Typical retinoids exert their antitumor activity mainly through the activation of RARs and RXRs, resulting in decreased expression of anti-apoptotic B-cell lymphoma 2 (Bcl-2) family genes^[45]. In contrast, fenretinide’s pro-apoptotic effect involves diverse signaling mediators activated by RAR-independent mechanisms, such as ROS, lipid second messengers, and stress-related proteins^[7-10,46]. Additional mechanisms contributing to fenretinide’s antitumor activity include induction of autophagy, cell cycle arrest and/or slowing, reduction of the CSCs content, depression of biosynthesis processes and inhibition of survival pathways^[8,12-18,47]. Transactivation assays have demonstrated that fenretinide acts as a potent transactivator of RARγ and a moderate activator of RARβ, while exhibiting no interaction with RARα and RXRα. This selective activation profile may explain its specific biological activities and favorable pharmaceutical properties^[48]. A study on hepatocellular carcinoma (HCC) cell lines demonstrated that the pro-apoptotic effect of fenretinide is mediated by direct activation of RARβ, whose expression also determines the susceptibility of HCC cells to fenretinide-induced apoptosis^[49]. In a subsequent study, the same authors identified nuclear receptor 4A1 (Nur77) as an additional mediator of fenretinide-induced apoptosis. Specifically, in Huh7 cells fenretinide promoted the translocation of Nur77 from the nucleus to the mitochondria, allowing its binding to Bcl-2 and inducing a conformational change of this protein toward a pro-apoptotic state^[50]. Accordingly, Nur77 knockdown prevented fenretinide-induced DNA double-strand breaks and caspase-3 cleavage^[50]. Similarly, in acute myeloid leukemia (AML) cells, fenretinide induced the translocation of Nur77 from the nucleus to the mitochondria, where it bound Bcl-2^[51]. Another study in HCC cells showed that Nur77 directly interacted with RARβ and that the induction of Nur77 expression was dependent upon the presence of RARβ, with which it co-localized in the cytoplasm^[52]. Consistent with these findings, in human neuroblastoma cells fenretinide-induced apoptosis was blocked by two RARβ/γ-specific antagonists, but not by a RARα-specific antagonist^[53]. However, a study on human epidermoid carcinoma A431 cells showed that treatment with RAR antagonists and caspase inhibitors provided only partial protection against fenretinide, suggesting that fenretinide-mediated apoptosis is a complex process that involves both RAR-dependent and RAR–independent pathways^[54]. Abnormal ROS generation plays a pivotal role in apoptosis signaling, leading to an increase in mitochondrial membrane permeability and subsequent release of pro-apoptotic proteins such as cytochrome c (CytC), which promotes cell death via caspase activation^[55]. Moreover, aberrant ROS production can lead to the generation of bioactive lipids such as ceramides, which trigger mitochondrial-mediated apoptosis through direct disruption of mitochondrial integrity^[56,57]. Elevated levels of ROS, particularly hydroperoxides, have been observed in a variety of cancer cell types following exposure to fenretinide. The ability of antioxidants to counteract fenretinide-induced apoptosis further supports a crucial role for oxidative stress in its cytotoxic effects^[7]. In multiple cancer cell lines, fenretinide has been shown to induce CytC release and to impair mitochondrial oxidative phosphorylation (OXPHOS), indicating that fenretinide-induced ROS generation mainly takes place in the mitochondria^[7,58]. Consistent with this mechanism, fenretinide has been reported to act as a functional antagonist of retinol in mitochondrial signaling by disrupting the protein kinase Cδ (PKCδ)-CytC signalosome, a crucial OXPHOS regulator^[55].

