Repurposing of drugs for triple negative breast cancer: an overview

Breast cancer (BC) is the most frequent cancer among women in the world and it remains a leading cause of cancer death in women globally. Among BCs, triple negative breast cancer (TNBC) is the most aggressive, and for its histochemical and molecular characteristics is also the one whose therapeutic opportunities are most limited. The REpurposing Drugs in Oncology (ReDO) project investigates the potential use of off patent non-cancer drugs as sources of new cancer therapies. Repurposing of old non-cancer drugs, clinically approved, off patent and with known targets into oncological indications, offers potentially cheaper effective and safe drugs. In line with this project, this article describes a comprehensive overview of preclinical or clinical evidence of drugs included in the ReDO database and/or PubMed for repurposing as anticancer drugs into TNBC therapeutic treatments.


Background
Breast cancer (BC) is the most frequent cancer among women in the world. Triple negative breast cancer (TNBC) is a type of BC that does not express oestrogen receptors, progesterone receptors and epidermal growth factor receptors-2/Neu (HER2) and accounts for the 16% of BCs approximatively [1,2]. Due to its lack of response to hormone and targeted therapies, the number of therapeutic opportunities is limited [3,4]. TNBC patients are difficult to treat, with unfavourable prognosis and are generally administered with the standard chemotherapy. At the moment, novel treatment approaches, such as immunotherapy, as well the repurposing of old drugs currently used for indications other than TNBC, is under investigation. In this context, we have previously reviewed the preclinical and clinical anticancer efficacy and safety of beta blockers in TNBC [5]. Table 1 shows all 17 clinical references collected (the article of Spera et al analyses two different retrospective studies on beta blockers efficacy and safety on TNBC [13], and the articles of Hagasewa et al [15] and Ishikawa et al [16] analysed the same cohort of patients). Clinical evidence on twelve licensed drugs was found, and of these drugs, eleven out of 268 (4.1%) were included in ReDO_DB. Eleven studies out of 18 were retrospective studies [10-13, 17, 19, 20, 22, 25, 26], six were phase II and [14-16, 18, 21, 23] one was a phase I clinical trial [24] (see Figure 1 for more details). Retrospective studies ranged from 1995 to 2016, and six out of eleven studies analysed a USA cohort of patients [10,12,19,20,25,26]. Eight studies were performed using medical records [10,12,17,19,20,22,25,26], one was based on disease registries [11] and two reported the results of previous clinical trials [13]. Of the 18 clinical studies collected, four analysed the efficacy of beta blockers (BB) [11][12][13], five of non-steroidal anti-inflammatory drugs (NSAIDs) [17][18][19][20][21], two of zoledronic acid [15,16], one of metformin [10], one of tetramolybdate [14], one of itraconazole [22], one of esomeprazole [23], one of mifepristone [24] and two of statins [25,26]. Outcomes retrieved from clinical studies were grouped, whenever possible, in pharmacological categories and summarised in Table 2.

Beta blockers (BBs)
BBs were evaluated on postmenopausal women with operated early primary TNBC, on women with invasive TNBC (receiving neoadjuvant chemotherapy), and on women with advanced or nodal positive TNBC. Study populations ranged from 35 patients to 1,417 patients. In the study of Melhem-Bertrandt et al [12], using medical chart and pharmacy data from the Breast Cancer Management System Database in the USA, women with invasive TNBC receiving neoadjuvant chemotherapy plus BBs were compared to patients receiving only neoadjuvant chemotherapy between 1995 and 2007. Hazard ratio of recurrence free survival for women administered with chemotherapy plus BBs was 0.30 (95% CI, 0.10-0.87; p = 0.027) and hazard ratio of overall survival was 0.35 (95% CI, 0.12-1.00; p = 0.05) [12]. Also, in the retrospective study of Botteri et al [11]

Metformin
The retrospective study of Bayraktar et al [10] using medical chart and pharmacy data from the Breast Cancer Management System Database compared women who received adjuvant chemotherapy with or without metformin in the USA between 1995 and 2007. In total, 1,448 patients (63 diabetic patients receiving metformin, 67 diabetic patients not receiving metformin and 1318 not diabetic patients). The 5 years survival estimates for distant metastasis free survival were 73% in the metformin group, 66% in the nonmetformin group and 60% in the non-diabetic group (p = 0.23). Overall survival was 67% in the metformin group, 69% in the non-metformin group and 66% in the non-diabetic group (p = 0.58). Recurrence free survival was 65% in the metformin group, 64% in the non-metformin group and 54% in the non-diabetic group (0.38). Also, after adjustments, no significant survival outcomes were obtained.

Tetramolybdate
The primary endpoint of phase II open label single arm study of Chan et al [14] was to assess the change in VEGFR2+ endothelial progenitor cells in women treated with tetrathiomolybdate. The study, performed on 36 women with stage II/III TNBC during adjuvant setting, showed that two year event free survival was 90%.

Zoledronic acid
The articles of Hasegawa et al [15] and Ishikawa [16]

Mifepristone
In the Phase I, randomised study of Nanda and colleagues performed in USA, four women with metastatic or locally advanced TNBC were analysed (those patients were allocated to mifepristone plus paclitaxel or placebo). Unfortunately, no information about patients allocation, nor any outcome information could be retrieved from this article [24].

