis and for testing significant independent variables. The risk hazard ratio (HR) is reported as a relative risk with a corresponding 95% confidence interval (CI). P < 0.05 was considered to indicate statistical significance. All of the statistical analyses were performed via SPSS 21.0 software.

Results

Clinicopathological Characteristics

A total of 287 patients who were diagnosed with triple-negative breast cancer and who underwent radical resection were included in the analysis. The median follow-up time is 73 months. The clinicopathological features of these patients are shown in Table 1. The median age of the participants was 51 years and ranged from 26-88 years. Approximately 4.5% of the patients were under the age of 35 years. The proportion of T1 patients in the low-HDL group was significantly lower than that in the high-HDL group, and 44.8% of these patients were in the T1 group, whereas 54% were in the high-HDL group (P = 0.000); however, there was no significant difference in lymph node stage, histological grade or Ki67 expression between the 2 groups. In the low-HDL group, 89.5% of the patients received chemotherapy (including adjuvant chemotherapy and neoadjuvant chemotherapy), whereas the chemotherapy rate in the high-HDL group was 88.7%. The proportion of patients receiving neoadjuvant chemotherapy in the low-HDL group was significantly greater than that in the high-HDL group (29.8% vs 12.3%, respectively). Breast preservation was more common in patients with high HDL levels than in patients with low HDL levels (39% vs 26.9%, respectively), but the difference was not statistically significant (Table 1).

Table 1.

Characteristics of HDL ≤1.08 and HDL >1.08

Characteristics	HDL ≤1.08 n = 67	HDL >1.08N = 220	P
Age (years), n (%)			0.981
<35	3 (4.4)	10 (4.5)
≥35	64 (95.6)	210 (95)
pT status, n (%)			0.000
T1	30 (44.8)	119 (54)
T2	32 (47.8)	99 (45)
T3+4	5 (7.5)	2 (1)
pN status, n (%)			0.335
N0	45	141
N1	12	59
N2	5	10
N3	5	10
Ki67%, n (%)			0.625
<20%	13	47
≥20%	54	183
Histological grade, n (%)			0.282
G1	0	6
G2	9	39
G3	42	129
Unknown	16	46
Chemotherapy, n (%)			0.003
Neoadjuvant chemotherapy	20 (29.8)	27 (12.3)
Adjuvant chemotherapy	40 (59.7)	168 (76.4)
None	7 (10.5)	25 (11.3)
Surgery, n (%)			0.068
Mastectomy ± reconstruction	49 (73.1)	134 (61)
Breast conserving	18 (26.9)	86 (39)

Cutoff Determination of Serum Lipids and Lipoproteins

Receiver operating characteristic (ROC) curve analysis revealed that the recommended threshold (combined with the maximum sensitivity and specificity) was 1.085 mmol/L (AUC: 0.530, 95% CI: 0.425-0.635) for measuring HDL cholesterol.

Patient Survival

The study had a median follow-up period of 73 months, with a range of 5-117 months.

The DFS and OS rates in the high-HDL subgroup were better than those in the low-HDL subgroup (DFS 5-year rate: 88% vs 77%, respectively, P = 0.013; OS 5-year rate: 93% vs 84%, respectively, P = 0.035) (Figure 1).

Figure 1.

Kaplan–Meier survival analysis for overall survival (A) and disease-free survival (B) according to high vs low HDL levels.

Low HDL levels, positive lymph nodes, histological grade G3 and mastectomy were prognostic factors for poor DFS according to the univariate analysis. According to the multivariate Cox regression analysis, lymph node positivity (HR: 2.188; 95% CI: 1.547-3.094; P = 0.000), histological grade G3 (HR: 4.643; 95% CI: 1.079-19.981; P = 0.039) and mastectomy (HR: 5.618; 95% CI: 1.660-19.012; P = 0.006) were found to be independent prognostic factors for poor DFS (Table 2). For OS, low HDL levels, positive lymph nodes and mastectomy were prognostic factors for poor OS according to the univariate analysis. The multivariate Cox regression analysis revealed that only positive lymph nodes (HR: 2.714; 95% CI: 1.935-3.806; P = 0.000) were an independent prognostic factor for poor OS (Table 3).

Table 2.

Univariate and Multivariate Analysis for Disease-free Survival.

