Background: Triple Negative Breast Cancer (TNBC) is a heterogeneous disease representing about 15% of all Breast Cancers (BC). TNBC are usually high-grade histological tumors, characterized by their aggressiveness and difficulty of treatment due to the lack of available targeted therapies. There is a strong association between achieving Pathological Complete Response (pCR) after neoadjuvant treatment and higher rates of Disease-Free Survival (DFS) and Overall Survival (OS).

Objectives: This prospective, observational, descriptive and analytical study attempts to identify new soluble immune biomarkers in early TNBC patients who received Neoadjuvant Chemotherapy (NCT) before surgery. Baseline histopathological and plasma biomarkers were analyzed and correlated with clinical features to assess their predictive and prognostic value.

Results: The expressions of Ki-67, EGFR, HER2 low and androgen receptors did not prove to have a predictive or prognostic value in early TNBC. An increase in the percentage of tumor-infiltrating lymphocytes (TILs) was correlated with better DFS (p-value 0.0047) and OS (p-value 0.031). Elevated plasma levels of certain cytokines and immune checkpoints like sGal-3 (p-value 0.03) and sGM-CSF (p-value 0.02), and lower levels of sCD86 (p-value 0.02) and sITAC (p-value 0.02), predicted poorer NCT response. Increased levels of sGal-3 (p-value 0.049), sMIP-3α (p-value 0.005), sIL-10 (p-value 0.002), sIL-13 (p-value 0.009), sIL-17A (p-value 0.049), and sTIM-3 (p-value 0.0049) were significantly associated with worse survival outcomes.

Conclusion: In early TNBC, diverse plasma cytokines and immune checkpoints were associated with a worse response to NCT and poor survival. The implementation of liquid biopsy analysis for the determination of soluble biomarkers could help for treatment planning and future therapeutic strategies.

Triple-Negative Breast Cancer (TNBC) represents approximately 15% of all BC [1]. TNBC is defined by ≤ 1% expression of estrogen and progesterone receptors and absence of Human Epidermal Growth Factor Receptor 2 (HER2) overexpression or amplification, typically presents with high histological grade, a high proliferation index, and abundant necrotic tissue [2]. More prevalent in women under 40 years, and often associated with germinal BRCA1/2 mutations (15-20%). TNBC exhibits aggressive biological behavior with higher rates of distant metastases, especially to the lungs and central nervous system, and a more aggressive clinical course with early recurrences (usually within 2-3 years) and higher mortality compared to hormone-dependent tumors [1,3]. Despite high response rates to Chemotherapy (CT), these responses are often short-lived due to early development of resistance mechanisms. The 5-year relative survival rates are 91% for localized TNBC, 65% for locally advanced TNBC, and only 11% for metastatic disease, highlighting the need to optimize treatments to minimize recurrences and improve survival [1-4]. TNBC consists of various subtypes with distinct biology, elucidated by histological and molecular classifications however similar diagnostic and therapeutic approaches [5,6].

The systemic treatment of early or locally advanced TNBC typically involves a combination of CT, surgery, and Radiotherapy (RT). Treatment decisions are personalized and made by a multidisciplinary team, considering tumor characteristics, patient age, comorbidities, and patient preferences. Similar outcomes concerning local control and survival have been obtained with preoperative CT (neoadjuvant) or adjuvant CT (postoperative). Neoadjuvant chemotherapy (NCT) before surgery helps increase resectability, achieves early control of micrometastatic disease, and assesses tumor chemosensitivity. Achieving a pathological complete response (pCR) after NCT is associated with improved prognosis and survival [7,8]. Various CT combinations, including anthracyclines, taxanes, and platinum salts, have been investigated to enhance pCR rates, with carboplatin and nab-paclitaxel showing notable efficacy [9,10]. Recent studies have explored the addition of immunotherapy to NCT, revealing promising results in improving pCR rates and survival outcomes, such us pembrolizumab or atezolizumab [11-14]. Additional adjuvant therapy, such as capecitabine or olaparib in gBRCA-mutated might be beneficial in patients who do not reach pCR [15,16]. Despite recent progress in treatment, further research with prospective trials based on specific immune biomarkers is needed to refine treatment approaches and improve outcomes in early TNBC.

The immune system plays a critical role in cancer eradication, with the potential to both destroy tumor cells specifically and create long-term immune memory to prevent recurrences. However, it can also provide a favorable microenvironment for tumor growth, leading to the dual functions of some immune molecules [17]. Tumor microenvironment is a complex ecosystem composed of cancer cells, inflammatory cells, stromal cells, and extracellular matrix, whose dynamic interactions dictate tumor development conditions. In TNBC, the levels of Tumor-Infiltrating Lymphocytes (TILs) have been consistently associated with implications for immunotherapy response [18-21].

Immune cells produce various molecules collectively termed soluble mediators of immunity. These include antibodies, cytokines, and other substances like complement, which act in inflammatory processes. Cytokines, low molecular weight proteins, play crucial roles in intercellular communication between immune cells and tumor stromal cells, serving functions in both innate and adaptive immunity. While pivotal in inflammatory processes, cytokines also regulate tumor progression processes such as proliferation, survival, cell migration, angiogenesis, epithelial-mesenchymal transition, activation of inflammatory cells, or promotion of cancer cell evasion from immune response. Their role in the tumor context is complex, as they can also exert antitumor actions depending on various factors in the tumor microenvironment [22-27]. The main types of cytokines are: Interleukins (IL), Interferons (IFN), tumor necrosis factors (TNF), Colony-Stimulating Factors (CSF), and Transforming Growth Factor (TGF) [28,29].

In addition to cytokines, immune mediators in tumor microenvironment include PD-1, PD-L1, PD-L2, CTLA-4, and LAG3, among others. Recent studies suggest that soluble forms, likely secreted by immune cells, influence immune regulation, cancer development, prognosis, and treatment response [30]. Liquid biopsy can analyze immune mediators or soluble immune biomarkers in the bloodstream, whether secreted by tumor cells or immune cells, which play roles in various biological functions and influence both innate and adaptive immune responses, Further research is needed to identify reliable biomarkers for improving therapy selection and prognostic assessment in TNBC. The study's main objective is to identify biomarkers by extensively characterizing the tumor microenvironment in TNBC patients undergoing NCT, with the goal of improving understanding of the disease and predicting treatment outcomes. This will be achieved through clinical-pathological characterization of the tumor, and lymphocytic infiltration to identify predictive and prognostic soluble biomarkers.

Study Design

A prospective, observational, non-interventional, cross-sectional, descriptive and analytical study was carried out in patients with early TNBC treated with NCT. The study included a total of 42 patients from the General University Hospital of Valencia (HGUV) and the Requena General Hospital between January 2016 and October 2021, with an average follow-up of 42 months. The study was approved by the HGUV Ethics and Research Committee (No. 7/2017) and was carried out in compliance with the Declaration of Helsinki and Spanish legislation on biomedical research and data protection. All participants signed an informed consent for sample acquisition for research purposes before the beginning of this study, ensuring confidentiality and ethical management of personal and clinical data.

Inclusion Criteria: Patients over 18 years of age diagnosed with localized or locally advanced -stages I, II and III- TNBC who had not initiated treatment with CT prior to the biopsy and analysis.

Exclusion Criteria: Patients under 18 years of age, those who refused to participate or could not understand the informed consent, patients with initial metastatic disease, those who did not receive NCT, and patients with a history of previous cancer.

Procedures and Measurements

Among the characteristics of the cohort of 42 patients with TNBC included in the study, we distinguish between baseline characteristics and post-surgical characteristics. Ultrasoundguided Core Needle Biopsies (CNB) were obtained from breast tissue and lymph nodes in cases of suspected involvement. Epidemiological and clinical variables were collected for each patient using LibreOffice and SPSS®. Epidemiological variables included date of birth, age, sex, performance status, menopausal status, and dates of initial diagnosis by CNB, start of treatment, and surgical intervention. Clinical and pathological variables of the tumor included tumor size, lymph node involvement, clinical stage, NCT regimen, type of surgical intervention, post-surgical tumor size, post-surgical node involvement, post-surgical pathological stage, adjuvant CT, adjuvant RT, details of tumor relapse (date and location), and death.

Patients received different NCT regimens (Table 1) followed by restaging using ultrasound and breast Magnetic Resonance Imaging (MRI) before surgery. Surgical techniques included conservative surgery or total mastectomy and RT, depending on response. In cases where no affected lymph nodes were initially detected, a selective sentinel lymph node biopsy was performed, and if results showed malignancy, a complete axillary lymphadenectomy was carried out. Post-surgical samples were analyzed by pathologists to evaluate the response to NCT using the Miller & Payne system.

