Currently, cold static storage (CS) is still the most commonly utilized method for liver preservation in LT. Meanwhile, due to the shortage of donor livers, extended criteria donor (ECD) liver grafts such as steatotic grafts, livers from older donors and donors after cardiac death (DCD) are utilized to expand the scarce donor pool (17). These ECD liver grafts are more susceptible to hepatic IRI and are linked to inferior post-transplant outcomes (18). Besides, the use of organs from DCD donors with prolonged WIT and severe steatotic donor livers was proved to increase the risk of HCC recurrence post LT (11, 19). In such a context, the limitations of CS become more obvious. As a future direction of graft preservation, there is increasing attention to machine perfusion (MP) due to its ability to preserve, evaluate and recondition such donor livers prior to transplantation (20).

Oldani et al. found that the use of ischemic rat liver graft accompanied by an increased serum pro-inflammatory cytokine profile increased the risk of cancer recurrence (21). Investigators approximated an in vivo normothermic perfusion by reperfusing the liver in vivo (animal alive) for two hours before flushing and retrieving. This method attenuated liver injury, measured by the release of liver enzymes (AST and ALT). Besides, prior-to-retrieval reperfusion decreased the recurrence and growth of HCC after transplantation. These effects were partly attributed to the improvement of serum inflammatory cytokine profile. In clinical practice, normothermic extracorporeal membrane oxygenation (NECMO) or NECMO-based normothermic regional perfusion (NRP) is the closest procedure to reproduce the studied model (22). As a technique that allows the in-situ perfusion of organs with oxygenated blood in DCD donors, NRP can reduce warm ischemia time and improve graft function (23). This investigation further revealed the possible potential of NRP in preventing hepatic IRI-associated HCC recurrence.

Hypothermic oxygenated perfusion (HOPE) is an emerging organ preservation strategy for marginal grafts (20). The use of artificial, cooled perfusion solutions with oxygenation applied in HOPE has been demonstrated to decrease oxidative stress and inflammatory damage in LT (24). To validate the protective effect of HOPE in HCC recurrence, Mueller et al. compared tumor recurrence rate between HOPE-treated DCD and unperfused DBD liver recipients after liver LT for HCC (25). The results showed that HOPE alleviated general inflammation (measured through plasma CRP of the recipient) and recovered liver functions. He also observed a significantly lower recurrence rate in patients who received end-ischemic HOPE treatment (5.7%, n = 4/70) compared to an unperfused DBD cohort (25.7%, 18/70), despite the use of high-risk DCD grafts (P=0.002). Besides, HOPE treatment improved recurrence-free survival and reduced tumor load significantly compared to an external, nonperfused DCD and DBD liver recipient cohort. HOPE before implantation protected the liver from reperfusion injury and subsequently prevented ongoing tissue inflammation and hypoxia, making the environment less favorable for tumor recurrence and metastasis.

Despite the ability to decrease damage and improve outcomes, hepatic IRI cannot be completely avoided using the techniques above. With the innovations of surgical techniques and NMP, the technique named ischemia-free liver transplantation (IFLT) has been created. Its application is able to procure, preserve, and implant liver grafts from DBD donors without stopping the blood and oxygen supply, thus entirely preventing hepatic IRI (26). A recent propensity-matched cohort study showed that IFLT recipients had a higher RFS rate at 1 and 3 years (92.2% and 86.7%) than conventional LT recipients (88.1% and 53.6%), with much lower liver damage (27). Though the oncological benefit of IFLT needs further observation and validation in a larger sample, this technique shows enormous potential not only for completely avoiding IRI but also for preventing associated HCC recurrence. However, this technique is not applicable to LDLT and DCD LT, which occupy the major types of LT.

