Current progress and remaining challenges of peptide–drug conjugates (PDCs): next generation of antibody-drug conjugates (ADCs)?

Abstract

Drug conjugates have emerged as a promising alternative delivery system designed to deliver an ultra-toxic payload directly to the target cancer cells, maximizing therapeutic efficacy while minimizing toxicity. Among these, antibody-drug conjugates (ADCs) have garnered significant attention from both academia and industry due to their great potential for cancer therapy. However, peptide-drug conjugates (PDCs) offer several advantages over ADCs, including more accessible industrial synthesis, versatile functionalization, high tissue penetration, and rapid clearance with low immunotoxicity. These factors position PDCs as up-and-coming drug candidates for future cancer therapy. Despite their potential, PDCs face challenges such as poor pharmacokinetic properties and low bioactivity, which hinder their clinical development. How to design PDCs to meet clinical needs is a big challenge and urgent to resolve. In this review, we first carefully analyzed the general consideration of successful PDC design learning from ADCs. Then, we summarised the basic functions of each component of a PDC construct, comprising of peptides, linkers and payloads. The peptides in PDCs were categorized into three types: tumor targeting peptides, cell penetrating peptide and self-assembling peptide. We then analyzed the potential of these peptides for drug delivery, such as overcoming drug resistance, controlling drug release and improving therapeutic efficacy with reduced non-specific toxicity. To better understand the potential druggability of PDCs, we discussed the pharmacokinetics of PDCs and also briefly introduced the current PDCs in clinical trials. Lastly, we discussed the future perspectives for the successful development of an oncology PDC. This review aimed to provide useful information for better construction of PDCs in future clinical applications.

Graphical abstract

Introduction

Chemotherapy remains the first-line treatment for various malignant cancers. However, its non-selective toxicity and acquired drug resistance pose significant challenges in clinical applications [1, 2]. To address these limitations, drug conjugates, comprising a tumor-targeting carrier, linker, and cytotoxic payloads, were designed to deliver toxic agents to tumor sites with increased therapeutic efficacy and lower side effects [3, 4]. As the most successful modality of drug conjugates, antibody-drug conjugates (ADCs) have been widely used for cancer treatments, exhibiting both high efficacy and favorable tolerability. Since the first ADC drug, mylotarg (gemtuzumab ozogamicin), was approved by the Food and Drug Administration (FDA) in 2000 for adults with acute myeloid leukemia (AML), ADCs have been attracting growing interest in the drug industry. By October 2024, 15 ADC drugs had been approved for hematological malignancies and solid tumors worldwide [5]. However, the high cost, low cellular penetration, undesired immunotoxicity, and emerging resistance undermine the development of ADCs [6]. The clinical limitation of ADCs has triggered various drug modalities for increased efficacy and lower toxicity. Nanobodies, single-domain antibodies, can deliver therapeutic drugs to target tissues with strong antigen-binding capacity and good tissue penetration, which have been evaluated in different stages of clinical trials for tumor diagnosis and therapeutics [7]. In recent years, the conjugation of nanobodies with functional moieties such as therapeutic drugs, radionuclides, and optical tracers has achieved specific tumor targeting, enhanced therapeutic efficacy, and reduced adverse effects [7,8,9]. Thus, the emergence of nanobodies could be a new drug modality that can overcome the limitation of ADCs.

Similar to ADCs, peptide-drug conjugates (PDCs) are defined as drugs covalently linked to peptides with certain functions via specialized linkers. PDCs can reach their target sites via homing peptides and release the toxic payload via linkers in targeted cells [10]. As a homing peptide, it must have specific targeting capabilities of protein receptors over-expressed at tumor tissues with strong binding affinity. The secondary structure of peptides plays a key role in the in vivo stability and affinity for targeted drug delivery [10]. As a linker, it must be stable in the circulation but efficiently release the toxic drug in target cells based on the specific tumor environment compared to normal cells such as low pH, over-expressed enzymes, et al. [11]. Meanwhile, the conjugation of PDCs should not affect peptide binding affinity and the toxicity of payloads. Understanding how PDCs work could help researchers design more efficient PDCs with good therapeutic efficiency and safety. PDCs are acknowledged to have many advantages over ADCs: (1) ease of synthesis by solid phase synthesis, liquid phase synthesis, or biosynthesis; (2) facile functionalization via different chemical modifications, such as cyclization, N-methylation, unnatural amino acid incorporation; (3) high tissue penetration; (4) low immunotoxicity; (5) overcoming drug resistance in tumor cells by altering the cell entry mechanism of small drugs [12, 13]. As of 2020, There were > 100 approved peptides with therapeutic or diagnostic applications, and hundreds of peptides are under clinical trials for the treatment of cancer, infection, diabetes, pain treatment, et al. [10]. Witnessing the success of peptides, PDCs, are now attracting increasing interest in drug discovery. However, PDCs have progressed slowly in clinical application since they were first reported in 1972 by Freer et al. [14]. Only several radioactive PDCs are approved for cancer diagnosis and therapy, such as ⁶⁸Ga-DOTATOC, ⁶⁸Ga-DOTATATE, ¹⁷⁷Lu-DOTATATE [15]. What possible barriers and challenges do PDCs meet when applying in the clinic? This review will first highlight the general consideration of successful PDC design learning from ADCs. Then, the basic function and choices of each component of a PDC construct will be summarized. The status of PDC clinical development will be introduced in the following section. Lastly, the future perspectives for the successful development of an oncology PDC will be discussed and concluded. The purpose of the review is to provide guidance for well construction of peptide-drug conjugates for clinical application.

General consideration of PDCs learning from ADCs

With the rapid progress of ADC candidates in clinical applications, major challenges such as undesired toxicity and drug resistance have also drawn much attention. The experiences of ADC construction are worth learning to circumvent the pitfalls encountered with ADCs. Before introducing the general consideration of PDCs from ADCs, we compared the properties of ADCs and PDCs, which could help better understand their differences, as shown in Table 1 [16]. As ADCs met a range of limitations, in the following section we will discuss some pitfalls of traditional ADC therapeutics and highlight ways that PDCs can overcome these challenges, concluded in Fig. 1 and the following introductions of PDCs.

Table 1 The comparison of PDC and ADCs

Off-site toxicity

A previous study showed that out of 79 ADC development programs terminated since 2000, 32 were discontinued due to safety concerns [20]. Therefore, effectively managing toxicity and enhancing safety are critical factors for developing ADCs successfully. The off-site toxicity of ADCs can be categorized into “on-target toxicity” and “off-target toxicity.” On-target toxicity occurs when ADC-mediated targeted uptake affects normal tissues expressing the same target. Off-target toxicity is caused by the pre-release in the circulation and non-specific uptake of payloads. The key to solving this problem of “On-target toxicity” is exploring novel targets expressed exclusively or predominantly in tumor cells but rarely or low in normal tissues. The optimization of targeting units to the selected targets is also crucial for a successful drug conjugate. Peptide ligands come from natural plants, animals, chemical synthesis, or de novo discovery approach. The rapid progress of peptide screening technologies accelerated the discovery of specific peptide ligands with high receptor affinity, including phase display [21], one-bead-one-compound (OBOC) method [22], mRNA display [23], split-intein circular ligation of peptides and proteins (SICLOPPS) et al. [24]. Compared to mAbs, peptides had lower molecular weight to overcome the drawback of low penetration along with rapid clearance from circulation. Peptide stabilization strategies have been widely developed to improve their physiological stability and pharmacokinetics, such as peptide stapling [25] and non-basic amino acid substitution [26, 27].

Thus, with this advanced technology in hand, the rapid identification of drug targets will greatly boost the development and clinical translation of drug conjugates with high efficiency and low toxicity.

Fig. 1

The challenges of ADCs and the chances for PDC in future drug development

The limited choice of payloads

In addition to on-target toxicity, bystander effects arise from non-target-mediated cellular uptake and catabolism in normal cells, leading to the release of the cytotoxic payload and/or its metabolites. This off-target toxicity is particularly evident in payload-associated platform toxicities, where ADCs utilizing the same linker-payload combination exhibit similar safety profiles, regardless of the target antigen [28]. Recent studies revealed that only 1–2% of ADC payloads reach the intracellular target. Thus, the potency of the payload must be high enough to eradicate tumor cells even at a low concentration [29]. However, the highly potent payloads might lead to severe side effects, especially a significant level of antigen expression on critical normal cells. Compared to ADCs, PDCs had more payload choices. Besides radionuclides and chemo-agents, toxic proteins, cytokines, PROTACs, and oligonucleotides could also be designed for the next generation of therapeutic PDC [30].

Drug resistance

The mechanisms of drug resistance are complicated and associated with enhanced efflux of drugs, elevated metabolism of xenobiotics, signaling pathway compensation, target change, cell death inhibition, et al. [31,32,33]. As for ADCs, efforts to overcome the resistance included: (1) choice of the payloads that are poor efflux substrates, such as trastuzumab deruxtecan, using a novel DNA topoisomerase I inhibitor, can overcome T-DM1 resistance caused by aberrant expression of ATP-binding cassette (ABC) transporters in HER2-positive gastric cancer [34]; (2) increasing the hydrophilicity via linker modification as MDR1 transports hydrophobic compounds more efficiently than hydrophilic compounds [35]; (3) developing of bispecific/ biparatopic ADCs [36]. These strategies could also be used for PDC construction. Moreover, cell penetration peptides could change the cellular uptake mechanism of chemo-agents, which could be an alternative method to overcome drug resistance. Meanwhile, due to the facile modification of peptides, two different drugs could be conjugated to peptides for combination therapy and avoiding drug resistance.

To sum up, the rapid development of ADCs also meets some limitations or challenges for their wide clinical application. The progress in PDCs is still in the early stages, but they have promising potential as they have unique advantages for targeted drug delivery. A better understanding of the design of ADCs and the components of PDCs could be beneficial for future PDCs’ construction.

