PROTACs: Current and Future Potential as a Precision Medicine Strategy to Combat Cancer

Abstract

Proteolysis targeting chimeras (PROTAC) are an emerging precision medicine strategy, which targets key proteins for proteolytic degradation to ultimately induce cancer cell killing. These hetero-bifunctional molecules hijack the ubiquitin proteasome system to selectively add polyubiquitin chains onto a specific protein target to induce proteolytic degradation. Importantly, PROTACs have the capacity to target virtually any intracellular and transmembrane protein for degradation, including oncoproteins previously considered undruggable, which strategically positions PROTACs at the crossroads of multiple cancer research areas. In this review, we present normal functions of the ubiquitin regulation proteins and describe the application of PROTACs to improve the efficacy of current broad-spectrum therapeutics. We subsequently present the potential for PROTACs to exploit specific cancer vulnerabilities through synthetic genetic approaches, which may expedite the development, translation, and utility of novel synthetic genetic therapies in cancer. Finally, we describe the challenges associated with PROTACs and the ongoing efforts to overcome these issues to streamline clinical translation. Ultimately, these efforts may lead to their routine clinical use, which is expected to revolutionize cancer treatment strategies, delay familial cancer onset, and ultimately improve the lives and outcomes of those living with cancer.

Introduction

Protein homeostasis is essential for normal cell physiology and function, and involves the precise regulation of protein activity, localization, and abundance (1–3). Protein ubiquitination is essential for each of these features, with the ubiquitin proteasome system encompassing various ubiquitination proteins that polyubiquitinate protein substrates to mark them for proteolytic degradation via the 26S proteasome to impact protein abundance and consequently, function (1, 4, 5). Understanding the precise mechanisms of proteolytic regulation has been of particular interest in cancer research, as ongoing efforts now seek to harness the proteolytic targeting capabilities of the ubiquitin proteasome system to selectively target and degrade proteins, particularly oncoproteins, for therapeutic benefit (6–9). Although there are numerous proteolytic approaches under development (reviewed in ref. 10), proteolysis targeting chimeras (PROTACs) are one of the most advanced proteolytic targeting approaches to date (8, 10).

In essence, PROTACs employ a linker to bring an endogenous ubiquitination protein into close spatial proximity of a protein substrate to mark it for proteolytic degradation (11). An inherent and desirable feature of PROTACs is its ability to target virtually any intracellular and transmembrane proteins, including those that are classically considered undruggable, such as constitutively active mutant and oncogenic proteins (8, 9, 11–13). Although there are many recent reviews that describe the targeting aspect of PROTACs (see refs. 6, 7, 9, 14, 15), this review is focused on presenting and discussing current and future PROTAC applications to begin to conceptually address some of the current limitations associated with persistent oncoprotein targeting. Accordingly, we begin by defining the normal functions of ubiquitin-regulating proteins before presenting ongoing areas of research aimed at harnessing the 26S proteasome system for cancer-specific targeting. We then discuss the unexploited potential for PROTACs in cancer therapy, the ongoing efforts to enhance cancer-specific delivery and the prospect for PROTACs to prevent and/or delay the onset of familial cancer syndromes.

The Ubiquitin Proteasome System and PROTACs

Ubiquitin is an 8.5 kDa globular protein, which can be covalently attached to or removed from protein substrates at lysine residues to regulate protein localization, abundance, and/or function (reviewed in refs. 1, 4, 16). Protein ubiquitination occurs through the concerted activities of E1 (activating), E2 (conjugating), and E3 (ligating) ubiquitination proteins. The human genome encodes two E1s, 35 E2s and ∼600 E3s, that confer protein substrate specificity and determine mono- or polyubiquitination assembly (Fig. 1A; refs. 17, 18). Polyubiquitination occurs by covalently attaching ubiquitin moieties onto one of seven lysine residues (K6, K11, K27, K29, K33, K48, K63) or the N-terminal methionine (M1) of the preceding ubiquitin moiety and may result in homotypic (i.e., same linkage) or heterotypic (i.e., different linkage) chains (1, 18, 19). The diverse chain conformations are frequently associated with distinct regulatory outcomes for the various protein substrates. For example, a well-studied role for polyubiquitin is in the targeting of proteins for proteasomal degradation, which primarily involves K48-linked chains (20, 21). Opposing ubiquitination is the activity of deubiquitination enzymes (Fig. 1A), and the balance of these contrasting families of proteins is essential for protein and cellular homeostasis (1, 16). Unsurprisingly, ubiquitination and deubiquitination genes are aberrantly expressed across a myriad of cancer types, as presented within The Cancer Genome Atlas (22–24) and the Catalog of Somatic Mutations in Cancer (COSMIC) databases (25, 26). Accordingly, E3s that are overexpressed in cancers provide an ideal system to exploit to develop novel targeted cancer therapeutic strategies (reviewed in refs. 7, 14, 27, 28).

