Breast cancer is one of the most common malignancies and causes of cancer-related death in women including South Korea [1]. Although the development of human epidermal growth factor receptor 2 (HER2) target therapy [2,3], endocrine therapy [4], treatment for endocrine therapy resistance [5], and immune checkpoint inhibitors [6,7] has dramatically increased the survival rate of breast cancer, breast cancer patients still suffer from late recurrences after treatment and show resistance to various treatments.

The median age at diagnosis in South Korea is approximately 10 years younger (52.8 years) than that in the United States (62 years) [1,8]. In addition, the proportion of young breast cancer patients diagnosed with breast cancer under the age of 40 is higher in South Korea (10.6 %) than in the United States (5%) [1,8]. Although the prevalence of BRCA1/2 mutation in South Korea is similar to that reported in Western countries [9,10], other predisposing germline mutations in young breast cancer patients with BRCA1/2-negative were quite different in distribution [9]. As the characteristics of breast cancer in South Korea tend to be different, it is necessary to accumulate sufficient information of breast cancer in South Korea and to establish the differentiation strategies of breast cancer treatment.

As molecular analysis advances, many tests that predict a poor prognosis for breast cancer have been widely developed and used. Multigene assays that use tumor RNA, such as Oncotype DX [11] and MammaPrint [12], are helping to predict the risk of recurrence and metastasis, along with the benefit of adjuvant chemotherapy, in hormone receptor (HR)–positive breast cancer. In addition, molecular analysis using tumor DNA has been used to predict prognosis by identifying molecular alterations associated with poor prognosis, including single nucleotide variants (SNVs), small nucleotide insertions and deletions (Indels), copy number alterations (CNAs), and gene rearrangement. Genetic alterations in various genes are associated with poor prognosis, such as PIK3CA (phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha) [13], ERBB2 [14], ESR1 (estrogen receptor 1) [15], and NF1 (neurofibromin 1) [16] and the genetic alterations have been widely analyzed by Sanger sequencing or real-time polymerase chain reaction. Next-generation sequencing (NGS) is an ideal tool for detecting molecular alterations associated with poor outcomes. It could help clinicians evaluate comprehensive molecular profiles of breast cancer and establish appropriate treatment plans. NGS also can provide numerous data about molecular alterations associated with target therapy.

The U.S. Food and Drug Administration (FDA) approved alpelisib for the treatment of patients with HR-positive/HER2-negative metastatic breast cancer and a PIK3CA mutation and olaparib for the treatment of patients with HR-positive/HER2-negative metastatic breast cancer with a germline BRCA mutation. Pembrolizumab was approved by the FDA to treat patients with microsatellite instability–high (MSI-H) solid tumors or solid tumors with a high tumor mutation burden (TMB). When treating patients with solid tumors that harbor NTRK gene fusion, larotrectinib and entrectinib can be administered. When the amount of information that can be provided is considered, an NGS study is cost-effective and has a relatively fast turnaround time because it can identify multiple target biomarkers at once. However, some researchers have suggested skeptical opinions about the need for NGS studies in daily practice, especially in breast cancer [17]. Therefore, it is crucial to accumulate real-world data of NGS studies and evaluate the effectiveness NGS in real-world clinical setting. In this study, we analyze real-world data about the prevalence of various genes used to prescribe targeted therapy for breast cancer and investigate how much target therapy has actually been applied.

NGS is an ideal method for comprehensive molecular profiling of solid tumors. Given the amount of information provided, NGS is relatively cost-effective and has a fast turnaround time. It has opened the era of precision medicine by detecting molecular alterations in a wide genomic region that can predict a patient’s prognosis or treatment effect and guide clinicians in administering an appropriate target therapy. In this study, we collected data from our institute to determine the effectiveness of NGS and present real-world data about precision medicine based on NGS results.

The drug-targetable biomarker most commonly found in our study was PIK3CA mutations, which were found in 43.1% (31/72) of HR-positive, HER2-negative breast cancer patients and 11.5% (11/96) of TNBC patients. The SOLAR-1 clinical trial showed that alpelisib plus fulvestrant significantly prolonged progression-free survival (PFS) in HR-positive, HER2-negative breast cancer patients with a PIK3CA mutation [25]. The SOLAR-1 trial included patients with PIK3CA hotspot mutations in the C2, helical, and kinase domains (corresponding to exons 7, 9, and 20, respectively) as found using QIAGEN’s therascreen PIK3CA RGQ kit, and the FDA approved that PCR kit as a companion diagnostic device for alpelisib. In this study, 72.5% (37/51) of cases with a PIK3CA mutation had PIK3CA SNVs included in that kit, and the other 27.5% (14/51) had mutations not included in the kit [24]. The N345K and Q546K mutations showed sensitivity to PI3K pathway inhibitors in preclinical models [26]. Therefore, additional studies are needed to determine whether breast cancer patients with other rare variants that are not included in the therascreen PIK3CA RGQ PCR kit show therapeutic benefit from receiving PI3K pathway inhibitors. A sufficient investigation should be conducted to allow more patients with advanced HR-positive, HER2-negative breast cancer and PIK3CA mutations to benefit from targeted therapy.

Another important drug-targetable biomarker is BRCA1/BRCA2 mutation. BRCA1 and BRCA2 are genes associated with the risk of ovarian cancer and breast cancer. Excluding cases in which germline BRCA1/BRCA2 mutations could not be confirmed, a BRCA1 germline mutation was found in 4.8% (4/84) of the patients in this study, and a BRCA2 germline mutation was found in 6.0% (5/84) of patients. No somatic mutations of BRCA1 were found, but somatic mutations of BRCA2 were found in 4.8% (4/84) of patients.