Key components of PDCs

PDCs are an emerging class of prodrugs formed through the covalent attachment of a specific peptide sequence to a drug via a cleavable linker. The diversity and facile construction of peptides allow for incorporating a great degree of functionality into PDCs, such as enhancing tumor targeting of chem-agents, circumventing multiple drug resistance, inducing self-assembling for drug delivery, and so on [37]. Besides peptide diversity, linker structure also plays a key role in manipulating the biological action of PDCs in different manners, for example, the cleavable linkers can be triggered by tumor-specific microenvironments such as pH, tripeptide glutathione (GSH) or enzymes to reach a tumor-targeted therapy, while the non-cleavable linkers can be stable in the circulation avoiding the release of payloads in blood and reduction of anti-cancer effect [38]. The payload selection is also important for PDC construction and application. In this part, we will systematically introduce the role of each component and how alternative approaches can improve the current limitations of this emerging modality.

Peptides in PDCs

Similar to antibodies, peptides are composed of amino acids linked together by peptide bonds. However, peptides have lower molecular weights and generally lack the complex tertiary structures found in antibodies. Given their ubiquitous presence in biological systems and their involvement in regulating various physiological processes, peptides provide an alternative approach to develop PDCs. The peptides significantly impact efficacy, pharmacokinetic/pharmacodynamic profile, and therapeutic index of PDCs as their influence on tumor targeting, in vivo stability, and cell penetration properties. Unlike mAbs, peptides can act as homing peptides for tumor targeting and can also be used as cell penetration enhancers to transport small drugs into cells. According to the cellular function, peptides in PDCs can be divided into tumor-targeting peptides, brain penetration peptides, cell penetration peptides, and self-assembling peptides (Fig. 2), which will be discussed in this part [10, 13, 39, 40].

Fig. 2

Multifunctional peptides used in PDCs, including cell penetration peptides, tumor targeting peptides, and self-assembling peptides

Tumor-targeting peptides

Tumor-targeting peptides have become an increasingly useful tool for targeted delivery of therapeutic and diagnostic agents into tumors. The ideal peptide for PDCs should have a strong target binding affinity, high selectivity to tumor cells, and in vivo stability. Herein, we focused on several potential tumor-targeting peptides used in PDCs and discussed their behaviors in delivering chemo-agents to tumor tissues, which helped distinguish the proper homing peptide for clinical application. To make this part comprehensive, we categorize the targets according to their function, including targeting immune checkpoint, targeting tumor vasculature, targeting hormone-associated receptors, targeting the tumor microenvironment, targeting the blood-brain barrier (BBB), and other targets (Table 2).

Table 2 Classification of tumor-targeting peptide in PDCs

Anti-angiogenic targets

Integrins Ανβ3

The tripeptide motif arginine-glycine-aspartic acid (RGD), identified in the 1980s within fibronectin, could specifically bind to integrins ανβ3 and ανβ5 and was frequently used as a peptide carrier for targeted cancer therapy [41]. The cyclic RGD peptide cyclo (RGDfK), which has good stability and target affinity, has gained considerable attention for different kinds of drug conjugation. Several radionuclide-RGD conjugates have entered clinical studies as positron emission tomography (PET) tracers including [18F] fluciclatide [41], [18F]RGD-K5 [42], ⁶⁸Ga-NOTA-bombesin(BBN)-RGD et al. [43]. However, despite the advantages of the RGD motif, cilengitide was discontinued in 2014, and three Phase II clinical trials of [18F] fluciclatide have been withdrawn. In 2023, ⁶⁸Ga-FAPI-RGD with the conjugation of fibroblast activation protein (FAP) inhibitors (FAPIs) and RGD have been evaluated in patients with FAP- and integrin αvβ3-positive tumors. The clinical results demonstrated an improved lesion detection rate and tumor delineation, particularly for the diagnosis of lymph node (99% vs. 91%) and bone (100% vs. 80%) metastases compared to ¹⁸F-FDG PET/CT (NCT37142301) [44]. Studies also revealed that RGD peptide can efficiently deliver chemo-agents to the tumor tissues with enhanced tumor accumulation and anti-cancer effect, including doxorubicin (DOX) [45], camptothecin [46], paclitaxel (PTX) [47], platinum (Pt) [48]. Despite the advantages of the RGD motif, cell biology research revealed that the combination of RGD with RGD-positive cells will induce cell deformation, which is conducive to the attachment, spreading, and migration of RGD-positive cells. Thus, when RGD is used as an integrin-specific binding ligand, it is necessary to fully consider the possibility of inducing the spread of tumor cells.

Mammalian aminopeptidase N (APN) (or CD13)

Asn-Gly-Arg (NGR) motifs have vasculature-homing properties via interactions with the aminopeptidase N (CD13) expressed on tumor neovasculature [49]. Many NGR-based conjugates have been reported for cancer therapy or imagine, such as ^99mTc-labeled NGR-chlorambucil conjugate [49], 99m Tc-HYNIC-CLB-c(NGR) [49], cyclic NGR peptide-daunomycin conjugates [50], RGD-NGR-NSAID(Non-steroidal anti-inflammatory drug) conjugates [51]. Most NGR-drug conjugates showed limited selectivity to CD13-positive cancer cells. Currently, only two fusion proteins conjugated with homing peptides are in clinical trials: NGR-human tumor necrosis factor (hTNF) and truncated tissue factor (tTF)-NGR [52]. NGR-hTNF was awarded “orphan drug status” for malignant pleural mesothelioma and hepatocellular carcinoma in the EU and the USA and is now in a Phase I clinical trial for cancer patients with solid tumors or lymphomas. The (tTF)-NGR targeting tissue factor to tumor vasculature can induce tumor infarction, which was reported to inhibit solid tumor growth in mice. Now, tTF-NGR is in a Phase I clinical trial for cancer patients with solid tumors or lymphomas [53].

CD133

CD133 is a transmembrane glycoprotein over-expressed in various solid tumors, which is not only the most prominent cancer stem cell (CSC) marker but also contributes to neovascularization in vivo [85, 86]. Sun et al. discovered a CD133 targeting linear peptide LS-7 (LQNAPRS) using phage display technology with good target affinity and selectivity [87]. Based on this study, Tao et al. designed a novel PDC by conjugating camptothecin (CPT) with peptide LS-7 via a succinyl linker [54]. This PDC could overcome the limitation of CPT with poor solubility and high systemic toxicity, which demonstrated improved biodistribution, good therapeutic activity, and a better safety profile in vivo. This discovery provides a new approach for the delivery of toxic CPT to targeted tumor tissues for the therapy of cancer stem-like cells.

VEGFR

Vascular endothelial growth factor (VEGF) is a well-characterized angiogenic factor that plays an important role in tumor-associated angiogenesis, tumor recurrence, and metastasis by binding to their receptors VEGFRs. Wang et al. discovered a VEGFR3 targeting peptide CP7 (CIQPFYP) by computational design, which exhibited high VEGFR3 targeting affinity and tumor-homing efficiency [88]. Then, they conjugated this homing peptide to PEG-b-PLL polymers to enhance the tumor accumulation and chemotherapeutic efficacy of DOX [89]. As linear peptide ligands could readily degrade, Ying et al. discovered a cyclic A7R peptide(cA7R) by computer-aided peptide design and head-to-tail cyclization [90]. This cyclic peptide ligand could enhance the delivery of doxorubicin (DOX) loaded liposomes to tumor tissues. Besides conjugated nanoparticles, VEGFR targeting PDCs were also discovered by us through the conjugation of VEGFR targeting peptide VEGF_{125 − 136} (QKRKRKKSRYKS) to a lytic peptide. This PDC could both suppress tumor angiogenesis and tumor proliferation in a TACE model for hepatocellular carcinoma therapy [55].

Prostate-specific membrane antigen (PSMA)

Prostate-specific membrane antigen (PSMA) is a protease over-expressed in prostate cancer cells and the endothelium of tumor vasculature but much less so in normal tissues [56]. PSMA processes PSMA-specific peptide substrate (Asp-γ-Glu-γ-Glu-γ-GluGluOH) is processed by PSMA to provide an analog that is now cell-permeable, cytotoxic, and extracellularly concentrated adjacent to cancerous prostate cells. G202, consisting of a PMSA-targeting peptide substrate and cytotoxic thapsigargin, was hypothesized to be effectively and efficiently cleaved by the extracellular enzymatic activity of PSMA, thereby delivering cytotoxic agents specifically to a tumor site. Nowadays, G202 has been entered into several Phase 2 clinical trials, including prostate cancer, renal cell carcinoma, hepatocellular carcinoma, and glioblastoma, but clinical results have not yet been reported [57].

Immune checkpoint targets

PD-L1

Immune checkpoint blockade (ICB) has provided significant clinical advances in various cancer types via monoclonal antibodies against programmed cell death protein 1 (PD-1), programmed death-ligand 1 (PD-L1), or cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) [91]. PD-L1 is overexpressed in cancer cells and plays a major role in suppressing T-cell activation. As the potential ability of PD-L1 on tumor progression, many PD-1/PD-L1 peptide antagonists had been rapidly developed and used as warheads for targeted drug delivery [92,93,94]. PDCs targeting PD-L1 are also designed and applied for cancer immunotherapy. For instance, Kim et al. developed peptide-oligonucleotide conjugate (Pep-21) consisting of a PD-L1-binding peptide (NYSKPTDRQYHF) covalently linked with an anti-miR-21 inhibitor via click chemistry, which was preferentially internalized in both cancer cells and tumor-associated macrophages (TAMs) [58]. This conjugate exhibited reduced tumor cell migration and restrained tumor progression. A shorter PD-L1 peptide(CVRARTR) used for PDC construction also showed a potent anti-cancer effect. This conjugate consists of an anti-PD-L1 peptide, cathepsin B-specific cleavable peptide (FRRG), and doxorubicin (DOX) and can be self-assembled into nanoparticles via intermolecular interactions [59]. This drug delivery system can achieve targeted tumor delivery of anti-PD-L1 peptide and DOX and efficiently inhibit tumor progression with minimal side effects. Zeng et al. designed a new PD-L1 degrader by conjugation of PMT degraders with an anti-PD-L1 peptide featuring disulfide linkers This conjugate achieved synergistic immunotherapeutic effect via both external PD-L1 blockade and internal PD-L1 degradation [93].