Figure 1.

The ubiquitination, deubiquitination, and polyubiquitination cycles of target proteins and the basis of PROTAC activity. A, The sequential activities of E1 (activating), E2 (conjugating), and E3 (ligating) ubiquitination proteins covalently attach ubiquitin to a lysine residue of a target protein, which may be released as a monoubiquitinated protein or remain bound to the E3 complex and be polyubiquitinated thorugh successive rounds of ubiquitination with the polyubiquitinated substrate potentially targeted for proteolytic degradation via the 26S proteasome. Alternatively, mono and polyubiquitin modifications may be removed by deubiquitination proteins. B, PROTACs bring the endogenous E3 and protein target into close spatial proximity through specific ligands that facilitate polyubiquitination and subsequent proteasomal degradation of the target. Once the target is polyubiquitinated and the PROTAC is released, the PROTAC is recycled and can polyubiquitinate additional targets for destruction. Figure generated using BioRender.

PROTACs are purposefully designed to hijack the endogenous E3-mediated degradation machinery and are an emerging proteolytic targeting approach for cancer therapy (reviewed in refs. 6–9, 14). Conceptually, they are hetero-bifunctional molecules comprised of two ligands specific to a protein of interest and a distinct E3-ligase that are joined by a chimeric linker to effectively capture and bring a protein of interest into close spatial proximity of the E3 (Fig. 1B). This artificial accessibility to the protein target promotes its polyubiquitination and subsequent degradation by the 26S proteasome. Unlike small molecular inhibitors, PROTACs activities are iterative (i.e., recycled) and as a direct result, are expected to function at lower doses relative to small molecular inhibitors (7, 9). Moreover, they can theoretically be developed to target any intracellular or transmembrane protein of interest, including those previously considered undruggable such as nonenzymatic proteins (6, 7, 13, 29). Finally, as PROTACs orchestrate protein target destruction, this perpetual degradation activity is expected to alleviate any nonenzymatic (e.g., structural) roles of the target protein, unlike small molecule inhibitors that classically only affect enzymatic activities (7, 9, 14). Despite these advantages, further research is critical to understand and mitigate off-target effects, which are not well understood.

Current PROTACs research has evolved significantly since the first proof-of-concept in 2001 (PROTAC1) that harnessed the SCF (SKP1, S-phase kinase-associated protein 1; CUL1, Cullin 1; F-box protein) complex, an E3 ligase, in Xenopus laevis extracts (11). Subsequent efforts have generated cell-penetrating PROTACs that evolved from simple peptide compositions to designs based on fully synthetic small molecules that also utilize alternative E3 complexes for proteasomal degradation (30–32). However, only thirteen of the ∼600 E3s have been employed in PROTAC designs, namely Arylhydrocarbon Receptor (AhR), Cereblon (CRBN), Cellular Inhibitor of Apoptosis 1 (cIAP1), DDB1 And CUL4 Associated Factor 11 (DCAF11), DCAF15, DCAF16, Fem-1 Homolog B (FEM1B), Kelch-like ECH-associated Protein 1 (KEAP1), Mouse Double Minute 2 Homolog (MDM2), Ring Finger Protein 4 (RNF4), RNF114, Von Hippel-Lindau Tumor Suppressor (VHL), and X-linked IAP (XIAP; ref. 33). The limited number of E3s currently employed in PROTAC strategies highlights a wealth of opportunities to expand the repertoire of targetable proteins (7–9, 14, 28, 33). However, the high molecular weight (0.6–1.3 kDa) relative to classical small molecule inhibitors (<0.5 kDa) and large polar surfaces of PROTACs remain a challenge in designing new PROTACs, as these biophysical characteristics restrict their solubility and permeability, and consequently reduce their bioavailability for clinical use and oral administration (14, 34, 35). Moreover, PROTACs rely on the ubiquitin proteasome system and are therefore limited to intracellular and transmembrane proteins, leaving the degradation of secretory and various monotypic membrane proteins unresolved by traditional PROTACs (29). Despite these limitations, more than 20 PROTACs have advanced to phase I and II clinical trials for both solid (e.g., ARV-766; KT-333) and hematologic (e.g., NX-2127; DT2216) malignancies (7). Furthermore, and although early, two PROTACs targeting the androgen (ARV-110; NCT03888612) and estrogen (ARV-471; NCT05654623) receptors have entered phase II and III trials for prostate cancer and breast cancer, respectively, underscoring the potential clinical utility of PROTACs as a novel therapeutic strategy in cancer (36–38).

