In this paper, we present the study assessing the safety and efficacy of cisplatin monotherapy in patients with HER2-negative BC and impeding or ongoing hepatic VC. We also discuss the factors that should be considered when determining the eligibility of patients for this particular treatment.

Systemic treatment of patients with BC and VC is the only situation in which guidelines recommend a different treatment over the first-line standard one. Interestingly, the phase II RIGHT Choice trial, presented at the 2022 San Antonio Breast Cancer Symposium, was the first prospective study to evaluate a Cyclin-dependent kinases 4 and 6 inhibitor (CDK4/6i) combined with an aromatase inhibitor (AI) versus combination chemotherapy (CTH) [18]. Careful analysis of the inclusion and exclusion criteria reveals that the study population comprised patients with clinically aggressive disease, not VC. Consequently, the study did not alter the management of HR + /HER2-negaive ABC with VC. CTH is favored over endocrine therapy (ET), and polychemotherapy is favored over single-agent CTH [3, 8, 19]. This recommendation is substantiated by data indicating a higher objective response rate (ORR) and longer PFS, with no significant benefit reported for OS [20, 21]. However, the preferred CTH regimen is not specified. The decision made by the physician is highly challenging, and its difficulties stem from several factors. Firstly, VC is a common exclusion criterion from clinical trials. Safety and treatment efficacy data are mainly obtained from retrospective series (see Table S1 in Supplementary Materials). Secondly, the decision-making process is based on a subjective evaluation of a constellation of patient and drug features, such as the toxicity profile, PS, and hepatic metabolism. National Comprehensive Cancer Network (NCCN), but not ESMO guidelines, recommend poly ADP-ribose polymerase inhibitors (PARPi) for BRCA mutation carriers in the case of VC as a first-line treatment, irrespective of the BC subtype [8]. The treatment of triple-negative breast cancer (TNBC) and HER2-positive ABC with VC aligns with guidelines [3, 8, 19]. For TNBC, it depends on the BRCA and programmed death-ligand 1 (PD-L1) status, tested by the combined positive score (CPS). Deciding on subsequent therapy lines becomes even more challenging due to the poor prognosis, compromised PS, and lack of specific recommendations; hence, the consideration of best supportive care should always be taken into account.

Cisplatin is not among preferred choices for HER2-negative metastatic breast cancer (mBC) as for ESMO or NCCN guidelines. According to NCCN guidelines cisplatin in HER2-negative BC is only recommended as an option (as category 1, preferred) in case of PD-L1 CPS < 10 and germline BRCA1/2 mutation [8]. ESMO ​Metastatic Breast Cancer Living Guideline for mTNBC do not mention cisplatin, while carboplatin stays as an option for first line combine treatment with pembrolizumab in patients with PD-L1 + status and platins are preferred among other chemotherapy agents in individuals with BRCA mutation. In the ESMO ​Metastatic Breast Cancer Living Guideline for metastatic hormone-positive HER2-negative BC platins are used as one of the options for patients with tumors that are endocrine resistant or in patients at risk of imminent organ failure [22, 23].

Cisplatin is generally not regarded as hepatotoxic. It has been linked to a low incidence of mild hepatic enzyme elevations during treatment. Instances of clinically evident liver injury attributed to cisplatin have been rarely reported [24]. Cisplatin has been used in BC setting for decades [25, 26], although a meta-analysis examining platinum-containing treatment approaches for metastatic breast cancer indicates that, in women without TNBC, there is substantial evidence suggesting minimal to no survival advantage and an increased risk of toxicity associated with platinum-based regimens. However, there is less robust evidence suggesting a moderate survival benefit for women with metastatic TNBC undergoing platinum-based treatments. That meta-analysis did not include patients with hepatic IVC/VC [27]. While it is claimed that cisplatin activity in ABC decreases with number of prior systemic treatment lines [25], this was not confirmed in our cohort. Data regarding the potentially effective dose were published more than 40 years ago, suggesting doses higher than 100 mg/m2 to be potentially active, with doses close to the ones used in our study 60 mg/m2 being inactive [26, 28]. However, a study by Isakoff and colleagues regarding the treatment of metastatic TNBC advocated for a dose of 75 mg/m2 to be effective, showing even with higher numerical response rates than carboplatin in first or second-line treatment [29].

