Upper GI and hepatobiliary cancers — hepatocellular carcinoma (screening, staging, transplant), gastric cancer (lymphadenectomy), pancreatic cancer (Whipple)

Definition / Overview

Hepatocellular carcinoma (HCC) is the most common primary malignancy of the liver and represents a major global health burden, ranking among the leading causes of cancer-related death worldwide. It arises from hepatocytes and occurs predominantly in the setting of chronic liver disease. The disease is characterised by a tightly linked relationship between the underlying hepatic dysfunction, tumour biology, and the limited therapeutic options imposed by portal hypertension and impaired liver reserve.

Epidemiology and Risk Factors

• Chronic hepatitis B, the dominant aetiology globally; risk exists even without cirrhosis
• Chronic hepatitis C, predominant cause in Western countries; cirrhosis is almost invariably present
• Alcohol-related cirrhosis, significant independent risk, compounded by co-existing viral hepatitis
• Non-alcoholic steatohepatitis (NASH), increasingly prevalent cause, particularly in metabolically obese patients
• Aflatoxin B1 exposure, potent hepatocarcinogen; prevalent in sub-Saharan Africa and Southeast Asia
• Haemochromatosis, primary biliary cholangitis, Wilson disease, less common but recognised associations
• Alpha-1 antitrypsin deficiency

Tumour Markers

• Alpha-fetoprotein (AFP): the most widely used serum marker; sensitivity for HCC is modest (~60%), with significant false-positive elevation in active hepatitis C, chronic hepatitis flares, and non-hepatic malignancies (gastric, gonadal germ cell tumours). A level $> 400\,\text{ng/mL}$ is highly specific.
• Des-gamma-carboxyprothrombin (DCP / PIVKA-II): an abnormal prothrombin produced in HCC; useful as a complementary marker, particularly in the transplant setting.
• Neither marker alone is sufficient for diagnosis; both are used in conjunction with imaging.

Screening and Surveillance

Rationale

Surveillance fulfils the established criteria for a screening programme: HCC has a recognisable preclinical phase, arises in a defined at-risk population, and early detection confers a survival advantage through access to curative therapies. Evidence demonstrates that surveillance in cirrhotic patients improves detection of early-stage disease and reduces cancer-specific mortality.

Target Population

Surveillance Modality

• Abdominal ultrasound (USS) every 6 months is the accepted standard, practical, non-invasive, and cost-effective.
• AFP alone is insufficient as a sole screening tool due to limited sensitivity and specificity.
• Surveillance with USS + AFP is superior to USS alone in detecting early lesions.
• MRI or CT are not recommended for routine surveillance but are used for characterisation of detected nodules.

Diagnostic Algorithm for a Detected Nodule

1. Nodule $< 1\,\text{cm}$: repeat USS in 3-4 months; if stable for 2 years, return to 6-monthly surveillance.
2. Nodule $\geq 1\,\text{cm}$: proceed to dynamic contrast-enhanced CT or MRI.
3. Hallmark imaging feature of HCC: arterial phase hyperenhancement followed by portal venous or delayed phase washout, this pattern on a single quality dynamic study is diagnostic without biopsy in cirrhotic patients.
4. If imaging is indeterminate: second-line dynamic study with the alternative modality (CT vs. MRI), or contrast-enhanced USS.
5. Biopsy is reserved for cases where imaging remains inconclusive; risks include false-negative sampling and, rarely, needle-tract seeding.

Staging

Staging in HCC must account for three interacting domains: tumour burden, hepatic functional reserve, and performance status. No single anatomic staging system captures all three.

AJCC/UICC Pathological Staging

Used for patients who undergo surgical resection; the only system validated in surgical cohorts.

• Microvascular invasion (MVI) is a critical pathological finding, it upstages prognosis and influences recurrence risk. Notably, MVI has minimal prognostic impact in tumours $\leq 2\,\text{cm}$.

Barcelona Clinic Liver Cancer (BCLC) Staging System

The most widely endorsed system for treatment allocation, incorporating tumour burden, Child-Pugh score, and Eastern Cooperative Oncology Group (ECOG) performance status.

• Candidates for resection are confined predominantly to BCLC 0 and A, though select BCLC B/C patients with good liver function may be resectable at experienced centres.

Child-Pugh Classification

Underpins hepatic functional assessment for staging and operative risk.

