Triterpenes are a diverse group of natural products with a complex structure and significant therapeutic potential. Although they can be found in marine animals or fungi, triterpenes are notably abundant in plants, with more than 20,000 unique structures identified so far (Goddard et al., 2024). Numerous plants play a role in traditional medicine for treating various health conditions, with their uses and preparations supported by scientific research (Joshi, 2023). Due to their diverse medicinal properties, triterpenes can exhibit anticancer, anti-inflammatory, antiviral, antioxidant, antibacterial, and antifungal activities (Gill and Kumar, 2016).

Triterpenes are widely investigated for their anticancer activity, as well as their potential use in cancer treatment and prevention (Nistor et al., 2023). In contrast to traditional chemotherapeutic drugs with high toxicity, relatively non-toxic phytochemicals like triterpenes are more suitable for long-term disease management (Li et al., 2020). Mechanisms that might be responsible for their anticancer activity are cytotoxicity, apoptosis regulation, interference with angiogenesis and dedifferentiation, metastasis inhibition and DNA polymerase blocking (Ghante and Jamkhande, 2019). Moreover, combining bioactive phytochemicals with chemotherapy has demonstrated a synergistic effect (Li et al., 2020).

Breast cancer is a major health threat to women worldwide. Its severity lies in the fact that it is frequently diagnosed in advanced stages, long after it has first developed. As a result, the disease spreads extensively, making treatment more challenging (Hussein et al., 2024). Breast cancer treatments, such as surgery, radiation, hormonal therapy, and chemotherapy, often come with significant side effects. Due to their limited effectiveness, relapse rates are rising, with varying morphological and molecular characteristics. Despite advances in research, challenges remain, especially with treatment side effects and drug resistance (El-Nashar et al., 2022). Moreover, there is an urgent need for natural treatments that can help limit tumor progression, enhance quality of life, and prolong patient survival (Aly et al., 2024).

According to scientific literature, the effectiveness of triterpenoids isolated from different plant species, such as masticadienolic acid, astragaloside IV (AS-IV), oleanolic acid (OA), ursolic acid (UA), ginsenoside Rh2 (Rh2), cucurbitacin B (CuB), saikosaponin D (SsD), asiatic acid (AA), and others (Table 1), has been extensively evaluated using various techniques. These have shown cytotoxicity against breast cancer, both in vitro and in vivo, through different mechanisms, including inhibiting cell proliferation, reducing mitochondrial membrane potential, suppressing glycolysis (downregulating LDH-A, c-Myc, and PDK1), inducing apoptosis (caspase activation, Bcl-2 modulation, ROS production, DNA damage), causing cell cycle arrest at the G0/G1 phase, modulating PI3K/AKT/mTOR and Beclin-1 pathways, reducing nitric oxide (NO) and TNF-α production, inhibiting STAT3, FAK, and MMPs, and enhancing Cav-1/NF-κB/c-Myc activity, supporting their potential as therapeutic agents (Aly et al., 2024).

According to the last ALLURE report issued by the European Union in 2022, approximately 250,000 rodents (74% of the total animals) were used for scientific purposes (European Commission, 2025). Despite the ethical dilemmas surrounding animal use in experiments (Kiani et al., 2022) and the available alternatives, such as computer models, in vitro cell and tissue cultures, alternative organisms like prokaryotes, protists, fungi, lower vertebrates (e.g., Danio rerio/zebrafish), invertebrates (e.g., Drosophila melanogaster) or microorganisms (e.g., Saccharomyces cerevisiae) (Doke and Dhawale, 2015), in chemico and ex vivo approaches (van der Zalm et al., 2022), animal experimentation remains invaluable to medical research, enabling the study of biological interactions within a living organism. Mice have revolutionized research by serving as a powerful model for evaluating gene function and understanding the molecular mechanisms underlying cancer pathogenesis in vivo. They offer several key advantages over other model systems: their small size, low maintenance costs, rapid reproduction, and the ability to undergo genetic manipulation, an essential feature for replicating the pathophysiology of human cancers (Cheon and Orsulic, 2011).

The current systematic review aims to investigate if the current evidence obtained from studies on breast cancer in rodents supports the further investigation of triterpenes as drug candidates in humans, or if new studies are needed to improve certain areas of knowledge. To the best of our knowledge, this is the first systematic review to analyze the in vivo evidence of triterpenes in breast cancer. This subject was only partially addressed by previous narrative reviews that focused on the in vitro and in vivo effects of only oleanane derivatives (Parikh et al., 2014) or included all triterpenes but for limited time frames: 2017–2023 (Aly et al., 2024) and until 2010 (Bishayee et al., 2011).