The close interplay between the intestine and liver in anatomy and function is known as the gut-liver axis. In the liver, two-thirds of total blood flow originates from the portal circulation, which contains gut-derived bacterial products (28). As a “microbial organ”, gut microbiota is involved in the development of many liver disorders through the gut-liver axis (29, 30). During LT and LR, portal vein clamping usually results in venous congestion and hypoperfusion of the intestines, followed by intestinal mucosa damage and increased permeability (31). In these cases, endotoxin/lipopolysaccharide (LPS), a cell wall component of Gram-negative bacteria, could translocate to the portal blood through the impaired intestinal mucosal barrier and intensify hepatic IRI (32).

A recent study has revealed the connection between hepatic IRI and HCC recurrence from a brand-new perspective of the gut-liver axis. The study found that pedicle clamping induced the congestive injury of the bowel wall and subsequent increase of LPS in portal circulation. Upon re-establishment of portal blood flow, bacterial LPS activated the Toll-like receptor 4 (TLR4) signaling pathway in the liver, leading to aggravated IRI, exacerbated inflammatory response, and increased tumor burden (33). In this process, hyperactivation of TLR4 by LPS results in the activation and over-expression of pro-inflammatory transcription factors (TFs) such as nuclear factor-kappaB (NF-κB), activator protein 1 (AP-1) and interferon regulatory factor 3 (IRF3) (34). Increased production of pro-inflammatory cytokines (IL-6, TNF-α, etc.) and chemokines (CXCL1, CXCL10, etc.) intensify hepatic IRI and upregulate the components involved in the progression and metastasis of HCC including angiogenesis cytokines and immunosuppressive cells such as Tregs (35).

The researchers also explored several therapeutic approaches to suppress tumor growth by targeting the gut–liver axis (33). They found that remote ischemic preconditioning (RIPC) induced by brief and repeated sequences of femoral vascular bundle clamping and declamping could prevent liver injury and associated HCC recurrence through dampening small bowel venous ischemia and preventing bacterial translocation. Despite promising animal evidence supporting the protective effect of RIPC on hepatic IRI, the results of several clinical trials are inconsistent, indicating that RIPC’s clinical role remains to be further confirmed (36–39). Besides RIPC, the study also showed that gut decontamination (antibiotics) and pharmacological TLR4 inhibition confer systemic protection against inflammation-mediated accelerated HCC growth. Their results indicated that modulating the gut–liver axis during liver surgery could be a potential target in combating HCC recurrence. Predictably, gut microecological therapy will gradually become an emerging treatment modality in preventing inflammation-induced HCC recurrence. Figure 1 displays the schematic diagram of the gut-liver axis as a target in inflammation-associated cancer recurrence induced by hepatic IRI.

Hepatic IRI is a dynamic process including two related stages: ischemic injury and inflammation-mediated reperfusion injury (40). In the phase of ischemia, hepatocytes are exposed to oxygen deprivation, ATP depletion, and pH decreasement (41). These changes result in the accumulation of reactive oxygen species (ROS), increasement of intracellular calcium concentrations which lead to hepatocyte damage and different cell death programs such as apoptosis, necrosis, ferroptosis and pyroptosis (13, 42). The following reperfusion triggers inflammatory cascade that aggravates hepatocyte injury through multiple mechanisms. In addition to cell death programs and metabolic disorders, innate immune activation plays a major role in this process (5). After sensing damage-associated molecular patterns (DAMPs) released from damaged or dead cells, Kupffer cells become activated through pattern-recognition receptors (PRRs) and release chemokines and cytokines to initiate a pro-inflammatory response (43).

The inflammatory cascade after reperfusion involves multiple inflammatory components and their interactions. Activated Kupffer cells release cytokines like IL-1β and chemokines such as CXCL1 and CXCL2 to activate and recruit neutrophils in damaged site (44). Platelets activated by DAMPs also serve as immune mediators which express CXCL4 and P-selectin and promote the recruitment of monocytes and neutrophils to the inflammation site (45–47). The expression of NKG2D (an activating receptor for NK cells) is significantly upregulated in the reperfused liver (48). Increased NK cells recruited to the liver can not only exacerbate IRI by producing IFNγ but also increase the production of IL-17, which promotes the recruitment of neutrophils (49, 50). Hepatic IRI also elicits a robust adaptive immune response in which CD4+ T cells mediate aggravated inflammatory damage (40).

