Colorectal Cancer

Meet the patient

A 67-year-old man lands in your clinic with a three-month story of looser, more frequent stools, a dragging left iliac fossa ache, and — twice last week — bright-red blood mixed into the motion. He has lost 4 kg. The GP sent him because his haemoglobin is 98 g/L, MCV 76, ferritin 8: a textbook iron-deficiency anaemia. He booked nothing for months because he decided it was "just piles."[1]

The single question that settles his next two weeks is the question that settles every colorectal-cancer stem: does a normal colonoscopy to the caecum explain these symptoms or not? Until that scope is done and reaches the caecum, the diagnosis stays open — and bright-red rectal bleeding mixed into the stool of a 67-year-old is cancer until proven otherwise.[1]

Right bleeds, left obstructs, rectum tenesmuses — the one spine of CRC

Site decides symptoms, and symptoms decide where you look. The right and left colon differ in embryology (midgut versus hindgut), luminal calibre, stool consistency and molecular biology — so a right-sided tumour behaves nothing like a left-sided one, and the rectum is its own problem again. Hold this triplet and the rest of the topic slots into place.[1]

The face-off a viva examiner wants reproduced verbatim — one discriminator line beneath each:[1]

[1]

The classic trap: bright-red blood on the surface of the stool and the toilet paper is haemorrhoidal; bright-red blood mixed into the stool is not. The discriminator is the stool, not the colour — and the move that costs patients is labelling any adult's rectal bleed "piles" without a proctoscope and a colonoscopy.[1]

APC, KRAS, TP53 — fifteen years to find a polyp (the adenoma-carcinoma sequence)

Every CRC is an adenocarcinoma that began as a benign polyp and climbed a stepwise genetic ladder over about a decade. Fearon and Vogelstein drew that ladder in 1990, and it is still the most reproduced model in GI oncology — because the long, silent climb is the biological basis of screening, polypectomy and surveillance.[3]

The four rungs, in order:[3]

• APC loss (chromosome 5q) — the gatekeeper. APC normally marks beta-catenin for destruction in the Wnt pathway; its loss frees beta-catenin to drive the earliest adenoma.
• KRAS activation — the oncogene that pushes the small adenoma into a larger, dysplastic one.
• Loss of heterozygosity at 18q (SMAD2/SMAD4/DCC) — progression to late adenoma.
• TP53 loss (17p) — the final gate. With TP53 gone, severe dysplasia crosses into frank invasive carcinoma.[1]

The whole ladder takes 10 to 15 years. That is why a colonoscopy today that finds and snips a polyp prevents a cancer in 2035 — and why most early tumours are cured by surgery alone.[1][3]

The consultant's reading of the model: it is not the gene names that earn marks, it is the sequence. APC first (you cannot get an adenoma without it), KRAS in the middle (growth), TP53 last (invasion). Swap the order in a viva and you have failed the stem.[3]

85-10-5 — name the cause

Eighty-five percent sporadic, ten percent hereditary, five percent inflammatory bowel disease. That triplet answers every aetiology question, and it tells you where to spend the history — on age, diet and lifestyle for the 85, and on the family tree for the 10.[1]

The modifiable and non-modifiable risks, in the order an examiner expects them:[1]

• Age — the dominant factor; incidence climbs steeply after 50, and roughly 9 in 10 cases occur at 50 or older.
• Diet — red and processed meat (haem iron, N-nitroso compounds, heterocyclic amines) raise risk; fibre, fruit, vegetables and dairy protect (fibre dilutes carcinogens and shortens transit).
• Lifestyle — smoking, alcohol, obesity (central adiposity, insulin resistance) and physical inactivity all raise risk.
• Family history — one first-degree relative roughly doubles risk; two, or one under 50, raises it further.
• Inflammatory bowel disease — long-standing ulcerative colitis (especially pancolitis) and Crohn's colitis raise risk after 8 to 10 years; risk tracks extent, severity and duration of inflammation.
• Hereditary syndromes — Lynch syndrome (about 3 percent of all CRC) and familial adenomatous polyposis (under 1 percent), detailed below.
• Personal history of CRC or adenomatous polyps, acromegaly, ureterosigmoidostomy, pelvic radiotherapy, and type 2 diabetes.[1]