Even though, one third of patients with IVC/VC in our study achieved PR or SD, it is still much worse effect than in most recent studies of pretreated patients with HER2-negative BC without IVC/VC on CTH and also in studies regarding patients with IVC/VC ([30, 31] see Table S1 in Supplementary Materials). However real-world data should not be compared with data obtained in randomized controlled trials. Furthermore, the response to treatment had not remained for a long time—PFS is also much shorter than it was reported in the literature (see Table S1 in Supplementary Materials).

Our patients presented with a very poor OS. Similar outcomes are confirmed in some studies [11], but other papers report much longer survival [4, 10]. The reason for these differences can be an unclear/changing hepatic IVC/VC definition, a heterogenous population that includes all types of VC or different BC subtypes included (see Table S1 in Supplementary Materials).

The safety assessment of the treatment did not uncover any new data in comparison to those already known from cisplatin product characteristics. Some of the AEs overlapped with ABC symptoms, and the relationship between cisplatin activity and a symptom was evaluated on a case-by-case basis [13,14,15,16, 24]. There is a question regarding chemotherapy cessation after obtaining maximal clinical benefit, observed in five patients in our population. All of them, except for one who initiated hormonal therapy, experienced prompt progression. While initiating maintenance hormonal therapy appears to be a common and reasonable clinical practice, data regarding cisplatin continuation or switching to another type of maintenance systemic treatment are scarce, with only retrospective studies addressing patients with BC receiving platinum doublets [33].

PS is frequently named as a factor influencing patients’ survival, also in patients with BC and VC, confirming our results [4]. There are efforts to incorporate more complex scoring systems in order to predict OS in patients with hepatic VC [34]. Interestingly, in our study, PS could not only serve as a prognostic factor but also differentiate between groups in terms of predicting the response to cisplatin treatment. Almost one-third of our patients were administered treatment in the last 30 days of life, which might have had a negative impact on patients' quality of life, causing unnecessary suffering and costs [35]. Avoiding such situations is an important but challenging task in daily clinical practice. While we suggest as simple parameter as PS to distinguish between patients that are more prone to experience better treatment response, there are case reports suggesting potential benefit in patients with poor PS, however with no known hepatic IVC/VC [32]. At the same time, none of the other tested factors served as a predictive parameter when tested using the CTree algorithm.

There are several primary limitations of this study: (1) it is a single-center analysis, (2) the population is relatively small, (3) it is retrospective in nature, (4) it includes patients with hepatic IVC, not exclusively those with VC and (5) the population is heterogenous as it includes patients with TNBC and luminal BC (mostly, hormone-resistant BC). We chose not to include patients treated before the end of 2016, as that population was managed differently before the era of CDK4/6 inhibitors, and the results might not accurately reflect the current everyday practice. However, despite these limitations, this study remains one of the largest evaluating the role of cisplatin in this cohort of patients with BC with an unfavorable prognosis. Notably, there are no studies (see Table S1 in Supplementary Materials) assessing the role of cisplatin monotherapy in this specific setting. Other published papers: (1) either do not specify which platinum agent was used or present highly heterogeneous treatment options with various platinum partners, (2) use various, outdated or unclear definitions for VC and IVC and (3) encompass populations of patients with both HER2-positive and negative BC collectively.

In our study, all patients identified with hepatic IVC had a recent diagnosis of rapid liver progression, with a majority of the organ involved. Additionally, ALT or AST was elevated (with minimum > 2 × ULN), and there were significant increases in LDH, alkaline phosphatase, or GGTP, along with the presence of disease symptoms. Thus, our criteria for IVC, not precisely defined in ABC 5, were strict [3]

For individuals with extensive or rapidly progressing VC, combination chemotherapy might be considered suitable as per available guidelines [3, 8]. In such cases, the anticipated higher response rates are weighed against the elevated risks of toxicity, primarily due to concerns about imminent organ dysfunction. Still, there is a lack of prospective data demonstrating that combination chemotherapy enhances OS when compared to the use of single-agent chemotherapy [36]. Moreover, there are practically no viable options for platinum partners in heavily pretreated patients with liver insufficiency.