Advances of Peptide-Drug Conjugates in the Treatment of Digestive System Tumors

Abstract

Digestive system tumors, including gastric cancer, hepatocellular carcinoma, pancreatic cancer, and colorectal cancer, represent one of the most prevalent and lethal malignancy groups worldwide, with high incidence and mortality rates. Despite the continuous development of chemotherapy in cancer treatment, this traditional therapy generally suffers from insufficient tumor specificity, drug resistance and limited efficacy. Precise targeting strategies such as antibody-drug conjugates (ADCs) and peptide-drug conjugates (PDCs) combine cell-targeting delivery with potent antitumor activity and have demonstrated therapeutic potential in digestive system tumors. Compared with ADCs, PDCs offer advantages such as lower molecular weight, improved tumor penetration and reduced immunogenicity risk. Preclinical studies suggest that PDCs can enhance tumor-specific drug accumulation and reduce damage to normal tissues, which may compensate for the shortcomings of traditional chemotherapy. In this review, we systematically outline recent progress in PDC research for digestive system tumors, focusing on target selection and PDC design strategies. Representative PDCs targeting integrins, GPC3, EGFR, HER2, transferrin receptor, and VEGFR in digestive system tumors are highlighted. Based on the preclinical findings and early clinical results, we provide insights on future directions for PDC treatment in digestive system tumors.

Introduction

Digestive system tumors rank among the most prevalent malignancies globally and pose a significant threat to public health. Epidemiological studies consistently show that digestive cancers—including gastric, colorectal, hepatocellular, pancreatic, and esophageal carcinomas—are associated with high incidence and mortality rates worldwide.1,2 Current treatments for digestive system tumors include surgical resection, chemotherapy, radiotherapy, targeted therapy and immunotherapy. Surgery remains the primary curative option for early-stage resectable tumors. Although considerable progress has been made in diagnostic and therapeutic approaches, a large proportion of patients are diagnosed at advanced stages, thereby missing the window for curative surgical intervention. This clinical challenge stems from the complex biological behavior of these tumors, their pronounced heterogeneity, difficulties in achieving early detection, propensity for metastasis and treatment resistance.

Conventional chemotherapy, while capable of partially controlling tumor growth, is hampered by its non-specific mechanism of action, resulting in systemic toxicity. This off-target effect damages rapidly dividing normal cells, leading to adverse events such as myelosuppression, gastrointestinal distress, and alopecia. These toxicities frequently limit dose intensity and treatment efficacy, and may also contribute to the development of drug resistance.3,4 Radiotherapy provides localized tumor control but carries inherent risks of damaging adjacent healthy tissues. The advent of targeted therapies and immunotherapies has improved outcomes for select patient subgroups; nevertheless, their effectiveness is often contingent upon specific genetic alterations, target expression levels, or the tumor immune microenvironment. A number of patients exhibit either primary or acquired resistance to these modalities,5 and immunotherapy may be accompanied by immune-related adverse events.6,7 Consequently, there is a pressing need to develop more effective and better-tolerated treatment strategies that can overcome drug resistance.

To address the poor selectivity and toxicity of traditional chemotherapeutic drugs, targeted drug delivery systems have been developed. These systems aim to selectively accumulate therapeutic agents within tumor tissues or cells, thereby enhancing efficacy while minimizing off-target effects.8,9 Prominent examples include ADCs, which utilize monoclonal antibodies specific to tumor-associated antigens to deliver potent cytotoxic payloads directly to cancer cells.3,10–12 The clinical success of ADCs has invigorated the field of targeted conjugates and offered key insights for the development of next-generation targeted delivery platforms.

Recently, PDCs have gained increasing attention as an important complementary or alternative strategy to ADCs.4,13–16 PDCs utilize tumor-homing peptides as carriers to deliver cytotoxic drugs to tumor sites. Compared with ADCs, PDCs exhibit several distinct advantages, such as lower molecular weight, improved tissue penetration, reduced synthesis costs, and lower immunogenicity.17 These characteristics may confer markedly improved properties to PDCs in certain scenarios, particularly in overcoming the permeability barriers associated with solid tumors.4,13–15

A PDC is a covalent conjugate in which a peptide (homing, receptor-targeting, or cell-penetrating) or peptidomimetic is connected through a linker to cytotoxic payload, such as a small-molecule cytotoxin, photosensitizer or chelated radionuclide (Figure 1).13 Lutathera (¹⁷⁷Lu-DOTATATE) is an approved peptide-based radiopharmaceutical for gastroenteropancreatic neuroendocrine tumors (GEP-NETs). It consists of the somatostatin analogue Tyr3-octreotate (TATE), a cyclic octapeptide (D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Thr) that binds somatostatin receptor subtype 2, conjugated to the macrocyclic chelator DOTA, which coordinates the β-emitting radionuclide lutetium-¹⁷⁷ (¹⁷⁷Lu). By contrast, Pluvicto (¹⁷⁷Lu-PSMA-617), approved for metastatic castration-resistant prostate cancer, is a small-molecule prostate-specific membrane antigen (PSMA)-targeted radioligand therapy, not a peptide-based PDC. It comprises the small-molecule PSMA-binding pharmacophore Glu-urea-Lys, a linker incorporating 2-naphthyl-L-alanine and trans-4-(aminomethyl)cyclohexanecarboxylic acid, and a DOTA-chelated ¹⁷⁷Lu moiety. Pluvicto is included in Figure 1 for structural comparison only. Although GEP-NET is a digestive system tumor, there are currently no other PDC reports besides Lutathera, so it will not be discussed in this review.

