Current status and advances in the treatment of colorectal cancer with liver metastases

Abstract

The incidence of colorectal cancer is gradually increasing, and a majority of patients are diagnosed with distant metastases at the time of initial diagnosis, with the liver being the most common site of metastasis. Unlike most malignant tumors, patients with distant metastases can still achieve favorable prognoses if both the primary tumor and liver metastases are surgically resected. With advances in systemic therapies, many patients with initially unresectable liver metastases from colorectal cancer can undergo systemic treatment to achieve conversion therapy, thereby gaining the opportunity for surgery. However, there is still no consensus on several issues, including the timing of systemic therapy before and after surgery, whether neoadjuvant therapy should be employed, and the choice between simultaneous or staged surgeries. This review aims to systematically describe the current treatment landscape for colorectal cancer with liver metastases and highlight several unresolved controversial issues, providing valuable insights for the diagnosis and treatment of colorectal liver metastases.

Key Words: Colorectal cancer; Liver metastases; Treatment strategies; Cancer; Review

Core Tip: Colorectal cancer liver metastases (CRLM) present a significant clinical challenge, yet surgical resection of both primary and metastatic lesions can yield favorable outcomes. Advances in systemic therapies enable conversion therapy for initially unresectable cases, offering potential surgical opportunities. However, key controversies persist, including optimal timing of systemic therapy, the role of neoadjuvant treatment, and the choice between simultaneous or staged surgeries. This review outlines current CRLM management strategies and highlights unresolved issues to guide clinical decision-making.

INTRODUCTION

Globally, colorectal cancer (CRC) has emerged as a major contributor to the cancer burden, with the second highest incidence of malignant tumors and the fifth highest cancer-related mortality[1,2]. In the world, the incidence and mortality rates of CRC have been increasing annually. The liver is the most common site of metastasis for CRC, with approximately 50% of patients developing liver metastasis occurs in disease progression[3,4]. At the time of diagnosis, 15% to 25% of patients with CRC already have liver metastases (synchronous liver metastases), and another 15% to 25% of patients with CRC will develop synchronous liver metastases after radical resection of the primary CRC lesion[5,6]. Initially, 80% to 90% of patients with colorectal liver metastases (CRLM) are deemed unresectable[7,8]. The median survival of patients with untreated CRLM is approximately 6.9 months, and the 5-year survival rate for unresectable patients treated with systemic therapy alone remains less than 5%[3,5,9]. Studies indicate that without aggressive treatment, the prognosis for CRLM patients is poor, highlighting the need for multidisciplinary approaches to improve outcomes[10].

Given the high prevalence of CRLM and their significant impact on patient prognosis, it is critical to understand the current treatment landscape[11]. Over the past decade, advancements in multidisciplinary approaches, including surgical resection, systemic therapies, and local-regional treatments, have transformed the management of CRLM. However, challenges remain in optimizing treatment strategies, particularly for patients with initially unresectable disease[6]. This section will provide a comprehensive review of current treatment modalities for CRLM, focusing on surgical interventions, systemic therapies, and emerging approaches, as well as exploring current controversies and future directions in the field.