Our analysis revealed that triterpenes, triterpenoids, their formulations or cotreatments with other drugs reduce tumor weight, volume and metastasis of breast cancer in rodent models. These effects are achieved through diverse mechanisms, showing reduced toxicity as well.

The eligible articles were published between 2003–2024, with a notable increase observed from 2016 onward, supporting the increased interest in medicinal plants, often used in ethnomedicine, as a rich source of compounds with therapeutic potential in cancer prevention and management (Hashim et al., 2024). The majority of the articles were published in China and India, two Asian countries where traditional medicine is well-preserved (Liu et al., 2023) and remains the main treatment option for the majority of the population (Liu, 2021; Yao et al., 2023), due to tradition and affordability (Oyebode et al., 2016). Unsurprisingly, a 2020 literature survey (Yeung et al., 2020) indicates that for the last 2 decades, these countries have been the leading contributors to ethnopharmacology, the interdisciplinary exploration of traditional remedies, including plants, minerals, fungi, microbes and animals (Pirintsos et al., 2022; Schultz and Garbe, 2023).

The rodent models used in breast cancer studies have specific characteristics that make them suitable for certain research purposes (Welsh, 2013). Inbred mice species, which were used by the majority of studies included in the analysis, are widely accessible, affordable and have a high genetic similarity to humans (Boggess et al., 2004); they also possess traits that make them relevant for breast cancer studies, such as the susceptibility of Balb/c mice to MMTV-induced mammary tumors (Cheon and Orsulic, 2011). The immunodeficient mice were the second most prevalent choice identified and, despite allowing the formation and study of human breast cancer xenografts, the tumor growth and drug response in mice might not entirely replicate the human response (Shultz et al., 2014). Genetically modified mice, which are considered a better option for target validation, tumor response assessment, and identification of pharmacodynamic drug markers (Sharpless and DePinho, 2006), were also used. However, even with specific genetic alterations made to mimic human disease, inherent differences between species are inevitable and have to be considered (Borowsky, 2011). A few studies have also used rats, which show better similarities to human breast cancer compared to mouse models (Miller et al., 2022), in particular for the hormone receptor-positive type, which accounts for 70% of the invasive cancer forms in humans (Nicotra et al., 2024). Two species of mice were used in six studies, enhancing our understanding of the mechanisms and effects of triterpenes in various cancer models. However, in the future, these extensive studies should be made on highly active compounds to minimize unnecessary animal use in experiments, aligning with international regulations such as the European Union’s “Three Rs” principles that promote replacement, reduction, and refinement in animal research (European Commission, 2025).

Female animals were predominantly used in the studies, reflecting the human incidence as women’s breast cancer cases make up 99% of the total (Kreiter et al., 2014), with only 1% affecting men (Gómez-Raposo et al., 2010). However, about 10% of the studies did not report the sex of the animals used. This could be a serious limitation as sex differences are considered an important factor in the pathogenesis of cancer and in the drug response to chemotherapeutics (Klein and Flanagan, 2016). The studies that reported the animals’ age varied between 3 and 10 weeks; however, a significant proportion of the studies failed to report the age of the animals or used unclear terms such as “adult”. The reasons for choosing the age of the rodents in studies seem to be linked more to practicality and cost, rather than rodent particularities in different development stages (Jackson et al., 2017). For instance, liver metabolism differs in young versus old rodents (Mori et al., 2007; Pibiri et al., 2015) and can seriously impact xenobiotic metabolism. Future studies need to report more accurately the information regarding the animals used for their experiments and take into consideration age-related particularities when designing an experiment.

In the evaluated articles, breast cancer has been mainly modeled using allografts and xenografts obtained through various methods of administration, such as subcutaneously, intratumorally, and intravenously (Supplementary Material). Allografts are created by inoculating cells from the same species, typically mice, into another immunocompetent host (Yang et al., 2017) - although they are easily obtained their similarity to human breast cancer is limited. Xenografts are formed by injecting human cancerous cells into immunodeficient mice, offering a better but not perfect resemblance to human conditions for drug testing (Liu C et al., 2021; Singhal et al., 2023). Some studies utilized genetically modified mice, better suited for studying targeted therapies, modeling drug resistance and target validation (Holen et al., 2017). DMBA (7,12-dimethylbenzantracene), which induces breast cancer by generating free radicals and DNA damage (Khazaei et al., 2018) was also used scarcely as this model has inherent limitations such as difficulty to predict the location and time necessary for tumor development (Boix-Montesinos et al., 2021).

