Chemotherapy for advanced oesophageal cancer improves survival when compared to best supportive care alone. In the UK, the combination of epirubicin, cisplatin and 5FU (ECF) has been used as the standard regime (Box 11.4). A randomized study (REAL-2) to explore the role of oxaliplatin instead of cisplatin and capecitabine instead of 5-FU (EOX regime) showed that toxicity of 5FU and capecitabine were similar; oxaliplatin resulted in increased neuropathy and diarrhoea but reduced renal toxicity, thromboembolism, alopecia and neutropenia when compared with cisplatin. Patients receiving EOX survived longer than ECF (38% vs. 47% at 1-year; p = 0.02).

Palliative treatment of dysphagia

Dysphagia is the predominant symptom which needs palliation. The various methods to relieve dysphagia are:

• Endoscopic dilatation – results in immediate but short lived (2–4 weeks) improvement and hence useful only for those with limited life expectancy.

• Stent – immediate and prolonged improvement. There are two types of stents: uncovered and covered (has risk of migration).

• Local therapies – laser, ethanol injection and photodynamic therapy. Response is short lived.

• Radiotherapy – is delivered by external beam radiotherapy or brachytherapy. Radiotherapy will not result in immediate relief of dysphagia and may cause initial deterioration.

• Palliation of other symptoms follows the principles that of any other cancer (p. 66).

Prognosis and survival

Stage, particularly depth of invasion of the oesophageal wall, is an important prognostic factor. Other prognostic factors include age, performance status and weight loss (>10% in 3 months). Grade and histologic subtypes have no prognostic importance.

The overall 5-year survival of oesophageal cancer is <10%. Less than a third of patients are suitable for surgical resection. Around 80% of patients with stage I disease are alive at 5 years whereas only 15% of stage IIa/III patients are alive at 5 years. The median survival of patients treated with palliative intent is 4 months.

Follow-up

All patients are followed up after curative treatment. The follow-up involves history and clinical examination, 4-monthly for one year, 6-monthly for two years and yearly thereafter. Investigations are done as clinically indicated.

Ongoing studies and new agents

New chemotherapy regimes and drugs are being tested to improve the efficacy of CRT. UK SCOPE 1 trial is testing the addition of cetuximab (an EGFR inhibitor) to standard CRT.

MRC OE05 study is comparing ECF chemotherapy followed by resection versus ECX chemotherapy followed by resection in patients with resectable adenocarcinoma of the oesophagus.

Along with various attempts to improve the loco-regional control of oesophageal cancer, studies are ongoing to assess the benefit of various biological agents. Gefinitib (500 mg daily), a tyrosine kinase inhibitor which acts by blocking epithelial growth factor receptor (EGFR), has been shown to have some benefit in patients with advanced oesophageal cancer (50% 6-month survival). Bortezomib, a proteasome inhibitor, is also being studied along with chemotherapy. Another agent under investigation is a monoclonal antibody, panitumumab (REAL 3 study).

Screening and surveillance

The role of screening is not fully established. In the UK, high-risk patients (tylosis and Plummer–Vinson syndrome) are regularly screened with endoscopy. Routine surveillance of chronic reflux (absolute individual risk of less than 1 in 1000 per annum) is not justified. Patients with Barrett’s oesophagus with mild dysplasia are treated with acid suppression therapy followed by a repeat endoscopy and multiple biopsies at 8–12 weeks. If this remains stable, six-monthly repeat biopsy is recommended. Patients with high-grade dysplasia have a 30–40% risk of having invasive adenocarcinoma and hence full staging is undertaken with a view to oesophagectomy.

Small cell carcinoma of oesophagus

This accounts for 0.5–2.4% of oesophageal cancers. The usual presentation is dysphagia (86%) of 1–3 months’ duration and presents in patients in their late 50s. Treatment is similar to small cell lung cancer with chemoradiotherapy. There are some reports suggesting that addition of surgical resection may improve long-term survival. The reported median survival ranges from 12–18 months.

Gastric cancer

Epidemiology

Gastric cancer is the second commonest cause of cancer death in the world. There are approximately 8200 new patients per year in the UK and over 26,000 per year in USA. Although the overall incidence has decreased significantly, there has been an exponential rise in tumours of the proximal stomach and cardia. The median age at diagnosis in men is 70 years and in women 74 years. It is common in men (male to female ratio 2.3 : 1).

Only 20% of patients present with localized disease and the overall 5-year survival is 15–20%.

Aetiology

The aetiology of gastric cancer is multifactorial. The following factors increase the risk:

• Diet: low consumption of fruits and vegetables and high intake of salts, nitrates and smoked or pickled foods increases the risk.

• Smoking.

• Helicobacter pylori – associated with gastric lymphoma and adenocarcoma (2.8-fold increase).

• Blood group A (20% increased risk for infiltrative type).

• Gastric resection, chronic atrophic gastritis and pernicious anaemia.

• Family history of gastric cancer (2–3-fold increase).

• Genetic syndromes: hereditary non-polyposis colorectal cancer, familial adenomatous polyposis, and Peutz–Jeghers syndrome.

High intake of fruits and vegetables (high in vitamin C and E, and antioxidants) and regular use of aspirin and non-steroidal anti-inflammatory drugs (NSAIDs) are associated with a reduced risk of gastric cancer.

Pathology

Gastric cancer arises through a multistage and multifactorial process. The intestinal type of gastric cancer evolves through the stages of metaplasia to dysplasia to carcinoma. However details of evolution of diffuse type of gastric cancer is not known.

• Adenocarcinoma accounts for 95% of gastric cancers. Histologically these are divided into intestinal or diffuse type. The intestinal subtype usually arises in distal stomach in older patients and the diffuse subtype is common in younger patients. The variant diffuse subtype which extensively infiltrates the stomach wall is known as linitis plastica.

• Primary lymphomas (MALT) are increasing.

• Gastrointestinal stromal tumours (GIST) (p. 261).

• Other tumours include small cell carcinoma, carcinoid and squamous cell carcinoma.

Clinical features

Most patients present with advanced disease. Presenting features are often nonspecific and typically include weight loss, persistent abdominal pain, nausea, anorexia, and early satiety. Dysphagia can occur with GOJ or proximal gastric tumours, whilst distal tumours can produce gastric outflow obstruction. 10–20% patients have haematemesis. Paraneoplastic features include dermatomyositis, acanthosis nigricans, diffuse seborrhoiec keratoses (sign of Leser–Trelat), circinate erythemas and microangiopathic haemolytic anaemia.

Evaluation

• History – detailed history including performance status, recent weight loss and family history.

• Clinical examination:

• General: anaemia, jaundice, cachexia, enlarged left supraclavicular lymph node (Virchow’s node).

• Abdomen: epigastric mass, hepatomegaly, periumbilical nodule (Sister Mary Joseph nodule), mass on anterior rectal wall on per rectal examination (Blumer’s shelf) and ovarian masses (Kruckenberg’s tumour).