Characteristics DFS	Univariate Analysis			Multivariate Analysis
Characteristics DFS	HR	95% CI	P	HR	95% CI	P
Age
<35 years	1.000
≥35 years	0.792	0.191-3.287	0.792
pT status
T1	1.000
T2	1.208	0.634-2.302	0.565
T3+4	2.913	0.676-12.557	0.152
pN status
N0	1.000			1.000
N1	0.859	0.312-2.363	0.768	0.684	0.239-1.964	0.481
N2	5.726	2.221-14.763	0.000	4.495	1.662-12.152	0.003
N3	21.180	9.828-45.643	0.000	8.701	3.231-23.430	0.000
Ki67%
<20	1.000
≥20	0.870	0.383-1.970	0.738
Histological grade
G1-2	1.000			1.000
G3	4.894	1.166-20.546	0.030	4.643	1.079-19.981	0.039
Chemotherapy
None	1.000
Adjuvant chemotherapy	0.502	0.203-1.243	0.136
Neoadjuvant chemotherapy	1.438	0.540-3.832	0.467
Surgery
Breast conserving	1.000			1.000
Mastectomy ± reconstruction	5.326	1.893-14.986	0.002	5.618	1.660-19.012	0.006
HDL
≤1.085	1.000			1.000
>1.085	0.452	0.237-0.862	0.016	0.818	0.375-1.785	0.613

Table 3.

Univariate and Multivariate Analysis for Overall Survival.

Characteristics OS	Univariate Analysis			Multivariate Analysis
Characteristics OS	HR	95% CI	P	HR	95% CI	P
Age (years)
<35	1.000
≥35	0.590	0.139-2.498	0.474
pT status
T1	1.000
T2	1.345	0.602-3.003	0.469
T3+4	5.057	1.120-22.841	0.035
pN status
N0	1.000			1.000
N1	0.568	0.123-2.630	0.470	0.500	0.108-2.328	0.500
N2	7.379	2.472-22.029	0.000	5.466	1.814-16.470	0.003
N3	20.072	8.124-45.590	0.000	15.230	6.054-38.311	0.000
Ki67%
<20	1.000
≥20	1.049	0.362-3.046	0.929
Histological grade
G1-2	1.000
G3	2.789	0.641-12.134	0.172
Chemotherapy
None	1.000
Adjuvant chemotherapy	0.293	0.100-0.857	0.025
Neoadjuvant chemotherapy	1.519	0.528-4.374	0.438
Surgery
Breast conserving	1.000			1.000
Mastectomy ± reconstruction	6.759	1.597-28.615	0.009	4.164	0.968-17.913	0.055
HDL
≤1.085	1.000			1.000
>1.085	0.438	0.199-0.968	0.041	0.697	0.310-1.567	0.383

Subgroup Survival

According to the subgroup analysis, for N0 patients, the OS and DFS of patients with high HDL levels were better than those of patients with low HDL levels (DFS 5-year rate: 94% vs 84%, respectively, P = 0.029; OS 5-year rate: 99% vs 86%, respectively, P = 0.001) (Figure 2(A) and (B)). For N1-3 patients, there was no difference in prognosis between the high-HDL subgroup and low-HDL subgroup (DFS 5-year rate: 79% vs 63%, respectively, P = 0.113; OS 5-year rate: 84% vs 79%, respectively, P = 0.719) (Figure 2(C) and (D)).

Figure 2.

Kaplan–Meier survival analysis for overall survival and disease-free survival in subgroups according to high vs low HDL levels. In lymph node negative patients, patients with high HDL level had a significantly better prognosis than patients with low HDL level (A) (B). In lymph node positive patients, there was no difference in prognosis between high and low HDL level (C) (D).

Discussion

The incidence of triple-negative breast cancer (TNBC) is greater among women under the age of 50 years, thus indicating a distinct propensity for enhanced metastatic potential and a higher pathological grade.^5,6,19,20 The prognosis for patients with TNBC is unfavorable, and it is characterized by a heightened risk of recurrence, with recurrence rates reaching as high as 25%. Furthermore, the survival rate within the 3-5 year period following diagnosis is significantly decreased.²¹ Compared with other molecular subtypes of breast cancer, the available treatment options are limited, with chemotherapy serving as the sole established adjuvant therapy.⁵ Hence, TNBC is regarded as the most unfavorable prognostic subtype among all breast cancer subgroups. Given the diverse treatment responses and survival rates observed across molecular subtypes of breast cancer, the investigation of breast cancer biomarkers and molecular classifications possesses significant importance in guiding clinical treatment decisions and prognostic assessments.⁹ Hence, the identification of prognostic factors in patients diagnosed with TNBC is important because it can lead to suitable clinical interventions and can potentially prolong patients’ overall survival. This study substantiates the association between low HDL levels and increased TNBC risk, particularly in N0 patients, whereas no such correlation was observed in N1-3 patients.