CRC is the third most commonly diagnosed cancer worldwide (behind breast and lung) and the second leading cause of cancer death — nearly 2 million new cases and over 900,000 deaths a year. The numbers examiners now test are the rising incidence in adults under 50, which is why the US screening start age dropped from 50 to 45.[1][2]

Three pathways, one carcinoma — CIN, MSI, serrated

Three molecularly distinct roads converge on the same endpoint: invasive adenocarcinoma. They differ in precursor lesion, driver mutation, biology and — increasingly — in therapy. Name all three and you have answered the molecular-biology stem.[1][3]

1. Chromosomal instability (CIN) — about 70 percent

This is the Vogelstein adenoma-carcinoma ladder above: APC to KRAS to TP53, over 10 to 15 years. It produces the conventional tubular and villous adenomas removed at colonoscopy and underlies most left-sided, chromosomally unstable tumours.[3]

2. Microsatellite instability (MSI) — about 15 percent

Defective DNA mismatch repair (MMR) — the enzymes that correct single-nucleotide and short-repeat errors during replication. Their loss produces MSI-H tumours carrying an enormous mutational burden. Two origins, and the split matters:[1][10]

• Sporadic (commoner) — epigenetic silencing (methylation) of the MLH1 promoter, usually with a BRAF V600E mutation; typically right-sided, mucinous, poorly differentiated tumours in older women.
• Lynch syndrome (inherited) — a germline mutation in MLH1, MSH2, MSH6 or PMS2; typically right-sided CRC with a tumour-infiltrating-lymphocyte-rich histology in the 40s or 50s.[1]

MSI-H tumours carry a better stage-for-stage prognosis and — the modern punchline — respond dramatically to immune checkpoint inhibitors (pembrolizumab, nivolumab). That single fact is why we now test MMR/MSI on every newly diagnosed CRC.[1]

3. Serrated pathway — about 15 to 30 percent

The third road, from sessile serrated lesions (SSLs) and traditional serrated adenomas in the right colon, not conventional adenomas. The drivers are an initiating BRAF V600E mutation, a CpG island methylator phenotype (CIMP) that silences genes including MLH1 (the bridge into the MSI pathway), and Wnt activation. The trap that costs lives: serrated lesions are flat, right-sided, pale and easily missed at colonoscopy — which is exactly why interval cancers still appear between clear scopes.[1]

How CRC spreads — the route tells you the prognosis. Lymphatic spread is the commonest and earliest route (epicolic to paracolic to intermediate to named-vessel nodes, then para-aortic), and node involvement is the single most important prognostic factor in non-metastatic disease. Haematogenous spread runs the portal vein to liver, then to lung — the order of the commonest metastatic sites. Transcoelomic spread seeds the peritoneum (omental cake, Krukenberg ovaries, ascites): a poor-prognosis pattern.[1]

Find it on colonoscopy — the two non-negotiable rules

A change in bowel habit, rectal bleeding, or iron-deficiency anaemia in an adult is colorectal cancer until a colonoscopy to the caecum says it is not. That is the spine of the whole topic. The two rules below are the ones a registrar forgets at 3am and a patient pays for:[1]

1. A change in bowel habit or new rectal bleeding in anyone over 40 mandates colonoscopy.
2. Iron-deficiency anaemia in an adult male or postmenopausal woman is CRC until excluded — urgent colonoscopy plus upper endoscopy. (In a premenopausal woman, investigate if the anaemia is significant or persistent.)[1]

The classic trap — calling it IBS: irritable bowel syndrome is a diagnosis of exclusion made only when weight loss, anaemia, bleeding and a family history are all absent and age-appropriate screening is normal. Labelling a 55-year-old's new bowel change as IBS without a colonoscopy is the recurring error that delays the diagnosis by a year.[1]

The mimics worth naming at the bedside, each with its one-line discriminator:[1]

[1]

The 3-2-1 rule — Lynch at the bedside

Lynch syndrome is the commonest hereditary CRC syndrome, and the one examiners test hardest. It is an autosomal dominant germline defect in a DNA mismatch repair gene — MLH1, MSH2, MSH6 or PMS2 (plus EPCAM deletions that silence MSH2). The tumours are MSI-H. Lifetime CRC risk is 40 to 70 percent, onset averages the mid-40s (a decade before sporadic CRC), and the extracolonic cancers are the half of the story candidates forget — endometrial (the sentinel cancer in many women), ovarian, gastric, urothelial, small bowel, biliary, brain (Turcot) and sebaceous (Muir-Torre).[10]