Investigations and staging

Evaluation of patients helps to establish histological diagnosis, to assess the extent of disease (stage) and assess fitness to undergo appropriate treatment.

Initial assessment

Flexible oesophagogastroduodenoscopy (OGD) is the diagnostic procedure of choice which permits direct visualization of tumour and biopsy of any suspicious lesions. However, the diagnosis of the diffuse-type gastric cancer can be difficult endoscopically as the overlying mucosa may appear normal.

Double contrast barium studies (Figure 11.5) complement endoscopy.

Figure 11.5 Barium study shows a large circumferential lesion in the body of the stomach (with permission).

Further assessment

Once histological diagnosis is made, staging investigations are necessary to determine the treatment options.

• CT scan of chest, abdomen and pelvis should be undertaken to establish loco-regional (sensitivity of 65–80%) and metastatic staging. High resolution dynamic two-phase multi-slice CT with contrast and water (600–800 ml) to distend the stomach has improved the diagnostic accuracy. The drawbacks of CT scan are that it is not very accurate for staging early cancer and up to 30% of peritoneal metastases will not be detectable.

• Endoscopic ultrasound (EUS) is useful in predicting depth of tumour invasion and helps to guide FNA of suspicious perigastric lymph nodes. The stomach is seen as five layers of alternating bright (hyperechoic) and dark (hypoechoic) bands.

• Laparoscopy – laparoscopy allows direct visualization of the liver surface, peritoneum and local lymph nodes. Laproscopy alters CT staging by up to 40% (up or down-staging) and is recommended in all patients with gastric cancer and GOJ cancer with a gastric component. Biopsy should be taken from all suspicious lesions. Intraoperative ultrasound may further increase the yield of liver metastasis and is useful in assessing lymph nodes. Cytology of peritoneal washings may be undertaken, but its sensitivity is limited by contamination with mesothelial cells.

• PET scan may be useful for establishing metastatic disease and to predict response to neoadjuvant treatment, particularly for intestinal type cancer.

Preoperative evaluation

Includes full blood count, biochemistry, coagulation profile, arterial blood gas, pulmonary function tests and ECG. All patients should undergo nutritional screening prior to surgery. A body mass index of <18.5, body weight <90% predicted, >20% weight loss and a low albumin predict an increased risk of perioperative complications.

Staging

Box 11.5 shows the pathological staging of gastric cancer. Sixteen nodal stations defined by the Japanese Research Society for Gastric Cancer help to define the extent of lymph node dissection in gastric cancer (Figure 11.6 and see below).

Figure 11.6 Upper abdominal nodal stations.

Management of gastric cancer (Figure 11.7)

All patients should be assessed in a multidisciplinary setting. Assessment of the performance status and co-morbidities should be done.

Figure 11.7 Management of gastric cancer.

Early gastric cancer (EGC)

By definition the tumour is limited to the gastric mucosa or submucosa (T1) irrespective of nodal involvement. There is a 10–20% risk of lymph node involvement, depending on the size of the tumour, histologic subtype and presence of submucosal invasion. With treatment the 5-year survival is >90%, whereas if untreated up to two-thirds progress to advanced cancer over 5 years. The treatment options include:

• Endoscopic mucosal resection (EMR) – is appropriate for small well-differentiated polypoidal or raised lesions.

• Gastric resection with lymphadenectomy.

Resectable gastric cancer

All patients should undergo laparoscopy with or without peritoneal washings for malignancy cells prior to open laparotomy to assess the extent of disease and resectability. Surgery is feasible only in less than half of the newly diagnosed patients and only 13–50% patients are cured with surgery.

Surgery

Surgery is aimed at complete removal of the tumour and lymph nodes. When performing gastric resection a 5 cm free margin is required for infiltrative tumours whereas 2 cm may be sufficient for expanding tumours. The pylorus seems to act as a barrier to extension of cancer and hence 2–3 cm surgical margin for pylorus may be necessary.

The extent of gastric resection depends on the size and location of the primary tumour (Figure 11.8).

• Subtotal gastrectomy – for an early or well circumscribed T2 cancer, >2 cm away from GOJ. A 5 cm margin is needed for more infiltrative lesions.

• Total gastrectomy – for tumour located <5 cm from GOJ or the tumour is diffuse with submucosal infiltration.

• Distal gastrectomy – is the option for distal gastric tumours which can be completely excised.

Figure 11.8 Gastric resections and reconstructions.

Lymphadenectomy with recovery of a minimum of 14, and an optimal of 25 lymph nodes is recommended. Based on the extent of lymphadenectomy, dissection is categorized as:

• D0 – gastric resection with incomplete dissection of level 1 nodes.

• D1 – gastric resection with complete dissection of level 1 nodes.

• D2 – gastric resection with complete dissection of level 1&2 nodes.

• D3 – gastric resection with complete dissection of levels1–3 nodes.

• D4 – gastric resection with complete dissection of level 1–4 nodes.

The current UK practice is to perform D2 lymphadenectomy without pancreatico-splenectomy. Laparoscopic gastrectomy and laparoscopy assisted D2 dissection are shown to be promising.

Adjuvant chemotherapy

Meta-analyses showed improved survival benefit with adjuvant chemotherapy (one study reported HR 0.85–95%; CI: 0.80–0.90); but there was no standard chemotherapy regime. Adjuvant chemotherapy is hence not offered except in a clinical trial.

Adjuvant chemoradiotherapy

A randomized trial showed better 3-year survival with post-operative chemoradiation (50% vs. 41%; p = 0.005) compared with surgery alone. However, it is not an accepted standard treatment in the UK and Europe because of the drawbacks of the study such as only 10% patients had D2 dissection, significant radiotherapy error in 30% patients and 30% patients did not complete chemoradiotherapy due to toxicity.

Perioperative treatment

A randomized study (UK MRC MAGIC) showed that three cycles of pre- and postoperative chemotherapy with epirubicin, cisplatin and continuous infusion 5 fluorouracil (ECF) significantly improved 5-year survival (36% vs. 23%; p = 0.009) compared with surgery alone in resectable gastric and lower oesophageal cancers (Box 11.6). This is the standard of care in most of the UK and parts of the Europe (Box 11.6).

Unresectable gastric cancer

The median survival of patients with unresectable non-metastatic cancer is 6 months without treatment. Treatment is aimed at improving the symptoms and quality of life and possible extension of life. The various palliative measures include:

• Bleeding – endoscopic laser photocoagulation or argon plasma coagulation is effective in controlling bleeding. If these measures fail, palliative resection may be considered.

• Obstruction – tumours of the proximal stomach may present with dysphagia whereas distal stomach tumours present with gastric outlet obstruction. Endoscopic stent placement offers effective palliative. However, 15–40% with develop recurrent symptoms. An alternative option is palliative bypass.

