TNBC is an aggressive subtype of breast cancer that is characterized by resistance to therapy and poor patient survival. Because of the lack of direct targets for treatment currently, it is especially important to identify gene mutations that can be used as therapeutic targets in patients with TNBC. To improve patient outcomes and to optimize treatment regimens, novel therapeutic targets need to be identified.
In the present study, using the NGS technique, we aimed to identify novel gene mutations in a cohort of 23 very-early-relapsed TNBC patients to identify new direct targets for treatments. Both the NGS platform and the cancer panel genes chosen in this study were previously used in several other studies on breast cancer  and other types of cancers .
We identified 137 CRGAs in 23 very-early-relapsed TNBCs, among which TP53, PTEN, RB1, PIK3CA, BRCA1, NOTCH1, MYC, and CCND1 were the most frequently mutated genes (Fig. 1) in our cohort. Compared with previous studies in TNBC, TP53 was still the most frequently mutated gene, while the mutation frequency of PTEN and RB1 was higher than that reported in other literature [12, 13]. It has been reported that those 3 tumor suppressors are also the most frequent drivers of metastasis in diverse types of solid human cancers, not just in breast cancer . Notably, our study found that 7 very-early-relapsed TNBC patients who were detected with RB1 mutations, including 4 frameshifts and 1 for each of missense mutation, loss, and splice site, had shorter DFS than patients without RB1 mutations; this difference was statistically significant. Thus, understanding the impact of these tumor suppressors on clinical outcomes could be valuable.
Of further note, 2 of 23 patients were detected with ERBB2 mutation. One patient detected with ERBB2 amplification was diagnosed with IIIC TNBC in May 2016, and lung metastasis occurred 5 months after surgery. ERBB2 amplification implied that the anti-Her2 theory might be correct. Another patient with ERBB2 IHC (1+) was found with an ERBB2-P780_Y781insGSP mutation, the third most common HER2 exon 20 insertions in lung cancer , which indicated that anti-Her2 therapies such as neratinib and trastuzumab ado-trastuzumab emtansine (T-DM1) might benefit patients. Notably, this insertion mutant is located in the Pkinase-Tyr sequence of ERBB2. Mutations in the ERBB2 kinase domain have been identified in about 2–5% of various human cancers . Lapatinib, which is known as a small molecule tyrosine kinase inhibitor (TKI), targeted the kinase domain of ERBB2-approved for breast cancer patients and may be resistant because of this insertion mutant . Another ERBB2 T798I mutation that occurs in the same kinase domain has been demonstrated to cause a strong lapatinib-resistance effect by in vitro study . However, whether the ERBB2 kinase domain mutation detected in our study could lead to clinical drug resistance or not has been validated by preclinical studies.
Rare ROS1 fusion in breast cancer
Comprehensive genomic profiling analysis revealed a novel ROS1-EPHA7 rearrangement. ROS1 fusion was detected in 2.59% of Chinese non-small cell lung cancer (NSCLC) patients  but has not been found in TNBC patients before. It has been reported that the objective response rate (ORR) of crizotinib in ROS1 fusion NSCLC patients was 83.3% . Notably, further IHC and FISH testing verified the existence of this novel ROS1 fusion on the RNA and protein level, suggesting that the ROS1 fusion may have a biological function. Unfortunately, the patient experienced disease progression after 5-month vinorelbine-capecitabine-combined chemotherapy as the first-line treatment and was then lost to follow-up. Thus, the response to crizotinib could not be observed in this patient.
We also detected a majority of mutations identified in only 1 patient (Fig. 1), which can be explained by the high heterogeneity of TNBC . These low-frequency mutations also have important clinical implications. For instance, ARID1A and MCL-1 have been related to chemotherapy sensitivity, ARID1A down-regulation has been associated with a poorer response to paclitaxel-based chemotherapy in patients with TNBC , and MCL, which is frequently co-amplified with MYC, has been associated with resistance to chemotherapy [22, 23] and decreased DFS. For in vitro studies, the role of IKBKE, IGF1R, NOTCH3, and MDM4 in tumorigenesis and tumor metastasis have been reported [25, 26, 27, 28] and have provided clinicians with potential insights for understanding the biological behavior of TNBC and exploring treatment strategies for heavily treated patients.