CD47

CD47 is over-expressed in many cancer types and induces immune evasion by sending out signals like “don’t eat me” via the CD47-SIRP α-axis [95]. CD47 has been an attractive target for cancer immunotherapy. In recent years, CD47 targeting peptides have been developed to inhibit CD47-SIRP α-axis, such as RS17 (RRYKQDGGWSHWSPWSS) [60]. As for PDC design, Pan et al. constructed a CD47 by conjugation of CD47-blocking peptide VK17 with pro-apoptotic VK30 peptide. This PDC can induce lung cancer cell apoptosis and macrophage phagocytosis and suppress lung cancer cell growth in mice. Another CD47 targeting peptide (CkRFYVVMWKk, k = D analog of lysine) was discovered to deliver gemcitabine to bladder cancer (BC) cells over-expressed with CD47 with high tumor accumulation [61]. Thus, the inhibition of the CD47-SIRPαinteraction by peptides or PDCs is a potential approach to cancer immunotherapy.

Hormone associated targets

SST2 receptor

Peptide hormone-based targeted therapy has been studied extensively. As a somatostatin receptor, the SST2 receptor subtype is highly expressed in various tumor cells and primary tumor tissues and has become an attractive target for SST receptor-mediated anti-cancer therapy [62]. Octreotide is a somatostatin analog with a high affinity for the SST2 receptor [62]. Octreotide-derived delivery strategies have been proposed and explored for tumor-targeted therapeutics or diagnostics in preclinical or clinical settings. Octreotide-based radioactive conjugates have been applied clinically to detect small neuroendocrine tumor sites, including ^111In-DTPA-octreotide with high renal clearance and less intestinal accumulation [63], ^99mTc-HYNIC/EDDA-3Tyr-octreotide for the diagnosis of neuroendocrine tumors [64], ^68Ga-DOTATOC, and ^68Ga-DOTATATE [65] et al. for gastroenteropancreatic neuroendocrine tumors. Meanwhile, Octreotide-small drug conjugates have also been designed, such as doxorubicin [66], paclitaxel [67], and camptothecin [68].

GnRH receptor

GnRH receptor, or LH-RH (Luteinizing hormone-releasing hormone) receptor, is expressed in many endocrine cancers, including breast, ovarian, endometrial, and prostate tumors [69]. Zoptarelin Doxorubicin (AN-152, AEZS-108, ZoptrexTM) is a peptide-drug conjugate composed of a GnRH analog ([D-Lys(6)]-GnRH) and doxorubicin The conjugate proved more effective than doxorubicin in inhibiting cell proliferation in GnRH receptor-positive cancer cell lines and several xenograft mouse tumor models. Meanwhile, this PDC successfully entered several clinical trials, which we discussed in the below part [69,70,71,72]. Besides chemo agents, an antiangiogenic compound (sunitinib) can be conjugated to the targeting peptide [D-Lys(6)]-GnRH, generating SAN1GSC, which had significant tumor growth delay compared with equimolar sunitinib alone in castration-resistant prostate cancer animal models [73].

Tumor microenvironment targets

Matrix metalloproteinase (MT1-MMP)

Overexpression of membrane type-1 matrix metalloproteinase (MT1-MMP) is closely associated with malignancies such as breast, lung, ovarian, and colon cancers [74]. BT1718 is a novel bicyclic peptide-drug conjugate targeting MT1-MMP to release its toxic payload DM1 [74, 75]. The bicyclic peptide binding to MT1-MMP is identified by phage display technology as the tumor-target peptide. The bicyclic peptide also has other inherent characteristics to minimize drug toxicity, such as short systemic half-life, rapid tumor penetration, low systemic exposure, and rapid renal clearance.

Blood-brain barrier targets

Low-density lipoprotein receptor-related protein 1 (LRP1)

LRP1 has been reported to be expressed on the blood-brain barrier (BBB) and to be responsible for the transportation of multiple ligands across the BBB [76]. ANG1005, the most typical PDCs targeting LRP1, has been reported to deliver paclitaxel to cross the blood-brain barrier for the treatment of metastatic breast cancer in clinical trials, which showed notable CNS and systemic treatment effects in all treated patients [77]. To treat primary and secondary brain cancers, two PDCs named ANG1007 and ANG1009, consisting of three doxorubicin molecules conjugated to Angiopep-2, exhibited dramatically higher BBB influx rate constants than unconjugated doxorubicin, demonstrating their potential on the treatment of brain cancers [78].

Other targets

Hepatocyte receptor A2 (EphA2)

Erythropoietin-producing hepatocellular receptor A2 (EphA2) is a receptor tyrosine kinase family member that regulates cell-cell interactions such as differentiation, adhesion, migration, and death [79, 80]. EphA2 is overexpressed in a variety of solid tumors and is related to poor prognosis in cancer patients. BT5528 is a bicyclic peptide drug conjugate, which comprises a bicyclic peptide as a highly active EphA2 ligand and toxic auristatin E. BT5528 can accumulate in tumor tissues with a very low plasma concentration, which increases the selectivity for killing tumor cells with low systematic toxicity [80]. YSA (YSAYPDSVPMMS) is another EphA2-targeting peptide identified by phage display that effectively delivers anti-cancer agents to cancer tumors. For instance, YSA-paclitaxel conjugate can target EphA2 over-expressing prostate cancer cells in vivo, inhibiting tumor growth in a prostate cancer xenograft model more effectively than paclitaxel alone [81].

In addition to the introduced targets, several other peptide receptors are also investigated as potential targets for anti-cancer drug delivery. This target included epidermal growth factor receptor (EGFR) [82,83,84], human Y1 receptor (hY1R) [96] et al., and these receptors were overexpressed in specific tumor cell surfaces, and the detailed information was concluded in Table 2.

Cell penetrating peptides(CPP)

The cell membrane poses a barrier that limits the transport of macromolecules, proteins, nucleic acids, and drugs, restricting their penetration. Therefore, developing drugs capable of traversing cancer cell membranes is critical to achieve intracellular destruction. CPPs, ranging from 5 to 30 amino acid residues, are rich in arginine and lysine, which carry positive charges and exhibit amphiphilic properties in a physiological environment [97]. These features make CPPs effective for delivering otherwise cell-impermeable compounds or drugs into cells [98]. The widely recognized mechanisms by which CPP penetrates cell membranes can be divided into two categories: endocytosis and direct translocation, and the penetration mechanisms could be influenced by the chemical structure of peptides [99, 100]. CPPs in PDCs can be used to (1) improve the solubility and toxicity of small drugs, (2) change the drug cell penetration mechanism to overcome drug resistance, and (3) reduce the non-specific toxicity of small drugs.

Improving anti-cancer effects

Chen et al. constructed three PDCs (NTD, d-NTD, and q-NTD) in which TAT peptide is covalently attached to one, two, or four doxorubicin. It was shown that q-NTD had the most potent anti-cancer effect with the slowest drug release efficiency, indicating that the efficiency of intracellular accumulation and the rate of free drug release are essential factors in predicting the efficacy of PDCs in vitro [101]. Additionally, CPPs have been designed to target lymphatic metastases via intravenous injection. For instance, nanoparticles modified with CPP have demonstrated a 63.3% inhibition rate of lymphatic metastasis in lung cancer and reduced tumor growth rates by 1.4-fold [102].

Overcoming drug resistance

Well-designed PDCs can also greatly assist in overcoming the emergence of multidrug resistance (MDR) in tumors. For example, one particular mechanism of drug resistance is the increased expression of membrane proteins that mediate drug efflux, such as P-glycoprotein (Pgp), while guanidinium-rich PDCs can help circumvent these efflux proteins as their internalization does not involve passive diffusion across the membrane [103, 104]. For example, Wender et al. conjugated paclitaxel (PTX) to an octaarginine peptide transporter with significantly improved activity against malignant cells that were resistant to the PTX alone [105].

Reducing non-specific toxicity

An activatable PDC with cell-penetrating peptide and DOX conjugation (DOX-CR8G3PK6-DMA) was designed to achieve tumor-targeted drug delivery and reduce the non-specific toxicity of small drugs [106]. At the tumor extracellular pH of 6.8, hydrolysis of DMA occurred, and the cell-penetrating function of R8 was recovered to enhance cellular uptake of DOX with decreased systematic toxicity. Besides reducing the non-specific toxicity of chemo-agents, we recently designed cyclic PDC conjugated with pan-HDAC inhibitor SAHA, shown in Fig. 3, which is the cyclic peptide derived from HDAC1-specific substrate H3K56 [107]. This PDC showed selective toxicity to HDAC1-sensitive cell lines but poor toxicity to normal cancer cells. Then, we also developed sulfonium-tethered lytic peptides to conjugate with SAHA, exhibiting a selective anti-cancer effect [108]. Despite the versatility of CPPs used for PDCs and the fact that these conjugates would be a valuable tool for the treatment of cancers, their highly charged nature poses problems for general application and translation into the clinic.

Fig. 3

Stabilized peptide drug conjugate targeting HDACs for cancer therapy. (A) The design of PDCs targeting HDAC. H3K56 peptide is derived from Histone 3, which is the substrate of HDAC1. The functional group hydroximic acid plays a key role for HDAC binding [107]. (B) Sulfonium-tethered PDCs targeting HDAC with decreased non-specific toxicity [108]. Copyright permission obtained with ref. no.83–84 from the author

Self-assembling peptide

The hydrophilic peptides combined with small hydrophobic drugs can observe overall amphiphilicity, introducing a tendency for self-aggregation in aqueous environments. The self-assembling PDCs can induce the formation of hydrogels or supramolecular nanostructures and are now receiving rapidly growing interest in the drug delivery community.

Controlling drug release

One important application of these self-assemblies is the sustaining drug release. The Stupp group has developed a range of peptide amphiphile (PA)-based hydrogels for biomedical applications [109, 110]. In one example, they conjugated the anti-inflammatory drug nabumetone to peptide amphiphiles through hydrazone bonds, and this conjugate could form a gel that release nabumetone in the low pH environment of the tumor [111]. The self-assembling PDCs can be used for the tumor-selective drug release activated by the tumor environment. For instance, the Xu group used the enzyme-instructed self-assembly strategy (EISA) to form nanofiber networks via a phosphorylated naphthalene-Phe-Phe-Lys-Tyr precursor conjugated with taxol [112]. Upon treatment with alkaline phosphatase (ALP) over-expressed in the tumor environment, the phosphate group was cleaved, and a translucent hydrogel spontaneously formed and exhibited comparable anti-cancer abilities with free taxol. To further improve the bio-stability, the Xu group developed D-amino acids-based hydrogels and could sustainably release free taxol for up to 1 month, leading to significant suppression of tumor growth [113].

