Inter-Relationship of Ki-67 and Triple-Negative Breast Cancer

Abstract

Triple-negative breast cancer (TNBC) is a heterogeneous group characterized by an early onset, aggressive course of the disease, a higher tendency of visceral metastases, and a poorer prognosis. It is also associated with basal-like phenotype and germline mutations for BRCA genes in 10–20% and somatic mutations in 3–5% of cases. Based on gene expression profiling, TNBC is divided into four tumor-specific subtypes (Basal-like 1, Basal-like 2, Mesenchymal, and Luminal androgen receptor) with different clinical, prognostic, and therapeutic implications. The Ki-67 antigen, a non-histone nuclear protein, is a surrogate marker to assess tumor proliferation. As TNBCs are expected to be highly proliferating tumors, a higher baseline Ki-67 level has been seen. Although a higher Ki-67 level is associated with a higher pathological complete response rate, the best cutoff point of this marker as a prognostic and predictive factor in TNBC remains unclear.

1. Introduction

According to GLOBOCAN 2020, female breast cancer surpassed lung cancer as the leading cause of cancer globally in 2020, with 2.3 million new cases worldwide [1]. Breast cancer is a heterogeneous disease encompassing different entities with distinct morphological features and clinical behaviors. The St. Gallen guidelines, the American Society of Clinical Oncology, and the College of American Pathology have defined triple-negative breast cancer (TNBC) as breast cancer with:

1. Less than 1% of tumor cells expressing ER and PR via IHC [2].

2. Her-2-neu negative: Immunohistochemistry (IHC) staining of 0 or 1 +, a Fluorescent in-situ hybridization (FISH) result of less than 4.0 HER2 gene copies per nucleus, or FISH ratio of less than 1.8 (FISH to be done in case IHC is 2+, equivocal) [3].

2. TNBC vs. Basal

TNBC is labeled based on IHC negativity of ER/PR/Her-2neu on tumor cells. However, based on gene expression profiles established through the 50-gene Prediction Analysis of Microarray (PAM50) assay, four “intrinsic subtypes” are defined: Luminal A, Luminal B, Basal-like, and Her-2-neu enriched. TNBC represents approximately 15–20% of all patients with breast cancer and shares various similarities with basal-like cancer [4]:

1. TNBC occurs in premenopausal young women under 40 years old

2. More aggressive disease course with a peak in recurrence between 1 and 3 years after diagnosis. Survival time is also shorter, and the mortality rate is 40% within the first 5 years after diagnosis

3. Approximately 46% of TNBC patients will have distant metastasis at presentation

4. The metastasis often involves the brain and visceral organs rather than the lungs or bones.

5. Due to the lack of targeted therapies, chemotherapy and surgery are the mainstays in treatment for TNBC

The basal-like subtype of breast cancer is characterized by a gene expression profile similar to that of the basal-myoepithelial layer of the normal breast; cytokeratins (5/6, 14, and 17), P-cadherin, EGFR17, and EGFR gene amplification (rarely) [5]. TP53 gene mutations are observed in up to 85% of cases [5, 6]. However, basal-like breast cancers, unlike “basal”/myoepithelial cells of normal breast, uniformly express cytokeratins 8 and/or 18 [5]. This questions microarray-based taxonomy of breast cancers that suggested that basal-like cancers would arise from basal/myoepithelial cells. This has been answered in a recent study with the possibility that a subgroup of basal-like breast cancers may originate from luminal progenitors rather than basal myoepithelial cells of the breast [7].

Although TNBC and basal-like share many similarities and the terms are very often used interchangeably, they are different. Not all basal-like cancers lack ER, PR, and HER2, and not all TNBCs show a basal-like phenotype by expression array analysis (25% discordance) (Figure 1) [8].

Figure 1.

Concordance between TNBC (IHC BASED) AND BASAL LIKE (genetic array analysis.