Sphingolipids are membrane lipids that, besides their structural roles, function as key signaling molecules regulating cell survival, metabolism, and stress responses^[59]. Ceramide (Cer), a bioactive lipid primarily produced through de novo sphingolipid biosynthesis, has attracted considerable attention for its ability to promote apoptosis in response to cellular stress^[60]. In various cancer cell models, fenretinide has been shown to increase Cer levels independently of caspase activation^[61,62], suggesting that it regulates de novo Cer biosynthesis and/or sphingomyelin hydrolysis. Several studies have demonstrated that fenretinide inhibits DES1, the enzyme that catalyzes the conversion of dhCers into Cer^[8,9,63-66]. In human neuroblastoma SMS-KCNR cells, DES1 silencing resulted in a marked accumulation of endogenous dhCers and subsequent inhibition of cell growth with cell cycle arrest in G0/G1. Moreover, fenretinide treatment inhibited desaturase activity in a dose-dependent manner without affecting messenger RNA (mRNA) or DES1 protein levels, suggesting that fenretinide may act as a direct enzymatic inhibitor^[63]. Consistent with these findings, treatment of lung and colorectal cancer (CRC) spheroid cultures with fenretinide led to significant accumulation of d18-dhCer, while pre-treatment with myriocin, an inhibitor of de novo sphingolipid biosynthesis, partially protected cells from fenretinide-induced cell death^[8]. Furthermore, a lipidomic study has shown that fenretinide-induced dhCers accumulation promotes autophagy by inducing autophagosome formation in prostate cancer cells^[64]. Subsequent studies demonstrated a role of dhCers in autophagy by showing that elevated dhCers levels lead to delayed G1/S cell cycle progression and endoplasmic reticulum (ER) stress induction^[60]. In line with this, a study on glioma cells revealed that DES1 inhibition increases the dhCer/Cer ratio in the ER membrane, thereby triggering ER stress and inhibiting the protein kinase B (Akt)/mTOR complex 1 (mTORC1) pathway, ultimately leading to autophagy^[67].

The serine/threonine kinase mTOR is crucial for the regulation of cell survival, metabolism, cell growth, and protein biosynthesis, and is aberrantly upregulated in more than 70% of cancers, contributing to tumorigenesis^[68]. Previous studies have shown that fenretinide targets the phosphoinositide 3-kinase (PI3K)/Akt/mTOR signaling pathway and that Akt phosphorylation at Ser473 is implicated in fenretinide-induced apoptosis^[69,70]. In 2012 mTOR was identified for the first time as a direct molecular target of fenretinide in both in vitro and in vivo models^[11]. Specifically, molecular docking analyses showed that fenretinide binds to the ATP-binding pocket of mTOR, forming hydrogen bonds with Ser2165 and Lys2187, as well as hydrophobic interactions with key residues for the binding of ATP. Direct inhibition of mTOR suppressed the activity of both mTORC1 and mTOR complex 2 (mTORC2), resulting in reduced proliferation of lung cancer cells in vitro and decreased tumor growth in mice^[11]. Interestingly, mTOR knockout in A549 cells decreased their sensitivity to fenretinide, suggesting that mTOR signaling is essential for their antitumor activity. However, no obvious apoptosis was observed in JB6 C141 cells and A549 cells after exposure to fenretinide at concentrations required to inhibit mTOR and cell proliferation, suggesting that mTOR targeting by fenretinide primarily contributes to its antiproliferative activity rather than to induction of apoptosis^[11]. Consistent with those findings, a recent study on human epidermal growth factor receptor 2 (HER2)/neu (neuT) transgenic mice, a model of spontaneously metastatic breast cancer (BC), demonstrated that fenretinide treatment decreased mTOR pathway activation and cell cycle progression, inhibiting the growth of primary and metastatic tumors in the absence of overt apoptosis^[12].

Autophagy is a tightly regulated catabolic process characterized by the degradation of cellular components, reorganization of intracellular membranes and vesicles, and lysosomal activity^[71]. In recent years, autophagic cell death has been increasingly recognized as a key component of cellular stress responses and tumorigenesis. Indeed, cancer cells can activate autophagy as a survival mechanism in response to stressors such as radiation and chemotherapy, thus contributing to chemoresistance^[13]. However, autophagy can also promote cell death, as excessive autophagic activity can lead to the digestion of the entire cell^[72]. As mentioned in section “Role of lipid second messengers in fenretinide-induced cytotoxicity”, fenretinide inhibits DES1, leading to increased dhCer levels. Mechanistically, alterations in dhCer levels modify the lipid composition of intracellular membranes, triggering stress responses, such as oxidative and ER stress, that accompany or precede autophagy signals^[60]. However, whether fenretinide-induced autophagy is a pro-survival or a pro-death response remains a matter of debate. Fenretinide’s ability to induce autophagy is context-dependent, with some studies supporting pro-survival effects and others linking autophagic signals to cytotoxic effects. For example, a study showed that fenretinide-induced ROS production led to the unfolded protein response (UPR) in cancer cells, and the efficiency of this response determined cell fate. An impaired UPR leads to insufficient autophagosome formation and promotes apoptosis, whereas a complete UPR response enhances autophagic clearance of protein aggregates, favoring cell survival^[73]. Fenretinide has been reported to trigger autophagic cell death when apoptotic pathways are deregulated, as in the case of caspase-3-defective Michigan Cancer Foundation-7 (MCF-7) cells. Conversely, an isogenic cancer cell line with functional caspase-3 did not exhibit autophagy features following fenretinide treatment^[13]. In CRC and lung cancer spheroids, fenretinide treatment increased the expression of both microtubule-associated protein 1A/1B-light chain 3 (LC3) and its lipid-modified form LC3-II, and the autophagy inhibitor 3-methyladenine attenuated its cytotoxic effects, indicating that autophagic signals contribute to fenretinide-induced cell death in these models^[8]. A recent review has proposed that the cell fate balance between apoptosis and autophagy following fenretinide exposure is dose-dependent and strictly linked to ROS levels. High concentrations of fenretinide induce elevated ROS levels, activating the p38-dependent apoptotic pathway and suppressing the PI3K/AKT/mTORC, DES1 and RARβ-Nur77 pathways, whereas lower concentrations lead to a modest increase of ROS levels, activating autophagy pathways^[41].