• Radiotherapy – is effective in controlling bleeding or obstruction in patients who are unsuitable candidates for other interventions. Radiotherapy is effective in controlling pain.

Systemic treatment of unresectable gastric cancer

Patients with unresectable and/or disseminated gastric cancer should be considered for a clinical trial, and those with good performance status (0–2) may be offered systemic therapy. Randomized trials suggest that chemotherapy may improve survival compared with best supportive care (9–11 months vs. 3–4 months HR 0.39) and quality of life. A meta-analysis showed that combination chemotherapy improves survival by 6 months and the best survivals are achieved by three-drug regimens containing 5-FU, anthracycline and cisplatin. The standard regime used in the UK is ECF (Box 11.6). A recent study (REAL 2) showed that ECF is comparable to EOX. This regimen has the advantage of avoiding the need for continuous infusion of 5-FU.

Other active agents include taxanes, and irinotecan.

Follow-up

There is no evidence that regular intensive follow-up improves outcome and hence symptom driven follow-up is recommended.

Management of recurrence

The role of second-line chemotherapy is not well-established. The commonly used agents include taxanes, irinotecan and oxalipatin. In the majority of patients management is essentially palliative aiming at best supportive care.

Prognosis

Important prognostic factors in resectable gastric cancer include depth of invasion, number of lymph nodes involved and positive resection margins. Estimated 5-year survival by stage is:

• Stage I: 70%.

• Stage II: 40%.

• Stage III: 20%.

• Stage IV: less than 5%.

Newer agents

Monoclonal antibodies and various agents targeting pathways of cellular proliferation are being evaluated. However data are still limited to early phase studies. Examples of molecules under study are bevacizumab (anti-VEGF), cetixumab (anti-EGFR) etc.

Screening

Routine screening is not recommended for gastric cancer. Role of screening of asymptomatic individuals for gastric cancer from countries with high incidence such as Japan shows inconsistent results of benefit.

Gastric stump cancer

By definition this is the cancer developing in the gastric remnant at least 5 years after surgery for benign peptic ulcer disease. Patients after distal gastric resection have a 4–7 fold increase in gastric cancer attributed mainly to gastroduodenal reflux. Treatment is complete removal of gastric remnants with a D2 lymphadenectomy. The pattern of lymph node metastases in these cancers may differ from primary gastric cancer and the prognosis is the same as primary gastric cancer.

Cancers of the liver

Introduction

Cancers of the liver can be primary (5%) or secondary (95%). The most common primary liver cancer is hepatocellular carcinoma (HCC) which accounts for 90% of primary liver cancers. There are more than 3000 new patients diagnosed in the UK. Males are more commonly affected (male: female ratio is 5 : 3). Median age of diagnosis is 64 years.

Aetiology

The risk factors include:

• Chronic liver disease – 80–90% HCC arising in patients with cirrhosis.

• Hepatitis infection – hepatitis B and C viruses account for 75% cases of HCC. HCV accounts for the recent rising trend.

• Alcohol accounts for 10% of cancers in Asia and 20% in Europe and USA.

• Non-alcoholic fatty liver disease (NAFLD) – associated with obesity and type 2 diabetes. 5% patients progress to cirrhosis and have an increased risk of HCC.

• Aflatoxin.

• Metabolic disorders – haemochromatosis, alpha-1-antitrypsin deficiency, and Wilson’s disease.

Pathology

HCC is the most common primary liver tumour and is multiple in >50% cases. They spread locally and invade blood vessels. Commonly metastasize to lungs. The other histologic types include cholangiocarcinoma, sarcomas and lymphoma.

Clinical features

Commonest presentation is an incidental mass on USS in cirrhotic patients. Other presentations include abdominal pain, weight loss, anorexia, fever, ascites, jaundice, and paraneoplastic syndromes (hypercalcaemia and hypoglycaemia). Spontaneous bleeding of HCC can occur in 5–15% patients who present with abdominal pain, haemorrhagic ascites or even shock.

Diagnosis and staging

In a patient with pre-existing cirrhosis, a mass of ≥2 cm has >95% of chance of being malignant. Serum AFP of ≥400 µg/l is diagnostic.

Imaging

Contrast enhanced CT scan and MRI scan are useful in establishing the diagnosis. Both tests have 80% accuracy. The characteristic feature of HCC is a specific vascular profile – an intense contrast intake during the early arterial phase followed by contrast washout during delayed/portal venous phase.

Tissue diagnosis

In patients with cirrhosis, >2 cm nodule with characteristic dynamic profile on one imaging, biopsy confirmation is unnecessary. In patients with a 1–2 cm nodule, if a characteristic vascular profile is seen in two imaging techniques, biopsy confirmation is needed. Lesions of >1 cm with non-specific vascular pattern need biopsy. However, biopsy can be delayed or avoided depending on the timing of definite treatment. Biopsy has the risk of tumour seeding. Nodules of <1 cm in size needs close follow-up.

Staging

There are two methods of staging: TNM and Okuda staging. Barcelona Clinic Liver Cancer group incorporate Okuda staging (Box 11.7) for staging classification and treatment recommendation.

Box 11.7
Okuda staging of hepatocellular carcinoma

Criteria	Positive	Negative
Tumor size*	>50 percent	<50 percent
Ascites	Clinically detectable	Clinically absent
Albumin	<3 mg/dL	>3 mg/dL
Bilirubin	>3 mg/dL	<3 mg/dL

Stage
I	No positive
II	One or two positives
III	Three or four positives

* Largest cross-sectional area of tumour to largest cross-sectional area of the liver.

Prognosis of untreated patients

• Stage I: 8.3 months

• Stage II: 2 months

• Stage III: 0.7 months

Management (Figure 11.9)

Treatment of HCC depends on tumour stage, liver function (Child–Pugh grading; Box 11.8), and performance status.

Figure 11.9 Staging classification and treatment of hepatocellular carcinoma

(Llovet JM, Burroughs A, Bruix J. LANCET 2003;362;1907–1917, fig 5, with permission). PST – performance status test, PEI – percutaneous ethanol injection.

Box 11.8
Child–Pugh grading of severity of liver disease

Stage O and stage A

Treated radically and treatment options include:

• Resection.

• Transplantation.

• Percutaneous tumour ablation.

Resection

Surgical resection is the treatment choice in HCC in non-cirrhotic patients. These tumours generally tend to be solitary and large. The 5-year survival is 50% with less than 1% mortality.

Resection is the appropriate treatment for only a subset of cirrhotic patients with HCC. Absence of portal hypertension and normal bilirubin are key factors in selecting the best candidates for resection. Hence the eligibility criteria include:

• Single tumour

• No jaundice

• No portal hypertension

• No extra-hepatic disease

With appropriate selection of patients a 5-year survival of 70% can be achieved with less than 5% operative mortality. There is no role for any adjuvant treatment after resection.