Expanding the Potential of PROTACs in Cancer Treatment

Advancing target specificity and improving the efficacies of established therapies

Undoubtedly, one of the most attractive features of PROTACs as a therapeutic strategy is their potential to target historically undruggable oncoproteins, such as Myc Proto-oncogene Protein (MYC) and Signal Transducer and Activator of Transcription 3 (STAT3; reviewed in refs. 6, 7, 14). However, PROTACs have additional advantages including increased target specificity and continued degradation of targeted proteins in cells that become resistant to small molecule inhibitors (7–9). For example, ARV-110 is effective at reducing androgen receptor abundance in patient xenografts harboring mutations that confer androgen receptor inhibitor resistance (39) and may offer clinical benefit for metastatic castration-resistant prostate cancer. However, as ARV-110 targets both mutant and wild-type androgen receptors, minor side effects have been observed and include nausea (41% with 1% severe), fatigue (27% with 1% severe), diarrhea (15% with 2% severe), and alopecia (11%). In contrast, Bond and colleagues (40) developed a KRAS^G12C-targeting PROTAC (LC-2) as an alternative to the small molecular inhibitor MRTX849, which selectively targets oncogenic KRAS^G12C (41). Indeed, LC-2 maintains KRAS^G12C-targeted destruction in various cell lines, including those that are MRTX849 resistant (40). Interestingly, LC-2 also appears to exhibit KRAS^G12C-specific degradation in both heterozygous (KRAS^WT/G12C) and homozygous (KRAS^G12C/G12C) cellular contexts, with no overt impact on wild-type KRAS abundance in the KRAS^WT/G13D cells as determined by Western blot analyses. Although LC-2 specificity will ideally need to be evaluated in appropriate, nonmalignant models, these findings suggest PROTACs can be designed to specifically target mutant oncogenic proteins. This possibility is further supported by biochemical studies demonstrating that PROTACs can also successfully target BRAF^V600E (42, 43). With minimal to no apparent effects on wild-type protein abundance, purposefully designed PROTACs targeting mutant oncoproteins appear to exhibit maximal on-target effects and potentially minimal to no side effects in nonmalignant, healthy cells with wild-type protein expression. Furthermore, the specific targeting of mutant oncoproteins, in combination with the iterative/recycling nature of PROTACs, may enable sustained activity and low effective doses with reduced off-target effects relative to traditional small molecule inhibitors. As a result, PROTACs may harbor the inherent ability to reduce adverse side-effects associated with many current chemotherapeutics.

The emergence of drug resistance is a relatively common occurrence for many people undergoing chemotherapy. Despite this continued obstacle in cancer treatment, efforts to improve therapeutic efficacy have expanded our understanding of both inherent and acquired resistance mechanisms (44). For example, the persistent targeting of oncogenic proteins can lead to the activation of drug resistance mechanisms including the upregulation and/or constitutive activation of downstream pathway members stemming from oncogenic addiction (45, 46). These resistance mechanisms may present an alternative application for PROTACs to pre-emptively target the established downstream mechanism(s) (i.e., hypermorphic expression and/or function) underlying initial mechanism(s) of drug resistance. This approach will be critical for cancers like BRAF^V600E inhibitor resistant metastatic colorectal cancer, where prognosis is often less than a year (47–49). Recent data have determined that genetic and pharmacogenetic targeting of mucin 1, cell surface associated (MUC1-C) suppresses the mitogen-activated protein kinase (MAPK) feedback pathway underlying BRAF^V600E inhibitor resistance in colorectal cancer (50). Targeting MUC1-C not only prevented BRAF^V600E inhibitor resistance in vitro and in tumor xenografts, but also restored sensitivity to cells with established resistance. More recently, a bromodomain-containing protein 4 (BRD4)-targeting PROTAC was utilized in mouse models to reduce BRD4 upregulation, which was proposed to drive doxorubicin resistance (51). The codelivery of doxorubicin and a BRD4–PROTAC to cells had a greater effect on reducing tumor volume than doxorubicin treatment alone. Although still in its infancy, these exciting findings support the potential of PROTACs to prevent resistance to established treatments, and present compelling evidence for their potential to restore drug sensitivity in various cancer contexts.