The balance of pro-inflammatory and anti-inflammatory components determines the outcome of inflammation during IRI (44). Multiple clinical and experimental studies have suggested that intense and continuous inflammation induced by hepatic IRI promoted tumor recurrence via activating cellular signaling for tumor cell proliferation, adhesion, invasion and angiogenesis (8, 51). Hypoxia-inducible factor 1α (HIF1α), matrix metalloproteinase 9 (MMP9), vascular endothelial growth factor (VEGF) and other key molecules play important roles in this process (52).

After the initiation of inflammation, some anti-inflammatory components and repair factors will be recruited into the inflammatory microenvironment to inhibit the progress of inflammation and promote injury repair. Anti-inflammatory cytokines, such as IL-4 or IL-13, and apoptotic cells promote in situ macrophage polarization and reprogramming and then initiate the resolution phase of inflammation (53–55). Some anti-inflammatory neutrophil subtypes exert anti-inflammatory and repair functions through inhibiting cytotoxic T cells (CTLs), promoting macrophage polarization and angiogenesis (44, 56). Besides, regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), endothelial progenitor cells (EPCs), platelets and many other factors also play critical roles in regulating inflammation and its resolution (44, 57, 58).

However, in the setting of oncology, these beneficial components may also result in worsened oncologic outcomes. Some components responding to inflammation have been proved to promote HCC recurrence through different mechanisms, including mediating the immune escape and angiogenesis of tumor as well as causing tumor cells to be more aggressive by triggering tumor cell adhesion, migration, and invasion (13). Targeting key inflammatory components in the tumor microenvironment may suppress tumor recurrence. Compared to machine perfusion, this method provides more targeted treatment for HCC patients and can be applied to both LR and LT. Figure 2 displays the key inflammatory components in the tumor microenvironment and relevant therapies.

Immune escape is a crucial characteristic during tumor growth and metastasis. MDSCs and Tregs have been identified as key immune cell subsets that mediate tumor immune escape (59). Of them, MDSCs are a heterogeneous population of immunosuppressive cells derived from myeloid progenitors, which can be upregulated by inflammatory mediators such as IL-2, GCSF, and so on (57). MDSCs suppress the immune system by various mechanisms including (i) inhibiting CTLs proliferation and activation by increased nitric oxide (NO), nitrotyrosine and ROS secretion, and decreased l-arginine production (ii) inducing immunosuppressive cells like T helper (Th) 17 cells and Tregs (60); Tregs are crucial suppressors of immune responses and are essential for maintaining immunological tolerance and controlling inflammatory diseases (61, 62). In the process of tumor immune escape, Tregs can suppress antigen presentation by DCs, CD4+ Th cell function and promote intratumoral T cell exhaustion through secreting TGF-β, IL-10 and IL-35 (63, 64). Recent findings also demonstrated the crosstalk between Tregs and MDSCs, which found that Tregs regulate MDSCs differentiation and function through TGF-β and the programmed death ligand 1 (B7-H1) (65, 66). In HCC patients, MDSCs and Tregs were found to be significantly elevated in the peripheral blood and tumor (67, 68). Besides, emerging tumor immunotherapy targeting MDSC and Tregs has effectively inhibited the progression of HCC (63, 69).

In addition to immune escape, tumor angiogenesis also plays a key role in tumor recurrence. EPCs are a kind of vascular progenitor with high proliferative potential that are derived from bone marrow (70). Inflammatory endothelial injury can trigger mobilization, homing, and transdifferentiation of EPCs, thereby contributing to the repair of injured endothelium (71). Recent studies have found that EPCs participate in neovascularization during acute ischemic injury (58). However, mobilization of EPCs also occurs in response to low oxygenation during tumor growth (72). They are able to facilitate the release of proangiogenic cytokines and promote vessel incorporation and stabilization (73). Multiple studies have shown that mobilization of circulating EPC results in tumor neovascularization and accelerates tumor growth and metastasis (74, 75). Besides, the inhibition of EPCs mobilization restrained angiogenesis and tumor progression (76).