Figure 1.

Basic structure and mechanism of action of PDCs, together with the chemical structures of Lutathera, a peptide-based radiopharmaceutical, and Pluvicto, a small-molecule PSMA-targeted radioligand therapy included for structural comparison. After binding to receptors on tumor cells, PDCs are internalized, and the payload is released from endosomes or lysosomes into the cytoplasm or nucleus, where it exerts its cytotoxic activity. For Lutathera, a prototypical radiopharmaceutical PDC, the chelated radionuclide (¹⁷⁷Lu-DOTA) is delivered to receptor-positive tumor cells and emits cytotoxic radiation locally; no cleavable linker is required. Pluvicto (shown for structural comparison; a small-molecule radioligand therapy, not a peptide-based PDC) uses the Glu-urea-Lys pharmacophore as its targeting moiety. The CTP is shown in blue, the payload in red, and the linker in black.

The CTP of PDCs binds to receptors overexpressed on cancer cells or markers in the tumor microenvironment, enabling selective accumulation at the tumor site. Nevertheless, PDCs are vulnerable to peptidase hydrolysis, resulting in short half-lives in the blood. In most cases, a short plasma half-life is insufficient to deliver a sufficient drug load to targeted organs and tissues, and circulatory instability and rapid renal clearance persist in clinical use. However, considerably rapid tissue distribution may compensate for the short half-life in blood. Several peptide-engineering strategies have therefore been developed to improve systemic proteolytic stability, including cyclization, peptide stapling, incorporation of D-amino acids or N-methylated residues, PEGylation, and albumin-binding modifications.17 Cyclization constrains the peptide into a rigid ring conformation, limiting protease accessibility and recognition. Peptide stapling locks peptides into stable α-helical conformations, improving both binding affinity and enzymatic stability. Incorporation of D-amino acids or N-methylated residues reduces recognition by proteases while maintaining target affinity.17

The linker connects the peptide with the cytotoxic payload, it is designed to be stable in circulation but cleavable within the target site—such as the acidic or enzymatic conditions of the tumor microenvironment or intracellular lysosomes—to ensure specific drug release.18 Enzyme-cleavable linkers (Val-Cit or Val-Ala) are cleaved intracellularly by the lysosomal cysteine protease cathepsin B after internalization, which limits premature release in circulation. Redox-sensitive linkers rely on disulfide bonds that are cleaved by the high intracellular concentration of glutathione, which greatly exceeds plasma levels and enables preferential cytosolic release. Non-cleavable linkers, composed of stable thioether or amide bonds, resist plasma peptidases and generally require the payload to retain activity after intracellular metabolism. For the systemically administered PDCs reviewed here, linker design is governed primarily by plasma stability and intracellular release kinetics rather than by the acidic and proteolytic environment of the gastrointestinal tract, which is relevant mainly for oral formulations. Thus, linker chemistry directly dictates pharmacokinetics, bioavailability, and therapeutic safety.18,19 The payload is usually a highly potent cytotoxic agent intended to eradicate tumor cells at minimal doses.

For oral administration—which does not apply to the PDCs reviewed here but is relevant to the broader peptide-drug field—additional strategies, including acid-stable coatings, gastrointestinal enzyme inhibitors, and mucus-penetrating peptides, have been explored to improve bioavailability, although further clinical validation is required.20

PDCs have been extensively studied in gastric cancer (GC), hepatocellular carcinoma (HCC), pancreatic cancer and colorectal cancer (CRC) (Table 1). However, esophageal cancers and cholangiocarcinoma remain relatively under-investigated for PDC therapy. This review focuses on PDCs evaluated in GC, HCC, pancreatic cancer, and CRC, the tumor types for which PDC-specific data are available. A systematic literature search was performed in PubMed, Web of Science, Scopus, and CNKI for articles published from database inception to June 2026, using combinations of the terms “peptide-drug conjugate”, “PDC”, “peptide–cytotoxin conjugate”, and “peptide–photosensitizer” together with each tumor type. Original research articles and clinical studies reporting the design or evaluation of peptide-based drug conjugates with in vitro or in vivo data in these tumor types were included; reviews were retained only as background sources. Studies of esophageal cancer and cholangiocarcinoma were screened but are not discussed in detail because no PDC-specific primary studies with therapeutic data were identified for these indications; these tumors are considered in the future perspectives.

Table 1.