Subsequently, the administration of adjuvant capecitabine was considered for patients without pCR or with a high risk of recurrence. Adjuvant RT was applied based on tumor and lymph node characteristics, with dosages adjusted accordingly. Follow-up frequency varied based on the pathological response obtained and individual patient needs, generally every 2-4 months during the first two years, and then every 6 months until the fifth post-surgical year. Imaging tests and additional analyses were requested to see if there were any signs or symptoms of recurrence.

Table 1: NCT schemes in the TNBC cohort of patients.

Histopathological Studies: The histopathological evaluation of initial biopsy included analysis of tumor histology, histological grade, percentage of Ki-67, expression of ER (estrogen receptor), PR (progesterone receptor), RA (androgen receptor), HER2, EGFR (Epidermal Growth Factor Receptor), and location of the tumor lymphocytic infiltrate (CD4+ and CD8+) (Table 2). A detailed IHC technique and hematoxylin-eosin staining were used for sample preparation and examination. The semiquantitative result was obtained using the following criteria: 0, negative (< 1% of stained cells); 1, weak (1-10% staining); 2, intermediate (10-50% staining); and 3, intense (> 50% staining). The semiquantitative result was applied to the expressions of ER, PR, AR, and EGFR. EGFR expression was considered positive when more than 1% of tumor cells were stained. Regarding the expression and/or amplification of HER2, it was considered positive by IHC when more than 10% of neoplastic cells showed intense and uniform membrane staining (3+) in more than 10% of neoplastic cells. By Fluorescence in Situ Hybridization (FISH), positive was defined as the identification of 6 or more copies of the gene per nucleus. HER2 was identified as negative by IHC when there was an absence of staining or weak, incomplete membrane expression in less than 10% of tumor cells (0+). The density of TILs was determined based on the recommendation of the International TIL Working Group [31], selecting both the intratumoral and peritumoral areas. The percentage of all mononuclear cells, including lymphocytes in the stromal area, was defined as TILs, excluding granulocytes and other polymorphonuclear leukocytes. TILs cut-off point was considered 30% (median TILs in the study). A high expression of TILs was considered when the quantification of intra- and peritumoral lymphocytes exceeded 30% of the analyzed tissue.

Immunoassay in Liquid Biopsy Samples: Peripheral blood samples were collected in EDTA-containing tubes before surgery, and plasma was separated by centrifugation (4000 rpm, 10min, 4° C) within 2 hours after blood collection. Plasma levels of soluble mediators such as cytokines and soluble immune checkpoints were analyzed using Luminex xMap technology (Luminex Corp, Austin, TX, USA), allowing detailed evaluation of multiple analytes simultaneously. A total of 39 analytes were measured with different commercial kits; MILLIPLEX® Human High Sensitivity T Cell Panel-Immunology Multiplex Assay, Cat #HSTCMAG-28SK; MILLIPLEX®Human Immuno-Oncology Checkpoint ProteinPanel 2-Immuno-Oncology Multiplex Assay, Cat #HCKP2-11K; and MILLIPLEX® Human Circulating Cancer Biomarker Magnetic Bead Panel 3 - Cancer Multiplex Assay, Cat #HCCBP3MAG-1K Kit. The analysis was conducted following the manufacturer’s recommendations using the Luminex 100-200® analyzer (Luminex Corp), a flow cytometry-based instrument integrating key detection components such as lasers, optics, advanced fluidics, and high-speed digital signal processors. Data acquisition and analysis were performed using BelysaTM software version 1.2 (Merck Millipore, Billerica, MA, USA). Cytokine concentrations were measured as mean fluorescence intensity. A 5-parameter standard curve based on 7 standard concentrations was used to convert optical density values into concentrations (pg/ml). Data for a minimum of 50 beads per cytokine were collected for each standard and sample. Values below the detectable limit were assigned the minimum value, and values above the detectable maximum were assigned the maximum value [32]. The median of soluble analyte levels was used as the cutoff point for categorization and subsequent statistical analysis.

Germline Genetic Analysis: Germline BRCA1/2 testing was conducted in patients under 50 years old and/or with a personal or family history indicative of hereditary breast and ovarian cancer syndrome. The genes analyzed by next-generation sequencing (NGS) included ATM, BRCA1, BRCA2, BRIP1, CDH1, CHEK2, EPCAM, MLH1, MSH2, MSH6, NBN, PALB2, PMS2, PTEN, RAD51C, RAD51D, STK11, and TP53.

Statistical Analysis

Kolmogorov Smirnov tests were carried out to evaluate the normality of the variables, applying non-parametric or parametric tests as appropriate. The comparison of median soluble levels of all tested biomarkers was conducted using non-parametric Mann-Whitney U test (two groups) and KruskalWallis (more than two groups) and parametric T student test (two groups) or ANOVA (more than two groups), to compare continuous variables. The Kaplan-Meier method was used to determine Disease-Free Survival (DFS) and Overall Survival (OS). Comparison between survival curves was performed using longrank contrast method. Hazard Ratio (HR) and 95% confidence interval (95%CI) were calculated using the univariate Cox proportional hazards regression model. Multivariate analysis was performed using Cox proportional hazards model (DFS and OS). The analysis was performed using R statistical software. P-values were considered significant if < 0.05.

Table 2: Characteristics of the antibodies used for IHC staining of the different markers in tumor tissue.

Characteristics of Patient’s Cohort with TNBC

Baseline Clinicopathological Characteristics: The median age at diagnosis was 55 years old [28-77]. All patients had good or acceptable performance status (PS 0-1). Most of them were peri- or postmenopausal (66.6%). Regarding the tumor, the mean size was 34.7 mm, with T2 being the most common; and 61.9% had lymph node involvement at the onset of the disease. Most patients were diagnosed at stage IIB (40.5%), followed by stage E-IIIC (31%) (Table 3).

Out of the 23 patients who underwent germline analysis of the BRCA1/2 gene, 6 cases of mutation in BRCA1 and 1 case in BRCA2 were detected.

Table 3: Baseline clinicopathological characteristics of the study patients, which include age, functional status, menopausal status, baseline CA 15-3, baseline LDH, tumor size, lymph node involvement, prognostic stage, and germline BRCA1/2.

Histopathological Baseline Tumor Biopsy Characteristics: Most of the histology corresponded to Non-Special Type (NST) in 95.2% of cases, followed by apocrine carcinoma in 4.8%. The most frequent histological grade was poorly differentiated (G3) (71.4%), followed by moderately differentiated (G2) (26.2%), with only one patient presenting a well-differentiated tumor (G1). The average proliferation index by IHC through nuclear staining of Ki-67 antigen was 49% [15-99%]. HER2-low expression was detected in 21.4% of the cohort by IHC. EGFR expression was positive in 73.8% of patients. Regarding AR expression, 5 out of 42 patients analyzed (11.9%) showed ≥ 1% positivity for nuclear staining of neoplastic cells (Table 4).

Table 4: Baseline anatomical-pathological characteristics of the study patients, which include histology, histological grade, and Ki-67 index, HER2, EGFR and AR.

NST; Non-Special Type; Ki-67: marker of proliferation Kiel 67; HER2: Human Epidermal Growth Factor Receptor 2; FISH: Fluorescence In Situ Hybridization; EGFR: Epidermal Growth Factor Receptor; AR: Androgen Receptor.

Analysis of TILs through hematoxylin-eosin staining revealed that 47.6% of patients had a high TIL percentage (defined as lymphocytic infiltration exceeding 30% of the analyzed tissue). The average TIL percentage was 30.7%, with variability ranging from 1% to 90%. Most lymphocytic infiltrates were located at the tumor periphery (around 68% of TILs [20-90%]), with only 32% intratumoral infiltrates [10-80%]. Further analysis showed that most lymphocytes were CD4+, representing 70% of total TILs with variation ranging from 50% to 85%, while only 30% were CD8+ [15-50%]. Due to the limited sample size, further determinations for a more detailed characterization could not be performed (Table 5) (Figure 1).

Figure 1: Histological images of TNBC at 40x magnification (left) and 200x magnification (right), showing a high pattern of tumor lymphocytic  infiltration (greater than 30%).

Table 5: Mean expression and range of TILs, as well as the proportions in the distribution of peri/intratumoral TILs and CD4+/CD8+ ratios in patients with early-stage TNBC.

Soluble Mediators Assessment

A total of 39 soluble biomarkers were analyzed from baseline blood samples taken prior to NCT infusion. 22 cytokines analyses were conducted in plasma (Supplementary Table S1). Among the results, it is noteworthy that the highest levels detected through multiparametric immunoassay were sGal-3 (median of 3205.6 pg/ml, range 4.23-16919.26 pg/ml), sIL-23 (median of 128.7 pg/ml, range 9.67-576.78 pg/ml), and fractalkine or sCX3CL1 (median of 102.91 pg/ml, range 9.56-232.98 pg/ml) (Figure 2).