Olaparib has been approved by the FDA for use as adjuvant chemotherapy in patients with germline BRCA mutations in high-risk, HER2-negative breast cancer or as palliative chemotherapy in patients with germline BRCA1/2 mutations in metastatic breast cancer. The National Comprehensive Cancer Network guideline also suggests that olaparib can be useful in some patients with somatic BRCA mutations in metastatic breast cancer. In addition to anti-cancer treatment, it has been reported that prophylactic mastectomy and salpingo-oophorectomy reduce the lifetime cancer risk and increase survival times in patients with a germline BRCA1/2 mutation [27]. Because a BRCA1/2 mutation is a dominant inheritance, familial history should be evaluated, and genetic consultations should be used to determine the lifetime cancer risk of family members. In breast cancer patients, BRCA1/2 mutation evaluations must be performed through germline BRCA1/2 sequencing or NGS.

In addition to drug-targetable biomarkers, NGS can provide information about molecular alterations that can predict therapy resistance. ER with a hotspot mutation in the ligandbinding domain (LBD) of ESR1 undergoes a conformational alteration to maintain an agonistic active conformation similar to estrogen-bound wild-type ER [28]. ER with genomic rearrangement of ESR1 thus creates a chimeric protein containing a proximal promoter domain without an LBD, resulting in the production of a constitutive promoter and ligand-independent tumor growth [28]. Therefore, an ESR1 mutation in HR-positive breast cancer can predict resistance to endocrine therapy [15,29,30]. ESR1 mutations are acquired after the administration of endocrine therapy, especially aromatase inhibitors, but some primary breast cancer without a history of endocrine therapy can also harbor an ESR1 mutation [28]. In our study population, 3 of 11 cases with ESR1 mutations had not received endocrine therapy before tissue sampling, suggesting that ESR1 mutations can be found in primary breast cancer without prior exposure to aromatase inhibitors.

An ESR1 mutation can be used as a predictive biomarker, as well as a drug-targetable biomarker. Recently, the EMERALD trial reported that an oral selective ER degrader, elacestrant, showed significant PFS improvement in patients with advanced/metastatic HR-positive/HER2-negative ESR1-mutant breast cancer who had progressed during prior treatment with endocrine therapy or CDK4/6 inhibitor therapy [31]. Therefore, an evaluation for an ESR1 mutation is crucial not only to predict endocrine therapy resistance, but also to enable the administration of appropriate targeted therapy.

TMB is another type of important information that NGS can provide. Many studies have reported that high TMB is associated with the production of many neoantigens and increases T cell reactivity, suggesting that it might be closely related to an improved immune checkpoint inhibitor (ICI) response [32,33]. The clinical benefit of ICI administration has been reported in patients with breast cancer, especially TNBC [6,7], and predictive biomarkers of an ICI response such as programmed death-ligand 1 are in the limelight as an essential evaluation in real clinical settings. However, TMB-high breast cancer was reported to be very rare [34], and our study confirms that only 6.1% of tumors had TMB ≥ 10. No MSI-H tumors were found.

For an accurate TMB evaluation, it is necessary to exclude germline variants by using matched normal tissues or blood. However, it is difficult to analyze both tumor tissue and matched normal tissues together in a real-world clinical setting because of financial and time constraints. TMB might be overestimated in a tumor-only NGS study unless germline variants or FFPE-induced deamination artefacts are properly excluded. Therefore, SNVs or Indels with a high probability of being germline variants should be carefully excluded using population frequencies in a well-known germline database [35]. Further studies should be conducted to establish guidelines for classifying variants as germline or somatic by setting appropriate cut-off values for population frequency. In addition, appropriate standards for total depth and variable frequency alleles should be established to exclude spurious variants generated by artefacts.

Recently, various groups have expressed skepticism about the need for NGS studies in daily practice. It is difficult to apply NGS to all patients because it is more time-consuming and expensive than other screening tools. For example, PIK3CA mutations in breast cancer, which are drug-targetable biomarkers, can be detected by real-time PCR; HER2 can be evaluated by IHC and in situ hybridization, and germline variants of BRCA1/2 can be evaluated by blood sequencing [17]. Our results indicate that only 7.2% of patients benefited from their NGS study, and were not significantly different from previous studies in South Korea [36]. Also 33% (3/9) of patients with a PIK3CA mutation had to undergo an additional companion diagnostic test. However, NGS studies can provide many predictive biomarkers can predict aggressive behavior by breast cancer, such as PIK3CA [13] and ERBB2 [14] mutations, or resistance to endocrine therapy, such as ESR1 [15] and NF1 [16] mutations. In addition, NGS examinations can provide new therapeutic opportunities to patients with metastatic or advanced breast cancer by detecting genomic alterations that match the requirements of clinical trials testing targeted therapies. Our study revealed that among 13 patients who benefited from precision medicine based on NGS study, 76.9 % (10/13) of patients received target therapy through clinical trial or expanded accecss of e-IND. These results demonstrated the effectiveness of NGS studies in enabling clinicians to provide target therapies as part of a clinical trials or e-IND to patients, especially advanced breast cancer patients who are refractory to most of standardized treatments and have extremely limited available treatment options.

In conclusion, our study has revealed that only 7.2% of breast cancer patients who underwent NGS in a real-world clinical setting benefited from the test. Given financial and time constraints, the application of NGS to all patients with breast cancer in daily practice might be excessive. However, it is possible to forecast prognoses of breast cancer by discovering predictive biomarkers that suggest aggressive behavior and resistance to endocrine or target therapy and by detecting potential drug-targetable biomarkers that can offer patients the chance to participate in a clinical trial. Therefore, the value of NGS in advanced or metastatic breast cancer could be increased through further studies.