Summary of PDCs for Digestive System Tumors

Types of Cancers	Target	PDC	Peptide	Linker	Payload	Clinical Status	Ref
Gastric Cancer	Integrin αvβ3	RGD@Micelles	RGD	Disulfide	Paclitaxel	Preclinical	[21]
	KK-LC-1	1131-MMAE	CKNTALTTC	MC-VC-PABC	MMAE	Preclinical	[22]
Hepatocellular Carcinoma	GPC3	PDC22	SNDRPPNILQKR (TJ12P2)	MC-VC-PABC	MMAE	Preclinical	[23]
		Peptide–Chlorin e6 conjugates	DHLASLWWGTEL	Amide	Chlorin e6	Preclinical	[24]
	TfR	BP9a-SS-DOX	CAHLHNRS	Disulfide	Doxorubicin	Preclinical	[25]
	VEGFR	QR-KLU	QKRKRKKSRYKS	Aminocaproic acid (Ahx)	KLU	Preclinical	[26]
	PSMA	G-202	Asp-γGlu-γGlu-γGlu-γGlu	Amide	Thapsigargin analog (12-ADT)	Phase II	[27]
Pancreatic Cancer	Integrin αvβ6	SG3299	A20FMDV2	Val-Ala	SG3199	Preclinical	[28]
	Integrin αvβ3	PDC-2	RGDfC	1,3,5-Triazine-based	FL118	Preclinical	[29]
	NTR	NT4–CA4	NT4	Ester	CA4	Preclinical	[30]
Colorectal Cancer	SLC1A5	P-LPK-CPT	LPKTVSSDMSLN	Triazole (click chemistry)	Camptothecin	Preclinical	[31]
	GnRH-R	GnRHIII[[4]Lys(Bu), [8]Lys(Dau=Aoa)]	GnRH-III[[4]Lys(Bu),[8]Lys(Dau=Aoa)]	Oxime	Daunorubicin	Preclinical	[32,33]
	TfR	LWJ-M30	GHKAKGPRKC (B6)	Disulfide	DM1	Preclinical	[34]
	EGFR	EBP	CMYIEALDKYAC	Ester	Doxorubicin	Preclinical	[35]
		P6-SN38	P6(disulfide-bridged cyclic nonapeptide)	Succinate ester	SN38	Preclinical	[36]
		P6-DA1	P6 (c(CHVPGSYIC))	Ahx (urea linkage)	Ahx-DA1	Preclinical	[37]

Abbreviations: PDC, peptide-drug conjugate; GC, gastric cancer; HCC, hepatocellular carcinoma; CRC, colorectal cancer; KK-LC-1, Kita-Kyushu lung cancer antigen 1; GPC3, glypican-3; TfR, transferrin receptor; VEGFR, vascular endothelial growth factor receptor; PSMA, prostate-specific membrane antigen; GnRH-R, gonadotropin-releasing hormone receptor; EGFR, epidermal growth factor receptor; MMAE, monomethyl auristatin E; DM1, maytansinoid; PTX, paclitaxel; DOX, doxorubicin; CPT, camptothecin; CA4, combretastatin A4; KLU, lytic peptide; PABC, para-aminobenzyloxycarbonyl; MC, maleimidocaproyl; VC, valine-citrulline;SN38,7-ethyl-10-hydroxycamptothecin; Ahx, 6-aminohexanoic acid.

Progress of PDCs in the Treatment of Digestive System Tumors

Gastric Cancer

GC is a prevalent malignancy worldwide, with particularly high incidence and mortality in East Asia.1 While early-stage GC is often manageable with surgical resection and carries a more favorable prognosis, the majority of patients are diagnosed at an advanced stage and face a poor prognosis. Chemotherapy remains the cornerstone of treatment for advanced disease, yet its efficacy is often limited and accompanied by significant toxicity. In recent years, targeted therapies have achieved notable advances in GC. Examples include trastuzumab for HER2-positive GC and, more recently, the antibody–drug conjugate trastuzumab deruxtecan (T-DXd).38,39 These developments underscore the promise of molecularly targeted approaches and emphasize the ongoing need for refined delivery strategies to improve treatment outcomes.

RGD Peptide-Modified Paclitaxel PDC

αvβ3 integrin is usually expressed at low levels in normal cells, but can be highly expressed in many types of cancer cells, including GC cells. Utilizing an RGD peptide that targets αvβ3 integrin, these PDCs can self-assemble into micelles (RGD@Micelles) (Figure 2A) with an average diameter of approximately 50 nm. The design incorporates a disulfide bond linking PEG to paclitaxel (PTX), enabling glutathione (GSH)-triggered disassembly and subsequent PTX release under weakly acidic intracellular conditions. In the presence of 10 mM GSH, approximately 85% of PTX was released over 72 hours. The conjugate exhibited significant cytotoxicity against both SGC7901 GC cells and their drug-resistant counterparts, inducing apoptosis in 58.13% of cells. In vivo, it achieved a tumor inhibition rate of 96.8%, considerably higher than that of free PTX (39.4%), without causing significant body weight loss, indicating a favorable tolerability profile in this mouse model.21

Figure 2.