OPERATIVE TREATMENT

Surgical resection remains the optimal approach for achieving long-term survival and the only potentially curative treatment in patients with CRLM, with those attaining complete resection of hepatic metastases (or achieving no evidence of disease) demonstrating a median survival of 35 months and 5-year survival rates of 35%-58%, its prognosis is affected by the following factors including: (1) Whether the primary lymph nodes are positive or not; (2) The number and size of liver tumors; (3) The interval between detection of the primary lesion and detection of the metastasis; and (4) The preoperative level of tumor markers, such as carcinoembryonic antigen (CEA), it was found that patients who had negative lymph nodes in the primary lesion, fewer and smaller number of liver metastases, longer interval between discovery of the primary lesion and detection of the liver metastasis, and lower level of tumor markers before surgery, had a better overall survival (OS). Patients with lower levels of tumor markers had better OS, which will of course be explained in more detail later[12-16]. However, approximately 80% to 90% of patients with liver metastases are deemed unresectable at diagnosis[9,17,18]. As a result, alternative locoregional strategies including thermal ablation (TA) (radiofrequency ablation or microwave ablation), cryoablation, and transarterial embolization/chemoembolization have been increasingly used[19,20]. Despite their adoption, these ablation therapies are associated with limitations, particularly for lesions > 3 cm in size, those with elevated preoperative CEA levels (> 30 ng/mL), or ablation margin widths < 5 mm, where local tumor progression may persist, resulting in a median survival of 36 months and a 5-year OS rate of 31%[21,22]. For some patients with unresectable liver metastases, liver transplantation therapy has also been used in clinical practice in recent years and is expected to improve the prognosis of patients with CRLM, with transplanted patients with unresectable liver metastases from CRC living significantly longer compared to chemotherapy alone[23]. However, before proceeding with surgery, the stage of the primary tumor (clinical or pathological, depending on prior therapy), the timing of liver metastases (i.e., synchronous or heterochronous), the timing of recurrence (in the case of heterochronous liver metastases), the number and size of metastases, the molecular characteristics of the tumor (i.e., RAS/BRAF mutation status and mismatch repair/microsatellite status, at a minimum), the level of preoperative CEA, the response to preoperative systemic therapy (if administered), and the type and amount of prior anticancer therapy are components that should be completely evaluated by our clinicians preoperatively.

Although the survival benefit for patients who undergo surgical treatment is influenced by a variety of factors, surgery remains the best option for CRLM cure at this time.

SYSTEMIC TREATMENT

Advances in systemic therapies have enabled a subset of patients with initially unresectable CRLM to achieve resectability through multidisciplinary treatment. Following conversion resection, both multidisciplinary management (MDT)-managed and non-MDT-managed cohorts showed improved survival outcomes, with the MDT group demonstrating superior median OS (42.0 months vs 37.0 months) and significantly higher hepatic metastasectomy rates compared to the non-MDT group (36.6 months vs 28.4 months)[16,24,25]. Systemic treatment options include chemotherapy, targeted therapy and immunotherapy. Neoadjuvant therapy is used primarily to shrink tumors, increase the likelihood of complete resection, and to assess tumor biology by evaluating response to therapy[3,24]. Additionally, it facilitates the identification of patients who may achieve long-term benefits from subsequent therapies. Neoadjuvant therapy creates a critical “window period” to monitor for the emergence of new unresectable metastases during treatment, thereby avoiding non-curative surgeries[9,26]. However, neoadjuvant therapies also face some specific challenges; for example, oxaliplatin-based regimens may lead to sinusoidal obstruction syndrome, and irinotecan causes hepatic steatosis and steatohepatitis, liver injuries that may increase the risk of complications after hepatectomy[27,28]. While chemotherapy-particularly FOLFOX, FOLFIRI, and CAPEOX regimens-remains foundational for advanced CRC, its objective response rates range between 40%-55%, with limitations in durable efficacy[11,17,29]. Targeted therapies, including anti-epidermal growth factor receptor (EGFR) drugs (e.g., cetuximab) and anti-vascular endothelial growth factor drugs (e.g., bevacizumab) for KRAS wild-type tumors, improved efficacy in combination with chemotherapy, with a conversion rate of 31%, compared with 20% when chemotherapy was used alone[30,31].

However, > 95% of CRCs exhibit proficient mismatch repair (pMMR), fostering an immune-excluded microenvironment marked by diminished cluster of differentiation 8 + T-cell infiltration and reduced programmed cell death ligand 1 (PD-L1) expression[32,33]. Immune checkpoint inhibitors (ICIs), such as pembrolizumab and nivolumab, are efficacious in 5%-10% of patients with defective mismatch repair (dMMR) or high microsatellite instability (MSI-H), e.g., median progression-free survival (PFS) of 16.5 months in the pembrolizumab group vs 8.2 months in the chemotherapy group[34,35]. In contrast, ICIs show limited benefit in pMMR tumors, where tumor mutational burden is typically low, no response was observed in patients with pMMR-MSI-L tumors in the key pembrolizumab study, consistent with the lack of efficacy of immunotherapy in early non-selective patient studies[33,36]. As studies have progressed, novel therapeutic strategies such as regorafenib and trifluridine/tipiracil hydrochloride mixture (TAS-102) have been incorporated into clinical practice, providing salvage options for refractory cases. However, their survival benefit remains modest, with studies reporting a median OS of 7.1 months in the TAS-102 group vs 5.3 months in the placebo group, and 1-year OS rates of 27% and 18%, respectively[37].