Figure 2: Representation of the median plasma levels measured in pg/ml of the various cytokines analyzed, with notably elevated levels of sGal-3.

Additionally, 17 immune checkpoint markers in plasma were analyzed (Supplementary Table S2). Notably, significantly elevated levels of sLAG-3 (median of 268384.2 pg/ml, range 52323-829873 pg/ml) were observed compared to other analytes, as indicated by multiparametric analysis (Figure 3). Other known biomarkers, such as PD-1 (median of 2653.69 pg/ ml, range 207.74-15151.58 pg/ml) and PD-L1 (median of 529.29 pg/ml, range 89.9-3496.6 pg/ml), showed modest levels in plasma.

Figure 3: Representation of the median plasma levels measured in pg/ml of the analyzed immune checkpoint markers, with notably elevated levels of sLAG-3.

Post-Surgical TNBC Patient Cohort Characteristics: The median age at the time of surgical intervention was 56 years [28- 77]. Out of the 42 surgeries performed, 27 patients underwent breast-conserving surgery (tumorectomy or quadrantectomy), while 15 underwent radical mastectomy. Complete axillary lymphadenectomy was performed in 21 patients (50% of the sample). Surgical specimen analysis revealed 31% pCR and 52.4% partial pathological responses, with only one case (2.4%) showing local progression during neoadjuvant therapy. Local pCR (G5) was observed in 40.5% of cases, with good local responses to NCT (G3 and G4) in 28.6% of cases (Table 6).

Table 6: Post-surgical characteristics of the study patients, which include pathological response (local and lymph node), adjuvant radiotherapy, adjuvant chemotherapy, relapse and death.

Regarding nodal pathological response, the majority (38.1%) showed negative lymph nodes with no chemotherapyrelated changes (A), followed by negative lymph nodes with post-chemotherapy changes (D) in 33.3% of cases. Post-NCT lymph node involvement (B and C) was observed in 28.6% of cases. Most patients received adjuvant RT (90.5%), with only 28.6% receiving adjuvant CT with capecitabine. The low rate of adjuvant CT administration, with only 1 in 3 patients receiving it, is likely due to the lack of data supporting its benefit until late 2017, when it became a therapeutic standard. Up to the follow-up cutoff (median follow-up of 68.5 months), a total of 13 disease recurrences were detected (31% of the total). The most common site of recurrence was the lung (7/13), followed by lymph nodes (5/13), liver (4/13), bones (4/13), local (3/13), and cerebral (3/13). 10 deaths occurred, representing 23.8% of the study cohort, with the cause of death in all cases being the underlying tumor.

Correlation Analysis

Correlation Analysis between Clinical Variables and Analytes: In our study, it was observed that tumors larger than 5cm (≥ T3) had higher levels of sGal-3 (> 3205.6 pg/ml, p-value 0.049), sMIP-3α (> 17 pg/ml, p-value 0.028), and sIL-8 (> 3.55 pg/ml, p-value 0.02) compared to tumors smaller than 5cm. Additionally, an increase in plasma levels of sMIP-1α (> 10.3 pg/ml, p-value 0.024), sPD-1 (> 2653.69 pg/ml, p-value 0.044), sLAG-3 (> 225737.54 pg/ml, p-value 0.035), sTIM-3 (> 3135.21 pg/ml, p-value 0.22), and sBTLA (> 2841.9 pg/ml, p-value 0.022) was associated with tumors larger than 2 cm (≥ T2). Patients with initial nodal involvement showed significantly higher levels of sTNFα (> 4.36 pg/ml, p-value 0.004) and sGal3 (3205.6 pg/ ml, p-value 0.01). This association was more evident with sGal-3, as higher levels of affected nodes correlated with higher plasma levels of sGal-3.

In the correlation analysis of tumor stage with the studied analytes, patients with locally advanced tumors (stage III) had lower plasma levels of sITAC (< 18.08 pg/ml, p-value 0.006) compared to localized tumors (stages I and II), with sITAC being the only soluble mediator showing lower levels with higher stage. However, higher levels of sGal-3 (> 3205.6 pg/ml, p-value 0.013), sIL-8 (> 3.55 pg/ml, p-value 0.02), sPD1 (> 2653.69 pg/ml, p-value 0.012), sCTLA-4 (> 46.58 pg/ml, p-value 0.042), sCD80/ B7-1 (> 636.54 pg/ml, p-value 0.032), sTIM-3 (> 3135.21 pg/ml, p-value 0.005), sBTLA (> 2841.9 pg/ml, p-value 0.007), sTLR2 (> 5880.16 pg/ml, p-value 0.034), sICOS (> 4913.88 pg/ml, p-value 0.004), sCD27 (> 1322.54 pg/ml, p-value 0.044), sCD28 (> 4769.95 pg/ml, p-value 0.007), and sGITRL (> 1008.42 pg/ ml, p-value 0.03) were associated with locally advanced stages (stages IIIB and IIIC) versus localized stages (stages I and II).

Correlation Analysis between Pathological Variables and Analytes: Poorly differentiated tumors (G3) were correlated with a significant increase in sPD-L2 levels (> 12627.9 pg/ml, p-value 0.039) compared to well (G1) or moderately differentiated (G2) tumors. Regarding the Ki-67 proliferation index measured by IHC, an expression of Ki-67 ≥ 50% was associated with higher levels of sGM-CSF (> 11.84 pg/ml, p-value 0.034), sIL-1β (> 8.44 pg/ml, p-value 0.008), and sIL-23 (> 128.7 pg/ml, p-value 0.006). In the correlation study of HER2 expression by IHC with different analytes, higher plasma levels of sPD-L2 (> 12627.9 pg/ ml, p-value 0.0016) and sTIM-3 (>3135.21 pg/ml, p-value 0.022) were observed in patients with low HER2 expression (HER2 1+ by IHC and HER2 2+ by IHC, but with non-amplified FISH) compared to patients who were HER2 negative (HER2 null or zero).

High expression of TILs, considered as 30% or more of the tumor biopsy occupied by lymphocytes, was correlated with lower plasma levels of sIL-6 (< 1.18 pg/ml, p-value 0.016). An increase in the ratio of peritumoral lymphocytes to intratumoral lymphocytes has been correlated with several soluble immune checkpoint markers associated with worse clinical characteristics, such as: sPD-1 (> 2653.69 pg/ml, p-value 0.031), sCTLA-4 (> 46.58 pg/ml, p-value 0.023), sCD80/B7-1 (> 636.54 pg/ml, p-value 0.03), sCD86/B7-2 (> 656.2 pg/ml, p-value 0.001), sLAG-3 (> 225737.54 pg/ml, p-value 0.007), sBTLA (> 2841.9 pg/ml, p-value 0.03), sTLR-2 (> 5880.16 pg/ml, p-value 0.014), sCD28 (> 4769.95 pg/ml, p-value 0.023), and sGITRL (> 1008.42 pg/ml, p-value 0.014). On the other hand, a higher proportion of CD4+ lymphocytes compared to CD8+ has been associated with lower plasma levels of some immune checkpoint markers, such as: sPD-1 (< 2653.69 pg/ml, p-value 0.006), sPD-L1 (< 529.29 pg/ml, p-value 0.012), sPD-L2 (< 12627.9 pg/ml, p-value 0.043), sCTLA-4 (< 46.58 pg/ml, p-value 0.003), sCD80/B7-1 (< 636.54 pg/ml, p-value 0.006), sTIM-3 (< 3135.21 pg/ml, p-value 0.012), sBTLA (< 2841.9 pg/ml, p-value 0.012), sTLR-2 (< 5880.16 pg/ ml, p-value 0.017), sCD27 (< 1322.54 pg/ml, p-value 0.012), sCD28 (< 4769.95 pg/ml, p-value 0.004), sGITR (< 296.05 pg/ ml, p-value 0.012), and sGITRL (< 1008.42 pg/ml, p-value 0.008). No other significant correlations were found in the rest of the variables analyzed.

Correlation Analysis between Germline Brca1/2 and Analytes: Among the 23 patients tested for BRCA1/2 germline detection, a sub-analysis found that in women with BRCA1/2 mutation showed lower levels of sTNFα (< 4.36 pg/ml, p-value 0.02) and higher plasma levels of sIL-2 (> 2.32 pg/ml, p-value 0.018) compared to non-mutated patients.