• Child-Pugh A (5-6): well-compensated; suitable for resection
• Child-Pugh B (7-9): intermediate; transplant preferred over resection
• Child-Pugh C (10-15): advanced decompensation; only transplant or palliation

Surgical Resection

Patient Selection

Optimal candidates for resection share these features:

• Preserved hepatic function: Child-Pugh A, no clinically significant portal hypertension
• Portal hypertension defined as: hepatic venous pressure gradient $\geq 10\,\text{mmHg}$, platelet count $< 100 \times 10^9/\text{L}$ with splenomegaly, or oesophageal varices
• Adequate future liver remnant (FLR): the volume of liver remaining after resection must be sufficient to sustain function
• No extrahepatic disease or macroscopic vascular invasion (relative contraindication)
• ECOG performance status 0

Future Liver Remnant (FLR) Thresholds

$$\text{sFLR} = \frac{\text{FLR volume (CT)}}{\text{standardised total liver volume}} \times 100\%$$

Standardised total liver volume is estimated from body surface area: $$\text{Total liver volume} \approx -794 + 1267 \times \text{BSA}\,(\text{m}^2)$$

Portal Vein Embolisation (PVE)

Used to induce hypertrophy of the FLR when it is below the safe threshold:

• Right PVE ± segment IV embolisation for extended right hepatectomy
• Assessed by CT volumetry 3-6 weeks after embolisation
• Degree of hypertrophy (DH) $< 5\%$ after PVE suggests inadequate hepatic reserve and predicts post-hepatectomy liver failure
• Contraindications include: main portal vein thrombosis, uncorrectable coagulopathy, inadequate access

ALPPS (Associating Liver Partition and Portal vein ligation for Staged hepatectomy)

• Stage 1: portal vein ligation + in-situ liver splitting
• Stage 2 (7-14 days later): completion hepatectomy after rapid FLR hypertrophy
• Achieves more rapid and pronounced FLR growth than PVE alone
• Associated with higher perioperative morbidity and mortality; reserved for selected patients at high-volume centres
• May rescue patients who fail to achieve adequate FLR with PVE alone

Principles of Resection

• Anatomical resection (removal of the portal territory containing the tumour) is oncologically preferable to non-anatomical resection, as it removes the portal drainage zone likely to harbour microsatellite deposits.
• A tumour-free resection margin is mandatory; a margin $> 1\,\text{cm}$ is ideal, though even a close negative margin confers benefit.
• Intraoperative ultrasound is essential to define tumour margins and identify satellite lesions.
• Laparoscopic hepatectomy is safe and feasible for peripheral tumours (segments 2, 3, 4b, 5, 6) in experienced hands; associated with reduced blood loss, shorter hospital stay, and preserved future transplant options.

Post-hepatectomy Liver Failure (PHLF)

• Defined by the International Study Group of Liver Surgery as an impaired ability to maintain synthetic, excretory, and detoxifying functions, manifest as elevated INR and bilirubin rising on postoperative day 5 or beyond.
• Risk factors: excessive FLR removal, underlying cirrhosis, cholestasis, portal hypertension, inadequate PVE response.
• Management: supportive; early recognition and escalation to hepatology/transplant team.

Liver Transplantation

Rationale

Transplantation uniquely treats both the HCC and the underlying cirrhotic liver, eliminating the risk of de-novo HCC from the diseased remnant. It is the preferred option for patients with HCC and significant hepatic dysfunction who would not tolerate resection.

Milan Criteria (Mazzaferro, 1996)

The internationally accepted benchmark for transplant listing:

• Single nodule $\leq 5\,\text{cm}$, OR
• Up to 3 nodules, each $\leq 3\,\text{cm}$
• Absence of macroscopic vascular invasion
• No extrahepatic disease

Outcomes within Milan criteria: overall 5-year survival exceeds 70-75%; recurrence rates are low ($< 15\%$).

Extended Criteria

Several expanded criteria have been proposed and validated:

• The UCSF criteria expand the eligible pool without significantly compromising survival outcomes.
• AFP trajectory (AFP slope) over time on the waiting list is emerging as an important dynamic predictor: a rising AFP despite locoregional therapy suggests aggressive tumour biology and poor post-transplant outcomes.

Bridging and Downstaging Therapy

• Bridging therapy maintains patients within criteria while awaiting a donor organ: TACE (transarterial chemoembolisation), radiofrequency ablation (RFA), or microwave ablation.
• Downstaging aims to reduce tumour burden to within transplant criteria: TACE is the most commonly used modality.
• Successful downstaging to within Milan followed by a period of stability ($\geq 3$ months) confers acceptable post-transplant outcomes.

UNOS Priority Allocation

• Patients with HCC within Milan criteria receive MELD exception points to account for the risk of tumour progression during the waiting period.
• MELD exception is not automatically granted; it requires confirmed HCC diagnosis and documentation of lesion characteristics.

