Preoperative Albumin-to-Fibrinogen Ratio as an Outcome Predictor in Triple-Negative Breast Cancer

By Shuo Wu¹, Weijie Li²

Affiliations

1. Department of Medical Oncology, Cancer Hospital of China Medical University, Liaoning Cancer Hospital and Institute, Liaoning, China
2. Department of Breast Surgery, Cancer Hospital of China Medical University, Liaoning Cancer Hospital and Institute, Liaoning, China

doi: 10.29271/jcpsp.2025.12.1557

ABSTRACT
Objective: To elucidate the prognostic value of the albumin-to-fibrinogen ratio (AFR) in patients with triple-negative breast cancer (TNBC).
Study Design: An observational study.
Place and Duration of the Study: Department of Medical Oncology, Cancer Hospital of China Medical University, Liaoning Cancer Hospital and Institute, Shenyang, Liaoning, China, from January 2010 to October 2022.
Methodology: A total of 195 patients diagnosed with TNBC were recruited and followed up till October 2022. The optimal cut-off value of AFR was assessed via receiver operating characteristic curve analysis, after which patients were categorised into low and high AFR groups accordingly. The relevance between TNBC and clinicopathological parameters were evaluated using χ² and Fisher’s exact tests. Survival time was analysed by using the Kaplan-Meier method and log-rank tests. Prognostic factors were identified via Cox regression model analysis. Nomograms were constructed, and calibration curve analysis (CCA) and decision curve analysis (DCA) were applied to evaluate their predictive clinical application.
Results: The optimal cut-off value of AFR was 15. The median survival time of patients with high-AFR values was higher than those with low-AFR values (DFS: 36.54 vs. 33.19 months, p = 0.004; OS: 56.45 vs. 53.29 months, p = 0.003). According to the Cox regression model, AFR was a significant predictor of disease-free survival [DFS; hazard ratio (HR): 0.595, 95% CI: 0.377-0.940, p = 0.026] and overall survival (OS; HR: 0.385, 95% CI: 0.221-0.670, p = 0.001). CCA and DCA indicated that AFR-based nomograms had good predictive clinical utility within the threshold probability range for different survival rates.
Conclusion: AFR is easy and inexpensive to conduct and has the potential to serve as a predictive factor for preoperative evaluation in clinical practice.

Key Words: Albumin-to-fibrinogen ratio, Triple-negative breast cancer, Albumin, Fibrinogen, Treatment.

INTRODUCTION

Breast cancer is the most common malignant tumour in females and a leading cause of cancer-associated mortality worldwide.¹ Triple-negative breast cancer (TNBC) is a high- grade subtype, accounting for 15-20% of all cases, with a higher prevalence in young females and a worse prognosis than HER2-enriched or luminal subtypes.² Survival rate analysis after preliminary treatment reveals a significantly high risk of recurrence and death. Due to a lack of effective therapeutic targets, chemotherapy remains the main treatment modality for  patients  with  TNBC.³

The systemic inflammatory response, which plays a prominent role in the tumour microenvironment, is associated with clinical outcomes and overall prognosis. Its indicators include single markers—such as neutrophils, monocytes, platelets, lymphocytes, albumin (ALB), and fibrinogen (FIB)—and combined marker ratios or scoring systems, such as lymphocyte-to- monocyte ratio (LMR), Glasgow prognostic score, neutrophil- to-lymphocyte ratio (NLR), and platelet-to-lymphocyte ratio (PLR).⁴ Coagulation dysfunction increases the risk of thrombosis during the development and progression of tumours by limiting the function of growth factors and natural killer cells.⁵ The nutritional status is a major factor affecting the prognosis of cancer patients. Previous research has confirmed that ALB was a critical predictor of nutritional status.⁶ FIB is a soluble serum glycoprotein that is generated by hepatic cells and involved in blood coagulation and platelet aggregation. The ALB-to-FIB ratio (AFR) is a promising prognostic indicator based on inflam- mation for certain tumours.⁷ However, there have been few studies on AFR and TNBC. Therefore, this study aimed to elucidate the prognostic value of AFR in the prediction of the sur- vival time of TNBC patients.

METHODOLOGY

A total of 195 female patients diagnosed with TNBC in the Department of Medical Oncology, Cancer Hospital of China Medical University, Liaoning Cancer Hospital and Institute, Liaoning, China, from January 2010 to October 2022, were included in the study. Clinicopathological and follow-up data were obtained from the electronic medical records. The study was approved by the Ethics Committee of Cancer Hospital of China Medical University, Liaoning Cancer Hospital and Institute, Liaoning, China, and was performed in compliance with the revised version of the Declaration of Helsinki. Individual patient information was protected and not disclosed in this study. All  enrolled  patients  provided  informed  consent.

Eligible participants were required to have histologically confirmed TNBC, complete clinical and pathological data, and available follow-up information. Preoperative blood samples were collected prior to any anticancer therapy. Patients were excluded if they presented with distant metastases (M1 stage) or concurrent malignancies unrelated to TNBC. Additional exclusion criteria comprised prior exposure to chemotherapy, radiotherapy, targeted therapy, or immunotherapy, as well as the presence of acute or chronic inflammatory diseases, including infections or autoimmune diseases. Individuals who had undergone blood transfusions within one month before surgery were also excluded to minimise confounding effects on haematological  parameters.