The bedside rule is the 3-2-1 — Amsterdam II, the clinical screen for Lynch:[10]

Everyone forgets: Amsterdam II catches only about 60 percent of Lynch families. Modern practice has moved past both Amsterdam and Bethesda at the diagnostic level — universal MMR/MSI immunohistochemistry on every newly diagnosed CRC is now standard, precisely because family-history rules miss the sporadic-looking case. Test the tumour, not the pedigree.[10]

The classic trap — missing young-onset Lynch: a 38-year-old with a new right-sided tumour is Lynch until MMR/MSI testing proves otherwise. Reach for the family tree and the IHC panel, not for "unlucky." Lynch tumours also evolve faster than the classic 10 to 15 year interval, so relatives are surveyed by colonoscopy from age 20 to 25, every 1 to 2 years.[10]

FAP — colectomy before the cancer

Familial adenomatous polyposis is autosomal dominant, on the APC gene at chromosome 5q, and the cancer is inevitable without surgery. Affected people grow hundreds to thousands of adenomatous polyps (over 100 by adolescence); without intervention, CRC is virtually inevitable by age 35 to 40. The operation is the treatment: prophylactic colectomy in the late teens to early twenties — proctocolectomy with ileal pouch-anal anastomosis, or total colectomy with ileorectal anastomosis and rectal surveillance.[1]

The variants examiners name: attenuated FAP (fewer than 100 polyps, later onset) and MUTYH-associated polyposis (autosomal recessive). After colectomy, surveillance continues — annual sigmoidoscopy of any retained rectum, annual upper-GI endoscopy for duodenal polyposis, and vigilance for desmoid tumours (a leading cause of death in FAP).[1]

Find it, stage it, watch it — colonoscopy, CT, MRI and CEA

The diagnostic and staging pathway has three jobs: confirm the diagnosis, stage the extent, and define the molecular biology that drives therapy. CEA is a monitoring tool, never a screening or sole diagnostic one.[1]

Confirm the diagnosis. Colonoscopy with biopsy is the gold standard — it visualises the tumour, provides tissue, finds and removes synchronous lesions (present in 3 to 5 percent), and tattoos small lesions for later surgical localisation. Complete the examination to the caecum (caecal photographs) and intubate the terminal ileum where possible. CT colonography is the alternative when colonoscopy is incomplete, contraindicated or refused — but it cannot biopsy or tattoo, so any positive still needs a scope.[1]

Stage the extent. CT chest/abdomen/pelvis is the standard staging modality for all CRC — it sets the M-stage (liver, lung, nodes, peritoneum) and decides curative versus palliative intent early. MRI of the rectum is essential and mandatory for rectal cancer: it maps T-stage, extramural vascular invasion, suspicious mesorectal nodes, and — the number that decides neoadjuvant therapy — the distance to the mesorectal fascia, where clearance under 1 mm predicts local recurrence. Endorectal ultrasound is an alternative for early (uT1/2) tumours when local excision is contemplated; PET-CT is not routine.[1]

Define the biology — and now do it on everyone. MMR/MSI testing (immunohistochemistry for MLH1, MSH2, MSH6, PMS2, and/or PCR) screens for Lynch, predicts a better prognosis, and flags candidates for checkpoint inhibitors. RAS (KRAS and NRAS) and BRAF testing is done on all metastatic CRC to select patients for anti-EGFR therapy, which helps only RAS wild-type tumours; a BRAF mutation predicts poor prognosis and no anti-EGFR benefit.[1][10]

The CEA trap — it is for monitoring, not screening. Carcinoembryonic antigen is a glycoprotein useful for a baseline before surgery, monitoring treatment response, and detecting recurrence after curative resection. It is not a screening test (poor sensitivity for early disease) and not diagnostic alone — it rises in smokers, IBD, pancreatitis, liver disease and other cancers. A rising or persistently elevated CEA after curative resection triggers imaging to find recurrence; serial measurement is the backbone of surveillance.[13]