Reasons for this discordance can be:

1. False positivity/false negativity of the IHC-based assays for determining the HR or HER2 status (inter-laboratory and inter-method discordance rates of 20%) [8]

2. Intra-tumor Heterogeneity: It is difficult that two different subtypes coexist in same tumor [8]

3. Some TNBCs do not express basal markers and are classified as normal breast-like. (Probably an artifact of gene expression profiling due to samples with disproportionately high content of stromal and normal breast epithelial cells) [5]

4. Multigene expression data using hundreds of genes better capture the accurate biological profile compared with three or four individual surrogate biomarkers used to label TNBC [6]

2.1 BRCA1-Associated TNBC

Basal-like tumors show similar molecular genetic profiles to tumors arising in BRCA1 carriers. Both sporadic basal-like tumors and tumors with BRCA 1 mutations express basal keratins, and both groups cluster together in gene expression profiling [9]. Germline mutations for BRCA genes occur in 10–20% of TNBC patients, and somatic mutations are seen in 3–5% [10]. Apart from somatic mutation in BRCA1 gene: [5, 6]

1. BRCA1 hypermethylation and/or loss of heterozygosity may give rise to a BRCA1-like molecular profile in wild-type TNBC

2. Sporadic invasive ductal carcinomas with basal-like phenotype express ID4, a negative regulator of BRCA15

3. Frequent loss of several other genes involved in BRCA1-dependent homologous recombination (HR) repair has been demonstrated in basal-like/triple-negative cancer [9]

BRCA1-like features are characterized by: [5]

1. Basal-like phenotype (associated with the BRCA1 phenotype but not with the BRCA2 phenotype)

2. Present as interval tumor

3. ER-negativity, EGFR expression, c-MYC amplification

4. TP53 mutations (85%) [5], loss of RAD51 focus formation

   1. Extreme genomic instability and sensitivity to DNA-crosslinking agents

5. Predominantly hematogenous spread over axillary nodes and bones

6. Sensitive to DNA-damaging agents such as platinum compounds, or poly (ADP-ribose) polymerase (PARP) inhibitors or their combination

2.2 TNBC subtypes

In 2011, Lehmann et al. performed gene expression profiling of tumor samples from 587 TNBC patients and divided TNBC into six subtypes: [9, 11]

1. Basal-like-1 (BL-1):

   1. Abnormal expression of cell cycle regulating and DNA repair-related genes (high amplification of MYC, PIK3CA, CDK6, AKT2, KRAS, IGF1R, and CDKN2A/B)

   2. High frequency of heterozygous or homozygous deletion of DNA repair-related genes such as BRCA2, PTEN, MDM2, RB1, and TP53

   3. A high Ki-67 mRNA expression is observed on nuclear Ki-67 staining (>70%).

   4. Nearly all of the cell lines with BRCA1 and BRCA2 mutations have gene expression patterns that correlate with this subtype [12]

2. Basal-like-2 (BL-2): Abnormal activation of growth factor signaling pathways such as the EGFR, MET, NGF, Wnt/β-catenin, and IGF-1R pathways

3. Mesenchymal-like subtype (M): Also called metaplastic breast cancer

   1. Highly activated cell migration-related signaling pathways, extracellular matrix-receptor interaction pathways, and differentiation pathways (Wnt pathway, anaplastic lymphoma kinase pathway, transforming growth factor (TGF)-β signaling)

   2. The M subtype has sarcoma-like or squamous epithelial cell-like tissue characteristics

   3. Prone to develop resistance to chemotherapeutic drugs

4. Mesenchymal stem-like subtype (MSL): Low levels of cell proliferation-related genes and high levels of stem cell-related genes

5. Immuno-modulatory subtype (IM):

   1. Characterized by increased expression of immune cell-associated genes and pathways such as the Th1/Th2 pathway, NK cell pathway, B cell receptor signaling pathway, dendritic cell (DC) pathway, T cell receptor signaling, interleukin (IL)-12 pathway, and IL-7 pathway.

   2. Substantially overlap with a gene signature for medullary breast cancer, high-grade histology with a favorable prognosis.

   3. This subtype has the best prognosis [9]

6. Luminal Androgen Receptor (LAR):

   1. Although the LAR subtype does not express ER receptors, it does have highly activated hormonal-related signaling pathways (including steroid synthesis, porphyrin metabolism, and androgen/estrogen metabolism).