Liver transplantation

Liver transplantation is a treatment option for HCC in cirrhotic patients. Indications for liver transplantation (only 10% are eligible) include (Milan criteria):

• Solitary HCC <5 cm.

• Up to three HCC all <3 cm.

In these highly selected patients the reported 5-year survival is >70%. Liver transplantation is not an acceptable option for HCC in non-cirrhotic patients because of the need for intense immunosuppression.

Percutaneous ablation

Percutaneous ablation is a treatment option for patients with stage 0 & A disease who are not suitable for resection or transplantation and those who are waiting for transplantation. Ablation is achieved either by:

• Chemical injection (ethanol, acetic acid and boiling saline).

• Thermal ablation (radiofrequency, laser and cryotherapy).

Percutaneous ethanol injection (PEI) and radiofrequency ablation (RFA) are equally effective for tumours of <2 cm whereas RAF is more effective for larger tumours.

Stage B & C

Treatment is palliative and options include:

• Transarterial chemoembolization (TACE) and transarterial embolization (TAE).

• Systemic treatment.

• Chemotherapy.

• Biological agents, e.g. sorafenib.

Chemotherapy results in low response rate and has no impact on survival. Sorafenib, a multi-kinase inhibitor, has been shown to prolong survival in patients with advanced HCC with good liver function compared with controls (10.7 vs. 7.9 months)

Stage D

Treatment is essentially symptomatic and supportive.

Management of recurrence

After surgical resection 70% of cases recur at 5 years. Some of these recurrences are true recurrences (intrahepatic metastases) and some are de novo tumours (30–40%). True recurrences typically occur within 2 years after resection, and are usually associated with vascular invasion, satellite nodules and poorly differentiated tumours. De novo tumours usually occur late.

The treatment options include re-resection (possible in 10–20%), salvage transplantation or palliative treatment.

Prognosis

Prognosis depends on cancer and cirrhosis. 50–60% patients die of cirrhosis, 30% with hepatic failure and 10% with gastrointestinal bleeding.

• Stage 0 & A after radical treatment – 5-year survival 50–75%.

• Stage B – median survival 16 months without treatment and with TACE 3-year survival is 50%.

• Stage C – median survival 6 months without treatment and with palliative treatment <10% survive for 3 years.

• Stage D – 1 year survival <10%.

Screening

Surveillance is offered to patients with cirrhosis and the two most common tests used for screening are serum alpha-foetoprotein (AFP) and ultrasonography.

Tumours of the biliary tract

Tumours of the biliary tract can occur at any point between the sphincter of Oddi and the gallbladder.

Carcinoma of the gallbladder

Introduction

Carcinoma of the gallbladder is the most common malignant lesion of the biliary tract. It is more common in females with a 4 : 1 female:male ratio.

Gallstones are considered one of the important risk factors, although the majority of patients with gallstones never do not develop cancer. 65–90% of patients have associated gallstones. Other reported risk factors include ‘porcelain’ gallbladder, gallbladder polyps of >10 mm diameter, and anomalous pancreaticobiliary duct junction (APBDI).

Pathology

90% of patients with gallbladder cancer have associated dysplasia and carcinoma-in-situ, suggesting a multi-step carcinogenesis. 90% tumours are adenocarcinoma and the remainder are squamous cell carcinoma. Spread of tumour is by local invasion (including liver segments IV and V), through lymphatics and venous blood (to liver).

Clinical presentation

The usual presentation is an incidental finding at cholecystectomy. Other clinical features are that of gallstone disease and unremitting jaundice in some patients.

Investigations and staging

Initial assessment is by ultrasound which may show a complex mass filling the lumen with localized thickening of gallbladder wall. There may be associated liver metastasis, ascites and dilated bile ducts.

• CT – useful in local staging, it can establish infiltration into adjacent tissues and vessels and nodal or distant metastases.

• MRI is useful in identifying invasion of the hepatoduodenal ligament and portal vein encasement, both features suggestive of an unresectable tumour.

• ERCP is not useful in diagnosing gallbladder cancer but can be useful in identifying growth in adjacent intrahepatic ducts or in the common bile duct.

Treatment

Patients with resectable cancer

Surgery is the only curative treatment and no more than 10–30% patients are eligible for surgery. Surgical resection is contraindicated in:

• Multiple liver metastases

• Ascites

• Multiple peritoneal metastases

• Extensive involvement of hepatoduodenal ligament

• Encasement or occlusion of major vessels

• Poor performance status

The surgical options include:

• Simple cholecystectomy

• Radical cholecystectomy (for tumours invading perimuscular connective tissue and beyond – ≥T2) (Figure 11.10)

• Radical cholecystectomy with liver resection – the extent of liver resection is controversial.

• Radical cholecystectomy with extensive node dissection – the role of node dissection is controversial. Para-aortic node dissection does not improve survival.

• Radical cholecystectomy with resection of the bile duct or pancreaticoduodenectomy.

Figure 11.10 Extent of dissection in radical cholecystectomy.

Adjuvant treatment

There is no proven role for adjuvant treatment. A small retrospective study suggests that in patients with completely resected gallbladder cancer adjuvant radiotherapy with 5-fluorouracil chemotherapy may improve survival (5-year survival rate of 64% compared with 33% of historical control).

Patients with unresectable and metastatic cancer

Treatment in this group of patients is essentially palliative. Jaundice is treated with stent placement or biliary bypass.

Palliative chemotherapy

5-fluorouracil is the most extensively studied regime. 5-fluorouracil based combination chemotherapy yields a response rate of 10–38%. Combination of gemcitabine and cisplatin gives a higher response of 61%.

Prognosis

5-year survival of gallbladder cancer is <5%. Stage at diagnosis is the most important determinant of prognosis. Median survival of early stage disease is 7–19 months whereas that of metastatic disease is 2 months.

Carcinoma of the bile ducts

Introduction

Cholangiocarcinoma (CCA), which comprises tumours arising from intrahepatic, perihilar and distal extrahepatic bile ducts, accounts for 3% of all gastrointestinal cancers. There is a slight male preponderance. Peak incidence occurs around 65–70 years.

Aetiology

Risk factors include:

• Primary sclerosing cholangitis (PSC; life time risk 5–20%) – occurs at an earlier age (30–50 years) and tends to present as a diffuse, multicentric tumour.

• Liver fluke infestation (Opisthorchis viverrini, Clonorchis sinensis) – 8–10% cases. These are endemic in Japan and Southeast Asia.

• Chronic typhoid carriers (6-fold increased risk).

• Chronic intraductal gallstones (10% risk).

• Choledochal cysts (life time risk 3–15%) – average age for development of CCA is 34 years. Higher risk on those with anomalous pancreatico-biliary ductal junction and mutations in p53 and Smad-4 tumour suppressor genes and K-ras oncogene.