Predictive Analysis Related to NCT Response

Patients receiving taxane and platinum-based NCT showed better objective response rates (p-value 0.002) both locally and nodally, and higher rates of pCR (58.83% vs. 12%, p-value 0.002) compared to those not receiving carboplatin. Elevated levels of sGM-CSF (>11.84 pg/ml) were predictive of a poorer response to NCT (p-value 0.02), with all patients having partial or complete response with levels below 11.84 pg/ml, while the 7 non-responsive patients had sGM-CSF levels above 11.84 pg/ ml. However, unlike to sGM-CSF, high levels of sCD86 (> 656.2 pg/ml) were predictive of a better response to NCT (p-value 0.02), with 6 out of 7 patients who did not achieve a pathological response having levels below 656.2 pg/ml. Additionally, pathological response at the nodal level was examined, with high levels of sGal-3 (> 3205.6 pg/ml, p-value 0.03) predicting a poorer nodal response. Specifically, 10 out of 12 patients with nodal involvement after neoadjuvant therapy had baseline levels of sGal-3 above 3205.6 pg/ml. However, unlike sGal-3, lower levels of sITAC (< 18.08 pg/ml, p-value 0.02) were detected in patients with greater nodal involvement. In 9 out of 12 patients with affected nodes after NCT, baseline levels of sITAC were below 18.08 pg/ml, suggesting that sITAC may be a predictive biomarker for nodal response (Table 7).

Table 7: Pathological response to NCT based on plasma levels of sGM-CSF and sCB86. Lymph node response to NCT based on plasma levels of sGal-3 and sITAC. Values ??of soluble analytes dichotomized according to the median. P value obtained in the chi-square test (X2).

Survival Analysis

Survival Analysis of Clinical Variables: Survival analysis revealed that a larger primary tumor size of 2 cm or greater (≥ T2) was associated with worse DFS (p-value 0.049) and lower OS (32 months vs. not reached, p-value 0.047). Patients with clinically detected nodal involvement at diagnosis had worse DFS (p-value 0.017) and lower OS (27 months vs. median not reached, p-value 0.0003) compared to patients without pathological lymph nodes. Moreover, a higher nodal involvement (N2 and N3) led to earlier relapses and poorer survival. Survival analyses showed that a higher clinical stage according to the 8th AJCC edition was associated with lower DFS (p-value 0.0034) and worse OS (32 months vs. median not reached, p-value 0.0047) (Supplementary Figure S1). Similarly to nodal involvement, patients with a higher clinical stage had a poorer prognosis in terms of both DFS (p-value 0.04) and OS (p-value 0.028), especially those with stages IIIB and IIIC. Patients achieving pCR after NCT had better DFS (31 months vs. median not reached, p-value 0.013) and better OS (median OS not reached in either arm, p-value 0.036) (Figure 4). It is noteworthy that none of the patients who achieved pCR experienced tumor recurrence.

Survival Analysis of Pathological Variables: Additionally, patients with well (G1) or moderately differentiated (G2) tumors had better DFS (32 months vs. median not reached, p-value 0.04) compared to poorly differentiated tumors (G3), with a trend towards increased OS (median not reached in either arm, p-value 0.08). Patients with a tumor biopsy showing TILs percentage of 30% or more had better DFS (28 months vs. median not reached, p-value 0.0047) and OS (32 months vs. median not reached, p-value 0.031), with a 2.25 times lower risk of death compared to patients with lower lymphocytic infiltration (Figure 5). No prognostic differences were found based on lymphocyte location (intra and peritumoral) or the proportion of CD4+ and CD8+ lymphocytes.

Survival Analysis of Soluble Biomarkers Variables: For survival analyses, patients were stratified by the median value of analytes. Higher plasma levels of sGal-3 (> 4143.66 pg/ ml) were associated with worse DFS (31 months vs. median not reached, p-value 0.049) and a trend towards lower OS (32 months vs. median not reached, p-value 0.06). Elevated levels of soluble Macrophage Inflammatory Protein 3α (sMIP-3α) (> 17.46 pg/ml) were correlated with a non-significant trend towards lower DFS (p-value 0.08) and significantly worse OS (32 months vs. median not reached, p-value 0.005). Higher levels of soluble Interleukin 10 (sIL-10) above 8.27 pg/ml were strongly associated with lower DFS (p-value 0.002) and worse OS (32 months vs. median not reached, p-value 0.006). Increased levels of soluble Interleukin 13 (sIL-13) (> 3.74 pg/ml) were associated with a trend towards worse DFS (p-value 0.06) and significantly worse OS (32 months vs. median not reached, p-value 0.009). Although soluble Interleukin 17A (sIL-17A) did not show a clear relationship with DFS, plasma levels below 8.17 pg/ml were associated with improved OS (32 months vs. median not reached, p-value 0.049).

The only soluble immune checkpoint marker associated with survival was soluble T Cell Immunoglobulin and Mucin Domain 3 (sTIM-3). Plasma concentrations of sTIM-3 above 3261.74 pg/ ml were linked to worse DFS (32 months vs. median not reached, p-value 0.049) and a non-significant trend towards lower OS (p-value 0.06).

No other significant correlations were found with the remaining variables analyzed. The results of this study confirm that in early-stage TNBC, adverse tumor characteristics such as larger size, nodal involvement, higher clinical stage, poorly differentiated carcinoma, low percentage of TILs, and failure to achieve pCR after NCT are associated with worse survival outcomes. Additionally, new insights are provided in the field of early TNBC by analyzing non-invasive biomarkers through liquid biopsy. This analysis revealed that higher plasma levels of certain cytokines and soluble immune checkpoints in patients with a higher tumor burden, indicating a potential increase in pro-inflammatory status leading to greater resistance to the anti-tumor activity of the immune system. Furthermore, elevated levels of sGal-3 and sGM-CSF, along with low levels of sCD86 and sITAC, were predictive of a poorer response to NCT, while increased plasma levels of sGal-3, sMIP-3α, sIL-10, sIL-13, sIL-17A, and sTIM-3 were associated with worse survival. These markers prove useful in determining disease prognosis and warrant further investigation to identify new therapeutic targets and more effective treatments against TNBC (Figure 6,7).

Figure 4: Kaplan-Meier curve for DFS and OS according to the type of pathological response (pCR vs. non-pCR). a) DFS b) OS. The yellow line symbolizes patients with pCR, while the blue line symbolizes patients with non-pCRP values obtained in the log-rank test.

Figure 5: Kaplan-Meier curve for DFS and OS according to TILs expression (≥ 30% vs. < 30%). a) DFS. b) OS. The blue line symbolizes patients with a  higher percentage of TILs, while the yellow line represents patients with a lower percentage of TILs. P values obtained in the log-rank test.

Figure 6: Kaplan-Meier curves for statistically significant DFS. a) sGal-3 b) sIL-10 c) sTIM-3. Soluble markers values were dichotomized with respect  to the median. The yellow line symbolizes lower levels of soluble biomarkers, while the blue line represents higher levels. P values obtained in the log-rank test.

Figure 7: Kaplan-Meier curves for statistically significant OS. a) sMIP-3α b) sIL-10 c) sIL-13 d) sIL-17A. Soluble markers values were dichotomized with  respect to the median. The yellow line symbolizes lower levels of soluble biomarkers, while the blue line represents higher levels. P values obtained in  the log-rank test.

In our study, the mean age at diagnosis was 55 years, with 57.5% of patients being postmenopausal, consistent with previous reports [33]. The average tumor size was 35 mm, with 64.3% of tumors between 2 and 5 cm, and 61.9% of patients presenting with lymph node involvement, similar to other early-stage TNBC cohorts [34]. Consistent with previous literature, larger tumor size, lymph node involvement, and consequently higher tumor stage confer a poorer prognosis, being the most important clinical prognostic factors for DFS and OS [35-37]. TNBC is commonly linked to BRCA1/2 mutations, and no significant prognostic differences observed between mutation carriers and noncarriers. Patients receiving taxane and platinum-based regimens had better local and nodal responses, with a notable increase in pCR rates compared to those without carboplatin [38]. Since pCR after NCT is considered a surrogate for survival, a trend toward improved outcomes was observed in the platinum-treated group. At a median follow-up of 68.5 months, tumor recurrence occurred in 31% of patients, and mortality was 23.8%.

In this study, NST was the predominant histological subtype, representing 95.2% of TNBC cases, aligning with other cohorts [39]. A small percentage (4.8%) of tumors were categorized as apocrine carcinoma with neuroendocrine differentiation, characterized by a poor response to NCT but generally considered to have a better prognosis [40]. Most tumors (71.4%) were poorly differentiated (G3), a common feature in TNBC. Although some studies associate G3 with lower OS, the majority do not find a clear association with survival. In this study, patients with well or moderately differentiated tumors (G1-2) showed better DFS compared to G3 and a trend towards better OS [35]. The proliferation marker Ki-67, with a mean index of 49%, reflects the high proliferative nature of TNBC [41,42].