Lipids play important roles in the development and progression of tumors. Proliferating tumor cells require continuous supplies of lipids to sustain membrane biogenesis, whereas nonproliferating tumor cells need abundant lipids to maintain signal transmission and to resist apoptosis.^22,23 Several changes in cancer cell lipid metabolism have been observed, and these changes have been found to be linked to important cellular processes, such as cell proliferation, differentiation, and motility.²⁴

According to a recent meta-analysis including both retrospective and prospective studies, patients with higher HDL levels had 37% lower death rates and 35% lower relapse rates than patients with lower HDL levels.²⁵ For lung cancer, nasopharyngeal carcinoma, colorectal cancer, and breast cancer, specific data are available with respect to these levels. A decreased plasma apoA-I level before therapy was associated with a poor prognosis in patients with non-small cell lung cancer. Mendelian randomization studies were also conducted to address the causal link between HDL cholesterol and cancer risk. As revealed by the Breast Cancer Association Consortium (BCAC), genetically elevated HDL levels are consistently linked to increased rates of breast cancer, regardless of the estrogen receptor status of the tumor.^26-29

Both in vitro and in vivo studies have demonstrated that HDL reduces the levels of cholesterol and its metabolites in cells, thus leading to an overall rewiring of cellular metabolism.¹⁰ HDL removes excess cholesterol from peripheral tissues and transports it to the liver and steroidogenic cells for catabolism.³⁰ Lower cholesterol levels in peripheral cells, which may be caused by a higher HDL level, hinder tumor growth and lymphatic dissemination.

The antioxidative and anti-inflammatory properties of HDL may have contributed to its beneficial effects. A previous study suggested that HDL can inhibit LDL oxidative damage to prevent lipid peroxidation.31 Additionally, various studies have shown that high HDL production is associated with increased production of anti-inflammatory cytokines such as interleukin 10, which may be protective against breast cancer.³²

HDL cholesterol is inversely associated with the activity of interleukin-6, which is a proinflammatory cytokine.³³ Additionally, IL-6 and IL-8 may promote the progression of TNBC.³⁴ HDL cholesterol has been proposed to possess antitumorigenic properties via the regulation of angiogenesis by decreasing VEGF expression.³⁵ TNBC patients with high VEGF expression are more likely to demonstrate metastatic potential, thus indicating a possible link between low HDL cholesterol levels and tumor progression.^36,37 In addition to helping to reverse immune escape, HDL can enhance the recruitment of M1 tumor-associated macrophages (TAMs) and cytotoxic CD8⁺ T cells.¹⁰

In the present study, we examined the prognostic impact of HDL levels on DFS and OS in patients with TNBC who underwent surgical intervention. Numerous studies have employed cutoff values to establish associations between HDL levels and survival outcomes. In the current investigation, ROC analysis was employed to assess survival prediction and to determine the most suitable cutoff points for HDL cholesterol. Our analysis revealed that the survival rate of patients with TNBC can be divided into 2 distinct groups: those with HDL levels <1.085 mmol/l and those with HDL levels >1.085 mmol/l. The precise functions of HDL in patients with TNBC remain uncertain. Consequently, our investigation revealed that HDL cholesterol levels exhibited superior prognostic capabilities in predicting survival outcomes among TNBC patients. Notably, there were fewer T1 patients, a greater proportion of T3-4 patients, and greater utilization of neoadjuvant chemotherapy in the low-HDL group than in the high-HDL group, thus suggesting that low-HDL TNBC may be more aggressive.

The prognosis of patients with high HDL levels was shown to be superior to that of patients with low HDL levels, which aligns with findings from previous studies.^9,16,36 However, none of these studies have examined the associations between lymph node metastases and patient outcomes. To address this knowledge gap, we conducted a subgroup analysis and reported that patients with high HDL levels in the lymph node-negative subgroup had a more favorable prognosis. Conversely, there was no significant difference in outcomes between patients with high HDL levels and those with low HDL levels in the lymph node-positive subgroup.