Body mass index (BMI) was evaluated based on the patient’s weight (kg) and height (m). The 8^th edition of the American Joint Committee on Cancer (AJCC) staging system and the TNM stage classification of the Union for International Cancer Control (UICC) were used for cancer staging. The time of menopause was defined as the final menstrual period one year before amenorrhea. The fifth edition of the Breast Imaging Reporting and Data System (BIRADS) was used. The P/T ratio was determined by positive lymph nodes divided by the total lymph nodes examined. CK5/6, P53, epidermal growth factor receptor (EGFR), TOP2A, Ki-67 expression, and lymphovascular invasion (LVI) were obtained by immunohistochemistry in the Department of Pathology. Blood samples were collected from a peripheral vein prior to treatment. Routine haematological parameters from peripheral venous blood were detected by an automated haematology analyser. AFR was calculated as AFR = ALB/FIB, where ALB and FIB represent the pretreatment peripheral ALB and FIB counts, respectively.

All enrolled patients received inpatient and outpatient services and made phone calls after surgery. The postoperative follow-up schedule was as follows: three months during the first and second years, every 6 months from the third to fifth years, and annually thereafter until death. Disease-free survival (DFS) was defined as the time from surgery to the occurrence of local recurrence or distant metastasis. Overall survival (OS) was defined as the time of surgery to death from any cause or to the date of the last follow-up. All statistical analyses were conducted using SPSS Statistics software version 22.0 (IBM Corp.) and GraphPad Prism software 8.0 (Dotma- tics). The Chi-square test and Fisher’s exact test were applied to compare nominal clinicopathologic variables [N (%)], such as marital status, menopause, chemotherapy, radiotherapy, targeted therapy, TNM stage, type of surgery, and histologic grade. The optimal cut-off value of the AFR was assessed using receiver operating characteristic curve (ROC) analysis. The Kaplan-Meier method was applied to compute the survival curves of DFS and OS, and the curves were analysed using the log-rank tests. The Cox model analyses were employed to assess potential independent factors. Nomograms were constructed, and calibration curve analysis (CCA) and decision curve analysis (DCA) were performed to evaluate the clinical utility of the predictive models. A p-value of less than 0.05 was considered statistically significant. The Youden index was calculated as sensitivity + specificity-1. Based on the highest value of the Youden index, the optimal cut-off value for AFR was obtained.

RESULTS

According to ROC analysis, the sensitivity and specificity were 69.35% and 58.65%, respectively. The optimal cut-off value for AFR was determined to be 15.0 (AUC: 0.697, p <0.0001) (Annexure 1). Based on this value, patients were classified into a low AFR group (AFR <15.0) and a high AFR group (AFR ≥15.0). There were 96 cases (49.23%) in the low AFR group and 99 cases (50.77%) in the high AFR group. Compared to the low AFR group, the high AFR group’s score was found to be associated with age (χ² = 4.329, p = 0.037), marital status (χ² = 5.425, p = 0.020), BMI (χ² = 4.934, p = 0.026), menopause (χ² = 5.596, p = 0.018), ultrasound (US) tumour size (χ² = 4.277, p = 0.039), and postoperative radiotherapy (χ² = 4.775, p = 0.029). Detailed patient data are shown in Table I.

According to the clinical stage at diagnosis, 111 (56.92%) cases were stage I or II, and 84 (43.08%) cases were stage III. There were significant differences in clinical T stage (χ² = 6.655, p = 0.010), histologic grade (χ² = 6.315, p = 0.012), positive lymph nodes (χ² = 3.881, p = 0.049), EGFR (χ² = 8.587, p = 0.003), and topoisomerase 2A (TOP2A; χ² = 10.412, p = 0.001) between the two groups (Table I).

In the current study, the reference ranges for these para- meters were as follows: white blood cells (3.5 x 10⁹-9.5 x 10⁹/l), red blood cells (3.8 x 10¹²-5.1 x 10¹²/l), platelets (125 x 10⁹-350 x 10⁹/l), haemoglobin (115-150 g/l), neutrophils (1.8 x 10⁹-6.3 x 10⁹/l), lymphocytes (1.1 x 10⁹-3.2 x 10⁹/l), C-reactive protein (CRP; 10-200 mg/l), ALB (35-55 g/l), alkaline phosphatase (ALP; 35-100 U/l), lactate dehydrogenase (LDH; 109-245 U/l), FIB (2-4 g/l), and D-dimer (D-D; 0-0.55 mg/l). Significant associations were observed for ABO blood type (χ² = 6.976, p = 0.008), red blood cells (χ² = 4.310, p = 0.038), ALB (χ² = 15.997, p <0.001), ALP (χ² = 10.345, p = 0.001), LDH (χ² = 4.957, p = 0.026), FIB (χ² = 60.897, p <0.001), and D-D (χ² = 6.253, p = 0.012). The associations between AFR and inflammatory indices are shown in Table I.

Table I: Demographic and clinicopathologic characteristics of TNBC patients.