PROTACs are now being designed to target multiple proteins for degradation within the same cell. Because of their increased specificity and ability to function at lower concentrations than traditional inhibitors, PROTACs are an emerging tool for targeting difficult to treat subpopulations of cells within a tumor (52–54). For instance, Jia and colleagues (52) devised a dual PROTAC (i.e., PROTAC 753B) capable of targeting B-cell lymphoma-extra large (BCL-XL) and BCL-2 to mitigate chemotherapy-induced senescence of acute myeloid leukemia cells. Importantly, PROTAC 753B utilizes VHL as the E3 ligand, which is normally minimally expressed in platelets and is therefore predicted to prevent/limit the thrombocytopenia (platelet deficiency) associated with Navitoclax (BCL-XL/BCL-2 inhibitor) treatments (52, 55). In addition to dual PROTACs, a novel multifunctional PROTAC (Y-PROTAC) was recently developed, which incorporates a glutathione-cleavable disulfide bond connecting a cyclin-dependent kinase 4/6 (CDK4/6) inhibitor to an anaplastic lymphoma kinase (ALK)-targeting PROTAC (56). The activity of the two Y-PROTAC components is sterically hindered until glutathione-mediated cleavage, which is elevated in numerous cancers relative to nonmalignant cells (57). This difference in glutathione expression may be particularly important for this Y-PROTAC in limiting off-target activity when treating glutathione-high tumors.

Multitarget PROTACs (e.g., PROTAC 753B; Y-PROTAC) present an invaluable tool for targeting difficult to treat cancers. Further development of these PROTACs may prove clinically useful to treat cancers exhibiting chromosome instability (CIN). CIN is an aberrant phenotype defined as an increase in the rate at which whole chromosomes, or large chromosome fragments, are gained or lost and is a driver of genetic and cell-to-cell heterogeneity (58, 59). CIN is also associated with early disease development (cellular transformation), intratumoral heterogeneity, metastases, and the acquisition of drug resistance (58, 60–64). As CIN-positive tumors are genetically heterogeneous, PROTACs may offer a more effective approach to therapeutically target cancers than traditional chemotherapeutics and small molecular inhibitors that are susceptible to ongoing genetic mutations/alterations. As the molecular determinants (i.e., genes, proteins, and pathways) underlying CIN are identified, PROTACs utilizing E3 ligases that are upregulated in cancer cells may represent invaluable tools to enhance drug sensitivity and efficacy. Swanton and colleagues (65) developed a treatment stratification plan based on specific genetic alterations occurring in CIN-positive breast and ovarian cancers, which identified patients most likely to benefit from taxane treatments. They determined that cancers exhibiting high CIN levels have increased expression of specific genes, including Nucleoporin 205 (NUP205), H2A Histone Family Member X (H2AFX), Cell Division Cycle 6 (CDC6), and Replication Protein A1 (RPA1), which when silenced, corresponded with enhanced taxane sensitivity. Although speculative, their findings support the future applicability for PROTACs to improve and/or restore sensitivities to established broad-spectrum therapeutics. The ability of PROTACs to target previously undruggable therapeutic targets further underlines their value in the development of advanced and more effective cancer treatment strategies, which may be particularly valuable for treating CIN-positive tumors.

Alternative therapeutic strategies have also sought to target (i.e., reduce) CIN in cancer and are reviewed in detail elsewhere (66). Briefly, one such protein target that may benefit from further PROTAC development is Cyclin E1. Cyclin E1 is a prototypic cell-cycle protein, whose abundance is dynamically regulated throughout the cell cycle (67, 68). Interestingly however, the Cyclin E1 gene (CCNE1) is amplified in numerous cancer types, which corresponds with increased protein abundance that is associated with CIN, cellular transformation (69, 70), and resistance to both cytotoxic and targeted therapies (71–73). More recent work has shown that Cyclin E1 knockdown or inhibition improves the efficacies of Regorafenib and Sorafenib (multikinase inhibitors), in a hepatocellular carcinoma context (74). Therefore, PROTAC targeting of Cyclin E1 in these contexts may reduce CIN and restore sensitivity to certain anticancer therapeutics that may also allow for the further reduction of effective drug concentrations that will potentially reduce adverse side effects.