   2. ESR1 (the gene encoding ERα) and other estrogen-regulated genes (PGR, FOXA, XBP1, GATA3) are present on micro-array profiling. Thus, there is molecular evidence of ER activation. However, they may be classified as “ER-negative” because <1% of these tumor cells express low levels of ER protein on IHC analysis

   3. Androgen receptor (AR) is highly expressed (mRNA level is nine times) as well as high expression of AR on IHC (10 times)

This classification was validated by Masuda et al. [13]. They further compared the TNBC subtypes between the PAM50 basal-like subtype and non-basal-like subtypes (other subtypes grouped). All tumors in the BL1 and BL2 subtypes belonged to the basal-like PAM50 subtype, and most tumors in the LAR subtype belonged to the non-basal-like PAM50 group. In the non-basal-like group, there were only three TNBC subtypes, LAR, MSL, and M; most of these tumors were the LAR subtype (59%). They further found that even though BL-1 and BL-2 are highly proliferative tumors, the BL-1 subtype had the highest pathological complete response (pCR) rate, and the BL2 subtype had the lowest pCR rate. Similarly, consistent with LAR’s low pCR rate, the luminal A and B intrinsic subtypes, hormonally regulated tumors, showed less response to chemotherapy. Therefore, the LAR group had delayed recurrences compared with the other groups and did not have the lowest OS rate despite having a low pCR rate.

Burstein et al. distinguished TNBC subtypes into four types only: luminal-AR (LAR), mesenchymal (MES), basal-like immune-suppressed (BLIS), and basal-like immune-activated (BLIA) [14]. However, in an updated analysis, Lehman et al. reported that transcripts in the previously described IM and MSL subtypes were contributed from infiltrating lymphocytes and tumor-associated stromal cells, respectively [15]. Therefore, in their new refined classification, TNBC molecular subtypes were reduced from six (TNBCtype-6) to four (TNBCtype-4) tumor-specific subtypes (BL1, BL2, M, and LAR) (Table 1). PAM50 subtype “calls” distribution among the TNBC subtypes showed that most BL1, BL2, and M were basal-like, while LAR was enriched in HER2 and luminal subtypes.

TNBC subtypes; Lehman et al. [11]	TNBC subtypes; Burstein et al. [14]	TNBC subtype-4; Lehman et al. [15]
Basal-like 1 (BL1)	Basal-like Immune suppressed (BLIS)	Basal-like 1 (BL1)
Basal-like 2 (BL2)	—	Basal-like 2 (BL2)
Immunomodulatory (IM)	Basal-like Immune-activated (BLIA)	—
Mesenchymal (M)	Mesenchymal (MES)	Mesenchymal (M)
Mesenchymal Stem-like (MSL)	—	—
Luminal androgen receptor (LAR)	Luminal androgen receptor (LAR)	Luminal androgen receptor (LAR)

Table 1.

Different classification of subtypes of TNBC.

2.3 Benefits of sub-classifying TNBC

1. All the subtypes of TNBC have different clinic-pathological features, affecting their prognosis.

   1. Age: Non-basal TNBC was reported in older patients than basal TNBC. LAR subtype was diagnosed in women of older age than all other subtypes.

   2. Grade: Basal TNBC tumors are more likely to be of a higher grade than non-basal TNBC. BL1 tumors are higher grade, and LAR tumors are lower grade. In contrast to lower histological grade, non-basal TNBC presents significantly more advanced clinical disease and a higher stage than basal TNBC

   3. Histopathology: BL1 tumors were mainly ductal carcinomas without notable atypical histology. In contrast, infiltrating lobular carcinomas were nearly exclusive to the LAR subtype. Medullary breast cancer histological types were present in BL1, BL2 and, LAR and absent in the M subtype.

   4. Regional nodes: Regional spread to lymph nodes was similar in basal (29%) and non-basal (31%). Approximately half (47%) of LAR TNBC patients have regional spread, whereas the node involvement was lower for the M TNBC subtype (21%).

   5. Distant Metastasis: The M subtype is prone to a higher frequency of lung metastasis (46%) than all other subtypes (25%). Whereas bone metastasis was significantly higher for the LAR subtype (46%) than all other subtypes (16%).

   6. pCR: pCR rates were similar in basal and non-basal subtypes. The BL-1 subtype had the highest pCR rate, and the BL2 and LAR subtypes had the lowest pCR rate. Moreover, BL1 patients had significantly higher pCR than all other subtypes (49% vs. 31%).

   7. Overall survival (OS): BL1 patients had significantly better OS than all other TNBCtype-4 subtypes combined. Moreover, BL1 patients displayed better relapse-free survival, with nearly 60% survival even at 10 years. The IM subtype displayed the best overall and relapse-free survival. The LAR subtypes had better survival despite a decreased response to neoadjuvant chemotherapy. The decreased response of AR-positive TNBC tumors to neoadjuvant chemotherapy has recently been validated with the report of significantly lower pCR [16].