The present study has several limitations that warrant consideration. First, the design of this study has inherent limitations, such as the use of retrospective data collection, which is a common constraint in all retrospective studies pertaining to TNBC. Second, the study is limited by its reliance on a retrospective analysis involving a relatively small sample size of only 287 eligible patients. Although detailed data were carefully recorded, a prospective study could offer a more comprehensive assessment of prognostic factors in TNBC patients. Consequently, it is imperative to validate these analyses within a larger cohort of patients.

Conclusions

In summary, HDL levels serve as a readily quantifiable biomarker of the prognosis of patients with TNBC. Low HDL levels are correlated with the progression of TNBC (particularly in patients classified as lymph node-negative ), but these HDL levels are not an independent prognostic factor. The monitoring and assessment of HDL levels before treatment can aid in predicting patient prognosis. However, it is important to note that this study represents the first investigation into the prognostic significance of HDL levels in early TNBC patients. Consequently, additional prospective studies are necessary to establish definitive cutoff values and validate our findings.

Footnotes

Acknowledgments

We thank the study participants; the study staff; and the doctors, nurses and other health care providers and health information sources that contributed to the study.

Author Contributions

YongCheng Liu and HeMing Zheng designed the study; YongCheng Liu wrote the manuscript text; Di Wu, HuiJuan Wang, and YongCheng Liu collected the patients’ clinicopathological data, followed up each patient, and performed statistical analysis.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Medical and Health Science and Technology Program of Zhejiang Province (Grant No. 2022RC193).

Ethical Statement

ORCID iDs

YongCheng Liu

HuiJuan Wang

Appendix

References

Kashyap

Pal

Sharma

, et al. Global increase in breast cancer incidence: risk factors and preventive measures. BioMed Res Int. 2022;2022:9605439.

Sung

Ferlay

Siegel

, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209-249.

Loibl

Poortmans

Morrow

Denkert

Curigliano

. Breast cancer. Lancet. 2021;397(10286):1750-1769.

Waks

Winer

. Breast cancer treatment: a review. JAMA. 2019;321(3):288-300.

Bou

Ghorayeb

Saliba

, et al. Triple negative breast cancer: updates on classification and treatment in 2021. Cancers. 2022;14(5):1253.

Denkert

Liedtke

Tutt

von Minckwitz

. Molecular alterations in triple-negative breast cancer-the road to new treatment strategies. Lancet. 2017;389(10087):2430-2442.

Tang

Wang

, et al. The effect of preoperative serum triglycerides and high-density lipoprotein-cholesterol levels on the prognosis of breast cancer. Breast. 2017;32:1-6.

Cedo

Reddy

Mato

Blanco-Vaca

Escola-Gil

. HDL and LDL: potential new players in breast cancer development. J Clin Med. 2019;8(6):853.

Dai

Chen

Wang

, et al. Pretreatment TG/HDL-C ratio is superior to triacylglycerol level as an independent prognostic factor for the survival of triple negative breast cancer patients. J Cancer. 2016;7(12):1747-1754.

10.

Ossoli

Wolska

Remaley

Gomaraschi

. High-density lipoproteins: a promising tool against cancer. Biochim Biophys Acta Mol Cell Biol Lipids. 2022;1867(1):159068.

11.

Loosen

Kostev

Luedde

Roderburg

. Low blood levels of high-density lipoprotein (HDL) cholesterol are positively associated with cancer. J Cancer Res Clin Oncol. 2022;148(11):3039-3046.

12.

Shen

Jin

Dong

Wang

Zhao

Shen

. Low level of serum high-density lipoprotein cholesterol in gastric cancer correlates with cancer progression but not survival. Transl Cancer Res. 2020;9(10):6206-6213.

13.

Paulette

Yiqing

Lin

, et al. Lipid biomarkers and long-term risk of cancer in the Women's Health Study. Am J Clin Nutr. 2016;103(6):1397.

14.

Borgquist

Butt

Almgren

, et al. Apolipoproteins, lipids and risk of cancer. Int J Cancer. 2016;138(11):2648-2656.

15.

Mathilde

Laureen

Fagherazzi

, et al. Associations between serum lipids and breast cancer incidence and survival in the E3N prospective cohort study. Cancer Causes & Control. 2016;28(1):77-88.

16.