CSCs were first identified in pioneering studies on AML, which demonstrated that only a rare subset of leukemic cells expressing stemness-associated markers was capable of initiating disease in immunodeficient mice^[74,75]. Subsequent studies identified CSCs in a variety of solid tumors, including brain^[76], prostate^[77], colorectal^[78], pancreatic^[79], breast^[80] and lung cancer^[81]. CSCs are characterized by self-renewal capacity, tumor-initiating potential and intrinsic resistance to cytotoxic agents, and therefore constitute key drivers of tumor relapse and therapeutic failure^[24]. They can evade the toxic effects of chemotherapy through a variety of mechanisms, including activation of pro-survival pathways such as PI3K/Akt/mTOR, upregulation of anti-apoptotic factors, and entry into a chemoresistant state of dormancy^[24,82,83]. Moreover, it has been demonstrated that CSCs can exploit autophagy as a response to stress, thereby promoting chemoresistance and subsequent disease progression^[84,85]. As mentioned in the introduction, the effectiveness of differentiating therapy was first demonstrated with ATRA in the targeted treatment of APL^[4,86,87], paving the way for the development of anti-CSCs therapeutic strategies. However, it has been demonstrated that differentiated cells can revert to an undifferentiated phenotype over time, highlighting the limitations of this approach^[88-91]. A recent study proposed the use of fenretinide as a potential alternative therapeutic strategy for APL in cases of resistance after protracted treatment with ATRA. In particular, the authors of this study demonstrated that the antitumor activity of fenretinide relies on apoptosis induction rather than differentiation^[92]. Consistent with this mechanism, a previous study in AML showed a selective cytotoxic activity on CD34⁺ leukemic stem/progenitor cells (LSCs) without affecting the normal counterparts^[18]. LSCs predominantly exhibit a quiescent phenotype that renders them resistant to conventional chemotherapeutic agents^[18]. Notably, fenretinide was found to inhibit the clonogenic capacity of primary AML CD34⁺ cells in vitro and reduce the engraftment of primary and secondary recipients in vivo, suggesting a specific effect on tumor-initiating cells. Mechanistically, fenretinide induces ROS production and inhibits the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) and Wnt/β-catenin pathways, both of which support LSCs survival and chemoresistance^[18] [Figure 2]. Similar findings have been reported in chronic myeloid leukemia, where fenretinide preferentially targets LSCs through the oxidative/ER stress-mediated pathway^[15]. Additional studies have demonstrated that fenretinide is also effective against CSCs in various solid tumors, such as medulloblastoma, colon, breast, and ovarian cancer^[8,16,17,47]. Using tumor spheroids enriched in stem-like cells, fenretinide was shown to preferentially target CSCs through ROS induction, ER stress, and inhibition of cell-cycle progression^[16,17]. Notably, low levels of ROS in CSCs are associated with enhanced tumorigenicity and resistance to radiotherapy^[93], suggesting that the stress-inducing effects of fenretinide may underlie its activity against these cells. In addition, fenretinide displayed marked efficacy against patient-derived CSCs both in vitro and in CSCs-derived xenografts of lung cancer, melanoma and CRC^[8,14]. Its anti-tumor activity correlated with reduced cell proliferation, induction of apoptosis, modulation of lipid metabolism, and downregulation of CSCs-specific markers. In primary lung cancer spheroids, fenretinide decreased the expression of stemness-related genes such as NANOG, SOX2 and POU5F1, whereas in CRC spheroids it reduced Wnt signaling activity^[8] [Figure 2]. In addition, reverse-phase proteomic analysis of fenretinide-treated primary lung cancer spheroids revealed a widespread repression of the mTOR pathway^[8] [Figure 2]. Consistently, a recent study in neuT mice showed that treatment with fenretinide prevented both initiation and progression of metastases through a combined induction of antiproliferative signals and inhibition of the mTOR pathway^[12]. Taken together, these findings strongly suggest that fenretinide preferentially targets CSCs through multiple mechanisms, including induction of ROS production, ER stress, and inhibition of key pathways that sustain CSCs survival and chemoresistance [Figure 2].

