Colorectal liver metastases

Section: Hepatobiliary Sub-section: Liver Curriculum: Curriculum, page 85

Pathology

• Metastatic cascade

Diagnosis

• Based on imaging. Typically no need for biopsy. FNA can be used (low risk of tract seeding)
• History, exam, blood tests e.g. CEA, and characteristic imaging are adequate to substantiate diagnosis

Investigations

• Important for:
  • Detect all mets and exact locations to maximise complete clearance
  • Characterise any benign lesions
  • Provide anatomical information necessary to perform complete and safe resection or ablative procedure.
• Ultrasound
  • Diagnostic sensitivity only 36-61% for lesions measuring 1-2 cm
  • Useful to guide FNA
    • However FNA not used in potentially curative therapy as risk of seeding metastases.
• CT
  • Multiphase – arterial, portal-venous and delayed phases
  • May show enhancing rim in arterial phase, and hypodense on PV phase
• MRI
  • Good to see small lesions
  • T2-weighted, unenhanced T1-weighted, dynamic contrast-enhanced T1 weighted and diffusion-weighted sequences
  • T2-weighted and DWI used for detection and characterisation of lesions e.g. haemangioma, liver cyst
  • With primovist contrast, on delayed phase, mets do not retain any contrast.
  • Higher sensitivity and accuracy compared to CT
• PET
  • More sensitive than USS, CT and MRI
  • Extra-hepatic disease detection
• Diagnostic laparoscopy and laparoscopic USS
  • Used in limited cases. Usually reserved for high risk patients (very high CEA, indeterminate imaging for peritoneal disease, etc.)
  • Used to detect small peritoneal mets or subcapsular liver mets, which can be missed on icross-sectional imaging
  • Addition of LUS may increase detection of small intrahepatic mets
• Liver biopsy
  • If diagnostic uncertainty or palliative setting
  • Small risk of tumour seeding

Resectability

Staging and assessment of resectability

• Criteria for resectability is evolving.

• In 2013 Americas Hepato-Pancreato-Biliary Association (AHPBA), consensus defined resectability as:

  • Tumour can be removed completely (R0 resection)
  • The predicted future liver remnant (FLR) function is adequate (>20% in normal liver)
    • Leaving two adjacent liver segments
  • Extrahepatic sites of the disease are controllable,
  • Primary tumour can be resected for cure

• Three factors

  • Patient
  • Anatomical
    • Ability to perform a R0 resection
    • FLR adequate
    • Adquate inflow and outflow
  • Tumour biology
    • Burden of liver metastases
    • Presence of extrahepatic disease
    • Disease-free interval and/or the response to neoadjuvant therapy

• Assessing potentially resectable disease:

  • Response to neoadjuvant chemotherapy
  • Embolisation – upsize FLR
  • Multimodal therapy w ablation & resection

Operative considerations

Margin status

• Aim for microscopically negative margin of >1 mm

Vessel involvement Main portal vein, common hepatic artery – non resectable disease

Extrahepatic metastatic disease

• Lungs, intra-abdominal lymph nodes and peritoneum most common sites of CRC mets after liver
• Presence of extrahepatic mets associated with poor outcomes.
• With recent advances in surgical technique and systemic medical treatment, hepatic resection can be considered if disease is amenable to surgical resection or long-term oncological control with adjuvant chemotherapy
• Managed in MDM setting, and generally treated with preoperative therapy to help define tumour biology
• Stage IVb and IVc generally incurable
• No survival benefit in resecting pulmonary disease
• Resectable local invasion – diaphragm, adrenal
• Resection of isolated adrenal mets possible