A significant proportion of the studies have investigated the effects of free triterpenes, naturally occurring compounds with anticancer potential in breast cancer (Şoica et al., 2021). However, due to their reduced bioavailability, they have a limited therapeutic effect (Atriya et al., 2023) Several strategies have been employed to overcome this limitation, as reflected in our search, where 60% of the articles utilized a wide range of approaches, including chemical derivatization, association with other drugs, and formulations for an enhanced anticancer effect.

Pentacyclic triterpenes and dammarane triterpenes (ginsenosides) were well-represented in the analyzed studies either in free form, derivatives, cotreatment or formulations due to their proven in vitro potential in breast cancer (Atriya et al., 2023; Deng et al., 2023).

Triterpenes were predominantly administered by i. p., i. v. and p. o. routes that are considered suitable for repeated administrations in rodents despite their disadvantages such as high first-pass metabolism (i.p. and p. o.), technical challenges (i.v.), need for sterility (i.p. and i. v.) and risk of injury if they are not properly used. In particular, i. p. administration is considered appropriate for proof-of-concept studies where the primary objective is to assess a candidate’s direct biological effect over pharmacokinetic investigations (Al Shoyaib et al., 2020), while p. o. and i. v routes are considered more suitable for clinical transition.

The triterpenes and triterpenoids were administered in doses ranging from 1.5 μg/kg (AECHL-1) to 250 mg/kg (BA). This wide range results from the structural diversity of this class, with more than 30,000 compounds isolated so far (Banerjee et al., 2019) and numerous semisynthetic derivatives (Nistor et al., 2022) that lead to different in vivo potencies. In preliminary studies, it is common for various compound doses to be tested to identify the maximum tolerated dose, an important parameter for toxicity assessment (Amorim et al., 2024). Two studies testing 2 and 4 mg/kg Cel (Raja et al., 2011) and 27 and 81 ppm CDDO-Me have found increased toxicity at the higher doses, providing valuable information regarding the safety profiles of these compounds. Several studies have incorporated the tested compounds into animal feed, while others have administered the compounds through gavage (forced feeding) (Table 2, Supplementary Material). While incorporating the compounds into the diet better simulates human intake and is more appropriate for chemoprevention studies, it also leads to unpredictable intake (Kapetanovic et al., 2011). On the other hand, gavage offers better drug intake accuracy but is considered more stressful for the animals and can lead to gavage-related reflux (Damsch et al., 2011) and esophageal injuries (Al Shoyaib et al., 2020).

The diversity of treatment duration across studies offers a broad perspective on the effect and toxicity of triterpenes, as single dose studies are important for preliminary evaluation of a new candidate (Organisation for Economic Co-operation and Development, 2002). Longer studies, which are better suited for the potential of clinical translation (Organisation for Economic Co-operation and Development, 2025), offer a more in-depth understanding on the effects of triterpenes. The treatment duration is limited by other factors such as costs involved and the need to reduce the animal pain in the late-phase of tumor development (Knight, 2011).

The studies that provided data on tumor weight and volume indicated that triterpenes, in all tested forms, reduced both weight and volume compared to negative controls. These values were either lowered or comparable to those in positive controls, with only a few studies reporting increased values compared to positive controls. Notably, one study (Song et al., 2020) employed only the positive control (β-actin) necessary to normalize the expression of certain proteins related to Rg3’s mechanism of action in breast cancer. This approach is justified here, as the anticancer effect of Rg3 in breast cancer has already been shown in previous studies (Liu Z et al., 2021), and the study focused on elucidating its mechanisms rather than validating its effectiveness. Although these results might not identically translate in human studies due to species particularities (Zheng et al., 2021), these consistent findings across diverse rodent models reinforce the efficacy of triterpenes in breast cancer. However, many articles reported neither one nor both of these outcomes. This might be due to the use of alternative outcome measurements such as tumor growth inhibition or tumor burden score, that were not collected for this review and are a limitation of our review design. Another limitation to be considered is the subjective interpretation of tumor weight and volume trends from the graphics, as the data was provided graphically rather than numerically. We have employed independent double screening and resolved the disagreements through discussions with a third reviewer to diminish this bias.