Statins
The retrospective study of Shaitelman et al [26] used medical records from the MD Anderson Cancer Centre to investigate if women with stage I-III TNBC receiving statins at any time from diagnosis. The authors showed that patients receiving statins did not get any advantage compared to the non-statin users group (0.82 (0.57-1.16); 0.70 (0.47-1.03) relative risk of recurrence and breast cancer death, respectively); when a multivariate analysis was performed (taking in consideration cholesterol and triglyceride values, stage and chemotherapy, the authors showed that statin use was predictive for OS (HR: 0.10, p = 0.026, 95% CI: 0.01-0.76).
The retrospective study of Lacerda et al [25] using Breast Cancer Management database at MD Anderson Cancer Centre in USA between 1995 and 2011, analysed the risk of loco-regional recurrence at 3 years associated to the use of statins, in patients with inflammatory breast cancer who received adjuvant post-mastectomy radiotherapy. 102 patients underwent post-mastectomy radiation (86 patients) or postmastectomy radiation plus statins (16 patients). Unfortunately no information about the outcome in TNBC patients was shown.

Clinicaltrials.gov
Searching the web site of clinicaltrials.gov (clinicaltrials.gov), we found only 17 drugs out of 286 presented in the ReDo_DB with ongoing or completed clinical trials for TNBC. Table 3 shows the list of trials and the recruitment status for each drug. As shown in Table 3, most part of the drugs present only one or few studies published on this website. In total, three studies are recruiting for the assessment of atorvastatin, two for metformin, two for mifepristone, and three for zoledronic acid.

Future directions
This review presents an overview of all the evidences about the repurposing of old, licensed, non-cancer-drugs in the treatment of TNBC, starting from preclinical evidence and going through current clinical trials. ReDO is an ambitious project aiming to investigate the repurposing of non-cancer-drugs in oncology, and ReDO_DB is a powerful tool that need to be dynamically implemented with recent findings, by adding to the database new drugs for which there are preclinical evidence, and by giving visitors a specific PubMed search string for each tumour and tumour subtypes. The ReDO approach is based on published literature and does not aim to identify new active compounds against cancer. Thus, the database does not include potential repurposing candidates identified through in silico modelling or other computational pharmacological approaches that, despite the interest for the research [28-31], unless validated by preclinical studies, represent only future hypothetical repurposed drugs and far from the aim of the ReDO project. The project, in particular, aims to drive scientist attention to investigate already approved non cancer-drugs in the oncology setting. Using this ReDO_DB, we found out that despite a lot of preclinical evidence was produced for drugs included in the database for the treatment of TNBC, only few of them were tested in clinical trials. Moreover, in clinical trials only few of the studies used a large sample of cases and gave explicit results on the repurposing of old drugs for TNBC. Some of the studies did not report any result for TNBC cohort when this is a part of a bigger BC cohort.
Beta Blockers (BBs) seem to be the more promising drugs in the repurposing for the treatment of TNBC. Three articles showed significant benefits of these drugs in women with advanced TNBC and in early primary TNBC patients treated with the combination of chemotherapy plus BBs [11][12][13]. Unfortunately, in clinicaltrials.gov we found no studies that specifically attempt to evaluate BBs within clinical trials for TNBC patients. One triple blinded phase II randomised trial evaluated the use of pre-operative propranolol (seven days before surgery) compared to placebo in 60 women with early stage surgically-resectable breast cancer.
[32]. The authors showed that the treatment with propranolol reduced intra-tumoral mesenchymal transition and promoted immune cell infiltration reducing biomarkers associated with metastatic potential. Unfortunately, authors did not present results stratified for breast cancer sub-type.
While BBs demonstrated to be beneficial in the treatment of TNBC, metformin, a promising molecule in preclinical studies, did not show any efficacy in the treatment of women with TNBC. Bayraktar et al [10] showed that metformin does not improve survival outcomes in a population of TNBC women when compared to not users. Of note, two studies on the use of metformin in clinicaltrials.gov on TNBC patients are ongoing.  For those drugs collected in ReDO_DB with favourable preclinical evidence or whose retrospective clinical trials were not so large to provide strong evidence, large retrospective cohort studies are needed to evaluate effectiveness. Further, as for BBs that have proven by retrospective studies to be effective in the treatment of TNBC patients, randomised clinical trials might be important to confirm the evidence of the repurposing.

Final remarks
Drug repurposing is a highly interesting novel strategy for the oncology community and ReDO_DB is a powerful tool that can give authors the opportunity to investigate weather non-anticancer drugs might be effective in cancer treatment. Some precision medicine studies, based on omics data, have included repurposed drugs and have reported interesting case reports of responses from patients [38, 39], however no one on TNBC. Due to the low number of therapeutic opportunities approved for TNBC, repurposing of old drugs seems a valuable approach for this particular type of cancer.
From the literature retrieved, BBs seemed to be the more promising drugs for the repurposing, while evidence about other drugs as NSAIDs still need to be assessed or proven for the treatment of TNBC.

Conflicts of interest
The authors declare that they have no conflict of interest

Authors' contributions
MZ and SC conceived the study. AS extracted the data. SD supervised the data extraction. MZ, SD, SC, AS, and PP contributed to the interpretation and discussion of study results. AS and SD drafted the manuscripts. All authors revised and approved the final version of the paper.        (2) (4)