Fan

Ding

Wang

, et al. Decreased serum HDL at initial diagnosis correlates with worse outcomes for triple-negative breast cancer but not non-TNBCs. Int J Biol Markers. 2015;30(2):e200-207.

17.

Hammond

E, H

Hayes

Dowsett

, et al. American society of clinical oncology/college of American pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. Journal of Oncology Practice. 2010;134(7):e48-72.

18.

McShane

Altman

Sauerbrei

, et al. REporting recommendations for tumour MARKer prognostic studies (REMARK). Br J Cancer. 2005;93(4):387-391.

19.

Dent

Trudeau

Pritchard

, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007;13(15 Pt 1):4429-4434.

20.

Haffty

Yang

Reiss

, et al. Locoregional relapse and distant metastasis in conservatively managed triple negative early-stage breast cancer. J Clin Oncol. 2006;18(4):401-402.

21.

Wang

Huang

, et al. Effect of capecitabine maintenance therapy using lower dosage and higher frequency vs observation on disease-free survival among patients with early-stage triple-negative breast cancer who had received standard treatment: the SYSUCC-001 randomized clinical trial. JAMA. 2021;325(1):50-58.

22.

Hsu

Sabatini

. Cancer cell metabolism: warburg and beyond. Cell. 2008;134(5):703.

23.

Rysman

Brusselmans

Scheys

, et al. De novo lipogenesis protects cancer cells from free radicals and chemotherapeutics by promoting membrane lipid saturation. Cancer Res. 2010;70(20):8117-8126.

24.

Santos

Schulze

. Lipid metabolism in cancer. FEBS J. 2012;279(15):2610-2623.

25.

Pingting

Liu

Chen

Xiao

. Prognostic role of serum total cholesterol and high-density lipoprotein cholesterol in cancer survivors: a systematic review and meta-analysis. Clin Chim Acta. 2018;477:94-104.

26.

Johnson

Siewert

Klarin

, et al. The relationship between circulating lipids and breast cancer risk: a Mendelian randomization study. PLoS Med. 2020;17(9):e1003302.

27.

Beeghly-Fadiel

Khankari

Delahanty

, et al. A Mendelian randomization analysis of circulating lipid traits and breast cancer risk. Int J Epidemiol. 2020;49(4):1117-1131.

28.

Nowak

Ärnlöv

. A Mendelian randomization study of the effects of blood lipids on breast cancer risk. Nat Commun. 2018;9(1):3957.

29.

Chatterjee

. Mendelian randomization analysis using mixture models for robust and efficient estimation of causal effects. Nat Commun. 2019;10(1):1941.

30.

Ganjali

Ricciuti

Pirro

, et al. High-density lipoprotein components and functionality in cancer: state-of-the-art - ScienceDirect. Trends Endocrinol Metabol. 2019;30(1):12-24.

31.

Ray

Husain

. Role of lipids, lipoproteins and vitamins in women with breast cancer. Clinical Biochemistry -Ottawa Then New York. THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, ENGLAND: Pergamon Press Elsevier Science; 2001.

32.

Tang

Wang

, et al. The effect of preoperative serum triglycerides and high-density lipoprotein-cholesterol levels on the prognosis of breast cancer. Breast. 2017;32(Complete):1-6.

33.

Holven

Retterstøl

Ueland

, et al. Subjects with low plasma HDL cholesterol levels are characterized by an inflammatory and oxidative phenotype. PLoS One. 2013;8(11):e78241.

34.

Hartman

Poage

Hollander

Tsimelzon

Brown

. Growth of triple-negative breast cancer cells relies upon coordinate autocrine expression of the proinflammatory cytokines IL-6 and IL-8. Cancer Res. 2013;73(11):3470.

35.

Tan

JTM

MKC

Bursill

. The role of high-density lipoproteins in the regulation of angiogenesis. Cardiovasc Res. 2015;106(2):184-193.

36.

Trygve

Mortensen

Hawa

, et al. Impact of pre-diagnostic triglycerides and HDL-cholesterol on breast cancer recurrence and survival by breast cancer subtypes. BMC Cancer. 2018;18(1):654.

37.

Roberti

Arriaga

Bianchini

, et al. Protein expression changes during human triple negative breast cancer cell line progression to lymph node metastasis in a xenografted model in nude mice. Cancer Biol Ther. 2012;13(11):1123-1140.