PROTACs in synthetic genetic therapeutic targeting

Beyond targeting undruggable oncoproteins, preventing resistance to broad-spectrum therapeutics and restoring drug sensitivities, PROTACs can conceivably be developed to therapeutically leverage oncogenic alterations that differentiate cancer from normal cells. Synthetic genetic approaches, such as synthetic lethality or synthetic dosage lethality, seek to exploit specific genetic alterations contained within a cancer cell (75). Although synthetic lethality is defined as a rare and lethal combination of two independently viable gene mutations/deletions, synthetic dosage lethality refers to a lethal combination in which one of the altered genes is overexpressed (oncogene/oncoprotein) whereas the other is mutated, deleted, or inhibited (76). Both of these synthetic genetic strategies seek to exploit a specific gene/protein defect (deletion/reduced expression or amplification/overexpression) by downregulating or inhibiting a second gene product to evoke highly selective killing within a cancer cell context. Accordingly, PROTACs may provide an ideal therapeutic advantage in specific cancer contexts, as they can be developed to target a synthetic genetic partner for proteolytic degradation to induce cancer cell killing.

Conceptually, synthetic genetic susceptibilities may arise through multiple mechanisms that traditionally include alterations at the level of the genome but may also include misregulation at the protein level. Indeed, it was recently shown that cells with CCNE1 amplification (i.e., Cyclin E1 overexpression) are selectively killed by downregulating protein kinase, membrane associated tyrosine/threonine 1 (PKMYT1) activity using a novel small molecule inhibitor (77). The SCF complex (E3 ligase) normally polyubiquitinates Cyclin E1 to target it for proteolytic degradation via the 26S proteasome (Fig. 2A), and defects in the SCF complex underlie increases in Cyclin E1 that are independent of genomic amplification (78–82). Interestingly, each of the core SCF complex genes (SKP1, CUL1, and RBX1) are somatically altered in a variety of cancer types (22–24), suggesting it may be a pathogenic event, or at least a cancer-specific alteration that could be therapeutically exploited using a similar synthetic genetic strategy (80–83), as members of the same biological pathway/complex frequently share synthetic genetic interactors (84–86). Although yet to be empirically established, these findings suggest that PROTACs targeting PKMYT1 may have therapeutic benefits beyond genomic amplification of CCNE1 and are predicted to include increased Cyclin E1 abundance stemming from its mis-regulation (i.e., lack of proteolytic degradation) due to genetic defects in SCF complex member genes (Fig. 2B). Furthermore, such robust synthetic dosage lethal interactions are expected to have increased efficacy across the molecular heterogeneity comprised within a tumor (87). The utility of PROTACs in targeting a synthetic genetic interactor, such as PKMYT1, may therefore expand the types of cancers and numbers of patients benefitting from a PROTACs-based strategy. As more synthetic genetic interactors (i.e., PROTAC targets) are identified, there will be an increasing need to develop novel therapeutics capable of targeting those proteins, which may be addressed through novel PROTAC development.

Figure 2.

Synthetic genetic PROTAC approaches to target Cyclin E1 overexpressing cells in cancer. A, The SCF complex, an E3 ligase, polyubiquitinates Cyclin E1 for subsequent proteasomal degradation. B, Genomic amplication of CCNE1 (Cyclin E1) or the loss of genes encoding the SCF complex (e.g., SKP1, CUL1, RBX1, or the 69 F-Box genes) lead to increases in Cyclin E1 abundance. Cells with CCNE1 accumulation are sensitive to PKMYT1 inhibition (77) and thus, each of these genetic contexts are predicted to be sensitive to PROTACs targeting PKMYT1. Figure generated using BioRender.

Challenges with Translating PROTACs into Clinical Settings

Improving cell permeability and tissue specificity

A fundamental challenge in ultimately realizing the clinical potential and promise of PROTACs is their delivery into the requisite tissues and cells within a patient. Their current biophysical properties (e.g., relatively large molecular mass and polarity) adversely impact cell permeability, and nonspecific PROTAC delivery and function are predicted to induce side effects, particularly when targeting overexpressed, wild-type oncogenic proteins (88, 89). Depending on the protein target of interest, there may be additional side effects including the potential to induce CIN or secondary malignancies, as both over- and reduced-expression of certain genes, such as Tumor Protein 53 (TP53) and RAD51 Protein A (RAD51), are pathogenic (90–94). Accordingly, there are numerous strategies currently under development aimed at addressing these issues, which include antibody-, ligand-, and nano-PROTACs (Fig. 3; reviewed in refs. 27, 88, 89), which are briefly described below.

Figure 3.

