Despite the substantial advancements in cancer research, breast invasive carcinoma (BRCA) still holds the record as the most prevalent neoplastic malignancy, especially in women.¹ It accounts for 25% of cancer cases and about 17% of cancer deaths.² The GLOBOCAN Cancer Tomorrow prediction tool estimates that the incidence of the disease will increase by over 55% by 2050.³ Mangone et al. pointed out that the 5-year relative survival rate of BRCA patients at stages I, II, III, and IV was 100%, 89.7%, 71.4%, and 29.1%, respectively (with 95% confidence intervals P-value <0.05).⁴ Therefore, detecting breast cancer at an early stage plays a pivotal role in reducing patient mortality and eradicating one of the biggest challenges in healthcare. BRCA is a genetically and clinically heterogeneous disease with several subtypes. These subtypes have been classified based on the expression of the following hormone receptors: estrogen (ER+), human epidermal growth factor (HER2+), and progesterone (PR+).⁵ In addition, there is also the triple-negative (TNBC) subtype, which is characterized by the lack of expression of these receptors.⁵ The expression status of these three receptors, Ki-67, a proliferative index, and p53, is used as molecular predictive and prognostic breast markers.⁶ Furthermore, other markers have been reported such as several gene alterations, miRNAs, urokinase-type plasminogen activator (uPA) and plasminogen activator inhibitor1 (PAI-1), programmed death ligand 1 (PD-L1), and programmed cell death protein 1 (PD-1) receptors as well as microsatellite repeat instability (MSI) for BRCA.^6-11 However, as tumor cells are highly heterogeneous, these biomarkers are unfortunately insufficient to classify BRCA patients and detect tumor formation at earlier stages.

In the search for reliable early detection methods, biomarkers have played an increasingly important role in recent years. They offer a non-invasive, cost-effective, and efficient means of early detection compared to traditional methods such as mammography, ultrasonography, and biopsy. Although biopsy is a reliable tool for a specific diagnosis compared to other methods, it is inefficient as it requires multiple procedures, usually taking 7 to 10 days to obtain a result. Therefore, finding the most efficient biomarker(s) to enable the earliest diagnosis, increase the possibility of personalized treatment, and facilitate monitoring the response to therapy is crucial.

It has been shown that the expression status of specific isoforms is more relevant and informative as a biomarker than overall gene expression for disease diagnosis, subtyping, prognosis, and drug response prediction.^12-15 Additionally, Zhleh et al. indicated that expression-based features were significantly more strongly associated with sensitivity to most drugs than mutations and copy number alterations.¹⁵ Several isoform-specific potential biomarkers have been reported as predictive outcome indicators in breast cancer, not only for detecting tumor formation or progression but also for drug response.^16-19 Nevertheless, none of them are considered predictive biomarkers for the early stage and are routinely used in clinics.

This study used web-based bioinformatics platforms to assess the transcript isoforms expressed in Stage I BRCA tissue samples. We further analyzed differentially expressed isoforms that are not functionally involved in BRCA and aimed to discover novel potential predictive isoform biomarkers that can be used for early diagnosis. As BRCA is a global health concern that requires the development of novel and highly efficient approaches for diagnosis and treatment, this study opens a pathway for the detection of potential early isoform biomarker(s) to improve diagnosis and personalized treatment.

Multiple transcript and protein isoforms are generated by alternative splicing events that significantly increase transcriptomic and proteomic diversity. Several studies revealed that cancer exhibits extensive RNA dysregulation, so in addition to gene signature differences, isoform signature differences can be used to distinguish normal tissue from cancer.^26-29 In line with this idea, this study uncovered the various differentially expressed transcript isoforms and investigated whether these spliced isoforms are relevant to the survival and diagnosis of stage I BRCA patients. As a result, our data demonstrated that elevated expression of NM_024037 (belongs to AUNIP gene), NM_001143782 (belongs to FKBP11 gene), and NM_021619 (belongs to PRDM12 gene) has a high diagnostic performance and could be considered as potential biomarkers for early-stage BRCA diagnosis.

This study showed a significant association between BRCA and the transcript isoform NM_024037, which belongs to the AUNIP gene. AUNIP (Aurora Kinase A and Ninein Interacting Protein), also known as AIBp, plays a crucial role in mitotic spindle assembly, the maintenance of centrosomal structure during the cell cycle process, and the repair of DNA double-strand breaks.^30,31 It has been shown that there is an interaction between AUNIP and Aurora-A, co-expressed in brain tumors.³⁰ Moreover, one of the studies reported that AUNIP expression was upregulated in oral squamous cell carcinoma (OSCC), and depletion of AUNIP expression inhibited cell proliferation in OSCC.³² AUNIP has also been associated with tumor infiltration and might serve as a prognostic and diagnostic biomarker for hepatocellular carcinoma (HCC) and lung adenocarcinoma (LUAD). ³³ In addition, a recent study indicated that AUNIP was one of the critical potential diagnostic and therapeutic targets for pancreatic cancer.³⁴