In the inflammatory cascade triggered by hepatic IRI, chemokines and chemokine receptors are involved in the recruitment of inflammation-responsive components discussed above (77). Selectively targeting these pathways has been heralded as a promising avenue for suppressing these cell subsets. Through a series of clinical analyses, in vitro and in vivo experiments, Liu et al. found that MDSCs recruited by CXCL10/TLR4 during acute phase inflammation played a critical role in tumor recurrence after LT. Targeting MDSC mobilization via CXCL10/TLR4 signaling could not only protect the liver graft from IRI but also reduce tumor recurrence after transplantation (78). Ling et al. identified that post-transplant enhanced CXCL10/CXCR3 signaling in small-for-size liver grafts directly induced EPC mobilization, differentiation, and neovessel formation, which further promotes tumor growth (79). Besides, CXCL10/CXCR3 signaling upregulated at liver graft injury directly induced the mobilization and intragraft recruitment of Tregs, which further promoted HCC recurrence after transplantation (80). Targeting CXCL10/CXCR3 signaling inhibited the mobilization of Treg and EPC, attenuated early-phase liver graft injury, and prevented late-phase tumor recurrence/metastasis after transplantation (79, 80).

In addition to genetic approaches and pharmacological inhibitors, some drugs can also act through the above pathways. FTY720 (fingolimod) is a sphingosine-1-phosphate (S1P) receptor agonist that has been approved by FDA as a treatment for multiple sclerosis (MS) (81). In models of lung, kidney, and liver IRI, FTY720 unfolds a demonstrated anti-inflammatory effect (82–84). Furthermore, FTY720 has exhibited a strong anti-tumor activity in liver cancer, breast cancer, bladder cancer, and so on (85–88). Li et al. found that FTY720 significantly attenuated hepatic IRI and tumor metastasis after LR through the downregulating CXCL10/CXCR3 signaling pathway. In their study, FTY720 treatment reduced the population of circulating EPCs and Tregs and thereby limited tumor angiogenesis and enhanced antitumor immune response (89, 90). A novel oxygen carrier called YQ23 played a similar role in suppressing liver tumor metastasis after major hepatectomy and partial hepatic I/R injury through increasing liver oxygenation and reducing the number of circulating EPCs and Tregs (91).

A number of studies have indicated that proinflammatory cytokines such as E-selectin, VEGF and MMP activated by hepatic IRI promoted tumor invasion and metastasis (92–94). Ligands of PPARγ, such as rosiglitazone can downregulate the expression of proinflammatory cytokines that are associated with tumor metastases through inhibiting the NF-κB signaling pathway (95–97). In an experimental mouse model of hepatic IRI-induced HCC metastasis, rosiglitazone exerts a protective effect on hepatic IRI and significantly inhibits tumor metastasis following that. As reported in this article, the dual action may be attributed to inhibited NF-κB signaling and reduced expression of proinflammatory cytokines in liver (98).

Prostaglandins (PGs) are products of arachidonic acid metabolism via the cyclooxygenase pathway, which are produced primarily by activated Kupfer cells during hepatic IRI (99). PGs exert anti-inflammatory effects by prevention of leucocyte migration and down-regulation of pro-inflammatory cytokines (100). The administration of alprostadil, a synthetic stable form of prostaglandin E1(PGE1), was shown to attenuate hepatic IRI and improve liver graft function (101). In a retrospective clinical study, Kornberg et al. found that treating hepatic IRI with alprostadil reduced systemic inflammation levels and the risk of early HCC recurrence following LT (102).

Besides these mentioned strategies above, glutathione peroxidase 3, thymoquinone, and zinc finger protein A20 also show the potential to be “one stone for two birds” strategies that attenuate hepatic IR injury and prevent tumor recurrence after liver surgery (103–105). However, their inhibition of hepatic IRI-associated recurrence needs further experimental research and clinical verification.