Screening — FIT is the backbone. The faecal immunochemical test (FIT) detects human haemoglobin in stool; it replaced the older guaiac faecal occult blood test because it is diet-independent, more sensitive and quantitative. A positive FIT prompts colonoscopy, and it is the engine of every population screening programme.[2]

[1] [2]

Staging — TNM-8 and Dukes

Staging uses the eighth edition of TNM (AJCC/UICC), with the older Dukes classification still widely used and heavily examined. Reproduce both.[1]

TNM-8 — Tumour (T)[1]

[1]

TNM-8 — Nodes (N) and Metastases (M)[1]

[1]

Stage grouping and the Dukes equivalent (with approximate 5-year survival):[1]

[1]

Surgery by site — en bloc, with the mesentery and twelve nodes

The principle is en-bloc resection of the tumour-bearing bowel segment with its mesentery and lymphatic drainage, a clear circumferential and distal margin, and a minimum of twelve lymph nodes for accurate staging. Treatment is decided by stage, site and molecular subtype at an MDT; the pillars are surgery (curative for localised disease), chemotherapy (adjuvant for stage III and high-risk II; first-line for metastatic), radiotherapy (neoadjuvant for locally advanced rectal), and biological therapy (for metastatic).[1]

Surgery by site, with the vascular ligation that defines each operation:[1]

• Right hemicolectomy (caecal, ascending, proximal transverse) — terminal ileum, caecum, ascending and right transverse colon; ligate the ileocolic, right colic and right branches of the middle colic vessels at their origin; ileocolic anastomosis.
• Extended right hemicolectomy — transverse colon and hepatic flexure tumours.
• Left hemicolectomy / sigmoid colectomy (descending, sigmoid) — ligation of the inferior mesenteric artery (high tie at its origin or below the left colic) and inferior mesenteric vein; colorectal anastomosis.
• Subtotal or total colectomy — synchronous tumours, an obstructing proximal cancer with unprepared bowel, or hereditary syndromes.
• Rectal cancer — low anterior resection (LAR) for upper and mid rectal tumours above the anorectal ring (sphincter-sparing, with TME and a circular-stapled colorectal or coloanal anastomosis, often protected by a temporary defunctioning loop ileostomy); abdominoperineal excision (APR) for low tumours involving the sphincter (rectum and anus removed via abdominal and perineal approaches, permanent end colostomy); Hartmann's procedure for emergency obstruction, perforation or the frail (end colostomy and closed rectal stump).[1]

TME — Heald's plane (rectal recurrence under 10 percent)

Total mesorectal excision is the sharp, anatomical, en-bloc removal of the rectum and its intact mesorectum within the avascular mesorectal fascial plane, ensuring the circumferential resection margin is clear. R.J. Heald described it in 1982, and it is the global surgical standard — because it took rectal cancer local recurrence from 20 to 30 percent down to under 10 percent when paired with neoadjuvant therapy.[9]

FOLFOX or CAPOX for six months — MOSAIC (stage III)

Adjuvant chemotherapy for colon cancer is decided by stage. Stage I gets surgery alone. Stage II gets surgery alone for standard-risk disease, with adjuvant 5-FU plus leucovorin or capecitabine considered only for high-risk features — T4 (stage IIC), poor differentiation, perforation or obstruction, fewer than 12 nodes sampled, lymphovascular or perineural invasion — in an individualised decision. Stage III gets the standard of care: adjuvant FOLFOX or CAPOX for 6 months.[1][6]

MOSAIC built the stage III standard. Adding oxaliplatin to 5-FU and leucovorin improved disease-free survival (André, NEJM 2004) and then overall survival in stages II and III colon cancer (André, JCO 2009) — the basis of FOLFOX. The IDEA collaboration later showed a 3-month course is acceptable for low-risk stage III (T1 to T3 N1), cutting cumulative oxaliplatin neurotoxicity, while 6 months stays standard for higher-risk disease.[4][5]

Short or long — neoadjuvant radiotherapy for the rectum

For locally advanced rectal cancer (broadly T3 or deeper, or node-positive on MRI), preoperative radiotherapy or chemoradiotherapy downsizes the tumour, sterilises the mesorectum and cuts local recurrence. Two regimens, each with a landmark trial, and examiners want both:[1][7][8]