Increasing drug loading and efficacy

Well-designed PDCs could also form nanostructures to provide targeted drug delivery and release. Cui et al. designed a novel class of PDCs that can form nanostructures by covalently conjugating one or more anti-cancer drugs to a rationally designed peptide through a biodegradable linker with a high drug loading and stability [114, 115]. Meanwhile, the positively charged peptide drug conjugates can also be self-assembled with negatively charged nucleic acid such as DNA or siRNA. We previously developed the HDAC-targeted PDC by the conjugation with a positive CPP and HDACi, which was then induced to be self-assembly by the interaction with anti-cancer DNA aptamer AS1411 [116]. This nanoparticle showed a synergetic anti-cancer effect through multiple signal pathways with low non-specific toxicity (Fig. 4).

Fig. 4

An assembly-inducing PDC deliver nucleic acid for cancer therapy. This PDC consists of the cell penetration peptide RW9, an HDAC inhibitor warhead (peptide C-terminus), and 5-FU (peptide N-terminus), which can coassemble with AS1411 to form nanospheres. This PDC showed unexpected synergy with AS1411 to augment the cancer cell suppression efficiency, exemplified by the downregulation of the stemness-related proteins and the upregulation of apoptosis-related proteins. Copyright permission obtained with ref. no 92 from The Royal Society of Chemistry

Improving drug bioavailability

Peptide polymers are biodegradable polymers that can be conveniently synthesized and widely used for conjugated drug delivery [117]. The polypeptides–drug conjugates for anti-cancer therapy had been systematically introduced by Chen et al. [117]. For instance, polypeptides-chemotherapeutic drug conjugates have been developed to improve the bioavailability, enhance the therapeutic effects, and reduce the systemic toxicity of chemotherapeutic drugs. Kataoka et al. first conjugated DOX to the side chains of PEG-b-PAsp by an acid-labile hydrazone (Hyd) bond that can be cleaved in the endolysosomes with an acidic pH of 5–6, which presented intracellular acidity-triggered DOX release behaviors, tumor-infiltrating permeability, and effective anti-cancer activity with extremely low toxicity [118].

Despite the self-assembling PDCs showing good potential for targeted drug delivery, they also meet many challenges before they enter the clinic, such as possible structural instability during self-assembly, unpredictability of drug release, and possible non-specific toxicity to normal cells. The key step to solving these problems is understanding exactly how observed structures are formed. To study the formation process, Cai et al. carried out a Brownian dynamics (BD) simulation on a model rod-coil block copolymer/rigid homopolymer binary system, which helped control the ordered structures of the self-assembly [119, 120]. To address the challenge of drug release behaviour of the self-assembly, Buxton and co-workers developed a new simulation method to simulate the drug release performance of nanoparticles and observed an increased drug release rate near tumors [121]. These simulation works provide molecular-level insight into drug delivery behaviors, which greatly benefits the design of desirable drug delivery systems. In recent years, Matthew et al. explored the specific contributions of the drug and its linker to PDC self-assembly, which provided design principles for the enhanced control of PDC nanomaterials as therapeutic drug carriers [122]. To sum up, with a better understanding of the mechanisms of the formation process and the contribution of peptides, linkers, and drug molecules in self-assembly, there will be good drug candidates suitable for clinical applications.

Linkers

In PDCs, the linker connects the peptide to the drug, playing a crucial role in preventing off-target or peripheral drug release and enhancing the efficacy of the PDC. A well-designed linker should reduce the drug’s activity during delivery to minimize off-target toxicity and release the drug fully active upon reaching the target cells. Also, the bond between the peptide and the linker should not influence the affinity of the peptide for its receptor. Based on their behaviors after cellular uptake or under in vivo conditions, these linkers can be broadly divided into four groups: enzyme-cleavable, acid-cleavable, reducible disulfide, and non-cleavable (Fig. 5 and Table 3). Kaur et al. gave a detailed review of all aspects of linkers used in the design of PDCs [11] (Fig. 6).

Fig. 5

Different linkers used in PDCs. The linkers used in PDCs included cleavable linkers and non-cleavable linkers

Table 3 Classification of different linkers used in PDC

Fig. 6

The design of VEGFR targeting peptide conjugate (QR-KLU) and the antineoplastic efficacy of peptide QR-KLU in vitro and in vivo. Peptide QR-KLU is a conjugation of VEGFR targeting peptide VEGF_{125 − 136} (QKRKRKKSRYKS) and a reported lytic peptide(KLUKLUKKLUKLUK). This PDC showed good in vivo anti-tumor effect than traditional drug DOX through TACE in VX2 rabbit tumor model, and efficiently inhibited angiogenesis in tumor tissues with good safety. Copyright permission obtained with ref. no [55]. from Springer nature

Enzyme cleavable linker

Enzyme cleavable linkers are popular choices for both ADCs and PDCs, as they can be selectively cleaved by enzymes over-expressed in the tumor microenvironment or the lysosomes, such as matrix metalloproteinases(MMP), cathepsin B [38, 123]. Studies revealed that some of the enzymes are usually inactive outside the cell due to environmental pH, which is beneficial for their stability in the circulation. MMP2 and MMP9, members of MMPs, are essential in metastatic tumor cells, facilitating tumor development and angiogenesis. You et al. developed a unique human epidermal growth factor receptor 2 (HER2) targeting PDC via an MMP-2 sensitive linker (PVGLIG) to conjugate the trastuzumab-derived peptide with DOX [123]. This PDC not only had a more potent in vivo anti-cancer effect but also had a longer half-life (17.6 h versus 7.7 h) and slower clearance compared to free DOX. To further facilitate the release of active drugs from the PDC, specific dipeptide or tripeptide sequences have been employed. The dipeptide (Val-Cit or VC) can be specifically cleavage by cathepsin B and used for cRGD and DOX conjugation, which displayed superior cellular uptake, cytotoxicity toward integrin αvβ3 overexpressing B16 cells in vivo compared to another PDC containing disulfide linker [38]. Another tripeptide sequence is the Ala-Ala-Asn (AAN) cleaved by enzyme legumain overexpressed in tumor cells and microenvironment [124]. This tripeptide was widely used for the design of prodrugs rather than PDCs, which could significantly reduce the non-specific toxicity of these prodrugs. However, the instability of such short peptide linkers could lead to the release of drugs in pre-clinical in vivo studies.

Other enzymes rich in intracellular compartments, like endosomes and lysosomes of cancer cells, are esterases and amidases [125,126,127]. Karampelas et al. designed a PDC utilizing an ester bond and an amide bond to attach gemcitabine and peptide ligand targeting gonadotropin-releasing hormone-receptor (GnRH-R) [128]. As expected, this PDC showed a significant anti-cancer effect on a GnRH-R-positive prostate cancer xenograft animal model compared to gemcitabine. However, pharmacokinetic studies revealed that this PDC administration in mice resulted in the presence of high levels of gemcitabine, suggesting the rapid cleavage of the conjugate at the ester site, releasing the free drug from the peptide-linker moiety.

A carbamate linker is a stable linker used for PDC construction. Sayyad et al. explored the use of a carbamate linker in the design of PDC and discovered that the carbamate linker was more stable in vivo and released the drug slowly compared to the PDCs with an ester linkage, resulting in less toxicity to cancer cells [129]. However, it is unclear which linker chemistry will lead to better efficacy for tumor reduction. Based on many studies, the order of in vivo linker stability from greatest to least is as follows: amide > carbamate > ester > carbonate.

Recently, Kim et al. constructed an albumin-binding caspase-3 cleavable peptide-drug conjugate by conjugating doxorubicin to KGDEVD peptide on an albumin-binding moiety [130]. The increased albumin metabolism could induce cancer cell apoptosis, and the upregulation of caspase-3 could cleave the DEVE peptide and release DOX to exert a potent anti-cancer effect. This discovery provides a promising method for the treatment of pan-KRAC mutant cancer cells. Later, they designed another albumin-binding peptide-docetaxel conjugate platform (MPD3) with this caspase-3 cleavable liner [131]. This PDC showed good tumoral accumulation of cytotoxic payloads and robust anti-tumor activities in both local and liver metastatic PDAC tumor models in mice. These results proved the efficacy of caspase-3 cleavable linker in PDC construction.

Acid cleavable linkers

The acid-labile bonds usually maintain stability in blood circulation (pH7.4) but get cleaved primarily in the acidic tumor microenvironment (pH 6.5 − 6.9) or the acidic cellular compartments, endosomes (pH 5.5 − 6.2) and lysosomes (pH 4.5-5.0), including acetals, imines, hydrazine and various metal-organic frameworks (MOFs) [132,133,134,135]. Hydrazone is a popular acid-sensitive linker utilized for controlled drug release, and it has been successfully used in an FDA-approved ADC gemtuzumb ozogamicin (Mylotarg) [132], for PDCs, most studies used this linker to conjugate a doxorubicin (DOX) with homing peptides. For example, a hybrid 24-mer peptide was linked with a DOX via the hydrazone linkage, in which the hybrid peptide consisted of an ERK peptide inhibitor and a peptide ligand of transferrin receptor (TfR) over-expressed in the tumor cell [136]. This PDC showed a more obvious in vivo anti-cancer effect and higher distribution in the mouse tumor compared to the free Dox-treated mice. However, another PDC containing triple-negative breast cancer (TNBC) cell targeting peptide and a DOX via hydrazone linker was not stable in media and human serum with a half-life of 6 h but had TNBC selective toxicity compared with normal cells [133]. To improve the in vivo application of these conjugates, Yousef pour et al. synthesized a PDC (20 or ABD-Dox) using aldoxorubicin and an albumin-binding peptide [134]. This peptide ligand strongly interacts with albumin by noncovalent binding, which would facilitate easier dissociation of the peptide for release into the tumor. This PDC had a longer plasma elimination with a half-life of 29.4 ± 0.8 h compared with free DOX, which had a half-life of minutes. Meanwhile, the Dox in the tumor was found to be ≈ 120-fold greater for the PDC-treated mice compared to the free Dox group. However, considering the in vivo stability and prerelease of free drugs, the acid cleavable linkers are not widely explored as enzyme-cleavable linkers.

