Germline mutations in the Breast Cancer gene-1 (BRCA1) and Breast Cancer gene-2 (BRCA2) are identified as significant risk factors for the development of solid tumors, including ovarian, breast, prostate and pancreatic cancers (Golan et al., 2019; Boussios et al., 2022). BRCA genes play a crucial role in DNA damage response because they encode proteins taking part in the repair process of double-strand DNA breaks by homologous recombination (Tobalina et al., 2021). The identification of these mutations has improved the development of target therapy based on poly-ADP ribose polymerase (PARP) inhibitors. PARP is a protein that supports damaged cells to repair themselves. Thus, PARP inhibitors, which currently include olaparib, rucaparib, niraparib and talazoparib, stop the PARP from doing its repair work in cancer cells leading to cancer cells’ death (Golan et al., 2019; Boussios et al., 2022).

Olaparib was the first PARP inhibitor to be approved in 2014 by the Food and Drug Administration (FDA) and the European Medicines Agency (EMA) for the oral treatment of BRCA-mutated advanced ovarian cancer in patients who had previously received three or more lines of chemotherapy (Kim et al., 2015). Throughout the years, the therapeutic indications of the drug were extended for the treatment of advanced epithelial ovarian, fallopian tube or primary peritoneal cancer after the first-line platinum-based chemotherapy, HER2-negative metastatic breast cancer, BRCA-mutated metastatic castration-resistant prostate cancer, pancreatic and endometrial cancer (Food and Drug Administration, 2024a; Food and Drug Administration, 2024b; Food and Drug Administration, 2024c; Fallah et al., 2024). Olaparib acts as a competitive inhibitor of NAD+ in the catalytic site of PARP1 and PARP2, which are involved in the repair process of DNA single-strand breaks (SSBs). Thus, olaparib leads to the accumulation of unrepaired SSBs and to the formation of dangerous double-strand breaks, which can be quickly repaired in healthy cells. In tumor cells, lacking in homologous recombination repair, the accumulation of unrepaired DNA damage is fatal, and the obstruction of replication forks leads to replication arrest. These multiple mechanisms explain the reason why tumor cells show sensitivity to olaparib and other PARP inhibitors (Bochum et al., 2018; Robson et al., 2017).

The results of both 5- and 7-year follow up analysis of phase III SOLO1 trial (Banerjee et al., 2021; DiSilvestro et al., 2023) and the evidence of a meta-analysis carried out by Guo et al. (2018) reported also the occurrence of neutropenia and thrombocytopenia. However, following the meeting held by the Pharmacovigilance Risk Assessment Committee (PRAC) on 31 July 2023 (European Medicines Agency, 2025), the EMA informed healthcare professionals (HCPs) about cases of hepatocellular damage and hepatitis occurring during the treatment with olaparib and provided special warnings and precautions for patients showing clinical symptoms or signs of hepatotoxicity or suspected Drug-Induced Liver Injury (DILI).

Generally, liver injury is identified as a common consequence of the administration of a variety of drugs, and in most cases, it requires the treatment discontinuation. When DILI is dose related, it is defined as intrinsic, direct or predictable, while when it is not dose-dependent, it can be defined as idiosyncratic (Scavone et al., 2020; Sow et al., 2021). Cytolytic liver damage represents a form of liver inflammation and injury characterized by the destruction of hepatocytes which is typically indicated by an increase in liver enzymes of five times the upper limit of normal, appearing within 90 days of drug initiation (Aithal et al., 2011; Garcia-Cortes et al., 2020). This liver damage can result from various causes, including medications. As reported by Zhu et al. (2024), DILI can be immune-mediated, resulting from the hepatic metabolism of the culprit drugs leading to reactive metabolites binding to hepatocytes and leading to autoantibodies reacting with liver-specific antigens as well as direct toxic effect on hepatocytes. Indeed, reactive metabolites may lead to hepatic damage by different mechanisms, such as oxidative stress, inflammation, cellular death due to apoptosis or necrosis and mitochondrial membrane alteration and organ dysfunction (Björnsson and Björnsson, 2022). Moreover, in some people genetically predisposed, stress can trigger innate immune responses which cause a co-stimulation for an adaptive immune response (Luo et al., 2017). Data from clinical trials revealed that almost 4% of patients receiving olaparib experienced abnormalities in routine liver tests (Rucaparib, 2017). Abnormal liver values, such as increasing ALT concentrations, have been reported also during clinical trials with rucaparib, another PARP inhibitor. On the other hand, no cases of hepatitis with jaundice or liver failure have been described for this drug. This means that rucaparib can lead to increased blood concentration of liver enzymes, but it has not been related to significant hepatotoxicity (Niraparib, 2017). The same is true for niraparib (Yang et al., 2021).