It is undeniable that systemic therapy is important for prolonging the survival of patients with CRLM, and it is undoubtedly beneficial to cancer treatment that some patients initially judged unresectable can be converted to resectable by systemic therapy. However, systemic therapy faces multiple drug choices, which increases the demands on clinical practitioners and demonstrates the importance of comprehensive MDT.

CONCURRENT SURGERY AND STAGED SURGERY

The choice between concurrent and staged surgery for CRLM remains a topic of debate. Synchronous resection, benefits include a single general anesthesia, shorter hospital stays, shorter systemic treatments, elimination of the risk of disease progression between surgical stages, and lower hospital costs. For selected patients with low tumor burden, this approach serves as a safe and effective therapeutic strategy, demonstrating comparable OS and disease-free survival to staged resection, with a complication rate similar to that of staged procedures[38,39]. Of note, the study showed a median disease-free survival of 19.1 months in the synchronous resection group compared to 8.8 months in the staged resection group, a difference that may be related to the shorter overall operative time of the synchronous approach[40]. However, in patients with high tumor burden, synchronous resection is associated with a significantly higher incidence of postoperative complications compared to staged resection (synchronous: 34.0% vs staged: 14.8%)[41-43]. However, with advances in surgical techniques and the concept of comprehensive MDT, the long-term survival rate is not significantly different from that of staged surgery, even for patients with a high tumor load.

In contrast, staged surgery, in which the primary tumor and liver metastases are resected in separate surgeries, allows for a more complete assessment of the patient’s response to neoadjuvant therapy and the opportunity to optimize the patient’s physiology prior to the second surgery[44,45]. While staged resection may reduce immediate surgical risks, it prolongs the overall treatment timeline, increases financial burdens on patients, and raises the risk of tumor progression during the interval between surgeries[46,47]. The colorectal-first approach may prevent primary CRC sequelae. However, in patients with a high burden of CRLM, this approach may carry the risk of progression of liver disease, and any complications from a one-stage procedure may delay liver surgery, which may affect overall prognosis. In contrast, the liver-first approach eliminates the risk of liver disease progression; furthermore, the incidence of complications and tumor progression associated with the primary tumor is typically lower. Despite these differences, current evidence suggests that there is no significant difference in long-term oncologic outcomes (e.g., OS) between the two approaches, and there was no significant difference in OS between the synchronous resection group and the staged resection group [hazard ratio (HR) = 1.13][48-50].

Staged surgery (either liver-first or colorectal-first) vs concurrent surgery, each has its own advantages and disadvantages, which should be taken into account in clinical decision-making. When deciding whether to use concurrent or staged treatment, it is important to assess patient comorbidities because patient-level risk factors are usually associated with an increased risk of postoperative complications. Depending on the patient’s characteristics and the complexity of the anticipated surgery, surgeons should routinely assess the risk of perioperative death and complications to select the optimal surgical approach[42,45]. The distribution and number of hepatic lobes of liver metastatic lesions has now been identified as a critical factor in surgical selection. It has been noted that for patients with isolated or single liver lobe multiple metastases, patient survival is mostly similar regardless of the surgical approach. However, for patients with multiple, multilobar metastases liver excellence resulted in better 3-year survival (65.9% vs 60.4% vs 54.4% in the liver-first, colorectal excellence, and concurrent surgery groups, respectively) (P < 0.001)[51].

LOCAL REGIONAL THERAPY

In addition to surgical resection and systemic therapy, advanced localized regional treatment modalities such as stereotactic body radiation therapy (SBRT) have become important in the treatment of metastatic lesions. In patients with CRLM unsuitable for surgery or other ablative therapies, SBRT demonstrated excellent local control and good survival outcomes, with a median PFS of 12.0 ± 4.2 months, and no patient experienced radiation-induced liver disease or grade ≥ 3 toxicity, highlighting the safety of SBRT[52]. Landmark randomized trials, including a phase II study in oligometastatic patients (≤ 5 lesions), demonstrated that integrating SBRT with systemic therapy improved median OS from 28 months to 41 months[53]. In addition, SBRT is able to limit radiation exposure to healthy liver tissue (average dose to the liver is less than 12 Gy), thereby reducing the risk of severe hepatotoxicity to less than 5%, even in patients with abnormal baseline liver function[54]. For patients treated with SBRT, the number, size, and location of CRC liver metastases (especially with luminal gastrointestinal structures, such as the duodenum, stomach, and bile ducts) should be adequately considered in clinical decision-making in order to maximize patient benefit.