To our knowledge, no association between AUNIP and BRCA development has been reported. In this study, ROC curve analysis revealed that one of the two transcript isoforms of the gene, NM_024037, had the most significant diagnostic ability among the other isoforms with a considerable AUC value (93%) and might be an excellent early predictor to detect tumor tissue. On the other hand, increased expression of NM_024037 did not correlate with survival in Stage I, since the early-stage survival ratio was very high up to 5 years. Moreover, the enrichment analysis showed that the isoform might be involved in pathways associated with the occurrence and progression of human malignant tumors. Overall, this data suggests that the NM_024037 isoform may play a crucial role as a promoter of breast tumor formation and can be used as a diagnostic isoform biomarker. However, functional and more detailed molecular studies are required to understand the function of the NM_024037 isoform in BRCA development.

The transcript isoform NM_001143782, which belongs to the FKBP11 gene, showed statistically significant results for predicting the diagnosis of Stage I BRCA patients. FKBP11 (FKBP Prolyl Isomerase 11) is a member of FK506-binding proteins (FKBPs), which belong to the family of conservative intracellular immunophilins.³⁵ The FKBP gene family comprises 16 members that play critical roles in various biological activities such as cellular homeostasis, metabolism, T-cell activation, and carcinogenesis.^36,37 However, several studies have demonstrated that increased expression of FKBP11 promotes tumorigenesis in OSCC, clear cell renal cell carcinoma (ccRCC), LUAD, and osteosarcoma. ^38-41 Therefore, these findings suggest that FKBP11 acts on the development of diverse cancer types; however, there is currently no data on FKBP11 to demonstrate its involvement with BRCA development.

Our results put the importance of FKBP11 at the isoform level (NM_001143782) for predicting the prognosis and diagnosis of BRCA. The increased expression of the NM_001143782 transcript isoform was associated with decreased survival when differences in the patient stage were not accounted for, as expected. In addition, survival analysis of this isoform showed no significance when Stage I BRCA patients were considered. The enrichment analysis also supports the idea that this isoform could contribute to the pathways involved in regulating gene expression, activating pro-tumorigenic signaling, and transforming healthy cells into cancer. Consequently, in vitro and in vivo studies should be performed to clarify the exact role of the NM_001143782 transcript isoform in the development of breast cancer. Our study also revealed that the transcript isoform NM_021619, which belongs to the PRDM12, has a high diagnostic capacity in Stage I BRCA patients. PRDM12 belongs to the PRDM [PRDI-BF1 (positive regulatory domain I-binding factor 1) and RIZ1 (Retinoblastoma Protein-Interacting Zinc Finger Gene 1) homologous domain containing] gene family that encodes for Kruppel-like zinc finger proteins. ⁴² Histone-modifying enzymes interact with PRDM proteins directly or indirectly.^42,43 Thus, these proteins were proposed to regulate gene expression. ⁴⁴ Sorrentino et al. demonstrated that mRNA expression of PRDM12 was increased in various cancers, such as breast, ovary, lung, colon, kidney, liver, prostate, and thyroid, compared to normal adult tissues.⁴⁵ However, several studies have indicated that derivative chromosome 9 deletion or rearrangements within a region containing the PRDM12 gene prompt the aggressive, chronic myeloid leukemia (CML) phenotype and correlate with poor survival of patients.^46,47 A recent study has reported that the combined expression of some genes, including PRDM12, is associated with poor progression-free survival in BRCA patients with lymph node invasion. ⁴⁸ Our results showed that high-level expression of PRDM12 transcript isoform, NM_021619, is linked to poorer survival for all BRCA stages, and that it has a significant ability to distinguish normal from Stage I cancer tissue. Besides, no significant prognostic impact of this isoform has been detected for Stage I. Notably, the interaction between PRDM12 and EZH2, which is also involved in BRCA growth and metastasis, demonstrates the functional relevance of PRDM12 in the development and progression of breast tumors.^49,50 Therefore, accurate functional data are essential to elucidate the possible role of NM_021619 in the early phase of breast tumorigenesis. Besides these findings, the main limitation of the study is that the data were obtained from bioinformatics analyses. For this reason, these potential early predictive BRCA isoform biomarkers need to be functionally analyzed exhaustively by performing both in vitro and in vivo assays to determine how their translation from bench to bedside will be impactful.

This study proposes the three isoforms NM_024037, NM_001143782, and NM_021619 as diagnostic biomarkers that can be used mainly for early-stage BRCA patients (i.e., Stage I). As far as we know, this study is the first to report the analysis of TCGA data based on transcript isoforms at the level of expression, diagnosis, and prognosis in BRCA Stage I. As previously emphasized, analysis at the isoform level is critical for discovering precious biomarkers to impede the misdiagnosis of cancers. Because of this, the present study provides valuable insight into the discovery of valid and reliable BRCA markers and the development of novel personalized therapeutic strategies for this intractable healthcare problem.

Ethical approval is not required as all data used in this study were obtained from a publicly accessible database.