Bilateral CRLM

• Depends on burden and location
• One stage atypical hepatic resection is safe for small and favourably positioned mets
• Hepatic sparing resection
  • Intraoperative USS key to identifying lesions
  • Metastectomy rather than anatomical
    • Does not compromise oncological outcomes provided R0 resection
    • Useful for superficial lesions
  • Additional ablations can be performed
• For extensive bilobar mets:
  • Parenchymal-sparing hepatectomy (PSH)
    • Safe without compromising oncological outcomes
    • Increases potential of salvage repeat hepatectomy for recurrent disease
  • Combination of ablation with repeat PSH
    • Does not compromise disease-specific survival
    • Sometimes associated with decreased blood loss, shorter hospital stay and less morbidity
  • Two stage hepatectomy (ALPPS or PVE)

Future liver remnant

Future liver remnant (FLR)

• 20-25% healthy liver (chemo <12 cycles)

• 30-35% NASH/post chemo/diabetics

• 40% cirrhosis

Calculation

• Functional
  • Biochemical - Child-Pugh Score
  • Imaging - ICG clearance
• Volume

Options

• May need PVE or ALPPS
• Embolisation
  • Portal vein embolization
    • Radiological or surgical occlusion of ipsilateral portal vein
    • Compensatory hypertrophy of contralateral lobe of liver (FLR) – can increase up to 15%.
    • Re-image 4 weeks, aim to operate 6 weeks
  • Hepatic vein embolization
    • Can be performed in addition to PVE
    • Can increase FLR with little addition to morbidity
• ALPPS (Associated liver partition and portal vein ligation for staged hepatectomy)
  • Very morbid procedure – high mortality
  • Stage 1: Ligation of portal vein + insitu split, tumour resections from remnant liver
  • Hypertrophy of remnant liver
  • Stage 2: Resection of the deportalised liver (majority of liver mets)

Resectable

1. Resectable Synchronous

• Resectable
  • Three options
    • Primary first approach
      • Preferred if symptomatic colonic disease e.g. impending obstruction or bleeding
    • Combined
      • Synchronous resection appropriate in certain settings
      • “Large liver, small bowel” or “large bowel, small liver”
    • Liver first approach
      • Benifit
        • Remove most prognostically worse disease first
        • Chemotherapy makes the liver worse
  • Preoperative chemo has potential to identify patients with aggressive tumour biology.
    • Downstage
    • Treat micromets
    • Determine biology
    • But causes liver injury
  • No evidence that any particular sequence is superior provided all treatment can be completed
  • Aim to treat greatest threat first
    • Symptomatic primary – bleeding, obstruction
    • High volume liver disease

2. Resectable metachronous CRLM

• Selection needs to be based on factors such as:
  • Tumour biology
  • Risk of underlying liver insufficiency
  • Difficulty of resection
• Chemotherapy pre or post
  • Advantages:
    • Facilitating resection of large tumours
    • Assessment of tumour response to chemotherapy
  • Disadvantages:
    • Progression of disease (however may avoid futile surgery)
    • Possible increased risk of post-resection complications and liver insufficiency.
    • Increase FLR required
  • Toxicity:
    • Irinotecan: Chemo associated steatohepatitis (CASH)
    • Oxaliplatin: Sinusoidal obstruction syndrome (SOS) – veno occlusive effect
  • Options
    • FOLFOX, FOLFIRI or CAPOX chemotherapy regimens are all acceptable
      • Optimal regimen has not been established
      • Duration is typically 4 – 6 weeks
      • Restage
      • Liver resection delayed at least 4 weeks after completion of chemo.
    • Bevacizumab (VEGFR inhibitor)
      • Improved pathologic response
      • May decrease SOS
    • Cetuximab and panitumumab
      • Should not be used in the perioperative setting for resectable patients given trials demonstrating OS detriment.