The overexpression or amplification of HER2, a transmembrane protein receptor involved in cell growth regulation, was also examined. HER2 expression was low in 21.4% of TNBC patients, consistent with literature reports, though it was not predictive of response to NCT or survival [43]. Similarly, 73.8% of patients exhibited positive EGFR staining, but EGFR expression did not correlate with treatment response or survival, in line with previous studies that question the efficacy of EGFR inhibitors in TNBC due to pathway activation and tumor heterogeneity [44- 46]. Overexpression of Androgen Receptor (AR) characterizes a distinct molecular subset of TNBC. AR expression is proposed to antagonize ER and act as an antiproliferative agent in ER-positive breast cancer but facilitates cell proliferation through androgen receptor-dependent pathways in ER-negative BC. AR expression was found in 11.9% of patients, consistent with the reported 10-40% prevalence in TNBC. Previous studies have associated AR presence with smaller tumor size, lower histological grade, higher percentage of apocrine morphology, and lower Ki-67; whereas the role of AR as a predictive and prognostic biomarker in TNBC remains controversial [47,48]. However, in this study, AR expression did not correlate with clinicopathological variables, nor survival, likely due to the small sample size.

TILs were examined as prognostic markers. TILs, primarily composed of CD4+ and CD8+ T lymphocytes, are well-established as prognostic indicators in TNBC, correlating with better DFS, OS, and higher pCR rates after NCT [21-50]. In our study, patients with TILs ≥ 30% showed improved DFS and OS, consistent with previous research. However, TILs were not predictive of NCT response. The specific roles of CD4+ and CD8+ subsets remain controversial, with studies yielding conflicting results on their prognostic value [18-51]. While some research associates CD8+ TILs with better prognosis, others report no significant correlation [52–54]. In line with previous literature, this study associated higher lymphocytic percentage (TILs ≥ 30%) with improvements in DFS and OS, with a 2.25 times lower risk of death compared to patients with low TIL percentage (< 30%). In our study, TILs were not predictive biomarkers for treatment response.

In TNBC, clinical-pathological markers and tumor tissue levels help in prognosis, but there’s limited data on noninvasive biomarkers from peripheral blood. Some small studies have started to highlight the importance of the tumor microenvironment and soluble analytes in neoplasms, including TNBC. Cytokines, small proteins that regulate cell function, are linked to diseases such as cancer and metastasis. ITAC is a cytokine highly expressed in peripheral blood leukocytes and various tissues, known for its role in inhibiting angiogenesis and tumor progression in preclinical trials. Its role in TNBC remains unclear [55,56]. This study found lower plasma levels of soluble ITAC (sITAC) in larger tumors, while higher sITAC levels predicted better nodal response to NCT, suggesting a potential link to favorable clinical outcomes. GM-CSF, secreted by immune cells, may promote tumor progression by inhibiting T cells [57,58]. In this study, elevated plasma sGM-CSF levels were linked to a higher proliferation index and poorer response to NCT in TNBC, indicating its potential as a predictive biomarker for early TNBC. However, sGM-CSF levels did not correlate with survival, raising concerns about its use in prophylactic therapy. IFN-γ, produced by CD4+ T cells and NK cells, has a controversial role in immune regulation and antitumor immunity. In this study, plasma IFN-γ levels were not linked to treatment response or prognosis, though other studies suggest its prognostic value in immunotherapy and chemotherapy in TNBC [59,60]. TNFα, a proinflammatory cytokine [61], was associated with nodal involvement but not survival. Elevated plasma Gal-3 levels correlated with higher tumor burden, worse nodal response, and poorer survival; consistent with previous studies [62-65]. Additionally, increased levels of MIP-3α and MIP-1α were linked to poorer survival outcomes, in accordance with findings in other cancers [66,67]

Interleukins are crucial in immune communication and play roles in tumor progression and metastasis [68,69]. In our study, elevated plasma levels of soluble IL-8 (sIL-8) were associated with larger tumor size and advanced stage, indicating a potential role as a prognostic marker. This aligns with previous research indicating that IL-8 can promote angiogenesis and tumor invasiveness, thereby contributing to tumor growth and dissemination [70]. Additionally, high sIL-6 levels were associated with decreased tumor lymphocytic infiltration, which is linked to poorer survival. This observation supports the role of IL-6 in creating a tumor microenvironment that favors immune evasion [71]. Elevated plasma levels of soluble IL-10 (sIL-10) and IL-13 (sIL-13) were observed in TNBC patients and were associated with poorer clinical outcomes, including lower pCR rates and decreased survival. Soluble IL-17A (sIL-17A) was elevated in TNBC and associated with worse prognosis.

Soluble immune checkpoints are key regulatory proteins that modulate the immune system, promoting self-tolerance and preventing auntoinmmunity. PD-1, a well-studied checkpoint, regulates immune responses by inducing apoptosis in antigenspecific T cells and preserving regulatory T cells [72]. In cancer, PD-1’s interaction with PD-L1 fosters an immunosuppressive environment that aids tumor growth. Soluble forms of PD-1 (sPD-1) and PD-L1 (sPD-L1) have been found in cancer patients’ plasma, with elevated levels linked to advanced disease and poor prognosis [73]. However, in our study, sPD-1/L1/L2 levels did not correlate with NCT response or survival. sCTLA-4 levels are elevated in some cancers, and it could be related to disease progression [74,75]. sCD80 and sCD86, co-stimulatory molecules, have roles in immune regulation, with their soluble forms showing associations with advanced cancer stages and altered immune cell ratios [76]. Soluble forms of other immune checkpoints like sICOS, sLAG3, sTIM-3, sTLR-2, sTNFα, sCD27, sGITR, and sCD40 have also been studied [77-82]. Their soluble forms may impact immune response and tumor progression, with varying associations observed with cancer stage, immune cell ratios, and prognosis. Our study shows similar data, since high levels of sICOS, sTLR-2, sTNFα and sCD27 were correlated with locally advanced tumors, while sLAG-3 and sTIM-3 were associated with tumors larger than 2 cm. Patients with nodal involvement at diagnosis had higher levels of sTNFα. Understanding the roles of cytokines and soluble immune checkpoints in cancer progression could identify TNBC patients with aggressive clinical features or poorer prognosis, potentially guiding more aggressive management or inform the development of novel therapeutic strategies targeting these pathways, improving treatment outcomes.

In early TNBC, patients who received platinum-based regimens within NCT achieved higher rates of pCR. Obtaining a pCR was identified as one of the most important factors for good prognosis in early TNBC. However, the expressions of Ki67, EGFR, HER2 low, and AR did not prove to have a predictive or prognostic value in this setting. The mean percentage of TILs was 30%, mostly peritumoral and CD4+. An increase in the percentage of TILs (regardless of distribution or subtype) was correlated with better survival. Additionally, patients with a higher tumor burden had elevated plasma levels of specific cytokines and immune checkpoints, suggesting that greater disease burden may induce immunosuppression within the tumor microenvironment, facilitating immune evasion in TNBC. High levels of sGal-3 and sGM-CSF, as well as low levels of sCD86 and sITAC, were predictive of a lower response to NCT. Increased plasma levels of sGal-3, sMIP-3α, sIL-10, sIL-13, sIL-17A, and sTIM-3 were related to worse survival; potentially strengthening the usefulness of these biomarkers in treatment planning and opening a range of possibilities in future therapeutic strategies in this subgroup of patients.

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors acknowledge grant CB16-12-00350 from CIBEROnc, the AMACMA foundation, and Lopez Trigo 2017.