(A) Schematic illustration of the self-assembly, cellular uptake, and intracellular drug delivery of RGD@Micelles; (B) chemical structure of 1131-MMAE. The linker between peptide 1131 and MMAE comprises a maleimidocaproyl (mc) spacer, a cathepsin B-cleavable valine-citrulline (vc) dipeptide, and a self-immolative para-aminobenzyloxycarbonyl (PABC) spacer. The CTP is shown in blue, the payload in red, and the linker in black.

KK-LC-1-Targeted Auristatin-Like PDC

Kita-Kyushu lung cancer antigen 1 (KK-LC-1) is exclusively expressed in several types of cancer including GC, representing a promising target for drug delivery. This study identified KK-LC-1 as a potential target for PDC development and introduced the first KK-LC-1-directed conjugate, 1131-MMAE. The molecule consists of a KK-LC-1-binding peptide linked to the antimitotic agent monomethyl auristatin E (MMAE) via an enzymatically cleavable linker. In KK-LC-1-positive cells, the conjugate showed potent cytotoxicity with an IC₅₀ of 3.87 nM, arrested the cell cycle at the G2/M phase—a radiosensitive stage—and enhanced sensitivity to radiotherapy. In vivo, it demonstrated favorable tumor accumulation and significant antitumor activity at a dose of 0.3 mg/kg. When combined with radiotherapy, it extended the survival of tumor-bearing mice to 58 days, supporting its potential for precision therapy (Figure 2B).22

Hepatocellular Carcinoma

HCC is a prevalent malignancy with a particularly high incidence and poor prognosis in Asia.1 The treatment of advanced HCC poses significant therapeutic challenges, which are compounded by the tumor’s high heterogeneity and the frequent coexistence of underlying cirrhosis, thereby limiting therapeutic options. Although targeted agents like sorafenib have improved outcomes for some patients, drug resistance remains a major clinical obstacle. PDCs have emerged as a targeted delivery strategy with distinct potential for advancing HCC treatment. Multiple targets and PDC designs have been explored in HCC.

GPC3-Targeted PDCs

GPC3, a cell surface proteoglycan, is overexpressed in HCC and less so in normal liver tissue, making it an attractive target for targeted therapy in HCC. Researchers have developed peptide-conjugated MMAE (PDC 22) targeting GPC3 that demonstrated efficient tumor cell internalization, drug release, G2/M cell cycle arrest, and induced immunogenic cell death of GPC3-positive tumor cells. In xenograft models, PDC 22 specifically accumulates in tumor regions and delays tumor growth with minimal toxicity. Notably, combining PDC 22 with radiotherapy resulted in improved tumor control compared with monotherapy, highlighting its potential for precision chemoradiotherapy in HCC.23 Another study developed a peptide-conjugated photosensitizer (chlorin e6) targeting GPC3 for targeted photodynamic therapy (PDT) in HCC, achieving tumor elimination in a HepG2 xenograft tumor model without significant damage to normal tissues (Figure 3A).24 These studies highlight the promise of GPC3 as an HCC target for PDC applications.

Figure 3.

Chemical structures of PDCs for the treatment of HCC: GPC3-targeting peptide–chlorin e6 (photosensitizer) conjugate (A), PDC-DOX2 (B), BP9a-SS-DOX (C), QR-KLU (D), and mipsagargin (G-202), in which the thapsigargin analog 12-ADT is masked by an Asp-γGlu-γGlu-γGlu-γGlu peptide substrate (E). The CTP is shown in blue, the payload in red, and the linker in black.

Integrin-Targeted PDCs

Integrins, particularly the αvβ3 and αvβ5 isoforms, are overexpressed on HCC cells and tumor vascular endothelial cells.40,41 RGD peptides and their derivatives are able to specifically bind these integrins. Applications of RGD peptides in cancer targeting are mentioned in the literature, including coupling to drug molecules or nanocarriers to improve drug delivery efficiency and reduce side effects.41 RGD peptide-based PDCs are expected to target HCC cells and tumor vasculature, exerting dual anti-tumor effects. For example, the use of iRGD—a tumor-penetrating RGD derivative—as a targeting ligand may improve drug distribution and tissue penetration within HCC. A new self-assembling amphiphilic peptide drug conjugate (SAAPDC) is fabricated as a “two-in-one” nanofiber system comprising a hexapeptide as a matrix metalloproteinases (MMP) inhibitor and doxorubicin (DOX) for the treatment of HCC. This nanofiber not only inhibits tumor growth in situ but also effectively prevents pulmonary metastasis in an SMMC7721 cell line–based mouse model. A synergistic effect was achieved by simultaneous inhibition of MMP activity and delivery of chemotherapeutic agents.42 Another novel PDC (PDC-DOX2) (Figure 3B), in which two DOX molecules are conjugated onto a short peptide (KIGLFRWR) with self-assembly function, was designed and synthesized. PDC-DOX2 with self-assembly properties forms a spherical structure under hydrophobic interaction in water. Hyaluronic acid (HA) was then coated on PDC-DOX2 micelles to form a HA-shelled, peptide-doxorubicin conjugate-cored nanomedicine (HA@PDC-DOX2). HA@PDC -DOX2 showed more potent anti-HCC activity than the PDC-DOX2 micelles and the free DOX both in vitro and in vivo.43