Variables	Number of cases			χ2	p-values
	Total (n = 195)	Low AFR group (<15.0) (n = 96)	High AFR group (≥15.0) (n = 99)
Demographic characteristics
Age, (years)				4.329	0.037
<48	96 (49.2%)	40 (41.7%)	56 (56.6%)
≥48	99 (50.8%)	56 (58.3%)	43 (43.4%)
Marital status				5.425	0.020a
Married	183 (93.8%)	94 (97.9%)	89 (89.9%)
Unmarried	12 (6.2%)	2 (2.1%)	10 (10.1%)
BMI				4.934	0.026
<23.50	97 (49.7%)	40 (41.7%)	57 (57.6%)
≥23.50	98 (50.3%)	56 (58.3%)	42 (42.4%)
Menopause				5.596	0.018
No	116 (59.5%)	49 (51.0%)	67 (67.7%)
Yes	79 (40.5%)	47 (49.0%)	32 (32.3%)
US-tumour size (cm)				4.277	0.039
≤2	83 (42.6%)	48 (50.0%)	35 (35.4%)
>2	112 (57.4%)	48 (50.0%)	64 (64.6%)
US-LNM				1.458	0.227
No	147 (75.4%)	76 (79.2%)	71 (71.7%)
Yes	48 (24.6%)	20 (20.8%)	28 (28.3%)
US-BIRADS				3.727	0.054
4 + 5	99 (50.8%)	42 (43.8%)	57 (57.6%)
6	96 (49.2%)	54 (56.3%)	42 (42.4%)
Postoperative complications				0.182	0.670
No	177 (90.8%)	88 (91.7%)	89 (89.9%)
Yes	18 (9.2%)	8 (8.3%)	10 (10.1%)
Postoperative chemotherapy				0.030	0.862
No	58 (29.7%)	28 (29.2%)	30 (30.3%)
Yes	137 (70.3%)	68 (70.8%)	69 (69.7%)
Postoperative radiotherapy				4.775	0.029
No	138 (70.8%)	61 (63.5%)	77 (77.8%)
Yes	57 (29.2%)	35 (36.5%)	22 (22.2%)
Postoperative targeted therapy				1.034	0.309
No	175 (89.7%)	84 (87.5%)	91 (91.9%)
Yes	20 (10.3%)	12 (12.5%)	8 (8.1%)
Pathological characteristics
Clinical T stage				6.655	0.010
T1 + T2	149 (76.4%)	81 (84.4%)	68 (68.7%)
T3 + T4	46 (23.6%)	15 (15.6%)	31 (31.3%)
Clinical N stage				0.211	0.646
N0	70 (35.9%)	36 (37.5%)	34 (34.3%)
N1 + N2 + N3	125 (64.1%)	60 (62.5%)	65 (65.7%)
Clinical TNM stage				0.941	0.332
I + II	111 (56.9%)	58 (60.4%)	53 (53.5%)
III	84 (43.1%)	38 (39.6%)	46 (46.5%)
Types of surgery				2.552	0.110
Mastectomy	151 (77.4%)	79 (82.3%)	72 (72.7%)
Breast-conserving surgery	44 (22.6%)	17 (17.7%)	27 (27.3%)
Histologic grade				6.315	0.012
I + II	100 (51.3%)	58 (60.4%)	42 (42.4%)
III	95 (48.7%)	38 (39.6%)	57 (57.6%)
Tumour size (cm)				1.126	0.289
≤2	105 (53.8%)	48 (50.0%)	57 (57.6%)
>2	90 (46.2%)	48 (50.0%)	42 (42.4%)
Pathological N stage				3.089	0.079
T1 + T2	122 (62.6%)	66 (68.8%)	56 (56.6%)
T3 + T4	73 (37.4%)	30 (31.3%)	43 (43.4%)
Pathological N stage				1.166	0.280
N0	137 (70.3%)	64 (66.7%)	73 (73.7%)
N1 + N2 + N3	58 (29.7%)	32 (33.3%)	26 (26.3%)
Pathological TNM stage				1.166	0.280
I + II	137 (70.3%)	64 (66.7%)	73 (73.7%)
III	58 (29.7%)	32 (33.3%)	26 (26.3%)
Continued…