1. Cardoso F, Kyriakides S, Ohno S, Penault-Llorca F, Poortmans P, Rubio IT, et al. Early breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up†. Ann Oncol. 2019; 30: 1194-1220. doi: 10.1093/annonc/mdz173. Erratum in: Ann Oncol. 2019 Oct 1;30(10):1674. doi: 10.1093/annonc/mdz189. Erratum in: Ann Oncol. 2021 Feb;32(2):284. doi: 10.1016/j.annonc.2020.08.2158. PMID: 31161190.
2. Wolff AC, Hammond ME, Hicks DG, Dowsett M, McShane LM, Allison KH, et al; American Society of Clinical Oncology; College of American Pathologists. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol. 2013; 31: 3997-4013. doi: 10.1200/JCO.2013.50.9984. Epub 2013 Oct 7. PMID: 24101045.
3. Carey LA, Perou CM, Livasy CA, Dressler LG, Cowan D, Conway K, et al, Moorman PG, Earp HS, Millikan RC. Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA. 2006; 295: 2492-2502. doi: 10.1001/jama.295.21.2492. PMID: 16757721.
4. Carey LA, Dees EC, Sawyer L, Gatti L, Moore DT, Collichio F, et al. The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res. 2007; 13: 2329-2334. doi: 10.1158/1078-0432.CCR-06-1109. PMID: 17438091.
5. Uscanga-Perales GI, Santuario-Facio SK, Ortiz-López R. Triple negative breast cancer: Deciphering the biology and heterogeneity. Med Univ. 2016; 18: 105–114.
6. Lehmann BD, Jovanovi? B, Chen X, Estrada MV, Johnson KN, Shyr Y, et al. Refinement of Triple-Negative Breast Cancer Molecular Subtypes: Implications for Neoadjuvant Chemotherapy Selection. PLoS One. 2016; 11: e0157368. doi: 10.1371/journal.pone.0157368. PMID: 27310713; PMCID: PMC4911051.
7. Liedtke C, Mazouni C, Hess KR, André F, Tordai A, Mejia JA, et al. Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol. 2008; 26: 1275-1281. doi: 10.1200/JCO.2007.14.4147. Epub 2008 Feb 4. PMID: 18250347.
8. Haque W, Verma V, Hatch S, Suzanne Klimberg V, Brian Butler E, Teh BS. Response rates and pathologic complete response by breast cancer molecular subtype following neoadjuvant chemotherapy. Breast Cancer Res Treat. 2018; 170: 559-567. doi: 10.1007/s10549-018-4801-3. Epub 2018 Apr 24. PMID: 29693228.
9. Shepherd JH, Ballman K, Polley MC, Campbell JD, Fan C, Selitsky S, et al. CALGB 40603 (Alliance): Long-Term Outcomes and Genomic Correlates of Response and Survival After Neoadjuvant Chemotherapy With or Without Carboplatin and Bevacizumab in Triple-Negative Breast Cancer. J Clin Oncol. 2022; 40: 1323-1334. doi: 10.1200/JCO.21.01506. Epub 2022 Jan 19. PMID: 35044810; PMCID: PMC9015203.
10. Loibl S, Weber KE, Timms KM, Elkin EP, Hahnen E, Fasching PA, et al. Survival analysis of carboplatin added to an anthracycline/taxane-based neoadjuvant chemotherapy and HRD score as predictor of response-final results from GeparSixto. Ann Oncol. 2018; 29: 2341-2347. doi: 10.1093/annonc/mdy460. PMID: 30335131.
11. Schmid P, Cortes J, Pusztai L, McArthur H, Kümmel S, Bergh J, et al. Pembrolizumab for Early Triple-Negative Breast Cancer. N Engl J Med. 2020; 382: 810-821. doi: 10.1056/NEJMoa1910549. PMID: 32101663.
12. Gianni L, Huang CS, Egle D, Bermejo B, Zamagni C, Thill M, et al. Pathologic complete response (pCR) to neoadjuvant treatment with or without atezolizumab in triple-negative, early high-risk and locally advanced breast cancer: NeoTRIP Michelangelo randomized study. Ann Oncol. 2022; 33: 534-543. doi: 10.1016/j.annonc.2022.02.004. Epub 2022 Feb 17. PMID: 35182721.
13. Goodman A. No Improved Pathologic Complete Response With Atezolizumab in Early Triple-Negative Breast Cancer. ASCO Post. 2020; 1: 1.
14. Mittendorf EA, Zhang H, Barrios CH, Saji S, Jung KH, Hegg R, et al. Neoadjuvant atezolizumab in combination with sequential nab-paclitaxel and anthracycline-based chemotherapy versus placebo and chemotherapy in patients with early-stage triple-negative breast cancer (IMpassion031): a randomised, double-blind, phase 3 trial. Lancet. 2020; 396: 1090-1100. doi: 10.1016/S0140-6736(20)31953-X. Epub 2020 Sep 20. PMID: 32966830.
15. Núñez Abad M, Calabuig-Fariñas S, Lobo de Mena M, José Godes Sanz de Bremond M, García González C, Torres Martínez S, et al. Update on systemic treatment in early triple negative breast cancer. Ther Adv Med Oncol. 2021; 13: 1758835920986749. doi: 10.1177/1758835920986749. PMID: 33613695; PMCID: PMC7871289.
16. Geyer CE Jr, Garber JE, Gelber RD, Yothers G, Taboada M, Ross L, et al. OlympiA Clinical Trial Steering Committee and Investigators. Overall survival in the OlympiA phase III trial of adjuvant olaparib in patients with germline pathogenic variants in BRCA1/2 and high-risk, early breast cancer. Ann Oncol. 2022; 33: 1250-1268. doi: 10.1016/j.annonc.2022.09.159. Epub 2022 Oct 10. PMID: 36228963; PMCID: PMC10207856.
17. Sanz JM, Lahoz AMG, Silva A. El sistema inmune y el microambiente tumoral: componentes y función. Med - Programa Form Médica Contin Acreditado. 2021; 13: 1932-1941.
18. Jamiyan T, Kuroda H, Yamaguchi R, Nakazato Y, Noda S, Onozaki M, et al. Prognostic impact of a tumor-infiltrating lymphocyte subtype in triple negative cancer of the breast. Breast Cancer. 2020; 27 :880-892. doi: 10.1007/s12282-020-01084-1. Epub 2020 Mar 28. PMID: 32222891; PMCID: PMC7438376.
19. Zgura A, Galesa L, Bratila E, Anghel R. Relationship between Tumor Infiltrating Lymphocytes and Progression in Breast Cancer. Maedica (Bucur). 2018; 13: 317-320. Greek, Modern. doi: 10.26574/maedica.2018.13.4.317. PMID: 30774731; PMCID: PMC6362880.
20. Kurozumi S, Matsumoto H, Kurosumi M, Inoue K, Fujii T, Horiguchi J, et al. Prognostic significance of tumour-infiltrating lymphocytes for oestrogen receptor-negative breast cancer without lymph node metastasis. Oncol Lett. 2019; 17: 2647-2656. doi: 10.3892/ol.2019.9938. Epub 2019 Jan 16. PMID: 30867728; PMCID: PMC6396218.
21. Denkert C, von Minckwitz G, Brase JC, Sinn BV, Gade S, Kronenwett R, et al. Tumor-infiltrating lymphocytes and response to neoadjuvant chemotherapy with or without carboplatin in human epidermal growth factor receptor 2-positive and triple-negative primary breast cancers. J Clin Oncol. 2015; 33: 983-991. doi: 10.1200/JCO.2014.58.1967. Epub 2014 Dec 22. PMID: 25534375.
22. Matkowski R, Gisterek I, Halon A, Lacko A, Szewczyk K, Staszek U, et al. The prognostic role of tumor-infiltrating CD4 and CD8 T lymphocytes in breast cancer. Anticancer Res. 2009; 29: 2445–2451.
23. Yang X, Ren H, Sun Y, Shao Y, Zhang L, Li H, et al. Prognostic significance of CD4/CD8 ratio in patients with breast cancer. Int J Clin Exp Pathol. 2017; 10: 4787–793.
24. Voutsadakis IA. Immune Blockade Inhibition in Breast Cancer. Anticancer Res. 2016; 36: 5607-5622. doi: 10.21873/anticanres.11145. PMID: 27793883.
25. Muenst S, Schaerli AR, Gao F, Däster S, Trella E, Droeser RA, et al. Expression of programmed death ligand 1 (PD-L1) is associated with poor prognosis in human breast cancer. Breast Cancer Res Treat. 2014; 146: 15-24. doi: 10.1007/s10549-014-2988-5. Epub 2014 May 20. PMID: 24842267; PMCID: PMC4180714.
26. Baptista MZ, Sarian LO, Derchain SF, Pinto GA, Vassallo J. Prognostic significance of PD-L1 and PD-L2 in breast cancer. Hum Pathol. 2016; 47: 78-84. doi: 10.1016/j.humpath.2015.09.006. Epub 2015 Sep 25. PMID: 26541326.