• Thorotrast (a radiological agent which is no longer used).

• Smoking increases the risk in patients with PSC.

Pathology

CCA can arise anywhere in the biliary tree. 90% of tumours are adenocarcinoma whereas tumours arising from choledochal cysts and or gallstones may be adenosquamous or squamous carcinomas.

Clinical features

The presentation of CCA depends on the anatomical location. 10% tumours arise from intrahepatic ducts, 50–60% from hilar (called Klatskin tumours) and 20–30% from the distal common bile duct. Intrahepatic tumours present as a mass lesion detected by abdominal imaging or with non-specific symptoms, whereas hilar or extrahepatic tumours present with jaundice (>90%), pruritus, weight loss and abdominal pain. Patients presenting with a triad of cholestasis, abdominal pain and weight loss should be evaluated for cancer.

Diagnosis of malignancy in patients with PSC is challenging. The new development of jaundice and a dominant bile duct stricture in an individual with previously stable PSC should raise the suspicion of malignancy.

Diagnosis and staging

Blood tests include raised markers of biliary obstruction. CA19.9 of >100 U/mL has a sensitivity of 89% and a specificity of 86%.

Imaging

• USS is used as a first line test in patients presenting with abnormal liver function tests, jaundice or abdominal pain. USS can differentiate obstructive from non-obstructive jaundice, detect a mass lesion in intra-hepatic CCA, and confirm ascites in patients with advanced disease. Further imaging is always needed to evaluate and stage the disease.

• CT of the chest, abdomen and pelvis is useful in locoregional and metastatic staging.

• MR cholangiography (MRCP) and endoscopic retrograde cholangiopancreatography (ERCP) are useful in accurate local staging.

• MRI is particularly useful for hilar CCA.

• Cholangiography assesses biliary drainage and can be used to obtain a biopsy.

Tissue diagnosis

Patients with resectable tumours do not need preoperative diagnosis whereas tissue diagnosis is needed for unresectable or metastatic disease. Tissue diagnosis can be obtained by biliary cytology, endoscopic ultrasound or CT guided biopsy depending on the location of tumour.

Laparoscopy and laparoscopic ultrasound are useful in identifying patients with superficial liver or peritoneal metastases and thus laparotomy can be avoided.

Staging

The Bismuth and Corlette classification of proximal bile duct cancers are shown in Figure 11.11.

Figure 11.11 Classification and treatment of biliary tumours.

Management of unresectable tumours

Management of unresectable tumours is essentially palliative and supportive. Jaundice is managed with stent or biliary bypass. Other evolving treatments include photodynamic therapy, high intensity focused ultrasound (HIFU), radiofrequency ablation, palliative radiotherapy and drug coated stents.

Palliative chemotherapy with gemcitabine alone or in combination with platinum agents or capecitabine may be considered for fit patients. The reported response rate is 15–40% with median survival of 4–16 months. A recent randomized phase III study (ABC-02) has shown that a combination of cisplatin and gemcitabine yields a better median overall survival compared with gemcitabine alone (11.7 months vs. 8.1 months p < 0.001).

New agents

Growth factor receptors HER-1 and HER-2 are overexpressed in biliary tract cancers. New agents studied include erlotinib (EGFR-1 inhibitor) and lapatinib (EGFR-1/HER-2 dual inhibitor) (p. 370).

Prognosis

CCA has a poor 5-year survival of <5%. The majority (75%) of patients die within 1 year. Surgery is the only curative treatment and outcome depends on completeness of surgical resection. R0 resection has a 5-year survival of 20–40%, R1 has 5–10% and R2 has 0%. The median survival after palliative chemotherapy is 4–16 months.

Pancreatic cancer

Introduction

Approximately 7400 new cases are diagnosed in the UK and 37,000 new cases in USA every year. The peak incidence is in the 65–75 year age group with 60% of patients being older than 65 years of age. The male:female ratio is 1.25 : 1. Only 10–15% patients present with early stage disease amenable to curative surgery, and 4% survive more than 5 years.

Aetiology

Exact aetiology is unknown and the suggested risk factors include:

• Cigarette smoking – increases the risk by two fold and accounts for 30% of cases.

• Increasing age – more than 50% pancreatic cancers occurring in individuals aged 70 years or older.

• Chronic pancreatitis – increases risk by 15–25%.

• Late onset diabetes mellitus.

• Hereditary pancreatitis – autosomal dominant condition, mutation of PRSS1 gene and results in a 70-fold increase.

• Cancer family syndromes – account for 10% of cases. Common syndromes associated with pancreatic cancer include Peutz–Jeghers, familial atypical multiple mole melanoma, familial adenomatous polyposis and Li–Fraumeni syndrome.

Pathology

Ductal adenocarcinoma develops through an adenoma–carcinoma sequence of epithelial preneoplastic lesions called pancreatic intra-epithelial neoplasia (PanIN).

Ductal adenocarcinoma is the most common malignant tumour of the pancreas. There are several variants such as undifferentiated (anaplastic), adenosquamous, signet ring cell carcinoma, and mucinous non-cystic. 65% tumours are located within the head, 15% in the body, 10% in the tail and 10% are multifocal.

Clinical features

Tumours of the head of the pancreas tend to present with obstructive jaundice or acute pancreatitis, but the onset is usually insidious. Tumours of the body and tail tend to present even later and are associated with a worse prognosis. Other presenting features include fatigue, back pain, weight loss, late onset diabetes mellitus, steatorrhoea and duodenal obstruction.

Signs

• General examination – anaemia, jaundice, cachexia, enlarged left supraclavicular node (Troisier’s sign).

• Abdominal examination – abdominal mass, ascites, hepatomegaly, umbilical nodule.

Diagnosis and staging

Initial investigations include abdominal ultrasound and CT scan.

• Abdominal ultrasound – is the initial evaluation which can detect biliary and pancreatic duct dilatations, tumours >2 cm and liver metastases.

• CT scan of the chest, abdomen and pelvis is helpful in local (Figure 11.12A) and distant staging. It can also help to determine the resectability of a tumour with 80–90% accuracy.

Figure 11.12 CT scan and ERCP in pancreatic cancer. Carcinoma pancreas (white arrow) leading biliary obstruction (black arrow) and duodenal obstruction with gastric dilatation (white asterix) (A) and ERCP shows strictures of common bile duct and pancreatic duct (‘double duct sign’) (B).

Further investigations include:

• Magnetic resonance cholangio-pancreatography (MRCP) – useful in assessing cystic tumours.

• Endoluminal ultrasonography (EUS) – useful in detecting small tumours and biopsy of small lesions.

• Endoscopic retrograde cholangiopancreatography (ERCP) – rarely used in the diagnosis except to insert biliary stent to relieve obstructive jaundice and obtain biopsy (Figure 11.12B).