Succinic acid (SA) is another pH-cleavable linker moiety that has been successfully used for PDC construction [137]. This PDC was established by conjugating a hydrophilic cell-penetrating peptide (CPP = c[RGDKLAK]) with paclitaxel via ester linkage with succinic acid. This PDC demonstrated excellent solubility in water, proper drug releases, and prominent tumor-growth inhibitory effects in vivo. This study provides a new approach to treating glioblastoma.

Reducible disulfide linkers

Glutathione (GSH) is found in a fourfold higher concentration in the tumor environment than in normal cells, which had the advantage of a controlled release of the payload. Liang et al. designed a PDC with a disulfide linker between Dox and cyclic RGD peptide, and studied their drug release properties [38]. They found Dox in PDCs was released slowly over with 39.2% drug released at 48 h, but the release rate was dramatically increased in the presence of DTT (50 mM) to almost completion (∼100%) in 48 h, indicating the selective drug release in reductive environment. BicycleTx Limited (Cambridge, UK) screened and proved that a bicyclic PDC containing EphA2 peptide ligand, a disulfide linker, and cytotoxic mertansine (DM1), had potent in vivo anti-cancer effect, fast renal elimination and low accumulation in liver. Which demonstrates great potential for clinical cancer therapy [80]. However, the disulfide linker could sometimes influence the cell toxicity of the free drug. Li et al. designed a reduction-responsive PDC where a peptide BP9a analogue (CAHLHNRS) was linked to Dox through disulfide linker [138]. This PDC showed cell-selective toxicity towards HepG2 liver cancer cells, which are highly expressing transferrin receptors (TfR), but low potency compared to free Dox. To improve the drug efficacy, the combination with other linkers might be a good direction, like a disulfide linker in combination with hydrazone successfully used in two of the FDA approved ADCs.

Noncleavable linkers

Noncleavable linkers are not activated through external stimuli, which present increased stability in circulation, such as succinimidyl thioether, oxime, and triazole [139,140,141]. The cleavage of PDCs begins at the most labile group, most probably peptide being metabolized, resulting in different forms of drugs, including free drugs, drugs with a portion of the linker, or drugs with linker and amino acid(s). Gabor Mezo et al. designed and screened an oxime-linked daunorubicin (Dau) − GnRH-III conjugate with the incorporation of unnatural and D-amino acids in the homing peptides [142]. These PDCs exhibited high binding affinity for the GnRH receptors and excellent stability in 90% human serum for at least 24 h. This oxime linker was found to be highly stable to chemical or enzymatic degradation, and the conjugate got cleaved at the peptide portion and released the drug in tumor cells. Meanwhile, this PDC had good and selective tumor cell penetration and anti-tumor effect on MCF-7 and HT-29 cells. Randelovic et al. continued this study and revealed that this conjugate had a significant tumor volume reduction as well as metastases inhibition effect with less toxic side effects compared to the free drug in colorectal cancer or breast cancer-bearing mouse models [143]. However, it is not always lucky enough to construct PDCs with good stability, cellular selectivity, and potent efficiency. Feni et al. utilized oxime to attach cytotoxic daunorubicin to a cell penetration peptide, and additional chemistries like triazole and amide were utilized to put an RGD on this PDC [143]. This PDC showed no selectivity toward cells with higher expression of integrin αvβ3 (U87) and much less toxic than the free daunorubicin. This proper peptide ligand or chemistry used for the PDC design greatly influenced the efficacy and cell selectivity.

The non-cleavable triazole formed by the alkyne − azide reaction provides stability to the PDC structure, and other bonds in the linker region likely facilitate the release of the drug from the PDC after cellular uptake. Zheng et al. explored and screened PDCs for cancer photodynamic therapy (PDT) and found the distances between the phthalocyanine and the peptide impacted the photocytotoxicity of these PDCs significantly [141]. They found one conjugate had selective toxicity towards HT29 cells compared to MCF-7. The in vivo study revealed that this PDC had selective accumulation at the tumor site and led to a reduction of tumor growth by approximately 75%. This case mentioned that both the linker stability and length might influence the drug behavior.

Payloads

Despite the potent ability for tumor killing, toxin drugs usually can not satisfy clinical demands, such as poor PK properties and target tissue selectivity. The drug conjugation with peptides allows for specific targeted therapy, consequently resulting in the enhancement of several properties, including the therapeutic window and safer therapeutic dosage. There are several requirements for cytotoxic drugs to use in PDCs, including a clear mechanism of cellular action, a small molecular weight, high cytotoxicity, and retained anti-tumor activity after chemical coupling to the peptides [10]. The advantage of conjugating traditional cytotoxic drugs with peptides lies in the ability to modify their chemical properties, including solubility, selectivity, and half-life, thereby enhancing the efficacy of conventional therapies [144]. Besides chemotherapeutic drugs and radiotracers [145], other different types of drugs can also be designed to conjugate with peptides to improve their therapeutic properties, including small targeted drugs [108], anti-cancer peptides [55], protein drugs [146], nucleic acid drugs [147] and gas molecules [148, 149] et al. These different kinds of payloads enriched the application of peptide-based therapeutics and promoted the possibility of clinical transformation.

The chemotherapeutic drugs are always the first-line drugs for cancer patients, such as DOX, PTX, CPT, and Pt, which are the most commonly used chemo-agents in clinical practices. However, these drugs usually suffer from poor solubility, short circulation half-lives, potential development of drug resistance, and severe side effects, which significantly reduce their therapeutic outcomes and limit their clinical application. PDCs have a great potential to address these problems for targeted therapy, and toxic drugs such as DOX [150, 151], PTX [152,153,154,155], CPT [156], Pt [157] have been conjugated in PDCs for better clinical application.

In addition to chemo-agents, radionucleotides can also be used as therapeutic or imaging agents. Nowadays, many different radionucleotides have been used as payloads in PDCs for cancer diagnosis or treatment, including gallium-68 (68Ga), copper-64 (64Cu) and fluorine-18 (18 F),¹²³I,^99mTc et al. For example, ¹¹¹In-DTPA-Octreotide (Octreoscan) was approved for the diagnosis of neuroendocrine tumors, and ¹⁷⁷Lu-dotatate showed an increased period for progression-free survival for cancer patients and had greater potential as a therapeutic [15].

The gas therapy, including nitric oxide (NO), hydrogen sulfide (H₂S), oxygen (O₂), carbon monoxide (CO), and sulfur dioxide (SO₂), is attracting great attention due to no drug-resistant, little side-effect, and no by-product, which showed great potential application in cancer treatment [10]. The lack of generators for the controlled release of gas limited their further application. Polypeptide-gas conjugates can achieve controlled release of gas and overcome multidrug resistance (MDR) and tumor recurrence [148, 149].

Small-molecule in targeted cancer therapy showed better tumor selectivity compared to chemotherapeutic drugs but also had limitations in the clinical applications, such as drug resistance, side effects, low efficiency, et al. [1]. Peptides can also be used as carriers to deliver this kind of drug to achieve high anti-cancer effects with low non-specific toxicity, such as peptide-HDACi conjugates by our group [107, 108]. In the future, peptide-targeted drug conjugation could be an alternative to overcome the challenges for clinical application.

Anti-cancer peptides can also be used as payloads for PDC construction, such as the most widely used lytic peptide [158] or pro-apoptotic peptide KLA [159, 160]. Like chemo-agents, this peptide can be conjugated with tumor-targeting peptide ligands to reduce the non-specific toxicity or conjugated with cell penetration peptides to increase their anti-cancer effects. We previously developed a VEGFR targeting peptide conjugate (QR-KLU), which showed good in vivo anti-tumor effect than traditional drug DOX through TACE in VX2 rabbit tumor model, and efficiently inhibited angiogenesis in tumor tissues with good safety, shown in Fig. 5 [55]. This work broadened the application of PDCs and may provide an alternative option for clinical HCC therapy via TACE combination.

Biomacromolecules such as protein drugs or nucleic acid drugs are novel and promising drug candidates for various refractoriness diseases. The main challenges of these drugs are poor cell penetration and tumor target ability. For example, NGR-human tumor necrosis factor (TNF) was designed to improve tumor targets and decrease their non-specific toxicity [52]. Small interfering RNA (siRNA) is regarded as a promising tool for cancer therapy because of its wide applicability to various cancer-related genes. Peptide-RNA conjugates could efficiently deliver RNA drugs to cancer cells to induce cell apoptosis [161]. However, due to the poor pharmaceutical properties of both peptide and RNA, this kind of conjugate is not suitable for clinical application. Nowadays, peptide and RNA are both encapsulated into various nanoparticles for targeted cancer therapy [162].

The pharmacokinetics of PDCs

A small molecule is considered “drug-like” and orally bioavailable if it adheres to Lipinski’s Rule of 5 (Ro5), which stipulates specific criteria for molecular weight, hydrogen bond donors, hydrogen bond acceptors, and log P [163, 164]. While peptides often do not meet these criteria, numerous marketed and clinical-stage peptide drugs demonstrate that Ro5 is not a definitive barrier to drug development. Understanding the ADME (Absorption, Distribution, Metabolism, Excretion) properties of peptides is crucial for designing effective PDCs for cancer therapy. The in vivo behavior of peptides significantly influences the pharmacokinetics (PK) of PDCs, ultimately impacting their therapeutic efficacy. However, the complex nature of ADME processes remains a subject of ongoing research. In this section, we will provide a concise overview of peptide administration, biodistribution, metabolism, and clearance based on current scientific knowledge. Our goal is to offer insights into the design of more effective PDC drugs for clinical application.

Absorption

Similar to ADCs, PDCs typically require parenteral administration to achieve optimal bioavailability and therapeutic efficacy. However, injections can be painful and may induce allergic reactions or serious side effects. Oral administration is the most convenient and preferred route for drug delivery. Nevertheless, the poor intestinal absorption of peptides due to enzymatic degradation and limited membrane permeability poses significant challenges for the oral delivery of PDCs.

Strategies developed for the oral delivery of peptide drugs can provide valuable insights for PDC oral delivery. Absorption enhancers, such as surfactants, bile acids, fatty acids, EDTA, and sodium salicylate, have been explored to improve the intestinal absorption of peptide drugs like insulin. The incomplete absorption of peptide drugs is often attributed to a combination of poor membrane permeability and metabolic degradation at the absorption site [165].