TfR-Targeted PDC

Transferrin receptor (TfR) is highly expressed in HCC cells and can be used as a target for drug delivery. DOX was conjugated to the TfR-targeting peptide BP9a via a disulfide-based linker, yielding a reduction-sensitive conjugate designated BP9a-SS-DOX (Figure 3C). This construct efficiently entered HepG2 cells through TfR-mediated endocytosis and subsequently released DOX intracellularly. It exhibited specific antiproliferative activity against HepG2 cells while showing reduced toxicity toward normal hepatocytes.25

VEGFR-Targeted PDC

Vascular endothelial growth factor (VEGF) is a key driver of angiogenesis. QR-KLU was developed by conjugating the VEGFR-targeting peptide VEGF₁₂₅–₁₃₆ (QR) with the lytic peptide KLU. This conjugate not only blocks VEGFR signaling but also exerts direct cytotoxic activity. In a rabbit liver tumor model, QR-KLU combined with transcatheter arterial chemoembolization (TACE) exhibited superior antitumor efficacy compared with conventional doxorubicin-based regimens. Furthermore, the combination therapy effectively suppressed tumor angiogenesis and demonstrated favourable tolerability in the reported animal models.44 QR-KLU also synergized with anti-PD-1 therapy by activating CD8⁺ T cells, reducing immunosuppressive factors, and improving survival in murine models (Figure 3D).26

PSMA-Targeted PDC

Mipsagargin (G-202) is an investigational prodrug in which the cytotoxic thapsigargin analog 8-O-(12-aminododecanoyl)-8-O-debutanoylthapsigargin (12-ADT) is linked to the masking peptide Asp-γGlu-γGlu-γGlu-γGlu. The peptide is cleaved specifically by prostate-specific membrane antigen (PSMA), releasing 12-ADT, which inhibits sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA). Although PSMA was initially identified in prostate cancer, it is also expressed on the tumor neovasculature of many solid tumors. In a Phase II, multicenter, single-arm study, mipsagargin was evaluated as second-line therapy after sorafenib in patients with progressive advanced HCC. It was well tolerated and produced stable disease in some patients, and dynamic contrast-enhanced MRI showed reduced blood flow to liver lesions, consistent with a tumor-vasculature-targeting mechanism.27 Mipsagargin remains investigational and has not received regulatory approval. This agent exemplifies a PDC strategy that targets the tumor vasculature rather than tumor cells directly (Figure 3E).

Other Peptide-Targeting Strategies

Although not strictly PDCs, some studies have explored the use of peptides as targeting ligands to modify nanocarriers for drug delivery in the treatment of HCC. For example, EGFR-targeting peptides conjugated to paclitaxel-loaded magnetic liposomes improved tumor accumulation and antitumor efficacy in an HCC xenograft model.45

Overall, HCC is among the most actively investigated digestive tumor types in PDC research. PDCs directed against GPC3, integrins, TfR, VEGFR, and PSMA have demonstrated antitumor activity with generally favorable tolerability in the reported preclinical models, while some agents have also entered clinical evaluation. PDC studies in HCC have extended beyond direct tumor-cell killing to strategies targeting the tumor vasculature and to combinations with radiotherapy or immunotherapy, reflecting attempts to address the therapeutic complexity of HCC.

Pancreatic Cancer

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive malignancy characterized by late diagnosis, aggressive progression, and resistance to conventional therapies. A key challenge is its complex tumor microenvironment, which features a dense, fibrotic stroma that forms a physical barrier, severely impeding drug delivery and efficacy. Given these challenges, PDCs present a promising therapeutic strategy for PDAC. Their small molecular size and enhanced tissue penetration capacity may allow them to overcome these stromal barriers more effectively than conventional agents.

Integrin αvβ6-Targeted PDC

Integrin αvβ6 is highly expressed in PDACs and expressed at low levels in normal pancreatic tissues, making it an ideal targeting marker.46 Researchers conjugated the DNA-binding pyrrolobenzodiazepine (PBD)-based payload SG3249 (tesirine) to an αvβ6-specific 20mer peptide from the VP1 coat protein of foot-and-mouth-disease virus (FMDV) (forming conjugate SG3299) or to a non-targeting peptide (forming conjugate SG3511). The αvβ6-targeted PDC SG3299 (Figure 4A) was more toxic (up to 78-fold) for αvβ6-expressing versus αvβ6-negative PDAC cell lines in vitro, and achieved higher toxicity at equal dose than the non-targeted PDC SG3511 (up to 15-fold better), and inhibited the sphere-forming ability of tumor stem cell-like cells. SG3299 eliminated established Capan-1 PDAC human xenografts, extending the lifespan of mice by a mechanism involving the induction of DNA damage, apoptosis (up-regulation of cleaved caspase 3), and inhibition of proliferation (decrease in Ki67). SG3299 had no significant toxicity to mouse normal tissues (eg, lungs, intestines), confirming its tumor specificity.28

Figure 4.