Variables	Number of cases			χ2	p-values
	Total (n = 195)	Low AFR group (<15.0), (n = 96)	High AFR group (≥15.0), (n = 99)
P/T ratio				3.812	0.051
<0.12	125 (64.1%)	55 (57.3%)	70 (70.7%)
≥0.12	70 (35.9%)	41 (42.7%)	29 (29.3%)
Total lymph nodes removed				0.004	0.947
<21	100 (51.3%)	49 (51.0%)	51 (51.5%)
≥21	95 (48.7%)	47 (49.0%)	48 (48.5%)
Positive lymph nodes				3.881	0.049
<2	129 (66.2%)	57 (59.4%)	72 (72.7%)
≥2	66 (33.8%)	39 (40.6%)	27 (27.3%)
CK5/6				1.006	0.316
Negative	119 (61.0%)	62 (64.6%)	57 (57.6%)
Positive	76 (39.0%)	34 (35.4%)	42 (42.4%)
EGFR				8.587	0.003
Negative	87 (44.6%)	53 (55.2%)	34 (34.3%)
Positive	108 (55.4%)	43 (44.8%)	65 (65.7%)
P53				0.598	0.439
Negative	105 (53.8%)	49 (51.0%)	56 (56.6%)
Positive	90 (46.2%)	47 (49.0%)	43 (43.4%)
TOP2A				10.412	0.001
Negative	83 (42.6%)	52 (54.2%)	31 (31.3%)
Positive	112 (57.4%)	44 (45.8%)	68 (68.7%)
Ki-67				3.055	0.080
Negative	37 (19.0%)	23 (24.0%)	14 (14.1%)
Positive	158 (81.0%)	73 (76.0%)	85 (85.9%)
LVI				0.536	0.464
Negative	150 (76.9%)	76 (79.2%)	74 (74.7%)
Positive	45 (23.1%)	20 (20.8%)	25 (25.3%)
Inflammation indices
ABO blood type				6.976	0.008
A + B	106 (54.4%)	43 (44.8%)	63 (63.6%)
O + AB	89 (45.6%)	53 (55.2%)	36 (36.4%)
White blood cells (x109/L)				0.004	0.951
<5.90	102 (52.3%)	50 (52.1%)	52 (52.5%)
≥5.90	93 (47.7%)	46 (47.9%)	47 (47.5%)
Red blood cells (x1012/L)				4.309	0.038
<4.36	97 (49.7%)	55 (57.3%)	42 (42.4%)
≥4.36	98 (50.3%)	41 (42.7%)	57 (57.6%)
Platelets (x109/L)				0.045	0.832
<237.00	99 (50.8%)	48 (50.0%)	51 (51.5%)
≥237.00	96 (49.2%)	48 (50.0%)	48 (48.5%)
Haemoglobin (g/L)				1.549	0.213
<131.00	107 (54.9%)	57 (59.4%)	50 (50.5%)
≥131.00	88 (45.1%)	39 (40.6%)	49 (49.5%)
Neutrophils (x109/L)				0.131	0.718
<3.60	99 (50.8%)	50 (52.1%)	49 (49.5%)
≥3.60	96 (49.2%)	46 (47.9%)	50 (50.5%)
Lymphocytes (x109/L)				1.157	0.282
<1.76	97 (49.7%)	44 (45.8%)	53 (53.5%)
≥1.76	98 (50.3%)	52 (54.2%)	46 (46.5%)
C-reactive protein (mg/L)				0.002	0.965
<0.40	110 (56.4%)	54 (56.3%)	56 (56.6%)
≥0.40	85 (43.6%)	42 (43.7%)	43 (43.4%)
ALB (g/L)				15.997	<0.001
<45.00	110 (56.4%)	68 (70.8%)	42 (42.4%)
≥45.00	85 (43.6%)	28 (29.2%)	57 (57.6%)
ALP (U/L)				10.345	0.001
<68.00	102 (52.3%)	39 (40.6%)	63 (63.6%)
≥68.00	93 (47.7%)	57 (59.4%)	36 (36.4%)
LDH (U/L)				4.957	0.026
<168.00	100 (51.3%)	57 (59.4%)	43 (43.4%)
≥168.00	95 (48.7%)	39 (40.6%)	56 (56.6%)
FIB (g/L)				60.897	<0.001
<2.88	100 (51.3%)	22 (22.9%)	78 (78.8%)
≥2.88	95 (48.7%)	74 (77.1%)	21 (21.2%)
D-D (mg/L)				6.253	0.012
<0.37	101 (51.8%)	41 (42.7%)	60 (60.6%)
≥0.37	94 (48.2%)	55 (57.3%)	39 (39.4%)
p-values by Chi-square test and aFisher exact test. AFR: Albumin-to-fibrinogen ratio; BMI: Body mass index; US: Ultrasound; LNM: Lymph node metastasis; BIRADS: Breast Imaging Reporting and Data System; P/T: Positive lymph nodes/total lymph nodes; CK: Cytokeratin; EGFR: Epidermal growth factor receptor; TOP2A: Topoisomerase 2A; ALB: Albumin; ALP: Alkaline phosphatase; LDH: Lactate dehydrogenase; FIB: Fibrinogen; D-D: D-dimer; LVI: Lymphovascular invasion.

Table II: Uni- and multivariate Cox-regression survival analyses of AFR for the prediction of DFS and OS in TNBC patients.
 