One major obstacle to the oral bioavailability of peptides is the extensive hydrolysis by digestive enzymes in the gastrointestinal tract. Protease inhibitors, such as aprotinin, trypsin, bacitracin, puromycin, and bestatin, can reduce peptide degradation at absorption sites, offering a potential strategy to improve peptide drug delivery [165]. Chemical modifications can enhance peptide oral absorption, including fatty acid conjugation [166], PEGylation [167, 168], cyclization, and the incorporation of unnatural amino acids [169,170,171].

CPPs can facilitate cell entry or membrane translocation, promising to increase peptide oral bioavailability [172]. Drug carrier systems, such as nanoparticles and hydrogels, can prolong gastric residence time and increase intimate contact with the intestinal epithelial layer, thereby improving drug absorption and oral bioavailability [173, 174]. However, despite these strategies, significant challenges remain in translating oral peptide drug delivery into clinical applications.

Biodistribution

The biodistribution of PDCs, like peptide drugs, is influenced by factors such as molecular weight, charge, protein binding, structure, and reliance on active transport. Conjugation to different molecules can alter the half-life, selective tissue distribution, or both of PDCs. Chemical modifications, which can enhance oral absorption, may also impact biodistribution.

Parisa et al. investigated the biodistribution and tumor localization of an ABD-Dox conjugate, where ABD is an albumin-binding peptide. This PDC exhibited approximately 4- and 2-fold greater tumor accumulation compared to free Dox and prodrug AlDox, respectively. Conversely, at 2 h post-administration, ABD-Dox accumulated at lower concentrations in the liver, spleen, kidney, and muscle compared to free Dox or AlDox [174]. These findings suggest that ABD-DOX may have reduced cardiotoxicity, hepatotoxicity, and nephrotoxicity, which are common side effects of free DOX. However, the mechanisms underlying PDC biodistribution are complex. Improving the enzymatic and chemical stability of peptides is a promising strategy to enhance targeted distribution.

Metabolism

PDCs, composed of peptides and small molecules, are susceptible to degradation by various proteases, including exopeptidases (such as aminopeptidases, carboxypeptidases, and dipeptidases) and endopeptidases (such as pepsin, trypsin, chymotrypsin, and elastase) [175]. These proteolytic enzymes play crucial roles in blood and cells, regulating peptide signaling by activating or deactivating proteins. The metabolism of small molecule drugs in PDCs is complex, as different molecules undergo diverse metabolic pathways catalyzed by various enzymes, including the cytochrome P450 family. While it is challenging to significantly alter the metabolic stability of small molecules significantly, stabilizing the peptide component can substantially impact the in vivo stability and efficacy of PDCs.

Renal clearance

While peptides often exhibit increased stability compared to small molecules, they can still be subject to rapid renal clearance. The in vivo clearance of peptides and PDCs is influenced by various factors, including molecular weight, charge, hydrophobicity, and enantiomeric purity [176]. It has been observed that peptides with a net negative charge tend to have longer half-lives compared to those with a net positive charge. The anionic nature of the glomerular membrane limits the filtration of negatively charged molecules into the urine [176]. However, negative charges can hinder cellular uptake, potentially affecting peptide function. Another strategy to prolong peptide half-life involves conjugation with larger molecules (> 450 kDa), such as PEGylation. This can increase lipophilicity and binding to plasma proteins like albumin, improving pharmacokinetic and pharmacodynamic properties [177]. For instance, PEGylation of the RGD-targeting HM-3 peptide significantly increased its half-life in rats. When this peptide was conjugated with a PEG linker attached at the N-terminus, the newly modified peptides had a 5.86-fold increase in half-life in male rat studies when compared to the unmodified HM-3 peptide [178]. The successful examples of prolonging peptide half-life in vivo are the remodeling of the insulin or GLP-1 analogs. In 2009, Liraglutide, a near analog of human GLP-1, had a fatty acid chain with a spacer joined to the main peptide backbone for binding to albumin. Liraglutide represented a significant improvement of GLP-1 with an extended IV half-life of 8–10 h and once-daily administration [179].

The recent advances of PDCs in clinical application

PDCs encompass a diverse range of conjugates utilizing various peptide types, including linear peptides with strong target affinity or cell penetration capabilities, cyclic or bicyclic peptides identified through phage display with enhanced stability and affinity, and peptide dendrimers or self-assembling peptides with favorable delivery properties. All three types of PDCs have entered clinical trials, demonstrating significant potential for clinical applications (Table 4).

Table 4 Recent advances in PDC clinical applications

Radiotracer peptide drug conjugates in clinical trails

Several radiolabeled PDCs have been approved for disease diagnosis or therapy. ^99mTc-EDDA/HYNIC-3Tyr-octreotide was approved in Poland in 2013 for the treatment of neuroendocrine tumors of bronchial and thymic origin [180, 181]. Lu-DOTATATE (Lutathera), approved in the USA in 2018, is the first peptide-drug conjugate for the treatment of somatostatin receptor-positive gastroenteropancreatic neuroendocrine tumors via intramuscular injection [182]. ⁶⁸Ga-DOTATOC (approved in the EU and USA in 2016–2019) and ⁶⁸Ga-DOTATATE (approved by the FDA in 2016) are approved for the diagnosis of neuroendocrine tumors. ⁶⁸Ga-DOTATOC, for example, exhibits 80% tumor accumulation within 30 min with a target-to-non-target ratio of 100:1 for central nervous system lesions compared to normal brain tissue [183, 184]. The tumor-targeting peptides used in these approved PDCs are Octreotide (the sequence D-Phe-Cys-Phe-D-Trp- Lys-Thr-Cys-Thr(ol) (fCFwKTCT-ol; 2–7 disulfide bond)) and its derivatives with good affinity to SST2 receptor subtype. Numerous other radiolabeled peptide-drug conjugates are currently undergoing clinical trials but will not be discussed further in this paper.

AEZS-108 (zoptarelin doxorubicin)

AEZS-108 (zoptarelin doxorubicin) is a cytotoxic hybrid molecule comprising doxorubicin covalently linked to a LHRH analog [71]. This design selectively targets doxorubicin to tumor cells expressing LHRH receptors. AEZS-108 has been evaluated in a Phase II trial in men with metastatic castrate-resistant prostate cancer who had progressed after taxane-based chemotherapy [71, 185]. Additionally, it is being tested in a Phase III trial in women with endometrial cancer resistant to platinum/taxane-based chemotherapy [186]. Preliminary data from the Phase II trial in men with metastatic castrate-resistant prostate cancer demonstrated activity in a challenging patient population and a favorable safety profile. However, in a large Phase III trial in advanced endometrial cancer, AEZS-108 failed to improve median overall survival or progression-free survival compared to standard doxorubicin therapy despite promising results in earlier Phase II studies in castration-resistant prostate cancer.

ANG1005 (Angiopep-2)

ANG1005, a novel PDC, consists of three paclitaxel molecules covalently linked to Angiopep-2. This design facilitates crossing the blood-brain and blood-cerebrospinal barriers and penetrating malignant cells via the LRP1 transport system [77]. A first-in-human study evaluated the safety, tolerability, pharmacokinetics, and efficacy of ANG1005 in patients with advanced solid tumors. The pharmacokinetics was dose linear, and the mean terminal-phase elimination half-life was 3.6 h. No evidence of accumulation was observed after repeat dosing [187]. No anti-GRN1005 antibodies were detected. The main dose-limiting toxicity was myelosuppression. Overall, ANG1005 demonstrated good tolerability and promising activity in heavily pretreated patients with advanced solid tumors, including those with brain metastases or prior failed taxane therapy. A subsequent open-label phase II study (NCT02048059) in adults with recurrent breast cancer brain metastases did not meet the primary endpoint of intracranial objective response rate (iORR) per independent radiology facility (IRF) evaluation. However, notable central nervous system (CNS) and systemic treatment effects were observed in all patients, including symptom improvement and prolonged overall survival compared to historical controls, particularly in the subset with leptomeningeal carcinomatosis (n = 28). Other registered clinical trials included ANG1005 in patients with recurrent high-grade glioma in the USA (NCT01967810) and ANG1005 alone or in combination with trastuzumab in breast cancer patients with brain metastases in the USA ( NCT01480583). However, no related results were posed on ClinicalTrials.gov.

CBX-12

CBX-12 is a clinically developed pH-sensitive peptide-exatecan conjugate (peptide sequence: ADDQNPWRAYLDLLFPDTLLLDLLW) that selectively targets cancer cells due to the acidic tumor microenvironment [188]. In vitro, plasma stability assays demonstrated robust linker stability and minimal warhead release in mouse, rat, dog, and human plasma samples at 24 h. CBX-12 effectively delivers high doses of exatecan directly to tumors, resulting in potent tumor cell killing across multiple xenograft models, including HCT116, MDA-MB-231, JIMT-1, and MKN45, with minimal to no bone marrow or life-threatening gastrointestinal toxicity.

The structure of CBX-12 reveals that it adopts the identical amino acid sequence as Variant 3 [189]. In vitro plasma stability tests conducted in mice, rats, dogs, and humans demonstrated excellent stability for CBX-12. No detectable toxin release was observed within 8 h, and only minimal toxin release was detected at 24 h, confirming its stability in blood circulation. Furthermore, in xenograft animal models, CBX-12 once again demonstrated both blood circulation stability and selectivity for tumor cells.

Most notably, compared to standalone Exatecan (DX-8951, Iritotecan), the peptide-conjugated CBX-12 exhibited significantly stronger tumor inhibitory effects across various tumor models [188, 190]. In treatments using unconjugated Exatecan, model animals consistently showed significant weight loss or fluctuations. In contrast, CBX-12 had a minimal impact on body weight, highlighting its tolerability advantage. Moreover, unconjugated Exatecan caused damage to multiple organs in the animals, including significant gastrointestinal (GI) toxicity observable under microscopy. In comparison, animals treated with CBX-12 showed no detectable GI toxicity.

A first-in-human, Phase 1/2 open-label, multicenter, dose-escalation study is currently recruiting subjects with advanced or metastatic refractory solid tumors to evaluate the safety, pharmacokinetics, and biomarker profile of CBX-12 (NCT04902872). We anticipate promising therapeutic efficacy from this clinical trial.