• Percutaneous transhepatic cholangiography (PTC) useful to relieve jaundice when ERCP has failed or not possible.

Tissue diagnosis

Tissue diagnosis is usually obtained by EUS with fine needle aspiration (FNA). The technique has a sensitivity of >90% and specificity of almost 100%. Percutaneous fine needle aspiration (FNA) biopsy is less sensitive (69%) and carries a risk of intraperitoneal seeding (16%).

Blood tests

• FBC, biochemistry and clotting profile.

• Serum CA19.9 is the commonly used tumour marker which has a sensitivity of 70–90% and specificity of 90%. It is useful to assess response and identify tumour recurrence. A level of <200 u/ml suggests longer survival.

Further tests prior to surgery

Laparoscopy with laparoscopic ultrasound alters the management of 15% of patients already assessed as resectable by CT. Selective laparoscopy based on the serum level of CA19-9 and large tumour, is a more efficient strategy, reducing the proportion of patients undergoing laparoscopic ultrasound from 100% to around 45% while increasing the yield from 15% to 25%.

Management of pancreatic cancer

Treatment of resectable cancer

Surgery

Patients with good performance status and resectable tumours are treated with radical resection. Criteria for resectability are given in Box 11.11. Though the aim of surgery is a complete microscopic resection (R0), 30–60% of resections result in microscopic residual disease (R1). Patients with jaundice may be considered for pre-operative biliary drainage. The type of surgery depends on location of the tumour:

• Tumours of head of the pancreas – pancreaticoduodenectomy (classic Whipple) or pylorus preserving pancreatoduodenectomy (pp Whipple) are considered curative. A meta-analysis showed that both procedures carry equal morbidities and survival. There is no survival advantage for extended radical lymphadenectomy in pancreatic cancer.

• Tumours of the body and tail of pancreas – left pancreatectomy which includes splenectomy and en bloc removal of the hilar lymph nodes.

Box 11.11
Criteria for resectable pancreatic cancer

• No coeliac, hepatic or superior mesenteric artery involvement

• A patent superior mesenteric-portal venous confluence

• Portal venous involvement of not more than 2 cm in length or more than 50% circumference

• No liver, peritoneal or other distant metastases

• Absence of portal hypertension and cirrhosis

• No severe co-morbidity to exclude surgery

The postoperative morbidity of radical surgery is around 40% and the mortality is <5%.

Neoadjuvant therapy

Neoadjuvant treatment using a combination of external beam radiotherapy with 5-FU or Gemcitabine based chemotherapy may improve resectability in a borderline resectable tumour. Several studies report a resection rate of 60% and negative resection margin of 90%. However, this is not routinely practised in the UK.

Postoperative treatment

• Adjuvant chemotherapy – a meta-analysis showed that adjuvant chemotherapy improves 5 year survival (19% vs. 12%) and median survival (19 months vs. 13.5 months) compared with no chemotherapy. The recently reported ESPAC-3 study comparing gemcitabine and 5-FU/FA in patients who had R0/R1 resection showed no difference in survival (median survival of 23.0 months with 5-FU/FA and 23.6 months with gemcitabine).

• Adjuvant chemoradiotherapy – a meta-analysis showed that chemo-radiotherapy did not improve survival compared with no treatment (median survival 15.8 months vs. 15.2 months).

• Adjuvant chemoradiotherapy and chemotherapy – RTOG 9704 trial reported no statistically significant survival benefit with gemcitabine to adjuvant 5-FU based chemoradiotherapy.

Unresectable and metastatic cancer

70% of patients will present with unresectable disease and treatment of these patients is essentially palliative. Treatment options include:

• Palliative chemotherapy (for locally advanced and metastatic cancer).

• Palliative chemoradiotherapy (locally advanced cancer).

• Symptom control (poor performance status).

Palliative chemotherapy – single agent response rate is seldom more than 10%. A meta-analysis showed that chemotherapy improves survival compared to best supportive care (HR 0.64; 95% CI, 0.42–0.98). Single agent 5-FU results in a median survival of 5–6 months. Single agent gemcitabine has a better median survival (5.7 vs. 4.4 months, p = 0.0025), 1-year survival (18% vs. 2%) and relatively mild toxicity compared with 5-FU. Hence the current drug of choice is gemcitabine.

A recent meta-analysis has confirmed that gemcitabine combination chemotherapy improves survival compared with gemcitabine alone (HR = 0.91; 95% CI, 0.85–0.97). Capecitabine and platinum are the commonly used drugs with gemcitabine.

Palliative chemoradiotherapy – the role of external beam radiotherapy given concurrent with 5-FU or gemcitabine is not clear. However, a meta-analysis showed that chemoradiotherapy increases survival compared with radiotherapy alone (HR 0.69, 95% CI 0.51–0.94). There was no survival difference between chemotherapy alone and chemoradiotherapy followed by chemotherapy.

Symptom control

• Pain control – measures include analgesics, celiac plexus block or unilateral thoracoscopic splanchnicectomy.

• Obstructive jaundice – managed with biliary stent or surgical bypass. Plastic stents are used for patients with metastatic disease and tumours of >3 cm diameter whereas self-expanding metal stents are used for patients with good performance status and locally advanced disease of <3 cm.

• Duodenal obstruction – treated with endoscopically placed stents or bypass surgery.

New agents

Early studies combining gemcitabine with EGFR inhibitor erlotinib has shown some activity but the exact role of EGFR inhibitors is still evolving.

Prognosis and outcome

The most important prognostic factors are resectability and performance status. Median survival following surgical resection ranges from 11–20 months. Without active treatment, metastatic pancreatic cancer has a median survival of 3–6 months and 6–10 months for locally advanced disease, which increases to around 11–15 months with surgical resection. Five-year survival ranges from 7–25%. Radical resection alone gives a 5-year survival rate of around 10%.

Pancreatic endocrine tumours

Pancreatic endocrine tumours arise from the islet cells of the pancreas and the main types are insulinoma, gastrinoma, glucagonoma and VIPoma. The mean age at presentation is 47 years. Genetics syndromes of MEN1, Von Hippel–Lindau, neurofibromatosis type 1 and tuberous sclerosis are associated with an increased risk of pancreatic endocrine tumours.

Clinical presentation is either due to tumour mass or from excessive peptide secretion.

• Insulinomas present with features of hypoglycaemia associated with fasting or vigorous exercise, and rapidly relieved by eating a snack or drinking a liquid rich in glucose. Diagnosis is based on a very low blood sugar (<2 mmol/l) and a high level of insulin and C-peptide (indicative of endogenous insulin).

• Gastrinoma presents with refractory multiple peptic ulcers and diarrhoea. It is diagnosed by high gastrin levels (>100 pg/ml, or >200 pg/ml after secretin stimulation).

