The Washington Manual of Oncology, 3 Ed.

Colorectal Cancer

Ashley Morton • Benjamin Tan

 I. PRESENTATION

1. Subjective. Colorectal cancer (CRC) is diagnosed as a result of a positive screening test or workup of a symptomatic patient. Symptoms associated with CRC include rectal bleeding, abdominal pain, changes in bowel habits, weight loss, fatigue, anorexia, and abdominal distention. Patients could also present with bowel obstruction, perforation, peritonitis, or fever. A thorough family history and past medical/oncologic history should be done to exclude any familial colorectal cancer syndromes.
2. Objective. Physical examination should always include a thorough abdominal and rectal examination and may reveal abdominal tenderness, hepatomegaly, ascites, a palpable mass, palpable adenopathy, and gross blood or heme-positive rectal examination. Extracolonic manifestations in patients with hereditary colorectal cancer syndromes should also be noted.
3. WORKUP AND STAGING
4. Initial evaluation. Patients found to have a pedunculated or sessile polyp or mass in the colon or rectum require careful pathologic review of the excised specimen and precise marking of the polyp site. If invasive cancer is detected, additional workup should include a complete blood count, chemistry panel with liver function tests, carcinoembryonic antigen (CEA), and computed tomography (CT) of the chest, abdomen, and pelvis. Although routine positron emission tomography (PET) scan is not standard, it may be valuable in assessing the extent of metastatic disease in a patient considered for potentially curative resection. For rectal cancer, important clinical factors such as tumor distance from anal verge, presence of circumferential, obstructing, and tethered tumors should be recorded.
5. Surgical principles. Surgery remains the primary modality for cure in patients with colorectal cancer. Select patients with synchronous or metachronous solitary or limited hepatic and pulmonary metastatic disease should still be considered for potentially curative resection of both the primary and metastatic sites. The extent, type, and timing of resection are dependent on the location of the tumor and presence of bleeding, obstruction, or perforation and the presence of polyposis. The number of lymph nodes harvested and examined (≥12 recommended) impacts staging accuracy and prognosis.

 Laparoscopic colectomy has been associated with higher cancer-related survival, reduced relapse rates, decreased morbidity, blood loss, and hospital stays compared with open colectomy (Lancet 2002;359:2224).

1. Pathology
2. Pathological review of the surgical or biopsy specimen should include histologic type, tumor size and differentiation, depth of invasion, number and location of lymph nodes involved and examined, presence of extranodal deposits, nonregional lymph nodes and distant metastases, perineural and lymphovascular invasion, and margins (proximal, distal, and radial).
3. Additional data. DNA mismatch repair (MMR) proficiency or deficiency determination and the presence of any KRAS, NRAS, BRAF, and other somatic mutations would be necessary for treatment options.
4. Staging. Five-year survival rates for 109,953 colorectal cancer patients using the 7th edition of the New American Joint Committee on Cancer (AJCC) based on Surveillance, Epidemiology, and End Results (SEER) data from 1992 to 2004 (J Clin Oncol 2010;28:265) are shown in Table 16-1.
5. Prognostic factors
6. Tumor grade. Low-grade tumors are significantly associated with better survival in stages II, III, and IV, but not in stage I, colon cancer.
7. Histologic subtype. Adenocarcinoma comprises >85% of CRC, while mucinous and signet ring types occur in 12% and 1%, respectively. Patients with signet ring cell carcinomas had the worse prognosis compared with adenocarcinomas and mucinous types (5-year overall survival (OS) 36%, 66%, and 62%, respectively).
8. Tumor location. Sigmoid colon cancers confer the best 5-year survivals (70%) compared with tumors in the right colon (64%), transverse colon (65%), and the left colon (65%).
9. Number of positive lymph nodes. Worse survival rates are seen in patients with more lymph nodes involved with cancer. Further subcategorizing lymph node involvement to N2 (4-5 LN+), N3 (6-8 LN+), and N4 (≥9 LN+) could provide a more accurate prognosis among stage III CRC patients with 5-year OS of 52%, 43%, and 27%, respectively.
10. Lymphovascular involvement is associated with worse prognosis.
11. Molecular markers including microsatellite instability, KRAS, NRAS, BRAF, and PIK3CA also carry predictive and prognostic information. The majority of KRAS mutations occur on codons 12 and 13 of exon 2, and are present in approximately 40% of colorectal cancers. KRAS is the best predictor for resistance for EGFR targeting agents. An additional 15% to 18% of CRC will harbor other KRAS and NRAS mutations (most NRAS mutations occur on codon 61 of exon 2), 5% to 8% of CRC will harbor BRAF V600E mutations, and 9% to 15% of CRC will have a PIK3CA mutation (68.5% located on exon 9 and 20.4% on exon 20) (PLoS One 2013;8:e81628; Lancet Oncol 2010;11:753). KRAS and BRAF are mutually exclusive and should be tested independently. KRAS mutations are strongly associated with PIK3CA mutations. Adjuvant aspirin therapy may improve overall survival in PIK3CA-mutated tumors (New Engl J Med 2012;367:1596). NRAS and PIK3CA mutations may inhibit response to anti-EGFR therapy, despite KRAS wild-type status. BRAF-mutated tumors are associated with poorer prognosis.