Adjuvant chemotherapy

• Goal is to reduce disease recurrence (which can occur in up to 60-70%)
• Most patients have already received systemic chemotherapy
• Not routinely given following R0 resection
• Systemic chemotherapy:
  • FOLFOX, FOLFIRI.
• Hepatic artery infusion (HAI) chemotherapy
  • Not routine due to:
    • Lack of long-term OS benefit
    • Potential hepatobiliary toxicity
    • Surgical complications of pump placement
    • Improved effectiveness of modern systemic regimens

Postop liver failure

• Volume does not equal function
• Symptoms:
  • Jaundice
  • Confusion
  • Hypoglycaemia
  • Coagulopathy
• Mortality from hepatic encephalopathy or bacterial/fungal infections
• International Study Group of Liver Surgery (ISGLS) classification
  • At 5 days
  • Grade A:
    • Abnormal liver function tests (INR, bilirubin)
    • No change in clinical management
  • Grade B:
    • Liver dysfunction requiring non-invasive supportive treatment (e.g., fluid management, diuretics, nutritional support)
    • No need for invasive interventions
  • Grade C:
    • Liver failure requiring invasive treatment (e.g., mechanical ventilation, renal replacement therapy, invasive monitoring)
    • May lead to multi-organ failure and death

Outcomes

• 5 year survival approx. 50% with R0 resection
• Many still develop recurrent disease
• Cure rate approx. 25-30%
• Perioperative factors as predictors of postoperative prognosis

Table: Fong et al. prognostic clinical risk score

Cumulative score	1 yr	2 yr	3 yr	4 yr	5 yr
0	93	79	72	60	60
1	91	76	66	54	44
2	89	73	60	51	40
3	86	67	42	25	20
4	70	45	38	29	25
5	71	45	27	14	14
3 or great suggests resection may not be a good idea

Prognostic factor	Score 0	Score 1
Node-positive primary	Negative	Positive
Disease-free interval	≥ 12 mth	< 12 mth
Number of liver metastases	1	> 1
Size of major liver metastases	≤ 5 cm	> 5 cm
CEA (ng/mL)	< 200 ng/mL	> 200 ng/mL
Predicts recurrence for colorectal cancer patients with liver metastasis after hepatic resection

• CEA > 200

• More than 1 colorectal metastatic lesion

• Largest lesion > 5cm

• Original colorectal cancer was lymph node positive

• Less than 1 year between detection of metastasis and original colorectal cancer diagnosis

• Patients with Fong Score 0-2 are considered “low risk” and may proceed with surgical intervention.

• Patients with Fong Score 3-5 are considered “high risk” and may benefit from close surveillance if surgery is deemed unacceptably high risk, additional preoperative workup/treatment before surgery, or additional postoperative treatment.

Other

• Biomarkers such as KRAS and BRAF V600E have been used as prognostic factors and to personalise treatment in CLRM.
• Some studies have shown associated between KRAS mutation and worse OS and RFS

Surveillance

• Peak recurrence 1.5 – 3 years
• Repeat resectiosn can be performed
• Typically follow colorectal cancer pathway:
  • CEA 3 monthly
  • CT every 1,3,5 years

Unresectable disease (synchronous/metachronous)

• Unresectable conversion therapy
  • Potential to downsize unresectable to resectable
  • FOLFOX and FOLFIRI
    • Two suggested regimens for induction therapy with high response rate
  • Addition of biologic has been recommended in selected patients
  • Delayed at least 4 weeks after completion of chemotherapy
  • Important to note that complete radiological response is not equivalent to complete pathological response (about 40-50% of viable tumour after radiological response)
• MSS, Right-sided and/or RAS or BRAF mutant → triplet (FOLFOXIRI) chemo + anti-VEGF (Bevicizumab)
  • Supported by CAIRO5 trial
• MSS, Left-sided RAS & BRAF wild-type → doublet (FOLFOX or FOLFIRI) + anti-EGFR (cetuximab or panitimumab)
  • Supported by PARADIGM trial
• Immunotherapy first-line
  • Pembro or Ipi + Nivo combo (both recommended 1L by NICE)
  • No particular washout period recommended – little evidence to guide this – unlike 5 weeks for Bev and 3-4 weeks for Cetuxi or Pani Systemic therapies
• Palliative chemotherapy can relieve symptoms, improve QOL and prolong survival.

