Tumor Microenvironment-Responsive Prodrugs Emerge as a Strategy to Overcome HCC Treatment Barriers
核心洞察
Hepatocellular carcinoma (搜索) is the sixth most common cancer globally and the third leading cause of cancer-related deaths, with a 5-year survival rate of approximately 18% and roughly 830,000 deaths annually.
A comprehensive review of 130 records (1956–2026) maps how prodrugs activated by acidity, hypoxia, glutathione, reactive oxygen species, and overexpressed enzymes could improve tumor selectivity and reduce systemic toxicity in HCC.
Early clinical signals include tirapazamine chemoembolization, which achieved a 60% complete response rate and median overall survival of 52 months in a first-in-human trial of intermediate unresectable HCC.
Hepatocellular carcinoma (搜索) (HCC) accounts for nearly 90% of the approximately 906,000 global liver cancer cases, yet it carries a relative 5-year survival rate of only about 18% and an incidence comparable to its mortality rate of roughly 830,000 deaths annually. Despite major advances in targeted therapy and immunotherapy, durable disease control remains out of reach for many patients, prompting a new review to examine whether tumor microenvironment (TME)-responsive prodrugs can address persistent challenges in selectivity, toxicity, and drug resistance.
The review, published in Drug Design, Development and Therapy, synthesizes 130 records dated 1956–2026, of which 81 (62.3%) were published between 2022 and 2026. It first maps the current treatment landscape for HCC and then systematically classifies prodrug activation mechanisms tied to the disease's distinctive microenvironment.
A Shifting Treatment Landscape with Persistent Gaps
Systemic therapy remains the treatment of choice for advanced HCC and for intermediate-stage patients ineligible for local therapy. Since 2017, the field has entered an era of sequential therapy, anchored by landmark trials. The SHARP study established sorafenib as the first targeted therapy, demonstrating an overall survival (OS) of 10.7 months versus 7.9 months with placebo. The IMbrave150 trial, combining the VEGF-A (搜索) antibody bevacizumab with the PD-L1 (搜索) antibody atezolizumab, delivered a median OS of 19.2 months—the longest reported for any first-line advanced HCC treatment to date—and marked a paradigm shift toward immune checkpoint inhibitor (ICI) therapy.
Subsequent studies have expanded the arsenal. The HIMALAYA trial's STRIDE regimen (a single priming dose of tremelimumab plus regular-interval durvalumab) produced 5-year OS rates of 19.6%, compared with 14.4% for durvalumab monotherapy and 9.4% for sorafenib. The CARES-310 trial reported a median OS of 23.8 months for camrelizumab plus rivoceranib versus 15.2 months for sorafenib. In 2025, nivolumab plus ipilimumab received regular FDA approval as first-line treatment for unresectable or metastatic HCC.
Despite this progress, the review identifies unresolved obstacles: insufficient early diagnosis, high recurrence rates, tumor heterogeneity, an immunosuppressive microenvironment, primary and secondary resistance to targeted therapy, limited liver function reserve, a lack of precise predictive biomarkers, and the toxicity and cost barriers of combination therapy.
Prodrugs as a Conditional-Activation Strategy
Prodrugs are pharmacologically inactive derivatives that undergo enzymatic or chemical conversion in vivo to release the active parent drug. The review argues that HCC's TME—characterized by weak acidity, hypoxia, elevated reactive oxygen species (ROS) and glutathione (GSH), dysregulated enzyme expression, and metabolic reprogramming—offers a potential "switch" for selective drug activation.
The review classifies activation mechanisms across several categories. Acid-sensitive prodrugs exploit the tumor's extracellular pH of 6.5–6.9 (versus roughly 7.4 in normal tissue), using hydrolytic bonds such as hydrazone, acetal/ketal, and cis-aconityl linkages. Reduction-sensitive prodrugs rely on GSH-mediated cleavage of disulfide or diselenium bonds, while oxidation-sensitive prodrugs leverage elevated ROS through cleavable linkers including boronic acids, thioketal, and diselenides. Hypoxia-activated prodrugs (HAPs) are reduced by enzymes such as nitroreductases and quinone oxidoreductases in oxygen-deprived regions. Enzyme-responsive prodrugs target overexpressed cathepsin B (搜索), matrix metalloproteinase 2/9 (MMP2/9 (搜索)), γ-glutamyltransferase (搜索) (GGT), and β-glucuronidase (搜索) (GUS).