Several studies have demonstrated that fenretinide can induce cell cycle arrest across various cancer types. However, the reported effects on the cell cycle are not consistent and appear to depend on both the experimental context and the tumor type. In a study on human CRC spheroids, treatment with fenretinide increased the proportion of cells in the G1 and G2 phases, with a corresponding decrease of the proportion of cells in the S phase. Moreover, fenretinide-treated cells exhibited downregulation of genes involved in G1/S and G2/M cell cycle transitions^[16]. In medulloblastoma cells, the impact of fenretinide on the cell cycle varied depending on the dose and the cell line tested. Specifically, fenretinide induced a dose-dependent increase in the proportion of DAOY cells in the S-phase, whereas ONS-76 cells showed an increased rate of apoptosis. Consistently, treatment with fenretinide caused a downregulation of cyclin D1 and cyclin-dependent kinase 4 (CDK4) in DAOY cells, while in ONS-76 only the highest dose tested resulted in a decrease in CDK4 expression. Notably, in both cell lines fenretinide induced checkpoint kinase 2 (CHK2) phosphorylation, which was associated with cell cycle arrest in S-phase^[47]. Similarly, in AML HL-60 cells fenretinide downregulated cyclin-dependent kinase 1 (CDK1) and retinoblastoma (RB) expression, while increasing phosphorylated RB (p-RB), ultimately leading to cell cycle arrest in the S-phase^[51]. In A2780 ovarian carcinoma cells, treatment with fenretinide caused a modest arrest in the G1 phase without altering the expression of the G1 regulators cyclin B1, cyclin D1, and cyclin E. Interestingly, the same study demonstrated that 4-oxo-fenretinide, an active metabolite of fenretinide, was able to induce G2/M arrest and alter the expression of the cell cycle regulatory proteins CDC25C and phosphorylated CDK1 (p-CDK1)^[94]. In primary lung and CRC spheroids, a novel fenretinide nanoformulation (NanoFEN) decreased the expression of cell cycle-related factors while increasing the levels of p16 and phosphorylated p38 (p-p38), suggesting a block at the G0/early G1 phase of the cell cycle [Figure 2]^[8]. This effect was further supported by the use of the mVenus p27K^- reporter, which labels cells in the G0 phase and at the G0/G1 transition^[95]. NanoFEN treatment markedly increased the number of Venus⁺ cells, indicating that it promotes entry into, or retention of, a quiescent state^[8]. These findings were subsequently validated in vivo in neuT mice treated with an orally bioavailable fenretinide nanoformulation (Bio-nFeR)^[12]. Bio-nFeR significantly reduced the growth of primary tumors and decreased the expression of the proliferation marker Ki67. Similarly, it reduced the number and size of lung metastases, which exhibited decreased levels of Ki67 and proliferating cell nuclear antigen (PCNA). In addition, primary tumors treated with Bio-nFeR showed a strongly increased expression of p-p38 and a decreased amount of phosphorylated extracellular signal-regulated kinase (pERK), where the low pERK/p-p38 ratio represents a key feature of cancer dormancy [Figure 2]^[96]. Treatment with Bio-nFeR also resulted in suppression of the mTOR pathway in murine breast tumors, consistent with the notion that mTOR inhibition promotes a state of cellular “hibernation” in both stem cells and cancer cells^[82]. Taken together, these results indicate that in neuT mice, fenretinide counteracts tumor growth and metastatic progression by inducing cellular dormancy alongside suppression of metabolic and biosynthetic pathways^[12]. In recent years, the concept of therapeutically inducing tumor dormancy has gained increasing attention as a strategy to transform cancer into a clinically manageable, non-aggressive disease^[97]. Emerging preclinical and clinical evidence supports the feasibility of dormancy-based therapeutic approaches, particularly in the context of disseminated tumor cells (DTCs) and minimal residual disease^[97-99]. In this framework, the ability of novel fenretinide formulations to induce tumor dormancy highlights their potential as chemopreventive agents in BC and possibly in other tumors. When considering fenretinide-induced dormancy, it is important to distinguish between cellular dormancy and tumor mass dormancy. Cellular dormancy refers to individual cancer cells entering a reversible G0/G1 arrest while maintaining metabolic activity without proliferation^[1,7,8], whereas tumor mass dormancy is defined as a dynamic equilibrium between cellular proliferation and apoptosis that results in stable tumor size without net expansion^[100,101]. In this context, fenretinide-induced modulation of p-p38/pERK balance, together with repression of the mTOR pathway, is consistent with the induction of cellular dormancy. Cancer cell dormancy is increasingly recognized as a clinically relevant state that enables long-term persistence of DTCs, thereby contributing to minimal residual disease and tumor relapse.