27. Núñez Abad M, Calabuig-Fariñas S, Lobo de Mena M, Torres-Martínez S, García González C, García García JÁ, et al. Programmed Death-Ligand 1 (PD-L1) as Immunotherapy Biomarker in Breast Cancer. Cancers (Basel). 2022; 14: 307. doi: 10.3390/cancers14020307. PMID: 35053471; PMCID: PMC8773553.
28. Fasoulakis Z, Kolios G, Papamanolis V, Kontomanolis EN. Interleukins Associated with Breast Cancer. Cureus. 2018; 10: e3549. doi: 10.7759/cureus.3549. PMID: 30648081; PMCID: PMC6324869.
29. Liu Z, Li M, Jiang Z, Wang X. A Comprehensive Immunologic Portrait of Triple-Negative Breast Cancer. Transl Oncol. 2018; 11: 311-329. doi: 10.1016/j.tranon.2018.01.011. Epub 2018 Feb 4. PMID: 29413765; PMCID: PMC5884188.
30. Gu D, Ao X, Yang Y, Chen Z, Xu X. Soluble immune checkpoints in cancer: production, function and biological significance. Soluble immune checkpoints cancer Prod Funct Biol significance. 2018; 6: 132.
31. Salgado R, Denkert C, Demaria S, Sirtaine N, Klauschen F, Pruneri G, et al. International TILs Working Group 2014. The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: recommendations by an International TILs Working Group 2014. Ann Oncol. 2015; 26: 259-271. doi: 10.1093/annonc/mdu450. Epub 2014 Sep 11. PMID: 25214542; PMCID: PMC6267863.
32. Merck KGaA. MILLIPLEX® Human High Sensitivity T Cell Magnetic Bead Panel 96-Well Plate Assay. In American Society of Clinical Oncology; 2022; 1–32.
33. Autier P, Boniol M, La Vecchia C, Vatten L, Gavin A, Héry C, et al. Disparities in breast cancer mortality trends between 30 European countries: retrospective trend analysis of WHO mortality database. BMJ. 2010; 341: c3620. doi: 10.1136/bmj.c3620. Erratum in: BMJ. 2010;341:c4480. LaVecchia, Carlo [corrected to La Vecchia, Carlo]. PMID: 20702548; PMCID: PMC2920378.
34. Te?i? Vuger A, Šeparovi? R, Vazdar L, Pavlovi? M, Lepeti? P, Šiti? S, et al. CHARACTERISTICS AND PROGNOSIS OF TRIPLE-NEGATIVE BREAST CANCER PATIENTS: A CROATIAN SINGLE INstitution RETROSPECTIVE COHORT STUDY. Acta Clin Croat. 2020; 59: 97-108. doi: 10.20471/acc.2020.59.01.12. PMID: 32724280; PMCID: PMC7382886.
35. Balkenhol MCA, Vreuls W, Wauters CAP, Mol SJJ, van der Laak JAWM, Bult P. Histological subtypes in triple negative breast cancer are associated with specific information on survival. Ann Diagn Pathol. 2020; 46: 151490. doi: 10.1016/j.anndiagpath.2020.151490. Epub 2020 Mar 3. PMID: 32179443.
36. Soares RF, Garcia AR, Monteiro AR, Macedo F, Pereira TC, Carvalho JC, et al. Prognostic factors for early relapse in non-metastatic triple negative breast cancer - real world data. Rep Pract Oncol Radiother. 2021; 26: 563-572. doi: 10.5603/RPOR.a2021.0073. PMID: 34434572; PMCID: PMC8382079.
37. Urru SAM, Gallus S, Bosetti C, Moi T, Medda R, Sollai E, et al. Clinical and pathological factors influencing survival in a large cohort of triple-negative breast cancer patients. BMC Cancer. 2018; 18: 56. doi: 10.1186/s12885-017-3969-y. PMID: 29310602; PMCID: PMC5759886.
38. Poggio F, Bruzzone M, Ceppi M, Pondé NF, La Valle G, Del Mastro L, et al. Platinum-based neoadjuvant chemotherapy in triple-negative breast cancer: a systematic review and meta-analysis. Ann Oncol. 2018; 29: 1497-1508. doi: 10.1093/annonc/mdy127. PMID: 29873695.
39. Lakhani S, Ellis I, Schnitt S, Tan P, van de Vijver M. WHO Classification of Tumours of the Breast, Fourth Edition. IARC WHO Classification of Tumours. 2012; 4.
40. Brumec M, Sobo?an M, Taka? I, Arko D. Clinical Implications of Androgen-Positive Triple-Negative Breast Cancer. Cancers (Basel). 2021; 13.
41. Zhou Q, Li W. Prognostic value of Ki-67 in patients with triple-negative breast cancer receiving neo-adjuvant or adjuvant chemotherapy: A systematic review and meta-analysis. Ann Oncol. 2019; 30: iii15.
42. Zhu X, Chen L, Huang B, Wang Y, Ji L, Wu J, et al. The prognostic and predictive potential of Ki-67 in triple-negative breast cancer. Sci Rep. 2020; 10: 225. doi: 10.1038/s41598-019-57094-3. PMID: 31937819; PMCID: PMC6959292.
43. de Moura Leite L, Cesca MG, Tavares MC, Santana DM, Saldanha EF, Guimarães PT, et al. HER2-low status and response to neoadjuvant chemotherapy in HER2 negative early breast cancer. Breast Cancer Res Treat. 2021; 190: 155-163. doi: 10.1007/s10549-021-06365-7. Epub 2021 Aug 18. PMID: 34409551.
44. Hashmi AA, Naz S, Hashmi SK, Irfan M, Hussain ZF, Khan EY, et al. Epidermal growth factor receptor (EGFR) overexpression in triple-negative breast cancer: association with clinicopathologic features and prognostic parameters. Surg Exp Pathol. 2019; 2: 1-7.
45. Ferraro DA, Gaborit N, Maron R, Cohen-Dvashi H, Porat Z, Pareja F, et al. Inhibition of triple-negative breast cancer models by combinations of antibodies to EGFR. Proc Natl Acad Sci U S A. 2013; 110: 1815–1820.
46. Lehmann BD, Bauer JA, Chen X, Sanders ME, Chakravarthy AB, Shyr Y, et al. Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest. 2011; 121: 2750-2767. doi: 10.1172/JCI45014. PMID: 21633166; PMCID: PMC3127435.
47. Gerratana L, Basile D, Buono G, De Placido S, Giuliano M, Minichillo S, Coinu A, Martorana F, et al. Androgen receptor in triple negative breast cancer: A potential target for the targetless subtype. Cancer Treat Rev. 2018; 68: 102-110. doi: 10.1016/j.ctrv.2018.06.005. Epub 2018 Jun 11. PMID: 29940524.
48. Hickey TE, Robinson JL, Carroll JS, Tilley WD. Minireview: The androgen receptor in breast tissues: growth inhibitor, tumor suppressor, oncogene? Mol Endocrinol. 2012; 26: 1252-1267. doi: 10.1210/me.2012-1107. Epub 2012 Jun 28. PMID: 22745190; PMCID: PMC3404296.
49. de Jong VMT, Wang Y, Ter Hoeve ND, Opdam M, Stathonikos N, Jó?wiak K, et al. Prognostic Value of Stromal Tumor-Infiltrating Lymphocytes in Young, Node-Negative, Triple-Negative Breast Cancer Patients Who Did Not Receive (neo)Adjuvant Systemic Therapy. J Clin Oncol. 2022; 40: 2361-2374. doi: 10.1200/JCO.21.01536. Epub 2022 Mar 30. PMID: 35353548; PMCID: PMC9287283.
50. Noske A, Möbus V, Weber K, Schmatloch S, Weichert W, Köhne CH, et al. Relevance of tumour-infiltrating lymphocytes, PD-1 and PD-L1 in patients with high-risk, nodal-metastasised breast cancer of the German Adjuvant Intergroup Node-positive study. Eur J Cancer. 2019; 114: 76-88. doi: 10.1016/j.ejca.2019.04.010. Epub 2019 May 7. PMID: 31075727.
51. Macchetti AH, Marana HR, Silva JS, de Andrade JM, Ribeiro-Silva A, Bighetti S. Tumor-infiltrating CD4+ T lymphocytes in early breast cancer reflect lymph node involvement. Clinics (Sao Paulo). 2006; 61: 203-208. doi: 10.1590/s1807-59322006000300004. Epub 2006 Jun 30. PMID: 16832552.
52. Van Berckelaer C, Rypens C, van Dam P, Pouillon L, Parizel M, Schats KA, et al. Infiltrating stromal immune cells in inflammatory breast cancer are associated with an improved outcome and increased PD-L1 expression. Breast Cancer Res. 2019; 21: 28. doi: 10.1186/s13058-019-1108-1. PMID: 30777104; PMCID: PMC6380068.
53. Matsumoto H, Thike AA, Li H, Yeong J, Koo SL, Dent RA, et al. Increased CD4 and CD8-positive T cell infiltrate signifies good prognosis in a subset of triple-negative breast cancer. Breast Cancer Res Treat. 2016; 156: 237-247. doi: 10.1007/s10549-016-3743-x. Epub 2016 Mar 9. PMID: 26960711.
54. Mahmoud SM, Paish EC, Powe DG, Macmillan RD, Grainge MJ, Lee AH, et al. Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol. 2011; 29: 1949-1955. doi: 10.1200/JCO.2010.30.5037. Epub 2011 Apr 11. PMID: 21483002.