 III. TREATMENT BY STAGE

1. Stage I colon cancer (T1–T2, N0 M0). Since surgical resection confers a high cure rate for patients with stage I cancer, no adjuvant therapy is recommended.
2. Stage II colon cancer (T3–T4, N0 M0). Although the American Society of Clinical Oncology does not recommend the routine use of adjuvant therapy for stage II CRC patients, “high-risk” patients including those with inadequately sampled nodes, T4 lesions, perforation, obstruction, or poorly differentiated histology should be considered for chemotherapy (J Clin Oncol 2004;22:3408).
3. The IMPACT B2 Study reported a 2% to 3% absolute difference in 5-year disease-free survival (DFS) (76% vs. 73%) and OS (82% vs. 80%) in patients treated with adjuvant 5-fluorouracil (5-FU)-leucovorin chemotherapy versus observation (J Clin Oncol 1999;17:1356).
4. The QUASAR study showed a significant improvement in recurrence rates (22.2% vs. 26.2%) and 5-year survival (80.3 vs. 77.4%, HR 0.83) with 5-FU versus observation.
5. A systematic review of published literature included 37 randomized studies, and 11 meta-analyses with 20,317 patients demonstrated a 5% to 10% absolute improvement in DFS associated with adjuvant therapy, although this did not achieve statistical significance (Cancer Prev Control 1997;1:379).
6. With more aggressive oxaliplatin-based adjuvant chemotherapy, patients with stage II colon cancer treated on the MOSAIC study had an absolute 4-year DFS benefit of 3.8% (85.1% vs. 81.3%) compared with those treated with 5-FU/leucovorin alone (N Engl J Med 2004;350:2343). In high-risk stage II patients, the benefit is higher at 5.4%. However, no overall survival advantage was observed with the addition of oxaliplatin.
7. Defective MMR is predictive of lack of efficacy of 5-FU adjuvant therapy in colon cancer (J Clin Oncol 2010;28:3219). A 12-gene recurrence score may predict recurrence risk in stage II and III colon cancer and potential benefit from oxaliplatin-based adjuvant therapy (J Clin Oncol 2013;31:4512). A careful assessment of risk of recurrence based on tumor and patient characteristics, and a thorough discussion with the patient regarding the absolute and relative benefits and toxicities are necessary during consideration of adjuvant therapy in patients with stage II CRC.
8. Stage III colon cancer (any T, N1–N2, M0). Adjuvant oxaliplatin-based chemotherapy is standard for patients with stage III CRC.
9. Stage III CRC patients treated with FOLFOX4 on the MOSAIC study had an 8.7% absolute benefit in 3-year DFS (77.9% vs. 72.8%) and 4-year DFS (69.7% vs. 61%) compared with 5-FU/leucovorin. For patients with N2 disease, the absolute benefit is 12%.
10. The NSABP C-07 confirmed the efficacy of oxaliplatin, with bolus 5-FU showing a significant improvement in 3-year DFS (76.5% vs. 71.5%) compared with bolus 5-FU/LV.
11. Capecitabine plus oxaliplatin is also another recommended option for adjuvant treatment. A large multicenter, randomized trial compared XELOX versus 5-FU/LV. The 3-year DFS rate was 70.9% with XELOX and 66.5% with 5-FU/LV; 5-year OS for XELOX and 5-FU/LF were 77.6% and 74.2%, respectively (J Clin Oncol 2011;29:146).
12. Elderly age should also be considered when choosing adjuvant treatment. In the NSABP-07 trial, patients <70 years old had improved OS with FLOX compared to FU/LV (HR 0.80), with 5-year OS 81.8% for FLOX and 78.8% with FU/LV. In patients >70 years old, OS did not vary significantly by treatment type, but nominal OS at 5 years was 4.7% worse with FLOX (71.6%) versus FULV (76.3%). Additionally, toxicity and dose intensity varied by age. Furthermore, MOSAIC included patients younger than 75 years of age.
13. For patients unsuitable for oxaliplatin-based therapy, capecitabine or 5-FU with leucovorin are alternative therapies. In the X-ACT study, adjuvant capecitabine resulted in a trend toward improved 3-year DFS (64.2% vs. 60.6%) and OS (77.6% vs. 81.3%) compared with Mayo Clinic schedule of 5-FU/LV (N Engl J Med 2005;353:2696).