A growing body of evidence shows that fenretinide is capable of modulating the TME through perturbation of signaling kinases and cell-extracellular matrix (ECM) interactions, inhibition of angiogenesis, and regulation of immune cells [Figure 3]. In a recent study on oral squamous cell carcinoma (OSCC), treatment with fenretinide induced several cancer-preventive effects including inhibition of basement membrane invasion, suppression of anchorage-independent growth, disruption of actin cytoskeletal components and inhibition of the focal adhesion kinase, a crucial protein that confers invasive properties^[20]. Using molecular docking and kinase assays, it was demonstrated that fenretinide binds with high affinity to the ATP-binding pocket of all three c-Jun N-terminal kinase (JNK) isoforms, thereby inhibiting of the JNK signaling pathway associated with epithelial–myoepithelial transition. Functionally, fenretinide treatment reduced OSCC cell adhesion to the ECM and activation of matrix metalloproteinases (MMPs) [Figure 3]. In an in vivo OSCC model, local treatment with biodegradable polymeric implants containing fenretinide significantly downregulated the expression of the proto-oncogene ERG, an epithelial marker strongly associated with angiogenesis^[20]. Consistent with these findings, other studies have demonstrated that fenretinide exerts robust anti-angiogenic effects through both direct and indirect mechanisms, reducing neovascularization and thereby enhancing antitumor efficacy^[19]. In multiple myeloma (MM), treatment with fenretinide inhibited cell growth of cancer cells in their bone marrow microenvironment, both directly through induction of apoptosis and indirectly through inhibition of osteoclastogenesis and angiogenesis. Consistently, fenretinide inhibited tube formation by human umbilical vein endothelial cells (HUVEC) stimulated by either a cocktail of pro-angiogenic growth factors or by conditioned media derived from MM cells^[102]. In a rat model of prostate cancer, fenretinide treatment significantly decreased carcinogenesis and reduced tumor vascularization^[103]. Similarly, in a xenograft model using Y79 RB cells, fenretinide inhibited tumor growth through a robust reduction of microvessel formation. These findings were confirmed in in vivo Matrigel sponge assays, where fenretinide inhibited neovascularization induced by pro-angiogenic factors^[104]. In Kaposi’s sarcoma, a highly vascularized tumor type, oral administration of fenretinide inhibited ectopic tumor growth and significantly reduced vascular density^[105]. At the molecular level, fenretinide exerted its anti-angiogenic effects through down-regulation of the vascular endothelial growth factor (VEGF) signaling axis, a crucial pathway that drives angiogenesis. Specifically, the compound was able to reduce the expression levels of both VEGF-A and its receptor VEGFR-2 in endothelial and Kaposi’s sarcoma cells^[105] [Figure 3]. Furthermore, in a mouse model of hepatocarcinogenesis VEGF-A was found downregulated in pre-neoplastic liver lesions following fenretinide treatment, thereby impairing progression of the lesions to cancer^[106].