TA is also a local curative therapy for CRLM, either alone or in combination with surgical resection, and is a safe and effective treatment for patients with specific small-volume lesions, especially in cases where surgical resection is difficult or where the lesion is small but the resection of the liver is extensive[55]. TA can be used as an alternative to surgical treatment or in combination with resection, as well as a salvage treatment option for recurrence of liver tumors after resection[56].

Interventional therapies are equally critical in CRLM management. Transarterial chemoembolization (TACE), combining cytotoxic agents (e.g., irinotecan or oxaliplatin) with vascular occlusion, has been used in the management of initially unresectable CRLM[57,58]. Moreover, compared with conventional TACE treatment, TACE with degradable starch microspheres showed better therapeutic effect, with a median survival time of 16 months. For the 13 months compared with the conventional TACE group, the survival time was prolonged[59]. Similarly, yttrium-90 radioembolization, a localized β-radiation therapy via intra-arterial microspheres, has been shown to be effective in controlling tumors and providing symptomatic relief in cases of unresectable secondary liver metastases due to CRC[60,61]. A randomized phase II analysis of a single treatment with yttrium-90 resin microspheres in a 5-fluorouracil/calcium folinate standard regimen for patients with CRC liver metastases found that the median OS was significantly longer than that of a single standard regimen (29.4 months vs 12.8 months, P = 0.02)[62]. Hepatic arterial infusion chemotherapy (HAIC) in combination with modern systemic regimens, such as FOLFOXIRI, has shown superior efficacy in phase II/III trials, improving tumor response rates and converting initially unresectable patients to resection, and patients with adjuvant HAIC after resection of CRLM had a superior composite HR for OS than those without adjuvant HAIC (OS HR of 0.77)[63,64]. These strategies are increasingly integrated into conversion protocols, with meta-analyses indicating surgical conversion rates of 22%-57% following interventional therapies (for the patients with previously evaluated unresectable CRLM, the surgical resection rate after conversion is 11%-38%), particularly when combined with biologic agents (e.g., bevacizumab or cetuximab)[65].

Various localized hepatic therapies are generally beneficial in improving the generation rate of patients, and should be selected flexibly in clinical practice according to the actual situation of the patient’s hepatic metastases (e.g., number, size, distribution, and vascular invasion, etc.) in order to maximize the benefit to the patient, and the patient’s economic level should be taken into account in the clinical decision-making process.

IMMUNE AND TARGETED THERAPY

Targeting the immune checkpoint pathway, therapeutic agents such as pembrolizumab and nivolumab (programmed cell death 1 blockers) in combination with ipilimumab (cytotoxic T-lymphocyte-associated protein 4 antagonist) have shown significant efficacy in advanced CRC, and this therapeutic paradigm has been particularly effective in tumors displaying specific molecular features, with optimal efficacy observed in malignancies that display MSI-H or defective DNA mismatch repair mechanisms (dMMR) or defective DNA mismatch repair mechanism (dMMR) malignancies[66,67]. For instance, pembrolizumab achieved Food and Drug Administration approval as first-line therapy for MSI-H metastatic CRC based on the KEYNOTE-177 trial, that preoperative administration combining a single ipilimumab infusion with dual nivolumab doses induces remarkable pathological remission in early colorectal carcinoma, with mismatch repair-deficient (dMMR) malignancies achieving complete response (100%) while mismatch repair-proficient (pMMR) cases showing partial response (27%) within approximately one month of therapeutic intervention[68]. However, many CRLM patients exhibit primary resistance to ICIs, necessitating more in-depth studies of tumor-intrinsic and microenvironmental resistance mechanisms[69]. The immunosuppressive tumor microenvironment (TME) in CRLM is critically shaped by chemokine signaling; the CXCL8-CXCR1/2 axis facilitates immune evasion and metastasis through activation of the phosphatidylinositol 3-kinase/protein kinase B pathway[70,71]. Novel strategies combining ICIs with CXCR2 antagonists or TME modulators [e.g., transforming growth factor-β (TGF-β) inhibitors] show promise for overcoming resistance preclinically[72,73]. Currently, triple chemotherapy (i.e., FOLFOXIRI or FOLFIRINOX) remains the treatment of choice for unresectable/non-ablative CRLM in patients with pMMR/microsatellite stability, RAS/BRAF wild type.