Although elucidating the mechanism of action is not essential for clinical translation or use, after the thalidomide incident, more resources were invested in ensuring drug safety, including mechanism investigation before drugs are approved by regulatory agencies (Greene and Podolsky, 2012). The most common mechanisms identified in our analysis were apoptosis induction and angiogenesis inhibition (Figure 4).

Apoptosis is a programmed cell death that can be induced through intrinsic and extrinsic pathways that lead to caspase activation and characteristic cell alterations such as DNA fragmentation and blebbing of the cell membrane (Singh and Lim, 2022). Inhibition of apoptosis is partly responsible for the anticancer effect of standard drugs used for breast cancer treatment, such as DOX (Nicoletto and Ofner, 2022), PTX (Zhao et al., 2022) and tamoxifen (Ahmed et al., 2021). Several studies in our analysis have also observed that triterpenes increased Bax/Bcl-2 ratio and cleaved caspase-3 levels, two apoptosis markers (Saddam et al., 2024), effects confirmed in human breast cancer cell lines such as MCF-7 and MDA-MB231 as well (Shahrestanaki et al., 2023).

Angiogenesis, the process of new blood vessel formation from pre-existing ones, has an important role in the development and dissemination of solid tumors, including breast cancer (Ayoub et al., 2022). The efficacy of angiogenesis inhibitors in breast cancer is limited due to tumor resistance by activation of pre-angiogenic factors (Ruan et al., 2022) and bevacizumab, a humanized anti-VEGF monoclonal antibody, previously approved through FDA’s Accelerated Approval Program for the treatment of metastatic breast cancer had this indication revoked in 2011 due to safety and efficacy concerns (Sasich and Sukkari, 2012). Our analysis showed that triterpenes decreased VEGF and CD31, two vital angiogenesis markers that have been measured in breast cancer patients where increased VEGF serum and tissue levels was associated with distant metastasis and poor outcome (Ali et al., 2011) and increased CD31 tissue levels were associated with high invasiveness (Sapino et al., 2001).

Angiogenesis is also involved in the metastatic process of breast cancer, promoting tumor dissemination to other organs (Ayoub et al., 2022; Harry and Ormiston, 2021). The collected data indicates that many triterpenes can reduce lung metastasis in rodents, with a limited number of studies assessing the other organs commonly affected by breast cancer metastasis, such as the bones, liver, and brain (Ibragimova et al., 2023). Wnt/β-catenin signaling plays an important role in cell proliferation, metastasis, immune microenvironment regulation, and therapeutic resistance in breast cancer (Xu et al., 2020). Currently, none of the Wnt/β-catenin signaling inhibitors studied have passed phase II trials, despite this being the most altered pathway in breast cancer and an important chemotherapeutic target (Mukherjee and Panda, 2020). Cell proliferation can also be assessed through specific markers such as PCNA, Ki-67, and minichromosome maintenance proteins (MCM), which can have prognostic value for breast cancer (Juríková et al., 2016). Triterpenes have inhibited the Wnt/β-catenin signaling pathway and decreased PCNA and Ki-67 levels, showing antiproliferative potential in rodents. Triterpenes have also decreased matrix metalloproteinases (MMP) in rodents, especially MMP-2 and -9, which seem to play a crucial role in breast cancer tumorigenesis, though their role is not completely understood. While both have been established in breast cancer patients, increased MMP-9 levels are associated with a poor prognosis, while increased MMP-2 levels are linked to prolonged survival time (Cheng et al., 2022). Currently, no MMP inhibitors have received FDA approval for breast cancer treatment, and many broad-spectrum MMP inhibitors have been halted in early-stage clinical trials due to a lack of specificity and toxicity (Kwon, 2023).

Other signaling pathways emerging as potential chemotherapeutic targets for breast cancer (Núñez-Marrero, 2019; Wood et al., 2018), including the IL-12/STAT4 pathway and the JNK1/2 pathway, were also altered by triterpenes, showcasing their versatile abilities to suppress breast cancer.

As some of these mechanisms have been identified in both in vitro studies of human breast cancer cells and in rodents, which are more complex models, this might support the clinical relevance of triterpenes for human studies. A comparison of standard therapeutic classes used in breast cancer and triterpenes’ mechanisms is presented in Table 7. Although triterpenes do not exert their effect in breast cancer exactly as standard drugs, their multi-target approach could lower cancer resistance and toxicity. Nonetheless, precautions must be taken concerning emerging mechanisms involved in breast cancer that, despite having tremendous potential as drug targets, also pose significant risks until more information is collected.