Chemical structures of PDCs for the treatment of pancreatic cancer: integrin αvβ6-targeted PDC SG3299 (A), integrin αvβ3-targeted PDC PDC-2 (B), neurotensin receptor (NTR)-targeted PDC NT4-CA4 (C), and gemcitabine-based PDC Gem-Cys-Pen (D). The CTP is shown in blue, the payload in red, and the linker in black.

Integrin αvβ3-Targeted PDC

Integrin αvβ3 is a cell adhesion receptor overexpressed in neovascular endothelial cells and aggressive tumors, which is a therapeutic target due to its role in angiogenesis and metastasis. Researchers have developed a novel peptide-drug conjugate, PDC-2 (Figure 4B), which links the potent camptothecin derivative FL118 to the targeting peptide c(RGDfC) through a stable triazine linker. It induced apoptosis of A549 and AsPC-1 cells and inhibited their adhesion, migration, and invasion in a concentration-dependent manner (IC₅₀ as low as 9.4 nM), while showing reduced toxicity to normal cells. Mechanistically, PDC-2 dually inhibits survivin protein expression and the PI3K/AKT/mTOR signaling pathway. In animal models, PDC-2 demonstrated tumor-targeting activity and tumor inhibition (>92% in pancreatic cancer models), while showing lower systemic toxicity than FL118. This is attributed to the long plasma half-life provided by its triazine linker (extended by 3.4-fold) and its efficient tumor tissue accumulation, highlighting its potential as a targeted therapy.29

Neurotensin Receptor-Targeted PDC

Neurotensin receptor is highly expressed in a variety of tumors such as pancreatic and prostate cancers, providing an important target for PDC. Researchers constructed a series of PDCs by linking neurotensin-derived peptides (eg, NT4) with the anti-microtubule drugs, combretastatinA4 (CA4) and monastrol, via ester or ether bonds. NT4-CA4 (Figure 4C) linked by ester bonds showed significant toxicity in PANC-1 pancreatic cancer cells (IC₅₀ = 5×10⁻7 M) and low toxicity to normal cells. The mechanism involves entry into the cell via receptor-mediated endocytosis, release of CA4 followed by inhibition of microtubule polymerisation and induction of apoptosis. Branching peptide design (eg, tetrabranched NT4) enhances multivalent binding to the receptor and improves the efficiency of internalization, while increasing drug loading to further enhance anti-tumor activity.30

Gemcitabine-Based PDCs

Gemcitabine is a first-line chemotherapeutic agent for pancreatic cancer, but is susceptible to degradation by cytidine deaminase and relies on the hENT1 transporter protein to enter the cell, leading to drug resistance. The pharmacokinetic properties of gemcitabine can be improved by conjugating it to a targeted peptide.47 Conjugates such as Gem-Cys-Pen and Gem-Cys-pVEC (Figure 4D) link gemcitabine to cell-penetrating peptides (Penetratin or pVEC) via a disulfide bond. These constructs show high in vitro stability, with a drug-release half-life of approximately 9.6 days and enhance cytotoxicity in multiple cancer cell lines, including MKN-28 and HT-29, with IC₅₀ values below 50 μM.47

Other PDC Targets

The EphA2 receptor is highly expressed in pancreatic cancer. Conjugating EphA2-targeted peptides (123B9) with gemcitabine inhibited tumor growth and prolonged survival in mice bearing MIA PaCa-2 xenografts, while demonstrating reduced toxicity compared to the free drug.48 Plectin-1, which is highly expressed in 92.9% of pancreatic cancer tissues, was utilized as a target for peptide-based nanoparticles (PTNPs) co-delivering olaparib (Ola) and the HR inhibitor JQ1. In this strategy, Ola induces DNA damage to “capture” the tumor, while JQ1 blocks DNA repair to “kill” it. PTNPs exhibited a four-fold higher tumor accumulation compared to the free drug and reduced tumor weight by 2.5-fold in an in situ model. In a patient-derived xenograft (PDX) model, the inhibition rate exceeded 30%, with no observed side effects such as liver injury.49

Additionally, the homing peptide SKAAKN (a cysteine-substituted variant of CKAAKN) was conjugated to doxorubicin (Dau) via oxime bonding. Among several derivatives, a branched conjugate bearing two Dau molecules (conjugate 4) displayed the highest cytotoxicity and tumor selectivity against PANC-1 cells in vitro. In vivo, this conjugate markedly inhibited the growth of subcutaneous PANC-1 tumors in SCID mice, with minimal systemic toxicity, suggesting a promising new approach for the targeted treatment of pancreatic ductal adenocarcinoma.50

Colorectal Cancer

Colorectal cancer (CRC) is one of the most common and lethal malignancies worldwide, and conventional chemotherapy is often limited by treatment-related toxicity and a lack of tumor specificity. PDCs have emerged as a promising targeted therapeutic strategy, designed to deliver cytotoxic agents precisely to tumor tissues by exploiting the binding affinity of tumor-homing peptides to receptors overexpressed on CRC cells. The following section summarizes the mechanisms and antitumor efficacy of representative PDCs currently under investigation for CRC treatment. PDCs achieve efficient delivery and precise drug release through the following mechanisms.