The genes that were of significant interest in our study could be enriched in key signaling pathways, like the PI3K/mTOR pathway, cell cycle pathway, growth factor receptors, or DNA repair, and alterations in these genes could be a potential therapeutic target. PI3K/mTOR pathway has the highest mutation frequency. In our study, the mutation of PI3K-AKT signaling pathway included the PI3K catalytic subunits (PIK3CA, PIK3CB), PI3K regulatory subunit (PIK3R1), AKT-independent mTOR pathway activator (STK11), and the loss of PTEN . In a preclinical study, TNBC cell lines of M and LAR subtypes preferentially responded to the dual PI3K/mTOR inhibitor NVP-BEZ235. The benefit of the pan-PI3K inhibitor BKM120 in metastatic TNBC, both in monotherapy and combination therapy with PARP inhibitors, is undergoing clinical research (NCT01629615; NCT01790932; NCT01623349) . The effectiveness of everolimus (the most studied blocking agent aimed at the mTOR kinase) in both primary and metastatic TNBC was confirmed by clinical trials [32, 33]. These promising data demonstrate that PI3K inhibitors or mTOR inhibitors may help select TNBC patients with activating mutations in the PI3K-AKT-mTOR pathway.
Some studies identified a subgroup of TNBC with a deficiency of DNA repair, mainly due to mutations or methylation of BRCA1/2, and other genes involved in DNA damage repair pathway [13, 34]. A clinical trial (NCT00494234) for a poly adenosine diphosphate-ribose polymerase (PARP) inhibitor, olaparib, in patients with BRCA1 or BRCA2 mutations and advanced breast cancer, provided an impressive ORR of 44% . A randomized, phase 3 trial in which olaparib monotherapy was compared with standard therapy in patients with a germline BRCA mutation and human epidermal growth factor receptor type 2 (HER2)-negative metastasis breast cancer, detected a longer progression-free survival (PFS) of 7.0 months in the olaparib group than the 4.2 months (HR = 0.58, 95%CI: 0.43–0.80,p < 0.001), but no statistically significant improvement in OS[36, 37]. Given that most BRCA1/2 carriers are attributed to TNBC , olaparib could provide a significant benefit among TNBC patients deficient in DNA damage repair. Except for BRCA1/2, many mutations associated with TNBC are mainly distributed in DNA damage repair pathway, including the above-mentioned PALB2, RAD21, and MSH2, along with some other genes that were not detected in our study. Therapies designed for these mutated genes are scarce. It is still unclear whether these mutated genes can be treatment targets or not, but the utility of DNA cross-linking agents in combination with targeted agents has been reported to improve the curative effect for patients with DNA damage repair .
Ras/MAPK activity can be aberrantly stimulated via the copy number alterations of KRAS and somatic alterations of NF1. Preclinical studies have demonstrated that basal type breast cancer cells have an activated RAS-like transcriptional program and are significantly more sensitive to MEK inhibitors compared with luminal and HER-2 amplified lines . Treatment with MEK inhibitor caused the up-regulation of PI3K signaling, and the dual inhibition of both pathways could achieve better anti-tumor effects both in vitro and in vivo . These studies provide a rational hypothesis for patient selection in clinical trials with the aim to evaluate the clinical effect of MEK and PI3K inhibitors in TNBC. Clinical trials of EGFR-targeted TKIs targeting EGFR amplification in TNBC failed in both TKI monotherapy and in combination with chemotherapy [42, 43]. It is still controversial if TNBC patients may respond to EGFR-TKI agents.
TNBCs are a highly proliferative group of tumors enriched for high expression of cell-cycle genes, although they are considered to be resistant to CDK4/6 inhibitors. As a heterogeneous disease, and early preclinical study has shown that the luminal androgen receptor (LAR) subtype of TNBC was highly sensitive to CDK4/6 inhibition both in vitro and in vivo in MDA-MB-453 LAR cell line xenografts compared with the basal-like subtype . Our study also illustrated that treatments that target the cell cycle pathway might be effective in selected TNBC patients.
Our study also has some limitations. Firstly, as a hospital-based retrospective study, the number of our samples was limited by sample quality and patient follow-up. Secondly, only 2 of 23 very-early-relapsed breast cancer patients were still under treatment but not with on-label targeted drugs; as a result, the efficacy of the drug predicted by FoundationOne CDx (F1CDx) cannot be determined in this study. Meanwhile, the patient with the rare ROS1 fusion was lost to follow-up, so whether crizotinib can benefit TNBC patients with ROS1 fusion was not validated in this study.