   8. Distant relapse-free survival (DRFS): Despite having better pCR to neoadjuvant chemotherapy (34% vs. 11%), TNBC patients had significantly worse DRFS survival than non-TNBC [15]. However, TNBC patients that achieved a pCR on chemotherapy had a far better DRFS than those patients that did not. BL2 patients have the worst outcome, with a median survival of 2.4 years. In contrast, the BL1 subtype had the best long-term DRFS, with 72% of patients relapse-free at a 7-year follow-up.

2. Treatment Options (Table 2)

   1. Basal Like: Cells have a complex DNA damage response and repair mechanisms to maintain genomic integrity. The most deleterious lesion, double-strand breaks are repaired by either HR (homologous recombination) or non-homologous end joining. Patients with mutations in the breast cancer susceptibility proteins BRCA1 and BRCA2A have cancers due to deficiency in HR repair. Moreover, these are dependent on other DNA repair mechanisms, the most prominent of which is the peroxisome proliferator-activated receptor (PARP)-based. Therefore, PARP inhibitors have significant antitumor effects on BRCA1/2-deficient tumors, and the inhibition effect on BRCA1-mutant tumors is 100–1000 times higher than in tumors without such mutations [4]. The basal-like subgroup has increased expression of proliferation-related genes and DNA repair genes; therefore, they may be sensitive to anti-mitotic drugs such as taxanes and platinum and PARP inhibitors such as olaparib and veliparib.

   2. Immune Check Point Regulators: Tumor cells can evade recognition and destruction by the host immune system through the immune checkpoint system. Under normal circumstances, the immune system reacts to foreign antigens that accumulate in the lymph nodes or spleen and promotes antigen-specific T-cell proliferation. Programmed cell death protein 1 (PD-1) binds to PD-L1 and can transmit signals to inhibit T cell proliferation and promote T cell depletion [17]. PD-L1 expression in tumor cells or its presence in the tumor microenvironment has been positively associated with triple-negative status in breast cancer [17]. Moreover, high PD-L1 levels have also been correlated with pCR after neoadjuvant chemotherapy and improved clinical outcomes in TNBC [17]. Pembrolizumab, a monoclonal anti PD-1, and Atezolizumab anti-PD-L1 antibody are under trials for their role in TNBC.

   3. LAR subtypes: The LAR subtype is characterized by high AR and an activating mutation in the kinase domain of PIK3CA. Antiandrogens, such as combination of bicalutamide with a PI3K inhibitor or enzalutamide, are being explored to target LAR subtypes.

   4. Epidermal growth factor receptor (EGFR): EGFR is expressed in 45–70% of TNBC and is associated with poor prognosis [10]. EGFR inhibitors are being evaluated in metastatic settings with not-so-promising results [18]. It has been seen that the EGFR downstream signaling pathways were still activated in most patients after EGFR-targeted treatment, suggesting that there might be other pathways involved in a bypass activation [4]. As a result, EGFR-targeted treatment alone cannot achieve significant efficacy. Use of growth factor inhibitors in BL-2, M, and MSL subtypes combined with other downstream signal transduction inhibitors might achieve better results [4].

Molecular subtypes	Cellular pathways	Therapeutic target
Basal-like 1 (BL1)	Cell cycle	PARP inhibitors
	DNA repair	Platinum agents
	Proliferation	Conventional chemotherapy
Basal-like 2 (BL2)	Growth factor pathways	mTOR inhibitors
	Metabolic pathways (glycolysis and gluconeogenesis)	Growth-factor inhibitors
Immunomodulatory (IM)	Immune cell processes	Immune-checkpoint inhibitors
Mesenchymal (M)	Cell motility, differentiation, and growth factor signaling	mTOR inhibitors
		EMT-targeted therapy
		CSC-targeted therapy
		AXL inhibitor
Mesenchymal Stem-like (MSL)	Low proliferation	PI3K inhibitors
	Angiogenesis	Antiangiogenic therapy
		SRC antagonist
Luminal androgen receptor (LAR)	Androgen receptor	Antiandrogen blockade
	Luminal gene expression	CDK4/6 inhibitors
	Molecular apocrine subtype	Immune-checkpoint inhibitors

Table 2.

Molecular pathways in TNBC subtypes as therapeutic targets.