Considering that olaparib was the first PARP inhibitor to be approved being used in clinical practice and given the relevance of hepatic AEs highlighted by the PRAC, we aimed to describe cases of olaparib-induced liver injury using data from the European spontaneous reporting system (EudraVigilance) and to stratify these cases according to the liver damage origin as cytolytic, cholestatic, mixed, not specified, and liver related investigation.

Using data from the EV database, we analyzed the safety profile of olaparib in terms of hepatobiliary AEs. This decision derived from a recent question highlighted by the PRAC about the risk of hepatocellular damage associated with this drug (Kim et al., 2015).

We found 344 DILI cases, ICSRs reporting at least one narrow PT related to the SMQs “Drug related hepatic disorders–comprehensive search” and olaparib as suspected drug. Patients experiencing olaparib-induced hepatotoxicity were mostly female. The higher female frequency is expected considering that ovarian cancer represents the first therapeutic indication olaparib was approved for, followed by HER2-negative metastatic breast cancer and, only in 2022, its indication was extended for the treatment of metastatic castration-resistant prostate cancer too (Food and Drug Administration, 2024a; Food and Drug Administration, 2024b; Food and Drug Administration, 2024c; Rydberg et al., 2018). Moreover, it is well known that some gender differences between women and men exist in terms of frequency of experiencing AEs. Women, for instance, tend to experience AEs more commonly than men probably because that sex hormones play in both pharmacokinetics and pharmacodynamics of drugs (Spoletini et al., 2012). In women menstrual cycle, modifications in the concentration of sexual steroids, the use of oral contraceptives or hormonal substitutive therapy are factors which may influence drug response. On the other hand, the impact of androgens in men is not so relevant thanks to the regular levels of these hormones in blood (Scavone et al., 2020; Velişcu et al., 2023; Nicoletti et al., 2023; Web et al., 2017).

We also found that DILI cases related to olaparib occurred more commonly in patients belonging to the age group 18–64 years followed by the age group 65–85 years. This is expected considering that, although ovarian cancer can be diagnosed in women of all ages, its occurrence is quite rare among women aged <40 years considering that the risk increases both with the age and after menopause. Epithelial tumors represent more than 90% above age 40 (Doubeni et al., 2016; Rahman and Helvie, 2022). Similarly, in breast cancer pathogenesis, age represents an important risk factor for its occurrence with a median age of new cases between 40 and 49 years. This is generally the age range in which the initiation of screening mammography is recommended (Tufoni et al., 2018).

The hepatic-related AEs were mainly classified as serious (above 60%), and 23 ICSRs reported fatal cases. When stratifying DILI cases by liver damage origin, more than 50% referred to cases of cytolytic origin and liver-related investigation. On the other hand, almost 3.5% of DILI cases were classified as of cholestatic origin, \while 1.5% were classified as mixed cases.