Early Clinical Signals and Preclinical Examples
The most advanced clinical evidence centers on tirapazamine (TPZ), a prodrug activated exclusively under hypoxic conditions. A first-in-human trial of TPZ chemoembolization in patients with intermediate unresectable treatment-naïve HCC reported a complete response (CR) rate of 60% within the targeted tumor field, a duration of response exceeding 12 months, and a median OS of 52 months. A separate Phase I study in an exclusively Asian population with broader inclusion criteria achieved a CR rate of 47.1%, an objective response rate of 64.7%, a median time to progression of 12.6 months, and a median OS of 29.3 months.
Evofosfamide (TH302), a nitroimidazole-based HAP that releases the DNA cross-linking agent bromoisophosphoramide mustard, was evaluated in a Phase IB study (NCCTG N1135) in combination with sorafenib, establishing a recommended Phase II dose with survival outcomes comparable to other second-line therapies.
Preclinical systems illustrate the breadth of the approach. The acid-responsive His–CA prodrug conjugated cinnamaldehyde to histidine, achieving 72.81% tumor inhibition in H22 tumor-bearing mice while prolonging circulation (half-life 2.55 hours). The GSH-cleavable FU–SS–IND conjugate linked 5-fluorouracil to an indoleamine-2,3-dioxygenase inhibitor, extending survival in 5-FU-resistant HCC-bearing mice and enabling synergistic chemo-immunotherapy with PD-L1 (搜索) blockade. The cathepsin B (搜索)-cleavable doxorubicin prodrug PDOX retained tumor-growth inhibition while reducing toxicity measures in an orthotopic HCC model.
The Central Challenge: Tumor Selectivity Against Diseased Liver
A defining theme of the review is that the proposed activation triggers are not tumor-exclusive. Acidification can occur in fibrotic liver, oxidative stress accompanies chronic liver injury, cathepsin B (搜索) participates in fibrosis and cirrhosis, GGT rises with liver injury, and lactate accumulation or lactylation can occur in activated hepatic stellate cells and fibrotic liver.
This biological overlap is underscored by conflicting data. One study found a twofold increase in GSH content in HCC tissues versus normal liver, while a separate LC-MS analysis of paired specimens found GSH significantly lower in tumors than in adjacent non-tumor tissue. Similarly, a non-invasive CEST MRI study of 15 patients confirmed lower pH in tumor tissue, but its small sample size and lack of stratification of adjacent tissues by fibrosis or cirrhosis status limit its conclusions.
"Evidence of trigger overlap alone does not prove off-target conversion; the risk must be tested by measuring intact prodrug, released active drug, and pharmacodynamic effects in paired tumor and diseased liver," the authors write. They emphasize that tumor selectivity "cannot be inferred from stimulus responsiveness or tumor accumulation alone."
Toward Quantitative, Biomarker-Guided Development
The review outlines three priorities for the field. First, quantitative, multi-input activation and biomarker-guided design, given that no single endogenous signal is tumor-exclusive. Second, disease-relevant preclinical models that reproduce underlying liver injury, fibrosis, or cirrhosis, hepatic immunity, and HCC heterogeneity—rather than tumor growth alone. Third, mechanism-based combinations that remain chemically and operationally feasible.
The authors also flag regulatory and manufacturing hurdles, noting that the applicable pathway—such as 505(b)(2) in the United States or the modified-new-drug framework in China—does not remove the need to demonstrate added clinical value. They caution that greater chemical complexity "is justified only when multi-input activation or combination therapy produces a measurable improvement in benefit–risk and remains manufacturable."
Ultimately, the review positions TME-responsive prodrugs as "potential adjuncts to, rather than replacements for, established clinical modalities," whose translational value will depend on achieving consistent activation across heterogeneous HCC lesions and adequate discrimination between tumor tissue and the surrounding diseased liver.