Parameters	DFS				OS
	Univariate analysis		Multivariate analysis		Univariate analysis		Multivariate analysis
	Hazard ratios (95% CI)	#p-values	Hazard ratios (95% CI)	##p-values	Hazard ratios (95%CI)	#p-values	Hazard ratios (95% CI)	##p-values
Age (<48 vs. ≥48 years)	1.155 (0.393-3.401)	0.793	-	-	0.931 (0.305-2.840)	0.901	-	-
Marital status (married vs. unmarried)	1.703 (0.322-9.005)	0.531	-	-	1.698 (0.315-9.156)	0.538	-	-
BMI (<23.50 vs. ≥23.50)	1.229 (0.482-3.132)	0.666	-	-	1.192 (0.467-3.041)	0.713	-	-
Menopause (No vs. Yes)	0.498 (0.313-0.791)	0.003	0.522 (0.328-0.831)	0.006	0.481 (0.151-1.539)	0.217	-	-
US-LNM (No vs. Yes)	3.417 (1.289-9.059)	0.013	3.071 (1.629-5.791)	0.001	5.209 (1.317-20.602)	0.019	2.426 (1.321-4.453)	0.004
US-BIRADS (4 + 5 vs. 6)	1.151 (0.534-2.482)	0.719	-	-	1.680 (0.736-3.833)	0.218	-	-
Postoperative chemotherapy (No vs. Yes)	0.141 (0.039-0.507)	0.003	0.298 (0.141-0.630)	0.002	0.243 (0.077-0.764)	0.015	0.400 (0.190-0.842)	0.016
Postoperative radiotherapy (No vs. Yes)	0.571 (0.233-1.402)	0.222	-	-	0.410 (0.149-1.128)	0.084	-	-
Postoperative targeted therapy (No vs. Yes)	0.335 (0.112-1.001)	0.050	-	-	0.648 (0.217-1.937)	0.437	-	-
Clinical T stage (T1 + T2 vs. T3 + T4)	1.684 (0.467-6.072)	0.425	-	-	2.449 (0.755-7.947)	0.136	-	-
Clinical N stage (N0vs. N1 + N2 + N3)	1.991 (0.305-12.985)	0.472	-	-	1.019 (0.181-5.734)	0.983	-	-
Clinical TNM stage (I + II vs. III)	1.053 (0.120-9.264)	0.963	-	-	2.988 (0.429-20.811)	0.269	-	-
Types of surgery (mastectomy vs. breast-conserving surgery)	0.343 (0.115-1.022)	0.055	-	-	0.397 (0.135-1.172)	0.094	-	-
Histologic grade (I + II vs. III)	1.093 (0.497-2.401)	0.826	-	-	1.927 (0.880-4.220)	0.101	-	-
Tumour size (≤2 vs. >2 cm)	1.265 (0.424-3.770)	0.673	-	-	1.343 (0.463-3.899)	0.587	-	-
Pathological T stage (T1 + T2 vs. T3 + T4)	1.108 (0.472-2.605)	0.813	-	-	1.160 (0.485-2.775)	0.739	-	-
Pathological N stage (N0 vs. N1 + N2 + N3)	2.221 (1.441-3.423)	0.0003	2.611 (1.670-4.085)	<9.0001	2.336 (1.517-3.597)	0.0001	2.163 (1.414-3.310)	0.0004
Pathological TNM stage (I + II vs. III)	2.661 (0.167-42.469)	0.489	-	-	3.834 (0.177-82.880)	0.391	-	-
P/T ratio (<0.12 vs. ≥0.12)	6.984 (0.886-55.038)	0.065	-	-	5.485 (0.612-49.136)	0.128	-	-
Total lymph nodes removed (<21 vs. ≥21)	1.339 (0.664-2.700)	0.414	-	-	1.846 (0.878-3.883)	0.106	-	-
Positive lymph nodes (<2 vs. ≥2)	5.540 (0.702-43.716)	0.104	-	-	2.459 (0.265-22.868)	0.429	-	-
CK5/6 (Negative vs. Positive)	3.991 (1.626-9.798)	0.003	-	-	3.806 (1.354-10.701)	0.011	-	-
EGFR (Negative vs. Positive)	1.606 (0.494-5.217)	0.431	-	-	1.379 (0.391-4.867)	0.617	-	-
P53 (Negative vs. Positive)	1.761 (0.686-4.522)	0.239	-	-	1.614 (0.652-3.991)	0.300	-	-
TOP2A (Negative vs. Positive)	2.874 (0.928-8.899)	0.067	-	-	2.100 (0.735-6.000)	0.166	-	-
Ki-67 (Negative vs. Positive)	0.912 (0.384-2.169)	0.836	---	-	1.063 (0.427-2.646)	0.896	-	-
LVI (Negative vs. Positive)	3.412 (1.800-6.469)	<0.0001	3.352 (1.572-7.145)	0.002	2.458 (1.221-4.948)	0.012	5.203 (2.569-10.536)	<0.0001
ABO blood type (A + B vs. O + AB)	1.128 (0.536-2.370)	0.751	-	-	1.838 (0.896-3.769)	0.097	-	-
White blood cells (<5.90 vs. ≥5.90 x109/L)	2.438 (0.743-7.998)	0.142	-	-	2.739 (0.760-9.872)	0.123	-	-
Red blood cells (<4.36 vs. ≥4.36 x1012/L)	1.890 (0.821-4.349)	0.134	-	-	2.309 (0.964-5.527)	0.060	-	-
Platelets (<237.00 vs. ≥237.00 x109/L)	1.220 (0.597-2.496)	0.585	-	-	1.122 (0.526-2.393)	0.766	-	-
Haemoglobin (<131.00 vs. ≥131.00 g/L)	1.315 (0.798-2.167)	0.283	-	-	1.141 (0.690-1.884)	0.608	-	-
Neutrophils (<3.60 vs. ≥3.60 x109/L)	1.249 (0.430-3.624)	0.683	-	-	1.358 (0.441-4.181)	0.594	-	-
Lymphocytes (<1.76 vs. ≥1.76 x109/L)	0.591 (0.288-1.209)	0.150	-	-	0.698 (0.325-1.500)	0.357	-	-
CRP (<0.40 vs. ≥0.40 mg/L)	1.966 (0.858-4.506)	0.110	-	-	1.622 (0.730-3.604)	0.235	-	-
ALB (<45.00 vs. ≥45.00 g/L)	0.182 (0.053-0.621)	0.007	0.507 (0.217-1.185)	0.117	0.464 (0.242-0.891)	0.021	0.578 (0.259-1.288)	0.180
ALP (<68.00 vs. ≥68.00 U/L)	1.141 (0.527-2.471)	0.737			1.250 (0.583-2.680)	0.566
LDH (<168.00 vs. ≥168.00 U/L)	3.443 (1.427-8.309)	0.006	2.247 (1.401-3.602)	0.001	1.560 (1.021-2.383)	0.040	2.201 (1.365-3.550)	0.001
FIB (<2.88 vs. ≥2.88 g/L)	1.811 (0.569-5.764)	0.315	-	-	2.507 (0.701-8.970)	0.158	-	-
D-D (<0.37vs. ≥0.37 mg/l)	1.061 (0.505-2.228)	0.876	-	-	1.425 (0.657-3.090)	0.369	-	-
AFR (<15.0 vs. ≥15.0)	0.627 (0.397-0.990)	0.045	0.595 (0.377-0.940)	0.026	0.238 (0.061-0.927)	0.039	0.385 (0.221-0.670)	0.001
#p-value by univariate Cox regression, ##p-value by multivariate Cox regression. AFR: Albumin-to-fibrinogen ratio; DFS: Disease-free survival; OS: Overall survival; BMI: Body mass index; US: Ultrasound; LNM: Lymph node metastasis; BIRADS: Breast imaging-reporting and data system; P/T: Positive lymph nodes / Total lymph nodes; CK: Cytokeratin; EGFR: Epidermal growth factor receptor; TOP2A: Topoisomerase 2A; CRP: C-reactive protein; ALB: Albumin; ALP: Alkaline phosphatase; LDH: Lactate dehydrogenase; FIB: Fibrinogen; D-D: D-dimer; LVI: Lymphovascular invasion.