• Short-course radiotherapy — 25 Gy in 5 daily fractions of 5 Gy over 1 week, then surgery within a week. Established by the Swedish Rectal Cancer Trial (Cedermark, 1997), which showed a survival benefit for preoperative radiotherapy in resectable rectal cancer. Favoured for convenience, lower cost and similar local control when immediate downstaging is not needed.[8]
• Long-course chemoradiotherapy — 45 to 50.4 Gy in 25 to 28 fractions over 5 to 6 weeks with concurrent radiosensitising chemotherapy (infusional 5-FU or oral capecitabine), then surgery 6 to 12 weeks later after a downsizing interval. The German CAO/ARO/AIO-94 trial (Sauer, 2004) proved preoperative chemoradiotherapy superior to postoperative — better local control, less acute and late toxicity, more sphincter preservation.[7]
• Total neoadjuvant therapy (TNT) — giving all chemotherapy and chemoradiotherapy before surgery is an emerging standard that raises pathological complete response rates and compliance.[1]

The discriminator examiners test: short course is quick (one week, then surgery); long course downsizes (5 to 6 weeks, then wait 6 to 12 weeks before surgery). Choose long course when you need the tumour to shrink; choose short course when you do not.[7][8]

Resuscitate first — the two surgical emergencies

Most CRC is managed electively, but two presentations are time-critical: acute large-bowel obstruction and perforation. Both are surgical emergencies masquerading as "constipation" or "septic abdomen" — miss either and the patient dies.[1]

[1]

Perforation with faecal peritonitis is the highest-mortality presentation — aggressive resuscitation, broad-spectrum antibiotics and emergency laparotomy with resection and stoma. Acute lower-GI bleeding is rarer in CRC than in diverticular disease: resuscitate and transfuse, localise, and control endoscopically or surgically.[1]

Metastatic disease — RAS first, then the biological

Metastatic CRC is treated with systemic chemotherapy plus a biological, with curative-intent resection considered for oligometastatic disease. The chemotherapy backbones are FOLFOX, FOLFIRI (interchangeable — one first, one second line) or CAPOX, and adding a biological improves response and survival. The biological is chosen by the molecular report, not by habit:[1][11][12]

• Bevacizumab — a monoclonal antibody against VEGF that chokes tumour angiogenesis, added to first- or second-line chemotherapy in any metastatic CRC; established by Hurwitz (NEJM 2004).[11]
• Anti-EGFR antibodies — cetuximab and panitumumab — added to chemotherapy (FOLFIRI or FOLFOX) in the first line, but only in RAS wild-type tumours. The CRYSTAL trial (Van Cutsem, NEJM 2009) showed cetuximab plus FOLFIRI improved response and progression-free survival in KRAS wild-type metastatic CRC. RAS mutation testing is mandatory before starting an anti-EGFR agent.[12]
• MSI-H metastatic disease — these tumours respond dramatically to immune checkpoint inhibitors (pembrolizumab, nivolumab), the modern paradigm of precision oncology.[1]

Metastasectomy — surgical resection of limited liver or lung metastases in selected patients — offers the only chance of long-term survival or cure in stage IV, with 5-year survival of 30 to 50 percent in well-chosen cases (staged or simultaneous resection, portal vein embolisation, radiofrequency or microwave ablation). Palliation — endoscopic colonic stenting, defunctioning stoma, bypass, best supportive care — manages obstruction, bleeding or pain when cure is not the goal.[1]

The classic trap — anti-EGFR in a RAS mutant: giving cetuximab to a KRAS-mutant tumour does nothing but add rash and cost. RAS status is checked before the first dose, every time. Bevacizumab, by contrast, works regardless of RAS.[12]

Surveillance — CEA every 3 to 6 months, colonoscopy at 1, 3, 5

Surveillance after curative-intent treatment aims to find resectable recurrence — local, metachronous cancer, or oligometastatic disease — while it is still curable. The backbone is serial CEA and imaging, topped and tailed by colonoscopy:[1][13]