Anti-EGFR (cetuximab) and anti-vascular endothelial growth factor (bevacizumab) agents, when integrated with cytotoxic regimens like FOLFOX or FOLFIRI, remain cornerstone therapies for CRLM[74,75]. Recent research advances have focused on dual targeting of oncogenic pathways and tumor tissue organ remodeling. Preclinical models have shown that TGF-β blockers (e.g., galanisertib) inhibit cancer-associated fibroblast-mediated immunosuppression, thereby enhancing the efficacy of chemotherapy[76]. Early-phase trials are evaluating combinations such as TGF-β inhibitors with PD-L1 blockers to potentiate antitumor immunity in metastatic CRC[77]. Additionally, genotype-directed therapies targeting BRAF V600E (encorafenib) and KRAS G12C (sotorasib) mutations are under investigation in combination with chemotherapy or ICIs[78,79].

Currently, targeted and immunotherapy for CRLM is actively applied in clinical practice, but it still faces challenges, such as existing biomarkers (e.g., MSI/MMR) can only screen part of the beneficiary population, the resistance mechanism is complex, and novel inhibitors still lack reliable clinical data, etc. In the future, we should strengthen the interface between basic research and clinical translation, and accelerate the individualized validation of innovative therapies (Table 1).

Table 1 Summary of treatment strategies for colorectal liver metastases.

Treatment category	Key findings
Surgical resection	Surgical resection is the only potentially curative approach. Patients achieving complete resection show a median survival of 35 months and 5-year survival rates of 35%-58%. Prognostic factors: Primary lymph node status, number/size of liver metastases, interval between primary and metastatic detection, preoperative CEA levels[12-16]. 80%-90% of patients are initially deemed unresectable[9,17,18]
Systemic therapy	MDT improves survival (MDT group: Median OS 42.0 months vs non-MDT 37.0 months)[16,24,25]. Neoadjuvant therapy aims to downsize tumors, assess biology, and select responders, but may induce hepatic injury (e.g., SOS or steatohepatitis)[3,24,27,28]. Chemotherapy ORR: 40%-55%, targeted therapies (anti-EGFR/anti-VEGF) combined with chemotherapy improve conversion rates (31% vs 20%)[11,17,29-31]
Concurrent vs staged surgery	Concurrent surgery: Advantages: Single anesthesia, shorter hospitalization, lower costs. Risks: Higher complication rates in high tumor burden patients (34.0% vs 14.8%)[38-43]. Staged surgery: Advantages: Optimizes patient status, assesses neoadjuvant response. Risks: Prolonged treatment timeline, tumor progression risk[44-47]. No significant difference in long-term survival (HR = 1.13)[48-50]
Local-regional therapy	Stereotactic body radiotherapy: Median PFS 12 months, no severe hepatic toxicity[52-54]. Thermal ablation: Effective for small lesions or combined with surgery[55,56]. Interventional therapies (TACE, Y90, HAIC): Improve conversion rates (11%-38%) and survival (e.g., Y90 median OS = 29.4 months)[57-65]
Immunotherapy/targeted therapy	Immunotherapy: PD-1/CTLA-4 inhibitors effective in dMMR/MSI-H patients (pembrolizumab median PFS 16.5 months vs chemotherapy 8.2 months)[66-68]. Targeted therapy: Anti-EGFR (cetuximab) and anti-VEGF (bevacizumab) remain foundational, emerging combinations (e.g., TGF-β inhibitors) under investigation[74-79]
Prognostic factors	Key predictors: Primary lymph node positivity, > 1 liver metastasis, largest tumor > 5 cm, interval < 12 months, CEA > 200 ng/mL[80-83]. Emerging biomarkers: CtDNA + (increased recurrence risk), KRAS mutations (median OS 26.6 vs 42.5 months), NLR > 4.6 (reduced OS)[84-87]
Future directions	Optimize biomarkers (e.g., TMB, PD-L1) for personalized therapy. Explore combination strategies (e.g., immunotherapy + TME modulators) to overcome resistance. Define optimal timing/sequencing of local-regional and systemic therapies[90,91]