Source: Silva D et al. [10] licensed under CC BY-NC 4.0.

2.4 Ki-67

The Ki-67 antigen, a nonhistone nuclear protein, was identified by Scholzer and Gerdes in 1983 in a Hodgkin lymphoma cell line [19]. The Ki-67 antigen encodes two protein isoforms with 345 and 395 kDa molecular weights [20]. This protein is expressed in the G1, S, G2, and M phase of the cell cycle but is absent in resting cells (G0) [20, 21]. Therefore, the nuclear expression of Ki-67 can be evaluated to assess tumor proliferation by IHC. The Ki-67 protein has a half-life of only 1–1.5 hours. Therefore, the quantity of Ki-67 present at any time during the cell cycle is regulated by a precise balance between synthesis and degradation [20].

2.5 Is Ki-67 a prognostic or a predictive marker?

A prognostic biomarker indicates the likely course of the disease in an untreated individual, and a predictive biomarker identifies subpopulations of patients most likely to respond to a given therapy. An increased Ki-67 is linked to a worse prognosis and an increased response to neoadjuvant chemotherapy. As Ki-67 represents proliferating tumor, a high level will translate to an increased response. However, same is not true for prognoses. Increased Ki-67 is an adverse prognostic factor in HR-positive tumors, and patients with low Ki-67 tumors have the best prognosis [21]. Whereas, in HR-negative tumors, low proliferating tumors have the worst prognosis. This phenomenon is also reported by Cortazar et al. in their meta-analysis, which shows that increased pCR rates are linked to better survival in the HR-negative subgroup [22]. At the same time, chemotherapy response does not affect the prognosis in HR-positive tumor. Based on this, Denkert et al. reported the biological plausibility of three different groups of tumors [21]:

1. Low proliferating tumors are not responding to chemotherapy but have a good prognosis (low Ki-67 linked to a good outcome)

2. High proliferating tumors are chemotherapy-sensitive, high Ki-67 is linked to an increased chance of pCR and improved survival (high Ki-67 linked to a good outcome)

3. High proliferating tumors are chemotherapy-resistant, increased Ki-67 is linked to reduced survival (high Ki-67 linked to a poor outcome)

2.6 Problems in Ki-67 assessment

1. Ki-67 cutoff as a measure of cell proliferation should be considered in the tumor’s histological type context: [6, 21] For example, a Ki-67 rate of 16% would indicate very high proliferation in a classic invasive lobular cancer; however, the same Ki-67 rate would indicate just about average proliferation in an Invasive Ductal Carcinoma [6]. Similarly, baseline Ki-67 values for TNBC are much higher than those for luminal tumors [23, 24]

2. Intratumoral heterogeneity [6, 21]:

   1. Spatial heterogeneity: the number of cells needed to be counted for consistent results might be much lower in tumors with low proliferation and high in tumors with high proliferation

   2. Temporal heterogeneity: This is commonly observed as a result of therapy. Several studies have shown that the short-term reduction of Ki-67 after 2 weeks of therapy is predictive of the outcome of endocrine therapy [5, 6, 21]

3. Inter-observer variability based on different evaluation approaches for different pathologists

2.7 Cutoff for TNBC

In the 2011 St. Gallen recommendations, a cutoff of 14% for the separation of luminal A and B tumors is suggested [25]. In 2013, the cutoff was revised to 20%, signifying 20% as the cutoff level for differentiating low proliferating Luminal A from high proliferating Luminal B [26]. However, the best cutoff point of this marker as a prognostic and predictive factor in TNBC remains unclear.

1. According to Zhu et al., the optimal cutoff value of Ki-67 for TNBC is 30% [27]. At a cutoff point of 30%, worse DFS and OS were observed in the Ki-67 high group.

2. Aleskandarany et al. reported the baseline mean Ki-67 value of Luminal cancer as 22 compared with 64.5 for TNBC tumors [23]. Therefore, 10% was the optimal cutoff in the luminal class separating low from moderate/high proliferative subgroups. In contrast, the cutoff for TNBC was found to be 70%. Moreover, there was no association of Ki-67 with survival compared to the luminal subtype.

3. In the study by Arafah et al., the median result of the KI-67 expression was 70%, with a range of 20–95% [24]. Moreover, High Ki-67 (>30%) was significantly associated with positive sentinel lymph node status, higher nuclear grade, diagnosis of an invasive tumor, advanced clinical stage, and adverse survival outcome.