Figure 1: DFS and OS in patients with TNBC stratified by AFR. Kaplan-Meier analysis of (A) DFS and (B) OS in patients with TNBC according to low or high AFR values. Kaplan-Meier analysis of (C) DFS and (D) OS in patients with recurrence and metastasis stratified by AFR values. Kaplan-Meier analysis of (E) DFS and (F) OS in patients who received chemotherapy after surgery by AFR values. Kaplan-Meier analysis of (G) DFS and (H) OS in patients receiving radiotherapy after surgery by AFR values. DFS: Disease-free survival; OS: Overall survival; TNBC: Triple-negative breast cancer; AFR: Albumin-to-fibrinogen ratio.

Figure 2: Nomograms established for predicting survival outcomes in patients with TNBC. Nomograms for (A) DFS and (B) OS.

Figure 3: CCA assessing the performance of the nomograms for practising survival time after surgery in patients with TNBC. CCA for (A) 1-year, (B) 3-year, and (C) 5-year DFS, and for (D) 1-year, (E) 3-year, and (F) 5-year OS.

Figure 4: DCA for assessing the clinical utility of the nomograms for predicting survival time after surgery in patients with TNBC. DCA for (A) 3-year and (B) 5-year DFS, and for (C) 3-year and (D) 5-year OS.

Multivariate analysis indicated that menopause (95% CI: 0.328-0.831, HR: 0.522; p = 0.006), US-lymph node metastasis (LNM; 95% CI: 1.629-5.791, HR: 3.071; p = 0.001), postoperative chemotherapy (95% CI: 0.141-0.630, HR: 0.298; p = 0.002), pathological N stage (95% CI: 1.670-4.085, HR: 2.611; p <0.0001), LVI (95% CI: 1.572-7.145, HR: 3.352; p = 0.002), LDH (95% CI: 1.401-3.602, HR: 2.247; p = 0.001), and AFR (95% CI: 0.377-0.940, HR: 0.595; p = 0.026) were found to be potential prognostic factors for the DFS. Ultrasound-lymph node metastasis (US-LNM; 95% CI: 1.321-4.453, HR: 2.426; p = 0.004), postoperative chemotherapy (95% CI: 0.190-0.842, HR: 0.400; p = 0.016), pathological N stage (95% CI: 1.414-3.310, HR: 2.163; p = 0.0004), LVI (95% CI: 2.569-10.536, HR: 5.203; p <0.0001), LDH (95% CI: 1.365-3.550, HR: 2.201; p = 0.001), and AFR (95% CI: 0.221-0.670, HR: 0.385; p = 0.001) were found to be potential prognostic factors for the OS (Table II).

According to multivariate Cox regression analysis, AFR had prognostic significance for both DFS and OS. For DFS, the hazard ratios were 0.627 (95% CI: 0.397-0.990; p = 0.045) and 0.595 (95% CI: 0.377-0.940; p = 0.026), respectively. For OS, the hazard ratios were 0.238 (95% CI: 0.061-0.927; p = 0.039) and 0.285 (95% CI: 0.221-0.670; p = 0.001), respectively. The median survival time of patients with high AFR values was higher than that of those with low AFR values (DFS: 36.54 vs. 33.19 months, p = 0.0042; OS: 56.45 vs. 53.29 months, p = 0.0031, respectively; Figure 1A and B). Among the 93 patients who had recurrence and metastasis, those with high AFR values also had a higher median survival time than those with low AFR values (DFS: 29.30 vs. 23.30 months, p = 0.0024; OS: 63.47 vs. 56.30 months, p = 0.0016, respectively; Figure 1C and D). In addition, 137 patients received chemo-therapy after surgery. For these patients, the median survival time of patients with high AFR values was higher than that of patients with low AFR values (DFS: 39.70 vs. 22.80 months, p = 0.0035; OS: 59.17 vs. 52.67 months, p = 0.0039, respectively; Figure 1E and F). Furthermore, 138 patients received radiotherapy after surgery, and among these patients, the median survival time of patients with high AFR values was also higher than that of those with low AFR values (DFS: 34.27 vs. 28.97 months, p = 0.0058; OS: 57.43 vs. 51.23 months, p = 0.0018, respectively; Figure 1G and H).

The variables identified by multivariate Cox analyses— menopause status, US-LNM, postoperative chemotherapy, pathological N stage, LVI, LDH, and AFR—were used to generate a DFS-predicting nomogram. Moreover, US-LNM, postoperative chemotherapy, pathological N stage, LVI, LDH, and AFR were applied to generate an OS-predicting nomogram (Figure 2).

CCA indicated a good agreement for 1-, 3-, and 5-year DFS and OS between predicted and actual probability (Figure 3). Furthermore, DCA indicated that the constructed nomogram models had an improved predictive clinical utility compared with AFR alone within the threshold probability (Figure 4).