CEA: Carcinoembryonic antigen; MDT: Multidisciplinary management; OS: Overall survival; SOS: Sinusoidal obstruction syndrome; ORR: Objective response rate; EGFR: Epidermal growth factor receptor; VEGF: Vascular endothelial growth factor; HR: Hazard ratio; PFS: Progression-free survival; TACE: Transarterial chemoembolization; Y90: Yttrium-90; HAIC: Hepatic arterial infusion chemotherapy; PD-1: Programmed cell death 1; CTLA-4: Cytotoxic T lymphocyte-associated antigen-4; dMMR: Defective mismatch repair; MSI-H: High microsatellite instability; TGF-β: Transforming growth factor-β; CtDNA: Circulating DNA; NLR: Neutrophil-lymphocyte ratio; TMB: Tumor mutational burden; PD-L1: Programmed cell death ligand 1; TME: Tumor microenvironment.

PROGNOSIS AND FUTURE DEVELOPMENT

Multiple clinicopathological factors influence outcomes in CRLM patients. Key determinants include: (1) Lymph node-positive primary; (2) Number of hepatic tumors (> 1); (3) Largest hepatic tumor size > 5 cm; (4) Disease-free interval from detection of primary to occurrence of metastasis < 12 months; and (5) CEA level > 200 ng/mL[80-83]. Patients having no more than 2 positive criteria had good postoperative OS. In addition, patients with multiple organ metastases other than liver, such as lung and peritoneal, had worse five-year survival, which was halved compared to patients with only liver metastases[81,82]. Circulating DNA (ctDNA) is considered a promising prognostic marker, and ctDNA-positive patients had a significantly shorter postoperative recurrence-free survival compared to ctDNA-negative patients (12.7 months vs 27.4 months)[84]. It has also been reported that after CRLM resection, ctDNA + patients are 2 times more likely to develop RAS + and TP53 co-mutations than their negative counterparts. Within 1 year after hepatectomy, recurrence was significantly higher in ctDNA + than in ctDNA - patients (94% vs 49%)[85].

Through genetic testing we can also learn about, for example, BRAF, KRAS and histologic student growth pattern, which also serves as one of the predictors of five-year OS. Patients with KRAS mutations had worse OS after undergoing surgical treatment compared to patients with wild-type RAS (median OS of 42.5 vs 26.6 months)[86].

Also in patients treated with yttrium-90, the neutrophil-lymphocyte ratio (NLR) has been recognized as a prognostic biomarker, with patients with a high NLR (more than 4.6) having a lower OS compared to those with a normal/low NLR (5.6 months vs 10.6 months)[87].

Emerging biomarkers such as preoperative albumin levels and circulating miRNA-145 overexpression (linked to enhanced metastatic potential via epithelial-mesenchymal transition pathways) may refine risk stratification[88,89].

Future research on CRLM treatment should focus on identifying predictive biomarkers for treatment response, optimizing neoadjuvant and adjuvant therapy regimens, and exploring novel therapies such as immunotherapy and targeted therapy[90]. Furthermore, additional studies are needed to determine the optimal timing and sequence of treatments, as well as the role of combining local-regional therapies with systemic therapies[91].

CONCLUSION

In recent years, considerable advancements have been achieved in managing hepatic metastases originating from CRC, where surgical resection maintains its pivotal role as the primary therapeutic approach for achieving cure. Advances in systemic therapy have expanded treatment options for patients with initially unresectable liver metastases, while local-regional therapies such as SBRT offer non-invasive alternatives for managing metastatic lesions. However, several issues remain unresolved, including the timing of systemic therapy, the role of neoadjuvant therapy, and the choice between concurrent and staged surgery. Further research is needed to address these controversies and improve outcomes for patients with CRLM.