Glutamine Transporter -Targeted PDC

A twelve-amino acid peptide (LPKTVSSDMSLN), designated P-LPK, was identified via phage display screening for its specific binding to CRC cells. This peptide targets the glutamine transporter SLC1A5, which is highly expressed in CRC, and internalizes into cells through clathrin-mediated endocytosis, enabling selective drug delivery.31 In vitro, P-LPK-CPT selectively inhibited the proliferation of CRC cells such as HCT116 and LoVo and showed low toxicity to normal intestinal epithelial cells. In vivo, the conjugate suppressed tumor growth in an xenograft model, with markedly reduced gastrointestinal toxicity compared to free CPT, and extended survival in mice (Figure 5A).31

Figure 5.

Chemical structures of P-LPK-CPT (A) and GnRH-III-daunorubicin (B); basic structure and mechanism of action of LWJ-M30 and DOX-EBP (C). The CTP is shown in blue, the payload in red, and the linker in black.

GnRH-R-Targeted PDC

The daunorubicin-GnRH-III bioconjugate [⁴Lys(Bu),⁸Lys(Dau=Aoa)], which delivers daunorubicin via GnRH-R-targeting. This conjugate features a butyryl modification at position 4 that enhances receptor binding affinity and cellular uptake compared to the acetylated version, while the oxime-linked daunorubicin at position 8 ensures stable drug delivery in human serum for at least 24 hours.32,33 The conjugate achieved 40–50% tumor growth inhibition in the HT-29 orthotopic model, it also reduces hepatotoxicity and cardiotoxicity and inhibits tumor angiogenesis, Moreover, the bioconjugate slightly decreases the tumor proliferation index and shows no significant toxicity on liver or heart tissues, although it does not prevent metastasis to other organs (Figure 5B).33

TfR-Targeted PDC

LWJ-M30, a conjugate of DM1 and B6 peptide, targeted transferrin receptors (TfRs) on the surface of the CRC cells, showing a potent anticancer effect. B6 peptide is derived from a phage display library with high specificity for TfR, and its mediated receptor-dependent endocytosis enhances drug accumulation in tumor cells. LWJ-M30 is characterized by easy synthesis, low cost, and high tumor penetration. B6 peptide is linked to DM1 through disulfide bonding, which allows for the stable delivery and efficient release of the drug in tumor microenvironments to avoid the nonspecific toxicity of DM1 (Figure 5C).34

EGFR-Targeted PDC

EGFR is overexpressed in many colorectal cancers and is an important therapeutic target. A PDC (DOX-EBP) conjugated to DOX using EGFR-binding peptide (EBP) as a targeting unit is described in the literature. This PDC showed higher cell accumulation and cytotoxicity in EGFR overexpressing tumor cells and demonstrated enhanced anti-tumor efficacy and reduced systemic toxicity in a tumor-bearing mouse model. Although this study was conducted on EGFR overexpressing tumor cells, the results support the feasibility of EGFR as a PDC target in colorectal cancer (Figure 5C).35 More recently, a non-canonical EGFR-targeting PDC (P6-SN38) has been developed to achieve tumor-selective payload delivery in KRAS-mutant colorectal cancer, offering a promising strategy to overcome resistance associated with this mutation type.36 Importantly, the Ahx‑DA1‑based P6‑DA1 conjugate also demonstrated potent antitumor activity in EGFR‑positive, KRAS‑mutant pancreatic and colorectal xenograft models. It achieved significant tumor suppression and favorable tolerability compared with cetuximab plus irinotecan, highlighting its potential to overcome intrinsic resistance in these aggressive malignancies.37

Research on PDCs for CRC is progressing rapidly, with major efforts directed toward overcoming chemoresistance and improving deep-tumor penetration. Future work should prioritize the discovery and validation of targeting peptides with enhanced specificity and internalization efficiency, the rational optimization of linkers and payloads, and the exploration of synergistic combinations between PDCs and established treatment modalities to advance therapeutic outcomes for CRC.

Conclusions and Future Perspectives

PDCs design must optimize the peptide (affinity, specificity, pharmacokinetics, low immunogenicity), linker (stability, cleavage efficiency, safety), and payload (potency, stability after conjugation).13,14,16 Targeting peptides should exhibit high affinity and specificity, favorable pharmacokinetic properties, and low immunogenicity. Linkers should remain stable in circulation while enabling efficient payload release at the target site, with minimal toxicity from cleavage products.18–20 Payloads need to be sufficiently potent and not affect their activity after conjugating. The choice of different components and the way they are conjugated can markedly affect the final performance of the PDCs. Finding the best combination requires extensive screening and optimization work, the peptide itself usually has a short plasma half-life, is readily degraded by proteases in plasma and tissues, and is rapidly cleared by the kidneys.12,14 This may result in the clearance of PDCs before they reach the tumor site, affecting the efficiency of tumor accumulation. Although conjugations can alter the PK properties of peptides, it remains a challenge to design PDCs to achieve the desired PK profile (sufficient plasma exposure to reach the tumor while avoiding excessive accumulation in normal tissues).