These results are in line with some case reports recently published. The first one describes the case of a 61-year-old woman with BRCA1 mutated ovarian cystadenocarcinoma and no previous history of abnormal liver function. After starting the therapy with olaparib, she has been hospitalized for noticeable hypertransaminasemia (Alshelleh et al., 2022). Another case report describes the clinical history of a 56-year-old woman with advanced endometrioid and serous carcinoma, firstly treated with neoadjuvant chemotherapy based on carboplatin and paclitaxel. After cancer relapsed, she started retreatment with carboplatin and doxorubicin for a total of 6 cycles. Olaparib was eventually administered because of BRCA1 mutation. Three months after the initiation of olaparib, a severe acute liver injury required the hospitalization of the patient. No use of alcohol was evidenced. Physical examination showed jaundice and no hepatic encephalopathy, while mild hepatomegaly with periportal edema was evidenced by computed tomography (Zhu et al., 2024). Lastly, Zhu K et al. reported the case of a 70-year-old woman who experienced significant a severe hepatocellular injury after initiating first line maintenance olaparib. Her clinical conditions improved after the culprit drug’s discontinuation but promptly worsened upon rechallenge (Kumachev and Wu, 2021).

Most of the fatal cases retrieved by our analysis were classified as “cytolytic” and as “liver neoplasm or nonspecific signs”. In contrast with our results, we found no fatal cases of hepatotoxicity related to olaparib treatment reported in literature. This evidence could suggest that the fatality of these cases may be partially explained by the presence of hepatic neoplasm or the progression of disease.

Therefore, according to data available from the literature and the results from our study, special attention is required regarding hepatobiliary disorders for olaparib, and further studies are needed to better evaluate the safety profile of the drug. According to the PRAC recommendations, HCPs should constantly monitor the liver function tests during treatment with olaparib. Moreover, in case of potential DILI develops, proper clinical measures should be taken, such as treatment interruption. In case DILI is confirmed, the treatment can be even discontinued.

Our study has some limitations and strengths. First, we used data from the EV database that, like any other spontaneous reporting system, suffers from the well-known under-reporting phenomena as well as the poor quality and lack of information listed in each ICSR. Indeed, it is known that many ADRs are never reported by healthcare professionals or patients, especially if they are mild or expected. Similarly, ICSRs may contain missing or incomplete or inconsistent clinical information (such as, for example, data related to the outcome that in our study was missing in more than 50% of retrieved ICSRs). For instance, we could not make any analysis concerning the time to onset of olaparib-induced liver injury and similarly we could not exclude the effects deriving from concomitant diseases or medications on the occurrence of DILI. Another important limitation is not knowing whether the reported patients experiencing “liver neoplasm or non-specific signs” have had liver metastases before starting treatment with olaparib. Thus, the overall results should be interpreted with caution. In addition, as any data based on the spontaneous reporting system, our data lack of denominator data and thus we do not know how many people actually used the drug, at what dose, or for how long. Therefore, incidence rates cannot be calculated. What’s more, we could not establish causality considering that EV is designed to detect signals, not to prove that a drug causes an event and we cannot rule out that many events could be related with the illness rather than the drug (leading to a confounding by indication). Lastly, since we classified retrieved cases according to the PT reported, the possibility of a misclassification bias (the wrong classification of cases, placed into the wrong categories, cytolytic or cholestatic) cannot be rules out. Considering these limitations, we are aware that the real hepatic safety profile of olaparib cannot be fully established but needs to be confirmed by the results obtained from ad hoc studies. On the other hand, we carried out a descriptive analysis of data derived from the EV database to identify cases of liver injury related to olaparib and to classify these cases according to their liver damage origin. A team of pharmacologists and clinicians with consolidated experience in pharmacovigilance performed this classification, with the aim to guarantee high quality standard for the overall methodological process. In addition, despite its limitations, the spontaneous reporting system represents a useful and inexpensive tool for the collection and analysis of medicine safety data for population excluded from premarketing clinical studies and for the identification of ADRs that are not detectable during the pre-marketing phase.

We found out that most DILI cases occurred in female patients belonging to the age group 18–64 years. DILI cases mainly were represented by cases of cytolytic origins and cases associated with liver related investigation. There were 23 fatal cases, mostly classified as “cytolytic” and “liver neoplasm or nonspecific signs”.

Additional pharmacovigilance studies assessing the hepatic safety profile of olaparib is required in light of the findings obtained from the EV. In addition, evaluation of the pathophysiological mechanism underlying the drug-induced liver damage could provide clinicians with practical resources for prevention and treatment.