DISCUSSION

Systemic inflammation has been recognised as a critical hallmark of tumours and is associated with tumour angiogenesis, recurrence, and metastasis.⁸ A low preoperative nutritional status has been associated with a prolonged hospitalisation after surgery.⁹ However, the potential mechanisms associated with systemic inflammation and nutritional status remain poorly understood. Although genetic markers, the TNM system, and histological subtypes have been used to predict survival, the same histological subtype or TNM stage can have significant heterogeneity, which greatly hampers their clinical application.¹⁰

ALB is involved in the maintenance of plasma colloidal osmotic pressure and the metabolism of toxic substances. Serum ALB plays a critical role in both acute and chronic inflammation in human body and acts as a critical nutritional reference index for patients with cancer.¹¹ A previous study showed that, in patients with endometrial cancer, low preoperative serum ALB levels were associated with postoperative complications and increased mortality.¹² In addition, Saito et al.¹³ reported that postoperative ALB level was a useful predictor of prognosis in older patients with gastric cancer. FIB is a clotting protein in the blood that is converted into fibrin and serves an important role in thrombosis.¹⁴ Moreover, FIB may be expressed abnormally in coagulation-associated diseases and is increased under certain conditions. Studies have also indicated that FIB may be a critical inflammatory regulator associated with malignant tumour cell metastasis, invasion, and proliferation.¹⁵ In a study performed by Hong et al.,¹⁶ a high level of FIB in blood predicted poor prognosis in patients with colon or rectal cancer. In the present study, although ALB and FIB were found to be relevant to treatment outcomes and prognosis in patients with breast cancer, the combined biomarker AFR provided a superior reflection of prognosis. Patients with low AFR values have been reported to exhibit higher mortality and recurrence rates across multiple malignancies.¹⁷ However, few studies have investigated the relationships between AFR and breast cancer, particularly in TNBC. In a study by Hwang et al.,¹⁸ breast cancer patients with a high preoperative FIB- to-ALB ratio had worse prognoses than those with a low ratio. Notably, the prognostic effect of FIB- to-ALB ratio was more prominent compared with that of either single marker alone. Besides, another study demonstrated that AFR was an independent predictor of oncological outcomes in patients with invasive ductal carcinoma of the breast.¹⁹

The present study’s results also indicated that AFR was a powerful and significant potential predictor. Using Cox regression and survival analyses, an independent association between AFR levels and DFS/OS was confirmed. Using the optimal cut-off value for AFR, a high AFR was corrected to several improved factors, including clinical T stage, histologic grade, positive lymph nodes, EGFR, TOP2A, ABO blood type, red blood cells, ALB, ALP, LDH, FIB, and D-D.

Through multivariate Cox analysis, factors such as US-LNM, postoperative chemotherapy, pathological N stage, LVI, and LDH were also found to be significant predictors for DFS and OS. A previous study reported that AFR is associated with tumour type and TNM stage and may act as a potential biomarker for evaluating prognosis in patients undergoing surgery for renal cell carcinoma.²⁰ LVI is commonly assessed to evaluate clinical and pathological outcomes in breast cancer.²¹ In a study by Kurozumi et al.,²² LVI was shown to be associated with the development and metastasis of invasive breast cancer, as well as with a specific transcriptomic profile. In addition, LDH has been found to be highly expressed in some tumours and plays a role in regulating glycolysis.²³

Some potential mechanisms have previously been described that may explain the prognostic value AFR in breast cancer. FIB has been indicated to mediate cell proliferation by binding to tumour cell surfaces, thereby also increasing cancer cell metastasis, while ALB is associated with the immune status of tumours and its expression is selectively inhibited by tumour necrosis factor-α.²⁴ In addition, a meta-analysis showed that a low AFR is relevant with increased cancer-related mortality and recurrence risk.²⁵

It is worth noting that this study had some constraints that should be considered. First, it was performed as a retrospective study; therefore, prospective studies are needed to further investigate the prognostic value of AFR. Second, the sample size was relatively small. Finally, heterogeneity among patients with breast cancer after surgery could contribute to a difference in clinical prognosis. Hence, additional research is necessary to externally verify the prognostic role of AFR.

CONCLUSION

The present study revealed that TNBC patients with a high AFR value had a lower rate of metastasis and improved survival rate compared with those who had a low AFR value. AFR is easy and inexpensive to determine and is potentially a predictive factor for preoperative evaluation in clinical practice.

ETHICAL APPROVAL:
The study was approved by the Ethics Committee of the Cancer Hospital and conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

PATIENTS’ CONSENT:
Informed consent was obtained from all the patients before the procedure.

COMPETING INTEREST:
The authors declared no conflict of interest.

AUTHORS’ CONTRIBUTION:
SW: Conception, study design, drafting, and critically revision of the manuscript.
WL: Data acquisition, data analysis, data interpretation, and drafting.
All authors approved the final version of the manuscript to be published.

Annexure 1: ROC analysis of AFR.