Digestive system tumors are molecularly and histologically heterogeneous, and target expression may vary both within and between tumors, potentially resulting in heterogeneous PDC distribution and tumor-cell escape. The tumor microenvironment, including a dense extracellular matrix, elevated interstitial pressure, abnormal vasculature, and local hypoxia, can further restrict penetration into deeper tumor regions.15 Resistance may arise through downregulation of target expression, impaired PDC internalization, altered lysosomal function, increased drug efflux, or activation of downstream signaling pathways.17,51 Future studies should therefore investigate multi-target strategies, combinations of PDCs with complementary mechanisms of action, and approaches capable of overcoming specific resistance mechanisms. Highly potent payloads may still affect normal tissues through target expression in nonmalignant organs or nonspecific uptake, particularly in the liver and kidneys.13,16 In addition, although peptide synthesis is relatively straightforward, conjugate production may require complex chemistry and stringent optimization to ensure efficiency, purity, and batch-to-batch consistency. Many PDCs remain at the preclinical stage, and successful translation will require rigorous characterization of pharmacokinetics, pharmacodynamics, tolerability, and efficacy in clinical studies.12,14 In response to these challenges, future research will focus on multiple directions. Future research should focus on identifying tumor-selective targets and optimizing targeting peptides with improved affinity, specificity, and stability through approaches such as phage display and computational design.14 Peptide-engineering strategies, including cyclization, D-amino acid substitution, incorporation of non-natural amino acids, PEGylation, and lipid modification, may improve enzymatic stability, plasma half-life, and membrane permeability.15,52 Further development of linkers responsive to tumor-associated stimuli, including hypoxia, acidity, reducing conditions, and enzymes, may improve site-specific payload release.53,54 Advances in cleavage mechanisms and conjugation chemistry may also facilitate the generation of more homogeneous and stable PDCs. Beyond conventional cytotoxic agents, future PDC payloads may increasingly incorporate compounds with alternative mechanisms of action.55

Combining PDCs with other therapeutic modalities may further improve antitumor activity. Potential strategies include combinations with chemotherapy, targeted therapy, radiotherapy, and immune checkpoint inhibitors.55,56 PDCs may also be combined with agents targeting the tumor microenvironment, such as anti-angiogenic agents or stromal-modifying therapies, to improve tumor penetration. In addition, integration with nanocarriers, including liposomes, polymeric nanoparticles, self-assembled nanostructures, and hydrogels, may improve pharmacokinetics, tumor accumulation, penetration, and controlled payload release. Nanocarriers may protect PDCs from enzymatic degradation and prolong systemic circulation.57,58 Bioinformatics and computational chemistry may further facilitate the prediction of peptide–target interactions, conjugate stability, and pharmacokinetic properties, thereby accelerating PDC design and optimization.

This review systematically compiles the research progress of PDCs in the treatment of major digestive system tumors such as gastric, liver, pancreatic and colorectal cancers. Researchers have designed and evaluated a variety of PDCs against targets overexpressed in these tumors, such as EGFR, HER2, integrins, GPC3, TfR, and VEGFR. Preclinical studies have shown that these PDCs can exert selective cytotoxic effects in vitro, accumulate at tumor sites, and inhibit tumor growth in animal models, with generally favorable tolerability reported in the evaluated models. In particular, PDCs have been more actively studied in areas such as hepatocellular carcinoma and pancreatic cancer, with PDCs targeting GPC3, VEGFR, integrins and other targets, as well as PDCs incorporating nano-delivery strategies, showing the potential to overcome the obstacles of the tumor microenvironment and to enhance the therapeutic efficacy.

Overall, PDCs hold promise for the treatment of digestive system tumors. With continued advances in peptide engineering, linker chemistry, and delivery platforms—and with rigorous evaluation of pharmacokinetics, safety, and efficacy in clinical trials—more innovative PDCs are anticipated to advance into clinical development, potentially offering improved outcomes for patients with digestive system tumors. Future research requires multidisciplinary collaboration, from basic research to clinical translation, to continuously optimize the design and application strategies of PDCs.

Funding Statement

The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. This research was supported by the National Natural Science Foundation of China (no. 82160129, no. 82473266) and Gansu Provincial Science and Technology Major Project (No. 24ZDFA011).

Disclosure

Yaping Ma is affiliated with Shenzhen DIVBIO Pharmaceutical. Shenzhen DIVBIO Pharmaceutical provided no funding or other support for this review and had no role in its conception, literature collection or interpretation, manuscript preparation or revision, or the decision to submit the manuscript for publication. Neither Yaping Ma nor Shenzhen DIVBIO Pharmaceutical holds patents or other intellectual property relevant to the subject matter of this review. The authors declare no competing financial interest.