TH1902

TH1902 is a docetaxel-peptide conjugate designed to selectively target cancer cells overexpressing the sortilin (SORT1) receptor [191]. SORT1, also known as neurotensin receptor 3, is a membrane-bound receptor belonging to the VPS10P receptor family [192]. It plays diverse roles in intracellular transport and sorting of various ligands, including neurotensin, granulin precursor, and apolipoprotein E. On the other hand, SORT1 upregulation has been observed in several types of human cancers, such as breast cancer, ovarian cancer, pancreatic cancer, melanoma, and pituitary adenomas [193, 194]. Notably, SORT1 is expressed in 59% of TNBC cases, making it a potential therapeutic target for TNBC. In vitro studies on TNBC-derived MDA-MB-231 cells demonstrated that TH1902 exhibited potent anti-proliferative and anti-migratory activities, inducing faster and more potent apoptotic cell death compared to unconjugated docetaxel [191]. In vivo, studies using MDA-MB-231 and HCC-70 murine xenograft models revealed that both intraperitoneal and intravenous administration of TH1902 led to greater tumor regression than docetaxel without inducing neutropenia. These results highlight the high in vivo efficacy and safety of TH1902. Additionally, studies demonstrated that weekly administration of TH1902 was better tolerated than equivalent doses of docetaxel and effectively inhibited tumor growth in ovarian and endometrial xenograft models. As a single agent, TH1902 exhibited superior in vivo efficacy compared to both unconjugated taxanes and carboplatin against SORT1-positive ovarian and endometrial cancers. Currently, an open-label, first-in-human study of TH1902 in solid tumors is active but not currently recruiting (NCT04706962). In early 2021, the FDA granted TH1902 Fast Track designation for monotherapy in treating sortilin-positive recurrent advanced solid tumors that are resistant to standard therapies.

In TH1902, the peptide molecule specifically targets the SORT1 protein expressed in multiple cancers. A cleavable linker connects the peptide to the toxic payload, docetaxel, enabling targeted therapy. In terms of its mechanism of action, TH1902 applies the “lock-and-key theory” to conjugated drugs by competitively binding to the target SORT1.

Another similarity between PDCs and ADCs lies in their ability to achieve precise payload release, minimizing exposure to normal tissues and enhancing both tolerability and efficacy. In a sense, conjugated drugs function as a type of drug delivery technology, and preclinical data for TH1902 support this characteristic.

TH1902 is constructed with a 2:1 ratio of docetaxel to a peptide molecule connected via a linker (succinic acid). It enters cells through a sortilin-dependent mechanism, effectively bypassing the P-gp efflux pump (MDR1), thereby increasing intracellular concentrations. This results in improved tumor migration inhibition and growth suppression.

EP-100

EP-100 is a novel anti-cancer PDC composed of a natural LHRH ligand linked to a cationic membrane-disrupting peptide [195]. EP-100 interacts with negatively charged tumor cell membranes, inducing cell death through rapid membrane lysis. Preclinical studies have demonstrated the anti-tumor efficacy of EP-100 against various human cancer cell lines that overexpress GnRH-R, both alone and in combination with paclitaxel. A Phase 1 clinical trial (NCT00949559) evaluated EP-100 in patients with diverse solid tumors, including breast, ovarian, endometrial, pancreatic, prostate, colon, and non-Hodgkin lymphoma. The drug was found to be safe and well-tolerated [196]. In 2021, a multicenter, open-label, randomized Phase II trial (NCT01485848) compared paclitaxel plus EP-100 to paclitaxel alone in patients with refractory or recurrent ovarian cancer. While the overall response rate was similar between the two groups, a subset of patients with liver metastases appeared to benefit from the combination therapy. Importantly, the addition of EP-100 did not significantly increase the adverse event profile of paclitaxel and was well-tolerated [195].

G-202

Mipsagargin (G-202) is a novel, targeted prodrug based on thapsigargin, a potent inhibitor of the sarcoplasmic/endoplasmic reticulum calcium ATPase (SERCA) pump. Mipsagargin is designed to be activated by prostate-specific membrane antigen (PSMA)-mediated cleavage of an inert masking peptide, leading to the release of active thapsigargin within tumor cells and tumor vasculature [57]. While thapsigargin itself is poorly water-soluble and lacks a therapeutic index, the peptide conjugation improves its water solubility, enabling systemic delivery and selective release within PSMA-expressing tumors. A Phase I clinical trial (NCT01056029) demonstrated acceptable tolerability and favorable pharmacokinetics in patients with solid tumors [56]. A subsequent Phase II study (NCT01777594) evaluated Mipsagargin in advanced hepatocellular carcinoma patients who had progressed on or were intolerant to sorafenib. The drug was well-tolerated and showed promising disease stabilization in this patient population. However, a larger clinical trial to further characterize Mipsagargin’s activity in advanced HCC was unfortunately canceled due to company reorganization.

Additionally, a Phase II study (NCT02067156) to evaluate the efficacy, safety, and CNS exposure of Mipsagargin in patients with recurrent or progressive glioblastoma was completed in 2017, but results have not been publicly disclosed. Another Phase II study (NCT02381236) was also completed to assess the impact of Mipsagargin on prostate perfusion and volume, but the results remain undisclosed.

Bycycic peptide drug conjugates

Bicyclic toxin peptides (BTPs) are a promising class of PDCs composed of 9–20 amino acids with three cysteine residues that form disulfide bonds, creating a rigid cyclic structure. These disulfide bonds link the peptide to small molecule toxins, enhancing stability and target specificity. In 2021, Bicycle Therapeutics announced three investigational BTPs (BTCs) for Phase II/III clinical trials: (1) BT1718, targets membrane type I matrix metalloproteinase (MT1-MMP) to release the cytotoxic payload DM1 [75]; (2) BT5528, targets EphA2, a transmembrane tyrosine kinase receptor overexpressed in various cancers, including breast, colon, uroepithelial, lung, cervical, and ovarian cancers [197]; (3) BT8009, targets nectin-4, a type I membrane protein overexpressed in many tumors, including uroepithelial, breast, pancreatic, and triple-negative breast cancers [198]. A Phase I/IIa trial of BT1718 in patients with advanced solid tumors (NCT03486730) was initiated with the aim to study their pharmacokinetics and find the maximum dose of BT1718 and their potential side effects. The results revealed that BT1718 is well tolerated, and the pharmacokinetics are consistent with the proposed preclinical mechanism of tumor-targeted toxin delivery [199]. A phase I/II study of BT5528 in patients with advanced malignancies associated with EphA2 expression (NCT04180371) was carried out to study the drug safety, pharmacokinetics, and preliminary clinical activity. The results revealed that BT5528 was well tolerated with controllable side effects. They demonstrated favorable and preliminary anti-tumor activity, particularly in urothelial cancer. The pharmacokinetics are generally dose-proportional, with a short half-life (0.4–0.7 h) [200]. These results support further development of BT5528 as monotherapy and in combination with nivolumab and potentially other agents in select tumor types [201]. Two clinical studies of BT8009 have been recruited in patients with locally advanced or metastatic urothelial cancer(NCT06225596) and patients with Nectin-4 expressing advanced malignancies(NCT04561362). Both of these two trials will assess the safety and tolerability of BT8009 alone and in combination with pembrolizumab in patients with select advanced solid tumors. However, the results have not been published. Overall, the promising anti-tumor activity in preclinical and early clinical studies highlights the potential of BTPs.

As we described above, PDCs have made some progress in certain clinical trials. However, most of them are in phase I/II clinical studies, and the clinical efficacy and safety have not been largely investigated. The most promising PDCs are radionuclide-peptide conjugates, which exhibited good tumor targeting and therapeutic efficacy, such as the approved Lutathera for the treatment of patients with somatostatin receptor (SSTR)-positive gastroenteropancreatic neuroendocrine tumors and the approved 177Lu-PSMA-617 (Pluvicto) for the treatment of adult patients with prostate-specific membrane antigen (PSMA)-positive metastatic castration-resistant prostate cancer (mCRPC). The success of these two approved PDCs accelerated the development of radionuclide-peptide conjugates for the treatment of various malignant cancers. Pepaxto (Melflufen) is another approved PDC for the treatment of heavily pretreated relapsed and refractory multiple myeloma. However, this drug was withdrawn from the market due to the unsatisfied clinical outcomes in phase III clinical trials. As for other PDCs tested in clinical trials, the safety has been widely proven, but the efficacy is not convincing enough to get approval. Among these PDCs, ANG1005 (Angiopep-2) showed potential for cancer patients due to the good clinical results that notable central nervous system (CNS) and systemic treatment effects were observed in all patients treated with ANG1005 in phase II clinical trial, including symptom improvement and prolonged overall survival with good tolerability in heavily pretreated patients with advanced solid tumors with brain metastases or prior failed taxane therapy. Although the results of ANG1005 in phase III trials have not been revealed, this kind of drug modality provided a new approach or direction for refractory metastatic tumors. The bicyclic peptide drug conjugates also demonstrated great potential for cancer therapy due to the controlled side effects, good pharmacokinetics, and therapeutic activity.

Current challenges and future directions of PDCs

Current challenges of PDCs

In order to achieve their full therapeutic potential, PDC drugs must overcome enzymatic degradation in serum and effectively penetrate target tissues to release their payloads. Two primary strategies can be employed to enhance stability and cell penetration: active formulations and covalent modifications. Active formulations, such as polypeptides, nanoparticles, liposomal formulations, polymers, dendrimers, quantum dots, and polymeric micelles, have expanded the clinical applications of PDCs [202]. For instance, the FDA-approved Sandostatin LAR encapsulates octreotide within a glucose-poly(lactide-co-glycolide) (Glu-PLGA) polymer, extending its release profile. Polypeptide carriers like PEGylated poly(aspartic acid) (PAsp) and PEGylated poly(l-glutamic acid) (PLG) can prolong blood circulation time, enhance therapeutic efficacy, and reduce systemic toxicity of DOX [203]. Opaxio (CT-2103), a PLG-doxorubicin conjugate, demonstrates improved stability, comparable efficacy to free doxorubicin, and reduced adverse effects [204]. The conjugation of short hydrophilic peptides with hydrophobic drugs can induce self-assembly into supramolecular nanostructures, such as nanofiber hydrogels [205]. These structures can significantly improve drug delivery and efficacy.