• CEA every 3 to 6 months for the first 3 years, then 6-monthly to 5 years; a rising CEA prompts cross-sectional imaging.
• CT chest/abdomen/pelvis annually for up to 3 years (3 to 5 years in higher-risk patients).
• Colonoscopy at 1 year after resection, then at 3 years, then every 5 years if normal (the perioperative clearing colonoscopy must be complete to the caecum).[1]

Prognosis — stage is destiny, sidedness and MSI modify it

Prognosis is stage-dependent, and beyond stage it is modified by grade, lymphovascular invasion, the circumferential resection margin, microsatellite status and sidedness. The survival curve by Dukes stage is the one table to reproduce:[1]

[1]

Favourable: early stage, well-differentiated tumour, no lymphovascular invasion, clear margins, MSI-H status (better stage-for-stage and dramatic immunotherapy response) and left-sided location. Adverse: advanced stage, BRAF mutation, perineural invasion, signet-ring histology, peritoneal metastases and right-sided tumours (which carry a poorer prognosis, partly reflecting their higher MSI/BRAF biology).[1]

Special populations — change the thresholds, not the spine

Long-standing IBD — surveillance colonoscopy with chromoendoscopy from 8 to 10 years of symptoms (from diagnosis for primary sclerosing cholangitis), every 1 to 5 years by risk; biopsy for dysplasia. CRC complicating IBD is more often multifocal and mucinous.[1]

Lynch and FAP — intensive, earlier surveillance from young adulthood, as above. Aspirin chemopreduction is supported by long-term data (notably CAPP2 in Lynch). If CRC develops in Lynch, subtotal colectomy may be preferred given the high metachronous risk.[10]

Elderly and frail — colonic stenting for obstruction, less aggressive chemotherapy, individualised decisions on neoadjuvant therapy and surgery; enhanced recovery after surgery (ERAS) pathways improve outcomes.[1]

Young-onset CRC (under 50) — rising incidence; always suspect a hereditary syndrome, test MMR/MSI, refer to genetics; these patients present more often with left-sided disease, advanced stage and symptoms.[1]

Complications — and the recurring pitfalls

Large-bowel obstruction is the commonest emergency presentation; complete obstruction risks caecal perforation (Laplace's law — the caecum is the widest, most distensible segment). Perforation spills faeces into the peritoneum (high mortality) or locally (abscess, fistula to bladder or vagina). Anastomotic leak typically presents around day 5 to 7 with fever, tachycardia, pain, ileus, leukocytosis and later septic shock — diagnose with water-soluble contrast enema or CT, manage with antibiotics, drainage and reoperation (often a defunctioning stoma); risk is higher after low pelvic anastomosis, in malnutrition and after neoadjuvant radiotherapy.[1]

The recurring pitfalls every candidate must name: attributing rectal bleeding to haemorrhoids without proctoscopy and colonoscopy; missing iron-deficiency anaemia as a red flag; not testing MMR/MSI on a young-onset tumour; under-staging the rectum by omitting MRI and so omitting neoadjuvant therapy; failing to confirm RAS status before an anti-EGFR agent; assuming a "normal" colonoscopy excludes a small right-sided lesion when the caecum was not reached; and not enrolling patients in surveillance.[1]

The trials that changed practice

These are the trials an examiner expects you to name, and what each changed. The doses are reproduced verbatim.[1]

[8] [7] [4] [5] [6] [11] [12]

Regional screening — 45 in the US, 50 to 74 in the UK, opportunistic in India

The diagnostic and management spine is globally consistent; the screening start age and the organised programme are not.[1]

Prevention, in one line each: screening and polypectomy is the single most effective intervention (it interrupts the sequence); a high-fibre, low-red-and-processed-meat diet, exercise, weight control, smoking cessation and moderate alcohol reduce risk; low-dose aspirin reduces CRC incidence and mortality (notably in Lynch, CAPP2), balanced against bleeding risk; and surveillance colonoscopy for post-polypectomy, post-resection, IBD, Lynch and FAP cohorts.[1][2]

The mantra, and the mnemonics

The mantra — say it on every ward round: right bleeds, left obstructs, rectum tenesmuses — find it on colonoscopy.[1]

[1] [3] [1]

Ward-round test — three stems, sixty seconds each

Stem 1 — the man from the top of the topic (answer)