REFERENCES

1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer Statistics, 2021. CA Cancer J Clin 2021; 71(1):7-33. doi: 10.3322/ caac.21654.
2. Nakhjavani M, Hardingham JE, Palethorpe HM, Price TJ, Townsend AR. Druggable molecular targets for the treatment of triple negative breast cancer. J Breast Cancer 2019; 22(3):341-61. doi: 10.4048/jbc.2019.22.e39.
3. Yin L, Duan JJ, Bian XW, Yu SC. Triple-negative breast cancer molecular subtyping and treatment progress. Breast Cancer Res 2020; 22(1):61. doi: 10.1186/s130 58-020-01296-5.
4. McMillan DC. The systemic inflammation-based Glasgow prognostic score: A decade of experience in patients with cancer. Cancer Treat Rev 2013; 39(5):534-40. doi: 10. 1016/j.ctrv.2012.08.003.
5. Falanga A, Marchetti M, Vignoli A. Coagulation and cancer: Biological and clinical aspects. J Thromb Haemost 2013; 11(2):223-33. doi: 10.1111/jth.12075.
6. Ho JW, Wu AH, Lee MW, Lau SY, Lam PS, Lau WS, et al. Malnutrition risk predicts surgical outcomes in patients undergoing gastrointestinal operations: Results of a prospective study. Clin Nutr 2014; 34(4):679-84. doi: 10. 1016/j.clnu.2014.07.012.
7. Huang L, Mo Z, Zhang L, Qin S, Qin S, Li S. Diagnostic value of albumin to fibrinogen ratio in cervical cancer. Int J Biol Markers 2020; 35(2):66-73. doi: 10.1177/1724600 820915916.
8. Iyengar NM, Gucalp A, Dannenberg AJ, Hudis CA. Obesity and cancer mechanisms: Tumor microenvironment and inflammation. J Clin Oncol 2016; 34(35):4270-6. doi: 10. 1200/JCO.2016.67.4283.
9. Arends J, Bachmann P, Baracos V, Barthelemy N, Bertz H, Bozzetti F, et al. ESPEN guidelines on nutrition in cancer patients. Clin Nutr 2017; 36(1):11-48. doi: 10.1016/j.clnu. 2016.07.015.
10. Bagante F, Spolverato G, Beal E, Merath K, Chen Q, Akgul O, et al. Impact of histological subtype on the prognosis of patients undergoing surgery for colon cancer. J Surg Oncol 2018; 117(7):1355-63. doi: 10.1002/jso.25044.
11. Deme D, Telekes A. [Prognostic importance of albumin in oncology]. Orv Hetil 2018; 159(3):96-106. doi: 10.1556/ 650.2018.30885.
12. Lei J, Wang Y, Guo X, Yan S, Ma D, Wang P, et al. Low preoperative serum ALB level is independently associated with poor overall survival in endometrial cancer patients. Future Oncol 2020; 16(8):307-16. doi: 10.2217/fon-2019- 0732.
13. Saito H, Kono Y, Murakami Y, Shishido Y, Kuroda H, Matsunaga T, et al. Postoperative serum albumin is a potential prognostic factor for older patients with gastric cancer. Yonago Acta Med 2018; 61(1):72-8. doi: 10.33160/yam. 2018.03.010.
14. Bai Y, Shi M, Yang X, Zhang W, Yang R, Wei X, et al. The value of FDP/FIB and D-dimer/FIB ratios in predicting high- risk APL-related thrombosis. Leuk Res 2019; 79:34-7. doi: 10.1016/j.leukres.2019.02.007.
15. Kavitha S, Sridhar MG, Satheesh S. Periprocedural plasma fibrinogen levels and coronary stent outcome. Indian Heart J 2015; 67(5):440-3. doi: 10.1016/j.ihj. 2015.06.006.
16. Hong T, Shen D, Chen X, Wu X, Hua D. Preoperative plasma fibrinogen, but not D-dimer might represent a prognostic factor in non-metastatic colorectal cancer: A prospective cohort study. Cancer Biomark 2017; 19(1): 103-11. doi: 10.3233/CBM-160510.
17. Liang Y, Wang W, Que Y, Guan Y, Xiao W, Fang C, et al. Prognostic value of the fibrinogen/albumin ratio (FAR) in patients with operable soft tissue sarcoma. BMC Cancer 2018; 18(1):942. doi: 10.1186/s12885-018-4856-x.
18. Hwang KT, Chung JK, Roh EY, Kim J, Oh S, Kim YA, et al. Prognostic influence of preoperative fibrinogen to albumin ratio for breast cancer. J Breast Cancer 2017; 20(3): 254-63. doi: 10.4048/jbc.2017.20.3.254.
19. Zheng L, Zhao Y, Liu F, Liu P, Li W, Yang Y, et al. Prognostic significance of preoperative albumin to fibrinogen ratio associated nomograms in patients with breast invasive ductal carcinoma. Medicine (Baltimore) 2020; 99 (26):e20681. doi: 10.1097/MD.0000000000020 681.
20. Ma C, Liu Q, Li C, Cheng J, Liu D, Yang Z, et al. Novel blood indicators of progression and prognosis in renal cell carcinoma: Red cell distribution width-to-lymphocyte ratio and albumin-to-fibrinogen ratio. J Oncol 2020; 2020:2895150. doi: 10.1155/2020/2895150.
21. Zhang S, Zhang D, Gong M, Wen L, Liao C, Zou L. High lymphatic vessel density and presence of lymphovascular invasion both predict poor prognosis in breast cancer. BMC Cancer 2017; 17(1):335. doi: 10.1186/s12 885-017-3338-x.
22. Kurozumi S, Joseph C, Sonbul S, Alsaeed S, Kariri Y, Aljohani A, et al. A key genomic subtype associated with lymphovascular invasion in invasive breast cancer. Br J Cancer 2019; 120(12):1129-36. doi: 10.1038/s414 16-019-0486-6.
23. Zhou Y, Niu W, Luo Y, Li H, Xie Y, Wang H, et al. p53/Lactate dehydrogenase A axis negatively regulates aerobic glycolysis and tumor progression in breast cancer expressing wild-type p53. Cancer Sci 2019; 110(3): 939-49. doi: 10.1111/cas.13928.
24. Mantzorou M, Koutelidakis A, Theocharis S, Giaginis C. Clinical value of nutritional status in cancer: What is its impact and how it affects disease progression and prognosis? Nutr Cancer 2017; 69(8):1151-76. doi: 10.1080/016 35581.2017.1367947.
25. Sun DW, An L, Lv GY. Albumin-fibrinogen ratio and fibrinogen-prealbumin ratio as promising prognostic markers for cancers: An updated meta-analysis. World J Surg Oncol 2020; 18(1):9. doi: 10.1186/s12957-020-1786-2.