• Glucagonoma presents with features of hyperglycaemia, stomatitis, weight loss, diarrhoea and psychiatric disturbances. Necrolytic migratory erythema of the skin is a characteristic feature. Diagnosis is by high plasma glucagon levels.

• Vasoactive intestinal polypeptide tumour (VIPoma) causes watery diarrhoea and hypokalaemia. Diagnosis is by high plasma levels of vasoactive intestinal polypeptide.

Staging investigations of pancreatic endocrine tumours include CT scan, MRI scan, EUS, ¹¹¹In-octreotide and selective portal/splenic venous sampling and intraoperative ultrasound.

Surgical resection is the treatment of choice which may be indicated even in metastatic disease. 5-year survival following surgical resection is 50–95%.

Cystic tumours of pancreas

These constitute around 15% of all pancreatic cystic masses. The common types are:

• Serous cystic neoplasms predominantly affect women, are found mostly in the head of the pancreas and represent 30% of cystic neoplasms. Conservative approach with regular imaging is the management.

• Mucinous cystic neoplasms are also common in women affecting the body and tail of the pancreas, and represent 40% of primary cystic neoplasms. These are treated with resection to avoid malignant transformation.

• Intraductal papillary mucinous neoplasms (IPMNs) commonly affect men and constitute 30% of pancreatic cysts. IPMNs arising from main duct should be resected whereas those from branch duct may be managed with regular follow-up imaging.

Carcinoid tumours

Carcinoid tumours are neuroendocrine neoplasms arising from enterochromaffin cells. These can be functioning or non-functioning. The mean age at presentation is 49 years and women are more commonly affected.

The clinical presentation of a carcinoid tumour can be a non-specific abdominal symptom, although carcinoid syndrome can occur in 25% of cases. CT is the diagnostic image of choice. Surgery is the treatment of choice for resectable tumours. However, 70–80% patients present with advanced disease. Treatment options for advanced disease include:

• Symptom control using octreotide or a long acting version.

• Radioisotope treatment using ¹³¹I-MIBG or ¹¹¹In/90Y octreotide in patients with positive scans.

• Palliative chemotherapy – streptozosin plus 5-FU/dacarbazine and doxorubicin plus 5-FU.

The overall five-year survival rate is 30–40% and the median survival of patients with metastatic disease is approximately 7 months.

Malignant tumours of the small intestine

Small intestinal malignancies constitute 2–3% of all malignant GI cancers. The ileum is the most frequent site of tumours and is more common in males.

Aetiology

The risk factors include:

• Familial adenomatous polyposis (FAP) – small intestinal adenocarcinoma is one of the common causes of death in patients with FAP after colectomy.

• Crohn’s disease (adenocarcinoma).

• Coeliac disease (adenocarcinoma and lymphoma) – type of lymphoma in coeliac disease is enteropathy associated T cell variant.

• Tropical sprue (lymphoma).

• MEN type-1 (gastrin producing tumour of duodenum and jejunum).

Pathology

Tumours of small intestine can be of epithelial, mesenchymal or lymphoid origin. The following are the common types of cancers occurring in the small intestine:

• Adenocarcinomas (45%) – occur in ampulla of Vater in the duodenum and jejunum.

• Carcinoids or neuroendocrine tumours (30%) – common in terminal ileum.

• Sarcomas (10%) – common in ileum, majority of tumours are GIST (p. 261).

• Lymphomas (15%) – B cell NHL is the most common.

• Metastatic cancers – are commoner than primary and usual primaries are GI tract, breast, uterus, ovary and melanoma.

Clinical features

Clinical presentation depends on the type of cancer, site and size. The majority of patients with small intestinal malignancies present with advanced disease. Nonspecific symptoms include abdominal pain, anaemia, nausea, bleeding, and weight loss. Patients can also present as a surgical emergency with perforation or intussusception. Duodenal tumours can present with obstruction as well as jaundice. Metastatic liver carcinoid can present with carcinoid syndrome. Lymphoma may present with pain, weight loss and features of malabsorption. GIST commonly presents with anaemia, sometimes with an abdominal mass.

Investigations

The initial investigations depend on the presenting symptoms.

Imaging

• Plain X-rays – useful in suspected obstruction.

• Abdominal ultrasound – useful to detect liver metastasis, ascites and biliary dilatation.

• CT scan of chest, abdomen and pelvis can delineate primary tumour and disease extent. Different histological types exhibit different features, e.g. lymphoma appears as diffuse segmental thickening of the small intestine and GIST as a well circumscribed mass.

• Small bowel follow-through.

• Endoscopic ultrasound (EUS).

• Small bowel endoscopy.

• ¹¹¹In-octreotide scan – in carcinoid to rule out metastatic disease.

Tissue biopsy

Biopsy confirmation is essential prior to definitive treatment. Tissue diagnosis can be obtained by endoscopic, CT guided or laparoscopic methods or by laparotomy.

Staging

Adenocarcinoma is staged similar to colonic cancer (p. 163). Lymphomas are staged using the Ann Arbor staging (p. 299).

Management

Small intestinal adenocarcinoma

In patients with localized adenocarcinoma, resection of visible disease and regional lymph nodes is done. In advanced disease, surgery may help with palliation of symptoms.

Chemotherapy regime is same as that of colonic cancer (p. 165). A small study reports a response rate of 50% and a median survival of 20 months with a combination of oxaliplatin with capecitabine.

Small intestinal lymphoma

Stage I and II B-cell NHL are treated surgically with or without chemotherapy, whereas III–IV disease is treated with primary chemotherapy with or without surgical debulking. Chemotherapy regimens are similar to NHL (p. 305).

GIST

Prognosis

Prognosis depends on the type of cancer and stage. Adenocarcinoma usually presents with advanced disease and the reported 5-year survival is 30%. Duodenal tumours have a better prognosis with 5-year survival of 50%.

Lymphomas have a 10-year survival of 60%. The prognosis of small bowel GIST depends on their size and mitotic rate (see p. 262 for NIH prognostic index).

Colorectal cancer

Epidemiology

Colorectal cancer (CRC) is the third commonest cancer in the UK after breast and lung. Over 36,000 new cases per year are diagnosed in the UK and 105,500 per year in the USA.

Over 80% of cases occur in people over 60 and is rare below the age of 40 (except in hereditary forms). The male:female incidence ratio is 1.2 : 1.0. In the UK, the lifetime risk is 1 in 18 for men and 1 in 20 for women. Two-thirds of primaries are in the colon and one third in the rectum. Left-sided cancers are more common than right (sigmoid and rectal cancers account for over half).

Aetiology

The majority of colon cancers are sporadic and not associated with known hereditary genetic mutations. Hereditary bowel cancer is characterized by early age at diagnosis, right-sided cancers, synchronous/metachronous colorectal tumours or other characteristic tumours in same individuals or families (see below). Up to 25% of patients with CRC give a positive family history, suggesting involvement of genetic factors.