4. Epithelial-mesenchymal transition (EMT) [24]: TNBC has also recently been linked to this phenomenon, which is characterized by the loss of the epithelial characteristics of the cells while they acquire a mesenchymal phenotype. This is a plausible explanation for distant metastases in breast cancer, skipping regional nodes. The high expression of KI-67 was correlated with an increased expression of Vimentin, a marker for EMT.

5. In the study by Keam et al., high Ki-67 (>10%) was associated with poor survival [28].

6. According to Wang et al., a high Ki-67 (>40%) level is associated with young age, higher grade, poor overall and recurrence-free survival [29].

7. A meta-analysis by Wu et al. reported the heterogeneity in cutoff value for Ki-67 for different studies, ranging from 10 to 50% [30]. Also, high Ki-67 is associated with worse overall and recurrence-free survival.

2.8 pCR

1. Arafah et al. reported a statistically significant association of high Ki-67 with the inability to achieve pCR [24].

2. Nishimura et al. reported that high Ki-67 levels and TNBC status were associated with higher pCR [31]. Moreover, no pathological responder in cases with Ki-67 < 25%

3. In the study by Keam et al., TNBC with high Ki-67 showed a higher pCR rate (18.2%) to neoadjuvant chemotherapy than TNBC with low Ki-67 (0.0%) [28].

2.9 Is there a triple-negative paradox?

Higher Ki-67 predicts a higher response to chemotherapy and a higher pCR rate, which generally means a good prognosis [21]. However, high Ki-67in TNBC is associated with poor recurrence-free or overall survival. This was termed as a “triple negative paradox.” However, this paradox can be explained by a higher likelihood of relapse in patients whose pCR was not achieved. In a study by Keam et al., only 18.2% of TNBC patients achieved pCR, and patients with high Ki-67 residual disease had statistically significant poor prognoses than patients with residual disease and low Ki-67 or patients with pCR with high Ki-67 [28]. Similarly, in a study by Carey et al., only 27% of patients with basal subtype achieved pCR [32]. Therefore, in patients with TNBC, the higher number of non-pCR patients tilt the results toward poor prognoses resulting in the so-called “triple negative paradox.”

Only 20–30% of patients with TNBC achieved pCR on neoadjuvant chemotherapy [21, 32], and pCR was strongly associated with prolonged overall survival [15, 33, 34]. Moreover, patients with TNBC who achieve pCR have the same prognosis as patients with non-TNBC [33]. However, among patients who did not achieve pCR, patients with TNBC have a significantly poorer outcome than patients with non-TNBC [33].

3. Conclusions

Although marred by intratumoral heterogeneity and inter-observer variability, true to the highly proliferative nature of TNBCs, higher baseline Ki-67 levels are seen as compared to luminal tumors. A higher Ki-67 is associated with a higher pCR rate in TNBC. However, the best cutoff point of this marker as a prognostic and predictive factor in TNBC remains to be seen even after many researchers have explored this idea. Moreover, the “triple negative paradox” concept is more of a myth arising from more non-pCR patients in the TNBC group.

Acknowledgments

Thanks to Dr. Biswajit Dubashi , Professor, Department of Medical Oncology, Jawaharlal Institute of Postgraduate Medical Education and Research (JIPMER), Puducherry, India, for reviewing the manuscript and suggesting improvements.

Conflict of interest

The authors declare no conflict of interest.

Acronyms and abbreviations

AXL	tyrosine-protein kinase receptor
BL	basal-like
CDK	cyclin-dependent kinase
CSC	cancer stem cells
DNA	deoxyribonucleic acid
EGFR	epidermal growth factor receptor
EMT	epithelial-mesenchymal transition
FGFR	fibroblast growth factor receptors
IGF-1R	insulin-like growth factor receptor
IL	interleukin
IM	immunomodulatory
LAR	luminal androgen receptor
MET	hepatocyte growth factor
MSL	mesenchymal stem like
mTOR	mammalian target of rapamycin
PARP	poly ADP-ribose polymerase
PD1	programmed cell death 1
PDGFR	platelet-derived growth factor receptors
PD-L1	programmed death-ligand 1
PI3K	phosphatidylinositol 3-kinase
SRC	Proto-oncogene tyrosine-protein kinase Src
TGFβ	transforming growth factor beta
TNBC	triple-negative breast cancer