Strategies aimed at enhancing PROTAC delivery to target cells. A, Antibodies targeting membrane proteins present on cancer cells can be fused to PROTACs to enhance their endocytosis-mediated entry. Antibodies are cleaved from PROTACs within the endosome and PROTACs are released. B, PROTACs can be fused to ligands that are recognized by membrane receptors exclusively expressed or upregulated on cancer cells. PROTACs enter the cell through endocytosis and are released into the cytoplasm. C, PROTACs are packaged into nanoparticles for improved pharmacokinetics and delivery into cancer cells. Nanoparticle delivery can be: (i) passive, using endocytic pathways; (ii) active, by fusing nanoparticles with appropriate antibodies; or (iii) active, by complexing nanoparticles with ligands that bind to membrane proteins or receptors located on cancer cell. In all cases, nanoparticle delivery undergoes endocytosis and cargo delivery to and release from endosomes. PROTACs escape the endosome and are released into the cytoplasm, where they function to target and degrade key target proteins. Figure generated using BioRender.

The overarching goal of antibody-PROTACs is to enable cell-specific delivery through antibody/antigen-dependent endocytosis and endosome-mediated release of the PROTAC to prevent its premature digestion by lysosomes (Fig. 3A; refs. 27, 89). This strategy requires a cancer-specific, cell surface antigen for which an antibody exists. For example, Trastuzumab (Herceptin) is an antibody-based therapy, which is specific to the tyrosine kinase receptor HER2 and was recently conjugated with a PROTAC to enhance its delivery into HER2-positive breast cancer cells (95). In an analogous fashion, the ligand-PROTAC approach employs a ligand/receptor pair (Fig. 3B) to enhance PROTAC delivery to cancer cells and has shown some initial promise. In 2021, Liu and colleagues (96) developed three folate-PROTACs that enhanced their delivery to cancer cells with increased folate receptor alpha (FOLR1) expression. As predicted, the folate-PROTACs induced targeted destruction that corresponded with FOLR1 abundance, with only minimal off-target destruction observed within control conditions. Overall, and although early, these initial results warrant further preclinical study before the true clinical utility of antibody- or ligand-PROTACs can be realized.

Antibody- and ligand-PROTACs exhibit enhanced targeting efficacies; however, such modifications also add bulkiness to an already large PROTAC molecule that may adversely impact their pharmacokinetic stability potentially limiting its clinical utility (88, 96). Rather than employing large antibody/ligand chimeric molecules, small peptides that mimic antibody/ligand recognition sites may be an attractive alternative (88, 97). In particular, stapled peptides contain cross-links (e.g., hydrocarbon) that maintain their α-helical structure and increase stability relative to unmodified peptides (98–100). Additional approaches are emerging that employ stapled peptides specific to the protein target and/or the E3 ligase to reduce the overall PROTAC size (101–104). Recently, Yokoo and colleagues developed a stapled peptide PROTAC (LCL-stPERML-R7) targeting estrogen receptor alpha to advance their previous nonstapled version (LCL-PERM3-R7; refs. 102). The stapled peptide alleviated stability issues encountered with the nonstapled version and improved the intracellular activity. Although the potential role of stapled peptides in PROTAC design is still being investigated, stapled peptides appear to mitigate current delivery issues that may limit the clinical utility of antibody/ligand approaches, which together, may produce a stable PROTAC with minimal passive delivery to off-target cells or tissues.

Nanoparticle technology has also garnered interest based on its ability to enhance drug delivery, specificity, and efficiency (7, 88, 89). On the basis of their mode of cell entry, nanoparticles can be classified into two general targeting groups: (i) passive or (ii) active (89, 105, 106). Briefly, passive targeting nanoparticles rely on existing malignant tumor-hypoxia and angiogenesis, which promote leaky vasculature and enable diffusion or endocytic entry into cancer cells, whereas active targeting nanoparticles rely on receptor-mediated endocytosis, like antibody-or ligand-PROTACs (Fig. 3C). In 2022, Zhang and colleagues (107) developed the first nanoparticle-PROTACs (nano-PROTAC) to successfully target Lewis lung carcinoma in vivo. More specifically, a BRD4-targeting PROTAC, dBET6, was packaged into a pH/glutathione reactive polymer nanoparticle (DS-PLGA). To further promote cancer-specific targeting, they generated two distinct nano-PROTACs using DS-PLGA coated with synthetically engineered Lewis lung carcinoma cell membranes with and without a macrophage-specific peptide (CRV) to also target tumor-associated macrophages. All three nano-PROTACs demonstrated prolonged circulation times in vivo. Importantly, all three nano-PROTACs induced greater reductions in tumor volumes (∼75–90%) than the unencapsulated dBET6 PROTAC (∼50% reduction). Nanoparticle distribution was also evaluated and revealed the membrane-bound nano-PROTACs had the greatest accumulation within tumor cells; however, this was exceeded by off-target accumulation within the spleen and liver. These off-target observations are consistent with other nanoparticle-based, in vivo studies (108, 109) and underscores the need for further optimization of in vivo target delivery.