The 67-year-old from the vignette: three months of looser stools, bright-red blood mixed into the motion, 4 kg weight loss, haemoglobin 98 with an MCV 76 and ferritin 8. What is the diagnosis, and what do you do next? Model: This is colorectal cancer until a colonoscopy to the caecum proves otherwise — the iron-deficiency anaemia in an adult male plus a change in bowel habit and blood mixed into the stool is a red-flag triad. The next step is urgent colonoscopy with biopsy, plus a staging CT chest/abdomen/pelvis and, if the tumour is in the rectum, an MRI rectum. Take a family history and, on every newly diagnosed CRC, send MMR/MSI testing. Do not accept "piles" as an explanation for blood mixed into the stool.[1]

Stem 2 — the 38-year-old with a right-sided tumour (answer)

A 38-year-old woman has a right-sided colonic adenocarcinoma. Her mother had endometrial cancer at 42 and a maternal aunt had bowel cancer at 48. What is the syndrome, the rule, and the surveillance for her relatives? Model: This is Lynch syndrome — an autosomal dominant germline mismatch-repair defect (MLH1, MSH2, MSH6 or PMS2). The family fits the Amsterdam II "3-2-1 rule" (three relatives with a Lynch-associated cancer, one a first-degree relative of the other two, across two generations, one under 50), but modern practice does not stop at the pedigree: test the tumour's MMR/MSI by immunohistochemistry regardless, and refer to genetics. At-risk relatives get colonoscopy from age 20 to 25, every 1 to 2 years, because Lynch tumours evolve faster than the classic 10 to 15 year interval; women are counselled on endometrial and ovarian risk. Consider subtotal colectomy given the high metachronous risk.[10]

Stem 3 — the rising CEA after a clear resection (answer)

Two years after a curative resection for stage III colon cancer, a 70-year-old's CEA rises from 3 to 14 over three serial measurements, but his CT chest/abdomen/pelvis is reported as normal. What is the right call? Model: A rising or persistently elevated CEA after curative resection is recurrence until proven otherwise — it triggers imaging even when the CT reads normal. The next step is repeat cross-sectional imaging (consider PET-CT) to find the recurrence, then MDT discussion for metastasectomy if disease is oligometastatic and resectable. Remember CEA is a monitoring tool, never a screening one — but in this context a rising trend is exactly the signal it was designed to give.[13]

References

1. [1]Dekker E, Tanis PJ, Vleugels JLA, et al. Colorectal cancer Lancet, 2019.PMID 31631858
2. [2]Davidson KW, Barry MJ, Mangione CM, et al. Screening for Colorectal Cancer: US Preventive Services Task Force Recommendation Statement JAMA, 2021.PMID 34003218
3. [3]Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis Cell, 1990.PMID 2188735
4. [4]André T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer N Engl J Med, 2004.PMID 15175436
5. [5]André T, Boni C, Navarro M, et al. Improved overall survival with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer in the MOSAIC trial J Clin Oncol, 2009.PMID 19451431
6. [6]Benson AB 3rd, Schrag D, Somerfield MR, et al. American Society of Clinical Oncology recommendations on adjuvant chemotherapy for stage II colon cancer J Clin Oncol, 2004.PMID 15199089
7. [7]Sauer R, Becker H, Hohenberger W, et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer N Engl J Med, 2004.PMID 15496622
8. [8]Cedermark B, Dahlberg M, Glimelius B, et al. Improved survival with preoperative radiotherapy in resectable rectal cancer N Engl J Med, 1997.PMID 9091798
9. [9]Votava J, Kachlik D, Hoch J Total mesorectal excision - 40 years of standard of rectal cancer surgery Acta Chir Belg, 2020.PMID 32200705
10. [10]Umar A, Boland CR, Terdiman JP, et al. Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability J Natl Cancer Inst, 2004.PMID 14970275
11. [11]Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer N Engl J Med, 2004.PMID 15175435
12. [12]Van Cutsem E, Köhne CH, Hitre E, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer N Engl J Med, 2009.PMID 19339720
13. [13]Shinkins B, Nicholson BD, Primrose J, et al. What carcinoembryonic antigen level should trigger further investigation during colorectal cancer follow-up? A systematic review and secondary analysis of a randomised controlled trial Health Technol Assess, 2017.PMID 28617240