Several studies have linked the anti-angiogenic properties of fenretinide to its interaction with RARs, particularly RARα and RARβ^[19]. Other studies have suggested that fenretinide is able to target the insulin-like growth factor-I (IGF-I)/IGF-I receptor (IGF-IR) axis, a key pathway promoting VEGF expression via hypoxia-inducible factor (HIF)-1α signaling. Consistent with this mechanism, treatment with fenretinide decreased plasma IGF-I levels in patients with breast and bladder cancer, downregulated IGF-I and IGF-IR in BC cells in vitro and suppressed IGF-I–mediated proliferation in meningioma and RB models^[19] [Figure 3]. Moreover, it has been demonstrated that fenretinide directly targets endothelial cells, interfering with multiple stages of neo-vessel formation. A mechanistic study showed that fenretinide reduces endothelial cell motility and invasiveness by inhibiting the activity of MMP-2, an enzyme essential for ECM degradation during vessel sprouting^[22]. In addition, fenretinide upregulates bone morphogenetic protein-2 (BMP-2) and macrophage inhibitory cytokine-1 (MIC-1), two multifunctional cytokines with anti-angiogenic activity in vitro and in vivo. Notably, treatment with antibodies against BMP-2 counteracted the inhibitory effects of fenretinide, suggesting that suppression of angiogenesis is a key contributor to its antitumor activity^[22]. Collectively, these findings indicate that fenretinide exerts a multifaceted inhibition of angiogenesis by acting on both tumor cells and components of the vascular TME while sparing the normal vasculature, as observed in clinical trials^[107]. Interestingly, fenretinide has been shown to suppress IL-4/IL-13-mediated M2 polarization of RAW264.7 macrophages. This effect was associated with inhibition of phosphorylation of signal transducer and activator of transcription 6 (STAT6), a key regulator of M2 polarization [Figure 3]. Moreover, fenretinide inhibited tube formation by HUVEC cells exposed to conditioned media derived from M2 macrophages^[21]. Consistent with these findings, in adenomatous polyposis coli (APC) transgenic mice, fenretinide inhibited colorectal tumorigenesis through a reduction of the M2-like macrophage population and a subsequent block of tumor angiogenesis^[21]. Figure 3 and Table 1 summarize the effects of fenretinide on the TME, including its anti-angiogenic properties, inhibition of macrophage M2 polarization and suppression of cell-ECM interactions. Collectively, these data indicate that fenretinide actions create an unfavorable microenvironment for tumor progression, further supporting its potential as a chemopreventive and antitumor agent.

There is no clear evidence to date that cancer cells develop resistance to fenretinide. While resistance to ATRA is well documented and involves diverse mechanisms, such as RARα mutations and activation of compensatory survival pathways, fenretinide has shown efficacy even in ATRA-resistant tumor cells^[130]. Nevertheless, potential adaptive or intrinsic mechanisms of resistance to fenretinide should be considered and evaluated in future studies. Such mechanisms may involve:

(a) Alterations in the apoptotic protein network, such as upregulation of anti-apoptotic Bcl-2 family proteins or suppression of Bcl-2 homology 3 domain (BH3)-only proteins. In this regard, preclinical studies on the synergistic activity of fenretinide in combination with BH3 mimetics suggest diverse and complementary actions of the two drug types^[131,132].

(b) Enhanced expression of ROS scavengers. Interestingly, a study on tumors of the Ewing sarcoma family demonstrated that the intracellular glutathione antioxidant system is a critical determinant of cellular sensitivity to fenretinide^[133].

(c) Bypassing mTOR inhibition. Since fenretinide inhibits mTOR activity, mechanisms of resistance to mTOR inhibitors, including activation of compensatory kinases, feedback activation of PI3K/AKT signaling or mTOR mutations affecting drug binding^[134], may theoretically block fenretinide effects on the mTOR pathway.

(d) Drug efflux mechanisms. Multidrug resistance efflux transporters and other mechanisms such as membrane rigidification can regulate intracellular fenretinide levels^[92]. However, in Ewing sarcoma cells fenretinide is not a substrate of the ATP-binding cassette (ABC) transporter multidrug resistance-associated protein 1 (MRP1), underscoring tumor–specific differences in the contribution of drug efflux proteins to fenretinide sensitivity^[135].

In summary, while the multifaceted mechanism of action of fenretinide makes the development of resistance less likely, tumor cells may theoretically still acquire escape mechanisms. Future studies evaluating novel fenretinide formulations will be crucial to identify and overcome potential resistance mechanisms.