14. Irinotecan-based therapies cannot be considered for the adjuvant treatment of stage III CRC based on three large randomized trials (ACCORD, PETACC-3, and CALGB 89803) showing no benefit over 5-FU alone.
15. Targeted agents in the adjuvant setting. Adjuvant bevacizumab or cetuximab were shown to be of no benefit in stage III colorectal cancer. NSABP-08 is a phase III trial that showed no benefit with the addition of bevacizumab to FOLFOX in the adjuvant setting (J Clin Oncol 2010;29:11), whereas N0147 demonstrated no benefit with the addition of cetuximab to adjuvant chemotherapy (JAMA 2012;307:1383).
16. Metastatic colorectal cancer—first-line therapy. The goal of therapy, whether potentially curative versus palliative, may help determine the choice of regimen used for initial therapy for patients with metastatic CRC. More aggressive regimens associated with the best response rates (RR) may be favored for patients with limited metastatic sites potentially amenable to resection or patients with rapidly growing symptomatic visceral metastases, whereas the toxicity profiles of equally effective regimens may determine the choice of palliative therapy for patients with widely metastatic CRC. KRAS and BRAF status will also be important in selecting appropriate therapy.
17. Chemotherapy backbone
18. Oxaliplatin-based (FOLFOX) or irinotecan-based 5-FU regimens (FOLFIRI) are appropriate first-line treatments for metastatic CRC. The N9741 study demonstrated the superiority of FOLFOX over irinotecan with bolus 5-FU (IFL) in terms of RR (45% vs. 31%) and OS (19.5 months vs. 14.8 months) (J Clin Oncol 2004;22:23). Two studies have shown equal efficacy between FOLFOX and FOLFIRI regimens (J Clin Oncol2004;22:229; J Clin Oncol 2005;22:4866). Capecitabine combination therapies have also been studied in the CAIRO2 and TREE-2 studies.
19. In a phase III trial, triplet therapy of FOLFOXIRI showed improved RR (60% vs. 34%), median PFS (9.8 months vs. 6.9 months; HR 0.63), and median OS (22.6 months vs. 16.7 months) compared with FOLFIRI in unresectable metastatic colorectal patients (J Clin Oncol 2007;25:1670).
20. Single-agent capecitabine or 5-FU may be appropriate initial therapy for those with significant comorbidities precluding more aggressive treatment.
21. Chemotherapy with targeted agents
22. Bevacizumab is the first antiangiogenic drug approved for cancer treatment. When added to chemotherapy, this anti-vascular endothelial growth factor (anti-VEGF) antibody significantly improved outcomes. Hurwitz demonstrated a 4.7-month survival advantage with bevacizumab + IFL compared with IFL alone (20.3 months vs. 15.6 months). Furthermore, RR (45% vs. 35%) and duration of response (10.4 months vs. 7.1 months) are superior in the bevacizumab arm. Moreover, when bevacizumab is combined with 5-FU/LV alone, median survival of 18.3 months was achieved with RR of 40% (J Clin Oncol2005;23:3502). Bevacizumab with FOLFOX or FOLFIRI also resulted in high RR and relatively long survivals (TREE-2 and BICC-C trials).
23. Comparison between XELOX and FOLFOX with bevacizumab versus placebo was evaluated in a phase III trial. Median PFS improved to 9.4 months with bevacizumab versus 8 months in placebo group (HR 0.83 to 0.72), and median OS 21.3 months versus 19.9 months (J Clin Oncol 2008;26:2013).
24. In KRAS wild-type tumors, cetuximab and panitumumab, anti-epidermal growth factor receptors, have also been combined with oxaliplatin and irinotecan-based chemotherapy in the first-line CRC setting.
25. CRYSTAL compared FOLFIRI 1 cetuximab versus FOLFIRI alone. In an all-randomized patient population, RR improved to 46.9% from 38.7%; median PFS 8.9 months versus 8.1 months; and median OS 19.6 months versus 18.5 months. Upon further review, a post hoc analysis for KRAS wild-type subpopulation was noted to have improved OS to 23.5 months versus 19.5 months with RR 57% versus 39% (J Clin Oncol2011;29:2011). OPUS assessed overall RR with FOLFOX4 + Cetuximab versus FOLFOX4 alone, and noted improvement (46% vs. 36%). In KRAS wild-type tumors, clinical significance was noted with the addition of cetuximab to 61% versus 37% (J Clin Oncol 2009;27:663).