55. Kistner L, Doll D, Holtorf A, Nitsche U, Janssen KP. Interferon-inducible CXC-chemokines are crucial immune modulators and survival predictors in colorectal cancer. Oncotarget. 2017; 8: 89998-90012. doi: 10.18632/oncotarget.21286. PMID: 29163806; PMCID: PMC5685727.
56. Nazari A, Ahmadi Z, Hassanshahi G, Abbasifard M, Taghipour Z, Falahati-Pour SK, et al. Effective Treatments for Bladder Cancer Affecting CXCL9/CXCL10/CXCL11/CXCR3 Axis: A Review. Oman Med J. 2020; 35: e103. doi: 10.5001/omj.2020.21. PMID: 32181005; PMCID: PMC7064791.
57. Hong IS. Stimulatory versus suppressive effects of GM-CSF on tumor progression in multiple cancer types. Exp Mol Med. 2016; 48: e242. doi: 10.1038/emm.2016.64. PMID: 27364892; PMCID: PMC4973317.
58. Liu L, Liu Y, Yan X, Zhou C, Xiong X. The role of granulocyte colony?stimulating factor in breast cancer development: A review. Mol Med Rep. 2020; 21: 2019-2029. doi: 10.3892/mmr.2020.11017. Epub 2020 Mar 10. PMID: 32186767; PMCID: PMC7115204.
59. Jorgovanovic D, Song M, Wang L, Zhang Y. Roles of IFN-γ in tumor progression and regression: a review. Biomark Res. 2020; 8: 49. doi: 10.1186/s40364-020-00228-x. PMID: 33005420; PMCID: PMC7526126.
60. Granqvist V, Holmgren C, Larsson C. Induction of interferon-β and interferon signaling by TRAIL and Smac mimetics via caspase-8 in breast cancer cells. PLoS One. 2021; 16: e0248175. doi: 10.1371/journal.pone.0248175. PMID: 33770100; PMCID: PMC7996988.
61. Wolczyk D, Zaremba-Czogalla M, Hryniewicz-Jankowska A, Tabola R, Grabowski K, Sikorski AF, et al. TNF-α promotes breast cancer cell migration and enhances the concentration of membrane-associated proteases in lipid rafts. Cell Oncol (Dordr). 2016; 39: 353-363. doi: 10.1007/s13402-016-0280-x. Epub 2016 Apr 4. PMID: 27042827; PMCID: PMC4972855.
62. Dong R, Zhang M, Hu Q, Zheng S, Soh A, Zheng Y, et al. Galectin-3 as a novel biomarker for disease diagnosis and a target for therapy (Review). Int J Mol Med. 2018; 41: 599-614. doi: 10.3892/ijmm.2017.3311. Epub 2017 Dec 5. PMID: 29207027; PMCID: PMC5752178.
63. Torres?Martínez S, Calabuig?Fariñas S, Moreno?Manuel A, Bertolini G, Herreros?Pomares A, Escorihuela E, et al. Soluble galectin?3 as a microenvironment?relevant immunoregulator with prognostic and predictive value in lung adenocarcinoma. Mol Oncol. 2023; 1–50.
64.  Torres-Martínez S, Calabuig-Fariñas S, Gallach S, Mosqueda M, Munera-Maravilla E, Sirera R, et al. Circulating Immune Proteins: Improving the Diagnosis and Clinical Outcome in Advanced Non-Small Cell Lung Cancer. Int J Mol Sci. 2023; 24: 17587. doi: 10.3390/ijms242417587. PMID: 38139416; PMCID: PMC10743468.
65.  Niang DGM, Ka S, Hendricks J, Diouf D, Gaba FM, Diouf A, et al. Profile of Plasma Galectin-3 Concentrations, Inflammatory Cytokines Levels and Lymphocytes Status in Breast Cancer under Chemotherapy. Open J Immunol. 2022; 12: 1–14.
66. Iwata T, Tanaka K, Inoue Y, Toiyama Y, Hiro J, Fujikawa H, et al. Macrophage inflammatory protein-3 alpha (MIP-3a) is a novel serum prognostic marker in patients with colorectal cancer. J Surg Oncol. 2013; 107: 160-166. doi: 10.1002/jso.23247. Epub 2012 Aug 23. PMID: 22926691.
67. Zhang X, Meng A, Wang H, Yan X. High serum macrophage inflammatory protein-3α is associated with the early recurrence or metastasis of non-small cell lung cancer following primary pulmonary resection. Oncol Lett. 2014; 8: 948-952. doi: 10.3892/ol.2014.2229. Epub 2014 Jun 5. PMID: 25013520; PMCID: PMC4081402.
68. Brocker C, Thompson D, Matsumoto A, Nebert DW, Vasiliou V. Evolutionary divergence and functions of the human interleukin (IL) gene family. Hum Genomics. 2010; 5: 30-55. doi: 10.1186/1479-7364-5-1-30. PMID: 21106488; PMCID: PMC3390169.
69. Koz?owski L, Zakrzewska I, Tokajuk P, Wojtukiewicz MZ. Concentration of interleukin-6 (IL-6), interleukin-8 (IL-8) and interleukin-10 (IL-10) in blood serum of breast cancer patients. Rocz Akad Med Bialymst. 2003; 48: 82-4. PMID: 14737948.
70. Fousek K, Horn LA, Palena C. Interleukin-8: A chemokine at the intersection of cancer plasticity, angiogenesis, and immune suppression. Pharmacol Ther. 2021; 219: 107692. doi: 10.1016/j.pharmthera.2020.107692. Epub 2020 Sep 24. PMID: 32980444; PMCID: PMC8344087.
71. Li W, Wu Z, Meng W, Zhang C, Cheng M, Chen Y, et al. Blockade of IL-6 inhibits tumor immune evasion and improves anti-PD-1 immunotherapy. Cytokine. 2022; 158: 155976. doi: 10.1016/j.cyto.2022.155976. Epub 2022 Jul 31. PMID: 35921790.
72. Khan M, Zhao Z, Arooj S, Fu Y, Liao G. Soluble PD-1: Predictive, Prognostic, and Therapeutic Value for Cancer Immunotherapy. Front Immunol. 2020; 11: 587460. doi: 10.3389/fimmu.2020.587460. PMID: 33329567; PMCID: PMC7710690.
73. Zhu X, Lang J. Soluble PD-1 and PD-L1: predictive and prognostic significance in cancer. Oncotarget. 2017; 8: 97671-97682. doi: 10.18632/oncotarget.18311. PMID: 29228642; PMCID: PMC5722594.
74. Erfani N, Razmkhah M, Ghaderi A. Circulating soluble CTLA4 (sCTLA4) is elevated in patients with breast cancer. Cancer Invest. 2010; 28: 828-832. doi: 10.3109/07357901003630934. PMID: 20482250.
75. Isitmangil G, Gurleyik G, Aker FV, Coskun C, Kucukhuseyin O, Arikan S, et al. Association of CTLA4 and CD28 Gene Variants and Circulating Levels of Their Proteins in Patients with Breast Cancer. In Vivo. 2016; 30: 485-493. PMID: 27381613.
76. Haile ST, Horn LA, Ostrand-Rosenberg S. A soluble form of CD80 enhances antitumor immunity by neutralizing programmed death ligand-1 and simultaneously providing costimulation. Cancer Immunol Res. 2014; 2: 610-615. doi: 10.1158/2326-6066.CIR-13-0204. Epub 2014 Apr 2. PMID: 24819296; PMCID: PMC4082431.
77. Faget J, Sisirak V, Blay JY, Caux C, Bendriss-Vermare N, Ménétrier-Caux C. ICOS is associated with poor prognosis in breast cancer as it promotes the amplification of immunosuppressive CD4+ T cells by plasmacytoid dendritic cells. Oncoimmunology. 2013; 2: e23185. doi: 10.4161/onci.23185. PMID: 23802069; PMCID: PMC3661154.
78. Chocarro L, Blanco E, Zuazo M, Arasanz H, Bocanegra A, Fernández-Rubio L, et al. Understanding LAG-3 Signaling. Int J Mol Sci. 2021; 22: 5282. doi: 10.3390/ijms22105282. PMID: 34067904; PMCID: PMC8156499.
79. Li N, Jilisihan B, Wang W, Tang Y, Keyoumu S. Soluble LAG3 acts as a potential prognostic marker of gastric cancer and its positive correlation with CD8+T cell frequency and secretion of IL-12 and INF-γ in peripheral blood. Cancer Biomark. 2018; 23: 341-351. doi: 10.3233/CBM-181278. PMID: 30223387.
80. Bailly C, Thuru X, Goossens L, Goossens JF. Soluble TIM-3 as a biomarker of progression and therapeutic response in cancers and other of human diseases. Biochem Pharmacol. 2023; 209: 115445. doi: 10.1016/j.bcp.2023.115445. Epub 2023 Feb 2. PMID: 36739094.
81. El-Kharashy G, Gowily A, Okda T, Houssen M. Association between serum soluble Toll-like receptor 2 and 4 and the risk of breast cancer. Mol Clin Oncol. 2021; 14: 38. doi: 10.3892/mco.2020.2200. Epub 2020 Dec 24. PMID: 33414918; PMCID: PMC7783720.
82. Huang J, Jochems C, Anderson AM, Talaie T, Jales A, Madan RA, et al. Soluble CD27-pool in humans may contribute to T cell activation and tumor immunity. J Immunol. 2013; 190: 6250-6258. doi: 10.4049/jimmunol.1300022. Epub 2013 May 15. PMID: 23677477; PMCID: PMC4084782.