Contraindications to Transplant for HCC

• Tumour burden beyond accepted criteria with unfavourable biology
• Active extrahepatic disease
• Macrovascular (main portal vein or hepatic vein) invasion
• AFP persistently $> 1000\,\text{ng/mL}$ (surrogate for aggressive biology at many centres)
• Patient unfit for major surgery (cardiorespiratory, sepsis)

Locoregional and Systemic Therapies

Ablation

• RFA: preferred for tumours $\leq 3\,\text{cm}$; heat generated by electrical current causes coagulative necrosis. Local recurrence is higher than surgery for larger lesions.
• Microwave ablation: faster ablation cycle; less susceptible to heat-sink effect near vessels.
• Percutaneous ethanol injection (PEI): largely superseded by thermal ablation except in challenging locations.

Transarterial Chemoembolisation (TACE)

• Based on the dual blood supply of HCC (predominantly arterial via hepatic artery; normal parenchyma relies on portal vein).
• Delivers chemotherapy (typically doxorubicin or cisplatin) with embolic material directly to the tumour vasculature.
• Indicated for BCLC B (intermediate) disease, multinodular, preserved liver function, no macrovascular invasion.
• Contraindicated in decompensated cirrhosis (Child-Pugh C), main portal vein thrombosis.

Systemic Therapy

• Sorafenib: oral multikinase inhibitor (VEGFR, PDGFR, Raf); first-line for advanced HCC (BCLC C). Modest survival benefit (~3 months improvement in median OS over placebo).
• Lenvatinib: non-inferior to sorafenib in first-line setting; increasingly used.
• Immunotherapy: atezolizumab + bevacizumab has superseded sorafenib as preferred first-line in eligible patients with good hepatic reserve.

Perioperative Considerations for the Surgical Trainee

Preoperative Assessment

• Full assessment of hepatic reserve: Child-Pugh, MELD score, volumetry, functional tests (where available, e.g. indocyanine green retention rate)
• Nutritional status: cirrhotic patients are frequently sarcopenic; nutritional optimisation reduces complications
• Portal hypertension assessment: endoscopy for varices, USS Doppler for portal flow
• Coagulopathy correction: vitamin K, FFP as needed perioperatively
• Cardiorespiratory fitness: standard surgical risk stratification; consider hyperdynamic circulatory state (high CO, low SVR) of advanced cirrhosis

Intraoperative Considerations

• Pringle manoeuvre (hepatoduodenal ligament occlusion): limits intraoperative haemorrhage; tolerated for up to 60 minutes in a normal liver, shorter in cirrhosis
• Maintain low central venous pressure (CVP $< 5\,\text{cmH}_2\text{O}$) during parenchymal transection to reduce hepatic venous back-bleeding
• Cell salvage controversial in oncological resections due to theoretical tumour dissemination risk
• Monitoring: arterial line, large-bore IV access, consider intraoperative USS

Postoperative Monitoring

• Daily LFTs, coagulation, and bilirubin, rising INR or bilirubin after day 5 signals PHLF
• Bile leak: persistent drain amylase or bilious output; managed conservatively or with ERCP
• Haemorrhage: most common in first 24 hours; low threshold for return to theatre
• Venous thromboembolism prophylaxis: balance between coagulopathy and DVT risk in cirrhotic patients

Key GSSE Summary Points

• HCC surveillance: 6-monthly USS ± AFP in cirrhotic patients; nodule $\geq 1\,\text{cm}$ requires dynamic CT/MRI.
• Diagnosis by imaging alone is valid when arterial enhancement + washout is demonstrated in a cirrhotic liver, biopsy not required.
• BCLC staging is the preferred clinical staging tool; AJCC is used post-resection.
• Resection requires Child-Pugh A, no significant portal hypertension, and adequate FLR (≥40% in cirrhosis).
• Milan criteria (single $\leq 5\,\text{cm}$ or $\leq 3$ nodules each $\leq 3\,\text{cm}$) govern transplant listing; 5-year survival $> 75\%$.
• PVE is indicated when FLR is insufficient; ALPPS is the surgical rescue option.
• AFP trajectory and DCP/PIVKA-II add prognostic information, especially in the transplant waitlist setting.
• TACE is the backbone of treatment for BCLC B (intermediate) disease.
• Advanced HCC (BCLC C) is managed with systemic therapy, atezolizumab/bevacizumab or sorafenib.

Sources

• NCCN clinical practice guidelines in oncology
• RACS position statements & curriculum (Royal Australasian College of Surgeons)