Developing PROTACs with appropriate E3s for optimal anticancer activities

Although PROTACs are a promising strategy for targeted proteolysis with the potential to revolutionize the cancer therapy landscape, they rely on ubiquitin-based, proteasome targeting. As ubiquitination genes are gained and/or lost in many cancer types (22–24), this dependency presents a critical limitation that must be carefully considered to fulfil their potential therapeutic promise (110). Resistance to VHL- and CRBN-based PROTACs has been observed in preclinical studies through mutations and diminished expression of these ubiquitination proteins and their interacting E3 complex partners (111, 112). Indeed, heterozygous and homozygous loss of genes encoding an array of E3 ubiquitin ligases occur in many cancers that are proposed to be pathogenic events (22, 23, 113). For example, the three genes encoding the core SCF complex members (SKP1, CUL1, and RBX1) exhibit copy-number losses in numerous cancer types (22–24, 80–83). This E3 ligase complex primarily polyubiquitinates protein substrates to target them for proteolytic degradation via the 26S proteasome. Moreover, each core member exhibits a distinct function—RBX1 recruits the E2 enzyme to the SCF complex, CUL1 exhibits a scaffolding function between RBX1 and SKP1, whereas SKP1 associates with one of 69 variable F-box proteins that impart their own distinct substrate specificities to the complex (114, 115). Because SKP1 and RBX1 also form additional E3 complexes (80, 116, 117), their loss in certain cancer contexts will undoubtedly disrupt PROTACs that hijack any E3s involved in those complexes. Moreover, overexpression of deubiquitination enzymes resulting from gene amplification in certain cancers may counteract and adversely impact PROTAC efficiency (118). Accordingly, genomic and epigenomic characterization of cancer cells will be essential for the appropriate selection of E3s for successful PROTAC design, particularly in CIN-positive cancers where chromosomes and their associated genes undergo continual gains and/or losses. As E3s are frequently altered in numerous cancer types, emerging work now seeks to expand the current repertoire of PROTAC E3s to ultimately broaden the clinical utility of PROTACs (reviewed in refs. 7, 8, 28).

For proteins to be efficiently ubiquitinated, its critical that the substrate lysine motif(s) be readily accessible to the PROTAC E3 ligase. As only a small fraction of E3s are currently utilized for PROTAC targeting, achieving this spatial compatibility will require a greater understanding of the 3D structure of both the E3 and the target protein (119–121). Additional considerations in the development of novel E3s should include an understanding of tissue-specific expression, the intracellular localization, associated half-life, and polyubiquitination capacity of each ligase, as well as the type of chain conformation (e.g., K48-linkage) employed (7, 8, 27, 28). PROTAC design will also need to consider sites of posttranslational modifications to the E3, essential E3-binding domains, and the spatial conformations of the E3 ligase with the E2 and its associated E3 complex members to ensure ubiquitination activity is maintained. Moreover, PROTAC therapeutics could include E3s that are essential and/or oncogenically upregulated specifically within cancer (28) as these E3s are likely to be maintained in cancer cell contexts, which may prevent or limit PROTAC resistance due to their subsequent loss or deletion (27, 28, 33). Hijacking oncogenically upregulated E3 ligases may also reduce the E3’s endogenous oncogenic activity in malignant cells, while additionally minimizing PROTAC activity in nonmalignant cells. Expanding the repetoire of PROTAC E3s used for therapeutic purposes will assuage the limitations of the small number of E3s currently under study and conceivably, will provide more therapeutic options within the clinic. Increasing the number of E3s amendable to PROTAC activity will also expand the structural diversity among PROTAC E3s to improve the spatial compatibility required for effective ubiquitination of a broader number of specific targets and ultimately benefit larger cohorts of patients from a wider spectrum of cancer types (reviewed in refs. 7, 8).

Future Considerations in Expanding the Utility of PROTACs as Long-term Anticancer Therapeutics

The development of PROTACs for therapeutic purposes is still in its infancy with the first clinical trials currently underway (36–38). Further translation of PROTACs into the clinic will require critical evaluation and further optimization of: (i) cancer cell–specific delivery to limit off-target effects; (ii) the identification of additional PROTAC E3 ligases for optimal target degradation, increased target diversity, and improved tissue specificity; and (iii) careful genomic/epigenomic understanding of the E3 ligase landscape within malignant cells. Only once these features are thoroughly evaluated in preclinical models and subjected to clinical trials will the full therapeutic utility of PROTACs be realized.