Table: Most commonly used regimens

Regimen	Irinotecan	Oxaliplatin	Leucovorin	Fluorouracil/capecitabine
FOLFIRI	✓		✓	✓
FOLFOX		✓	✓	✓
XELOX		✓		✓
FOLFOXIRI	✓	✓	✓	✓

• Addition of target therapies
  • Anti- VEGF - Bevacizumab
  • Anti-EGFR- Cetuximab or panitumumab

Liver directed therapy

Hepatic artery infusion chemotherapy

• Delivers high-dose chemotherapy (typically FUDR/floxuridine) directly into the hepatic artery via an implanted pump
• Significant interest especially in the US (Memorial Sloan Kettering experience)
• Adjuvant HAIP post-resection associated with improved OS vs modern systemic chemotherapy alone
• Growing evidence for use in unresectable CRLM for conversion, particularly in patients refractory to systemic therapy
• Complications of:
  • Hepatic artery thrombosis
  • Biliary toxicity
  • Incomplete perfusion of liver
  • Misperfusion to the stomach or duodenum with ulcers

Liver Ablation

• Radiofrequency ablation (RFA) was most common – can open, laparoscopic or percutaneous
  • Surgical resection superior to RFA for patients with potentially resectable CRLM.
• Microwave has become more popular.
  • Uses thermal energy, overcoming RFA’s heat-sink effect and allowing treatment of larger lesions.
• Indications:
  • Can be curative alone, or in combination with resection
  • Efficacy decreases >3cm
  • ESMO guidelines recommend thermal ablation for small CRLM (especially <3 cm) and for recurrence after prior resection
• Caution
  • <5mm from bile ducts – stricturing with inadequate drainage and cholangitis
  • Vessels – heat sink, decreases efficacy
• Evidence
  • The COLLISION RCT (Lancet Oncology, 2024) demonstrated non-inferiority of thermal ablation versus surgical resection for small-size CRLM, with fewer adverse events in the ablation arm
  • Local tumour control rates for RFA/MWA range 50–90%; 5-year survival rates range 20–60%, influenced heavily by tumour size, number, and ablation margins

Transarterial chemoembolization (TACE)

• Delivers chemotherapy (conventional TACE or drug-eluting beads — DEB-TACE) directly to the tumour via hepatic artery, combined with embolization
• In CRLM, generally considered palliative rather than curative intent
• ESMO guidelines: TACE should be used in CRLM for non-curative intent in patients with multiple metastases
• Best used in patients not suitable for ablation or SBRT, or as a bridge to other therapies

Stereotactic body radiotherapy (SBRT) or Stereotactive ablative radiotherapy (SABR)

• Effective antitumour dose of traditional external beam radiation therapy is very toxic to normal liver parenchyma.
• Useful for lesions not amenable for resection or ablation
• Offers non-invasive option for local control, especially in patients who are not fit/suitable for surgery or other treatments. This might include those who have had ablation or surgery previously, who now have local recurrence.
• SBRT vs RFA: similar local control for lesions <2 cm; SBRT may be preferred for larger tumours or lesions in difficult locations (perihilar, subcapsular, periampullary) where ablation carries higher risk
• CRC liver metastases are considered relatively radioresistant vs other primaries — dose escalation appears important for local control
• ESMO SABR review (2025): SABR leads to good LC and survival and is well-tolerated, especially with favourable tumour characteristics

Selective internal radiation treatment (SIRT) also called transarterial radioembolization (TARE)

• SIRT delivers yttrium-90 (Y-90) resin or glass microspheres via the hepatic artery, exploiting the dual blood supply of the liver.
• SIRFLOX, FOXFIRE, FOXFIRE Global randomised trials: Y-90 + chemotherapy did not improve OS vs chemotherapy alone in first-line mCRC, though hepatic progression-free survival improved
• Role is better established in chemotherapy-refractory patients with liver-dominant disease
• Can downsize disease to enable subsequent resection or ablation in selected patients
• Used as a bridge or salvage option, often in heavily pre-treated patients
• Best evidence in liver-dominant, chemotherapy-refractory disease
• Useful for conversion/downsizing, especially prior to surgery or ablation