The data presented in the majority of the studies suggest that triterpenes have a promising safety profile with no observed or significant toxicity in rodents. However, several studies have observed significant weight loss and in rare cases, leukopenia, hepatotoxicity and mortality at certain doses. Specifically, CuB in association with gemcitabine led to leukopenia at doses of 1 mg/kg (Aribi et al., 2013) and mortality at 3 mg/kg dose (Suebsakwong et al., 2019), effects that are absent when a smaller dose or a CuB prodrug is used. Triterpenes also reduced the toxicity of standard drugs such as DOX when used in combination, suggesting the benefit of their use in associations. However, a significant gap exists due to lack of toxicity assessment in several studies analyzed, impediments an objective toxicity assessment of these compounds, which further studies must address.

The development of nanotechnology has had a major impact on the creation of pharmaceutical formulations with improved bioavailability, safety, and patient compliance, allowing for the delivery of highly lipophilic or chemically unstable drugs (Milan et al., 2022). The results discussed in one review support the strong antitumor potential of delivery systems incorporating pentacyclic triterpenoids. Among these, self-assembling nanocarriers with inherent anticancer activity have emerged as a particularly promising direction for future research. Such systems not only reduce the risk of adverse drug events but also enhance drug accumulation at the tumor site (Kaps et al., 2021). For instance, nanoformulations containing BA, such as NPs, liposomes, polymeric micelles, nanotubes, nanogels, nanofibers, and nanosuspensions have demonstrated excellent therapeutic effects across a wide range of cancers, including breast cancer. In particular, they have substantially improved the solubility and bioavailability of BA, while also enhancing its anticancer efficacy through strategies such as controlled release and targeted delivery (Wang et al., 2024). Furthermore, the data analyzed in our review supports that the inclusion of free triterpenic derivatives in different types of nanocarriers might reduce toxic effects, leading to reduced mortality rates, no significant weight loss, no modifications in the rodents’ blood tests and no observed histopathological changes.

In the risk of bias analysis, we did not identify any study that checked all the criteria assessed, showing that the methodological rigor in animal studies can be improved in the future. As no study reported an allocation concealment or blinding of the research personnel regarding the treatment and control groups, the results reported and our analysis might be negatively impacted and should be taken into account. Moreover, our analysis revealed that a significant number of studies had unclear risk of bias for both sequence generation (96.93%) and random housing (75.46%), posing significant concerns that in the absence of proper animal randomization, the effects of triterpenes in breast cancer models might be overestimated. Other limitations of the review to be considered are the absence of a quality assessment due to the extensive number of included studies, publication bias (exclusion of abstracts, conferences, or proceedings), language bias (English-language only), database bias (only searching three databases), positive results bias (positive results are more likely to be published) and articles that could not be retrieved and were excluded from the analysis as they might contain relevant information on the subject.

Our primary vote-counting analysis showed that triterpenes reduce tumor weight (52.76%) and volume (76.68%) in a significant number of studies. However, as a considerable number of studies, (23.32% for tumor volume and 47.24% for tumor weight), failed to measure and report these metrics, these encouraging findings should be interpreted with caution, taking into account the methodological shortcomings identified in the risk of bias assessment. Our primary results are also supported by a sensitivity analysis of articles with more rigorous methodology. The analysis showed that triterpenes have a positive effect on both tumor weight (100%) and volume (50%), indicating that the overall findings are not only a result of studies with methodological weaknesses. Nevertheless, the reduced sample size used for this analysis (n = 2) limits our ability to draw a definitive conclusion. Further research with more meticulously designed protocols, aligned with renowned guidelines such ARRIVE guidelines (Percie du Sert et al., 2020), are needed to confirm these promising results.

We performed a vote-counting analysis on predefined subgroups to explore potential sources of heterogeneity in our assessment. Our subgroup analysis revealed that free triterpenes performed better than triterpenoids on tumor weight and volume, results consistent in both monotherapy and cotreatment settings. These results come in contrast with the results of many in vitro studies, which often report superior anticancer effects for triterpenoids (Lombrea et al., 2023; Ye et al., 2017). This suggests that in rodent models, the anticancer effect is highly dependent on the chemical structure of native triterpenes, a factor that may be related to differences in the mechanism of action and bioavailability. Our analysis also showed that in the monotherapy group, the use of carriers has a potent effect, particularly in reducing tumor volume, while the effect on tumor weight was similar to that of free compounds. The same consistent results were observed in the cotreatment group, but in this group, the differences were more pronounced, suggesting the use of carriers in combinational therapy might exacerbate the antitumor effect of triterpenes.