Beyond the promise of cancer treatments, PROTACs may also prove effective in delaying onset in individuals at risk for familial cancer syndromes such as Li–Fraumeni syndrome (122), Lynch syndrome (123, 124), and familial adenomatous polyposis (FAP; refs. 124, 125). For example, individuals with FAP harbor germline mutations in adenomatous polyposis coli (APC) that inactivates this classical tumor suppressor gene (detailed below; refs. 124, 125). As a result, individuals with FAP develop hundreds to thousands of precancerous polyps within their colon often by their mid-teens (126) and have a 100% lifetime risk to develop colorectal cancer. Thus, early detection, ongoing surveillance, and appropriate therapeutic intervention is critical to improve outcomes for these individuals (125, 126). APC normally functions as a core component of the APC destruction complex along with serine/threonine kinases glycogen synthase kinase 3 (GSK-3), casein kinase 1 (CK1), and Axin. The APC destruction complex normally sequesters and phosphorylates β-catenin as a crucial requirement for its recognition and polyubiquitination by the βTrCP-containing (F-box/WD repeat-containing protein 1A) SCF complex (125, 127). Accordingly, APC is essential for  β-catenin degradation to regulate the Wnt/β-catenin pathway, which encompasses the transcriptional control of proto-oncogenes such as MYC (125).

The current strategy to delay onset and/or prevent colorectal cancer in patients with FAP is routine monitoring by colonoscopy and prophylactic polypectomy (polyp removal) or colectomy (colon resection) as appropriate (125, 128); however, PROTACs may prove to be an effective alternative strategy. For example, a stapled β-catenin-PROTAC (xStAx-VHLL) was recently developed that employs a chemically stapled peptide (xStAx), in which the stapled peptide was modeled after the β-catenin Axin binding motif and linked to a VHL ligand (103). The xSTAx-VHLL PROTAC exhibited sustained β-catenin degradation over 24 hours and prevented Wnt/β-catenin signaling in colorectal cancer cell lines (103). Furthermore, proliferation was impeded in subcutaneously engrafted tumors in mice, whereas xSTAx-VHLL treatment increased killing of 11 of 12 colorectal cancer patient-derived organoids by >50% relative to a negative control. Although these results are encouraging, additional and long-term preclinical studies are highly warranted. In any case, future therapies may include β-catenin-targeting PROTACs at the onset of hyperpolyp formation as means to delay and/or prevent colorectal cancer formation to better preserve the colon and improve the quality of life of those with FAP. Despite the potential for PROTACs as a preventative strategy, such treatments would naturally require long-term administration, and like many therapies, resistance may become an issue. Nevertheless, additional preclinical studies are required to ultimately determine the clinical utility of PROTACs in preventing or delaying FAP and other familial cancer syndromes.

Conclusions

PROTACs are at the crossroads of many cancer research areas with a particular focus on targeting historically undruggable proteins. As additional PROTAC E3 ligases are developed, it is anticipated that PROTACs will advance tumor-specific targeting and improve current cancer treatments by providing novel precision medicine strategies. Furthermore, PROTACs that hijack oncogenically upregulated E3s may promote context-specific target degradation (i.e., cancer cell–specific) with minimal impact in normal tissues, while those targeting downstream mechanisms of resistance may mitigate issues associated with classical and persistent oncoprotein targeting by small molecule inhibitors. In addition, PROTACs may become a clinically relevant tool in precision medicine by targeting synthetic genetic interactors and may prove useful in preventing or delaying familial cancer syndromes by targeting the underlying mutant proteins. Ongoing research is actively tackling the various technical, biophysical, and clinical challenges associated with PROTACs, including their delivery (e.g., nanoparticle technologies) and the number of E3 ligases employed, to ultimately increase the diversity of protein substrates that can be efficiently targeted for therapeutic purposes. Only once these challenges are surmounted, will the true clinical utility of PROTACs be realized.

Authors' Disclosures

K.A. Rutherford reports other support from CIHR CGS-M (CIHR Master's Scholarship), NSERC CGS-D (NSERC PhD Scholarship), and Research Manitoba/CancerCare Manitoba PhD in Health Research Studentship Award outside the submitted work. K.J. McManus reports grants from Canadian Institutes of Health Research and CancerCare Manitoba Foundation; other support from Canadian Institutes of Health Research Graduate Scholarship and Research Manitoba/CancerCare Manitoba Studentship during the conduct of the study.