• Genetic (p. 50):

• Familial adenomatous polyposis (FAP) accounts for 1% of all cases of CRC.

• Hereditary non polyposis colorectal cancer (HNPCC) accounts for about 5% of all CRC cases.

• Familial – those with an affected first-degree relative are at increased risk of developing CRC (relative risk 2.25).

• Inflammatory bowel disease – increases the risk. In ulcerative colitis the cumulative risk of developing CRC is estimated to be 8% at 20 years and 18% at 30 years.

• Diet and supplements – high intake of red meat, calorie and high body mass index increase the risk whereas high vegetable intake and high-fibre diet are protective.

• Drugs – females on hormone replacement therapy and those using regular aspirin are at low risk of developing CRC.

• Lifestyle – long-term smokers are at high risk of adenomas and CRC mortality. Regular physical activity reduces the risk of CRC, particularly in men.

Pathogenesis and pathology

Almost all colorectal cancers are adenocarcinomas and arise in adenomatous polyps through a multi-step process (Figure 11.13). Most adenocarcinomas are moderate to well differentiated with typical morphology. About 80% of primary colorectal adenocarcinomas and 70% of metastases will be CK7− CK20+. Rare histologic types include squamous cell carcinoma, small cell carcinoma, adenosquamous carcinoma and medullary carcinoma.

Figure 11.13 Pathogenesis of colorectal cancer.

Presentation

About 85% of patients with bowel cancer have these symptoms at the time of diagnosis. Approximately one-third presents with altered bowel habit or obstructive symptoms (usually left sided tumours as the distal bowel is narrower and faecal content more solid) and one-third with iron deficiency anaemia (usually right-sided tumours). For rectal cancers other symptoms include faecal incontinence, passage of mucus and tenesmus. Up to 30% present with acute colonic obstruction. More advanced tumours may cause weight loss, nausea and anorexia, and abdominal pain and distension from ascites or hepatomegaly.

Investigations and staging

Initial investigations include double contrast barium enema (DCBE), sigmoidoscopy (allows visualization up to splenic flexure at 60 cm) and colonoscopy (Figure 11.14A & B). The whole colon should be imaged as approximately 5% of patients have synchronous cancers and up to 40% have synchronous adenomas. CT colonography is an alternative DCBE (Figure 11.14C). Endoscopy also allows biopsy confirmation.

Figure 11.14 Double contrast barium enema shows a polypoid cancer with an irregular indrawn base (A), colonoscopy showing T1 high rectal cancer (B) and CT colonography showing an annular carcinoma (C).

Parts A and C from Adam A, et al: Grainger & Allison’s Diagnostic radiology, 5th Edition, Volume 1, 2009 (Elsevier), with permission.

Those patients presenting with suspected intestinal obstruction should have cross-sectional imaging preoperatively unless there are signs of peritonitis when emergency surgery will take precedence.

Further investigations

• Blood – full blood count, liver function tests, U&E and serum carcinoembryonic antigen (CEA). The majority of patients will have an elevated CEA preoperatively, and this can be used to detect early recurrence in patients who may be appropriate for further curative surgery. It is also useful in monitoring response to treatment in recurrent and metastatic disease. It is less likely to be elevated in poorly differentiated carcinomas.

• All patients with confirmed CRC need CT scan of the chest, abdomen and pelvis for staging.

• Patients with rectal cancer need locoregional staging with pelvic MRI (Figure 11.15); including T2W image MRI helps to define the extent of the tumour and to identify mesorectal margin invasion by tumour which helps to decide need for neoadjuvant treatment.

• If MRI suggests a small (<3 cm) early T1 lesion endorectal ultrasound is useful to access feasibility of local excision and exclude tumour penetration into the muscle of the bowel wall (T2 disease) with an accuracy of 87%.

• PET scan is not routinely used in CRC but may be useful in the context of a rising CEA or clinical suspicion of metastatic disease or local recurrence, where CT is negative or equivocal and confirmation of relapse would alter management. PET scan may also have a role in potentially resectable metastatic disease such as lung to rule out extensive metastasis.

• In patients with resectable or potentially resectable liver metastasis, MRI of the liver is done to exclude more extensive disease prior to perioperative chemotherapy (Figure 11.16).

Figure 11.15 T2W MRI of mid rectal cancer shows threatened mesorectal margin.

Figure 11.16 CT scan of liver in portal venous phase shows resectable liver metastases (A) and MRI liver with hepatocyte specific contrast reveals more extensive metastases (B).

Staging

Figure 11.17 shows TNM and Dukes’ staging which are postoperative staging. After radical resections, 12 or more nodes should be examined as this correlates with prognosis for Dukes’ B cancers. Pathology reports should also inform about presence or absence of tumour perforation, extramural vascular invasion, radial margins, lymph nodes (number examined and involved) and assessment of completeness of resection.

Figure 11.17 Staging for colorectal cancer.

Management of colon cancer (Figure 11.18)

Stage I (Dukes’ A)

• Surgery

Stage II (Dukes’ B)

• Surgery

• Consider adjuvant chemotherapy in patients with PS0-2 if:

• high risk factors (vascular invasion, peritoneal involvement or T4 tumours, perforation, poorly differentiated adenocarcinoma and surgical margins with tumour cells).

• or young age (<70 years).

Stage III (Dukes’ C)

• Surgery.

• Consider adjuvant chemotherapy in all patients with PS 0–2.

Stage IV

• Consider radical surgery with perioperative chemotherapy if resectable liver +/− lung metastases.

• Palliative chemotherapy

• Palliative radiotherapy

• Other palliative treatments

• Active symptom control

Figure 11.18 Management of colon cancer.

Surgery

Surgery is curative and several studies have shown equal survival with better cosmesis and shorter hospital stays for laparoscopic versus open surgery for colectomies. Emergency surgery has higher perioperative mortality than that of elective surgery. Hence emergency stenting may be useful to temporarily decompress an obstructed lesion prior to definitive surgery.

Adjuvant chemotherapy

Table 11.1 shows current recommendations for adjuvant chemotherapy. There is no benefit of adjuvant chemotherapy in Dukes’ A disease. In stage III (Dukes’ C) disease, 5-fluorouracil/folinic acid (5FU/FA) chemotherapy gives an absolute survival benefit of between 8–13%, with 6 months chemotherapy as effective as 1 year. In stage II disease, chemotherapy results in a similar proportional reduction (approximately 33%) in recurrence, but absolute benefits are less because of lower mortality rates (QUASAR 1 showed an absolute increase in 5-year OS of 3.6% for patients less than 70 years). Patients with stage II disease with adverse features have a higher disease-related mortality rate (with >1 risk factor this approaches that of Dukes’ C cancers) and hence absolute benefits of adjuvant chemotherapy are greater.

Table 11.1 Chemotherapy regimens for colorectal cancer