Covalent modifications, including peptide cyclization, incorporation of unnatural amino acids or N-methylation, and conjugation with macromolecules, can enhance PDC stability and function [10]. The conformational constraints enforced by cyclization can render cyclic peptides highly resistant to proteolysis by endogenous proteases compared to their linear counterparts [206]. Various cyclization methods, such as disulfide bonds, lactam bridges, and stapling, have been explored [207]. For example, N-terminal crosslinked aspartic acid (TD) stapling offers a simple and cost-effective approach to stabilize peptides and improve their selectivity [107]. Besides improving peptide binding affinity and cellular uptake, peptide cyclization can enhance binding affinity and cellular uptake, as well as tissue penetration. Pentelute et al. demonstrated that perfluoroalkyl-cysteine SNAr chemistry could be used to create macrocyclic peptides with improved blood-brain barrier penetration. These macrocyclic peptides exhibited increased brain parenchyma penetration following intravenous administration in mice, suggesting potential applications for PDCs and peptide drugs in brain-related diseases [208]. Bicycles are another strategy involving the cross-linking of three cysteine residues to a trifunctional linker, forming a rigid cyclic structure. This approach can enhance the stability, selectivity, and potency of peptide-based drugs. The modern technologies for bioactive cyclic peptide screening accelerate the development of optimal drug candidates for binding targets with high affinity and specificity, such as phase display [40], one-bead-one-compound (OBOC) method [209], mRNA display [210], split-intein circular ligation of peptides and proteins (SICLOPPS) et al. [211].

D-amino acids can reduce susceptibility to proteolytic enzymes, increasing peptide stability. Octreotide, a peptide hormone containing two D-amino acids, exhibits a significantly longer half-life (1.5 h) compared to the unmodified peptide somatostatin (a few minutes) [181]. N-methylation can enhance peptide stability by increasing steric hindrance and altering peptide conformation. Cyclosporin, a hepta-N-methylated cyclic undecapeptide, is a well-known example of a highly stable and orally bioavailable peptide drug [212]. PEGylation is a common strategy to improve peptide pharmacokinetic properties [213,214,215]. PEGylation can increase molecular weight, reduce renal clearance, and prolong circulation time. For instance, PEGylation of the HM-3 peptide significantly extended its half-life [178]. Lipid conjugation is another approach to enhance peptide stability and pharmacokinetic properties. Semaglutide, a GLP-1 analog, is conjugated to a fatty acid chain, resulting in a significantly longer half-life and enabling once-weekly administration [179]. Thus, the chemical modification of peptides can not only improve PDC’s pharmacokinetic and pharmacodynamic properties for absorption but also have a pronounced effect on cell membrane permeability and target affinity to reach intracellular targets.

Besides the peptide stability, effective drug release is crucial for therapeutic efficacy, which depends on linker structures to break off in specific environments (temperature, pH, enzymes, redox, etc.). Despite different kinds of linkers being designed, toxic drugs sometimes can not be released as a prototype drug or even not be released at targeted tissues, which demonstrates weaker effects than prototype drugs. Alas et al. reviewed different linkers used in PDCs between the peptide and the linker and between the drugs and the linkers for cancer therapy [11]. The ideal linkers conjugate to homing peptides should not affect the binding affinity for its receptor, and the linkers attached to drugs can be cleaved to release the unmodified drug at the target site. Thus, linker chemistry plays a major role in the success of a PDC and in improving efficacy.

In summary, the chemical structure of both peptides and linkers used in the design and development of PDCs play key roles in their clinical applications.

Future directions of PDCs

Bicycle-toxin conjugates (BTCs)

Contrary to the common belief that cyclic peptides are inherently more stable than linear peptides, they can still be susceptible to rapid in vivo degradation [216]. Bicyclic peptides, with their additional cyclic constraint, offer enhanced proteolytic stability. BTCs have emerged as promising drug delivery systems (Fig. 7). These conjugates typically consist of a 9–20 amino acid bicyclic peptide linked to a small molecule toxin [217]. The rigid, cyclic structure of the peptide ensures the toxin’s conformation remains intact, optimizing its activity [198]. BTCs offer several advantages over ADCs, including deeper tumor penetration, rapid extravasation, and slower renal clearance. Bicycle Therapeutics is at the forefront of BTC development, with several compounds, including BT1718, BT5528, and BT8009, currently in clinical trials. These BTCs target specific tumor markers and have shown promising anti-tumor activity.

Advanced screening technologies, such as phage display, mRNA display, split intein circular ligation of peptides, and in silico screening, have revolutionized cyclic peptide drug discovery [218]. These technologies enable the rapid generation and screening of diverse cyclic peptide libraries, facilitating the identification of potent drug candidates. Structure-based drug design is another powerful approach to developing bicyclic peptides. By understanding the molecular interactions between a target protein and a peptide ligand, researchers can design bicyclic peptides with high affinity and specificity. For example, Grossmann et al. designed a library of bicyclic β-sheet mimetics that target β-catenin, a key player in Wnt signaling [219]. These peptides could serve as valuable building blocks for the development of potent BTCs.

Fig. 7

Three different kinds of PDCs in the future development

Peptide dendrimer drug conjugates

Dendrimers are characterized by their highly branched, multilayered architecture, consisting of a core, multiple generations of branching, and a functionalized outer shell (Fig. 7) [220, 221]. Unlike linear peptides and polymers, dendrimers possess unique properties, including thermal stability, viscosity, and encapsulation capabilities [222]. Peptide dendrimers can be classified into two main types: covalent and noncovalent. Covalent peptide dendrimers are constructed using natural or unnatural amino acids, while noncovalent peptide dendrimers rely on noncovalent interactions to form their structure [223]. Peptide dendrimers have emerged as promising drug delivery vehicles. They can be used to deliver drugs through covalent conjugation or noncovalent encapsulation. For instance, Gu and colleagues demonstrated the successful use of peptide dendrimers to selectively deliver doxorubicin to cancer cells [224]. The incorporation of polyethylene glycol (PEG) into dendrimer nanoparticles can further enhance biocompatibility, reduce toxicity, and improve drug delivery efficiency.

Self-assembling peptide drug conjugates

Self-assembling peptide-based nanostructures offer significant advantages, including sustained drug release, precise targeting, overcoming drug resistance, and reducing off-target toxicity(Fig. 7) [122, 225]. As discussed in the Part 3.1.3 “Self-assembling peptide” of this review, various PDCs can self-assemble into nanostructures to enhance drug delivery. In order to address the challenges associated with clinical translation, innovative approaches are being developed to optimize the design and formulation of self-assembling peptide-based drug delivery systems. For example, the self-assembly of CXCR4 antagonist peptide-docetaxel conjugates, could self-assemble to nanoparticles, targeting CXCR4-upregulated metastatic tumor cells and enhancing the DTX efficacy. This self-assembly achieved promising efficacy on inhibiting both bone-specific metastasis and lung metastasis of triple-negative breast cancer [226]. These advancements hold the potential to revolutionize the treatment of various diseases.

Conclusion and discussion

PDCs hold immense potential as next-generation targeted therapeutics, offering advantages over ADCs in terms of synthesis, immunogenicity, and potency. However, challenges such as poor stability, low bioavailability, and lengthy development timelines hinder their clinical translation. PDCs and ADCs share fundamental similarities, and the clinical success of ADCs can inform the development of PDCs. Key challenges in ADC development, including on-target toxicity, linker instability, immune-related toxicity, receptor-mediated toxicity, and drug resistance, can guide the optimization of PDC design. In order to minimize off-target toxicity, ideal targets for PDCs should be selectively expressed on tumor cells. Recent advancements in gene sequencing and antigen discovery have enabled the identification of neoantigens and tumor-specific biomarkers. Additionally, targeting stromal cells, which share common markers across different tumor types, offers a promising strategy for broad-spectrum cancer therapy. Machine learning-based biology analysis can further accelerate the discovery of novel therapeutic targets.

Besides targeting the tumor-specific proteins to improve tumor selective toxicity, various strategies can be employed to enhance peptide stability and extend their circulation time: (1) Peptide Cyclization: Improves stability and resistance to proteolytic degradation. (2) Unnatural Amino Acid Incorporation: Enhances stability and alters peptide properties. (3) Macromolecular Conjugation: Increases molecular weight, reduces renal clearance, and prolongs circulation time (e.g., PEGylation). The choice of payload and linker is crucial for PDC efficacy and safety. Potent cytotoxic agents, such as small molecule drugs and toxins, can be conjugated to peptides. Additionally, emerging modalities like PROTACs and oligonucleotides offer exciting opportunities for PDC development. PDCs can be designed to address drug resistance mechanisms, such as increased drug efflux, altered target expression, and activated downstream signaling pathways. For example, targeting HSP90α, a protein involved in multiple drug resistance pathways, can sensitize tumor cells to therapy.

PDCs are currently limited to intravenous administration and cannot be delivered orally, posing a significant challenge for patients who are unable to attend regular follow-up visits. Recent studies have reported a novel method for achieving oral delivery of bioactive peptides using anionic nanoparticles [227]. While the use of nanoparticles is an attractive strategy to address the challenges of low oral bioavailability and stability, further testing is needed to assess the feasibility of this approach in future clinical trials. Additionally, alternative strategies, such as acid-resistant coatings, gut enzyme inhibitors, and mucus-penetrating peptides, have been proposed to improve the oral availability of PDCs [26]. These advancements promise to expand the accessibility and ease of administration of PDC therapies.

Meanwhile, the industrialization challenges of ADCs such as production costs and quality control also promote the development of PDCs. ADC production is expensive, time-consuming, and requires specialized facilities. Scaling up ADC production while maintaining consistency is difficult and costly. PDCs have lower production costs as peptides are chemically synthesized, relatively inexpensive process compared to antibody production. The conjugation of peptides, linkers, and payloads is simpler and more cost-effective than ADC conjugation [18]. ADCs are inherently heterogeneous due to variations in Drug-to-Antibody Ratio (DAR), conjugation sites, and antibody glycosylation patterns [19]. This requires highly stringent quality control (QC). While, PDCs are more homogeneous than ADCs as peptides are smaller and easier to characterize [18]. QC for PDCs are less stringent relying on simpler techniques like HPLC and mass spectrometry.

In conclusion, while the field of PDC development is still in its early stages, significant progress has been made. By addressing challenges in stability, delivery, and efficacy, PDCs have the potential to revolutionize cancer therapy. As our understanding of cancer biology and drug delivery mechanisms deepens, we can expect to see increasingly sophisticated PDC designs that offer improved therapeutic outcomes.