Taking into account the utilized tumor models, the analysis revealed that the positive effects were mostly observed in 4T1, MBA-MB-231 and MCF-7 tumor models. A solid explanation for this predilection might be supported by the large variety of preclinical studies involving these types of cell lines (Atiya et al., 2019; Lee et al., 2015; Sahu et al., 2024; Tao et al., 2008). The highly observed percentages for the positive effects (50%–62% for tumor weight and 85%–90% for tumor volume) might suggest a robust efficacy tendency in these established models. It has been well established that these cell lines represent standard models for triple-negative mammary cancer (4T1 and MBA-MB-231) and luminal (MCF-7) cancer (Garbar et al., 2017), hence offering translational relevance. However, in chemical-induced (DMBA) or transgenic models, the results showed less effective results (33% and 16%). These results could be explained by the fact that these rodent models are limited by their low heterogeneity compared to the highly heterogeneous human tumors (Lampreht Tratar et al., 2018) hence, the anticancer effects of triterpenes and triterpenoids could be stronger in a highly heterogeneous tumor microenvironment (Tonello et al., 2025). These results strongly suggest that more complex models that mimic the human tumoral environment are required in future studies.

Comparing the results in the route of administration subgroup, we found that all routes showed similar results for tumor weight, showing that the long-term effect of triterpenes is not dependent on the route of administration. In contrast, the route of administration had a significant effect on the tumor volume, a marker for early tumor shrinkage in breast cancer (Che et al., 2020). Interestingly, the routes that avoid the first-pass metabolism, i. v. and i. p. had a strong positive effect in reducing tumor volume, while the i. g. administration, which also avoids first-pass metabolism, had shown a lower efficacy, suggesting that the influence of the gastric environment might reduce triterpenes’ bioavailability and efficacy. These findings are also supported by the moderate positive effect on tumor volume of triterpenes administered orally.

Current evidence of triterpenes in rodent models of breast cancer is promising but to achieve clinical translation, future investigations are needed. Several areas of improvement are the following:

optimization of preclinical models: stringent methodologies (baseline characteristics, allocation concealment, blinding of personnel, ethical approval), methodological transparency, use of patient-xenograft models for better translatability;

Our analysis showed that triterpenes were tested in free form, chemical derivatives, delivery systems (liposomes, micelles, NPs) and association with other drugs in different rodent models of breast cancer, yielding promising results in terms of tumor weight and volume reduction. Our primary vote-counting analysis results have shown that triterpenes have positive effects on key outcomes, such as tumor weight and volume, findings confirmed by a sensitivity analysis. The subgroup analysis highlighted key findings: free triterpenes were more effective than triterpenoids in both monotherapy and combination treatments; delivery systems significantly improved early tumor volume reduction, especially in cotreatment scenarios; greater positive effects were seen in well-established tumor models (4T1, MDA-MB-231, and MCF-7), while chemically-induced or transgenic models were less effective; all administration routes demonstrated similar long-term effects on tumor weight, but methods bypassing first-pass metabolism (i.v. and i. p.) led to greater early tumor volume reduction.

Triterpenes exerted their effects primarily through the induction of apoptosis, along with angiogenesis and metastasis inhibition; however, other mechanisms, including the inhibition of the Wnt/β-catenin signaling pathway, the activation of the IL-12/STAT4 pathway, and the activation of the JNK1/2 pathway, were also observed. Determining the structure-activity relationship could accelerate the understanding of the triterpenes’ mechanism of action.

Reduced toxicity has been observed in the majority of the studies, with only a few reporting significant weight loss that might indicate some degree of toxicity. Notable side effects, such as hepatotoxicity, mortality and leukopenia were also observed in a limited number of studies for certain triterpenes used in higher doses; toxicity that could be reduced by the use of nanocarriers. Major limitations that hamper our assessment are the lack of data on the mechanisms of action or safety in a substantial number of studies and the methodological shortcomings identified in the risk of bias assessment (lack of randomization and blinding), which could have contributed to the observed positive effects of triterpenes.

Future research must employ more rigorously designed studies aligned with international guidelines to ensure transparency and reproducibility that are needed to validate these results. We conclude that while the evidence for the effects of triterpenes in rodent models of breast cancer is positive, it currently does not support clinical translation until further understanding is obtained regarding the mechanisms of action, the structure-activity relationship, and safety.