26. PRIME evaluated panitumumab with FOLFOX4 and noted improved PFS when compared with FOLFOX4 alone (10.1 months vs. 7.9 months). OS was 26 months with panitumumab arm and 20.2 months with FOLFOX alone. But mutated KRAS tumors had a significant reduction of PFS with the combination therapy. Interestingly, 17% of the patients in this trial had KRAS wild type (WT) tumors on exon 2, but were noted to have other RAS mutations and were associated with inferior PFS and OS with the panitumumab + FOLFOX (N Engl J Med 2013;369:1023).
27. Maintenance therapy
28. Once maximal response to 1st line therapy has been achieved, maintenance therapy should be considered. Optimox1 and Optimox2 evaluated a “stop-and-go” approach to chemotherapy administration. Duration of disease control, resulted in Opitmox2, was longer within the maintenance arm of 5FU/LV at 13.1 months, while the chemotherapy-free interval arm was 9.2 months. OS improved by 4.3 months with maintenance 5FU (JCO December 1, 2009 vol. 27 no. 34 5727-5733).
29. CAIRO3 supports maintenance therapy with cabecitabine + bevacizumab after XELOX versus observation with PFS 7.4 months versus 4.1 (Koopman, M. JCO 31, 2013 (suppl; abstr 3502).
30. Metastatic colorectal cancer—second-line and subsequent therapy
31. For patients who received oxaliplatin-based therapy, irinotecan-based therapy can be considered after progression, and vice versa, with a median OS of 21 months (J Clin Oncol 2004;22:229). Bevacizumab is approved for second-line therapy when combined with FOLFOX. This regimen improved OS (12.5 months vs. 10.7 months) compared with FOLFOX alone in patients who progressed after first-line therapy with irinotecan (J Clin Oncol 2007;25:1539).
32. Bevacizumab beyond first progression (ML18147) is associated with significantly prolonged PFS and OS (5.9 months and 11.9 months), when compared with patients who had chemotherapy alone (4.3 months and 10.6 months) (Ann Oncol 2013;24:2342).
33. Ziv-Aflibercept, a VEGF Trap, prevents activation of VEGF receptors, therefore inhibiting angiogenesis. VELOUR tested second-line aflibercept with FOLFIRI in patients who failed oxaliplatin-based regimens (irinotecan-naive patients). FOLFIRI + ziv-aflibercept had OS of 13.5 months versus 12.1 months in FOLFIRI + placebo; RR 19.8% and 11.1% (Eur J Cancer 2014;50:320).
34. Single-agent cetuximab or panitumumab have been compared with best supportive care and are appropriate third or later line therapies for those with KRAS wild-type tumors. Regorafenib, an oral multiple kinase inhibitor, can be considered if all other standard therapies fail. In refractory colon cancer, the phase III CORRECT trial compared regorafenib versus placebo and found that the OS was slightly improved by 1.4 months (6.4 months vs. 5.0 months). PFS was also modest at 1.9 months versus 1.7 months (Lancet 2013;381:303).
35. Liver dominant or liver-only metastatic disease may be amenable to regional therapies
36. Principles. Liver metastases can be found in approximately 80% of CRC patients; about 25% to 50% are encountered at initial presentation. Alternative treatment regimens should be considered to prolong survival rates.
37. Isolated liver lesions may be amendable to surgical resection (usually up to five, in one lobe of the liver) and is the standard curative treatment.
38. Systemic chemotherapy and/or transarterial administration of chemotherapy (with either vascular occlusive agents or hepatic chemoperfusion) may produce further tumor response and improved survival.
39. Rectal cancer. A multimodality approach, including colorectal surgery, radiation oncology, and medical oncology, is necessary for the optimal treatment of patients with rectal cancer. Initial workup should include a transrectal ultrasound to assess depth of invasion and lymph node involvement, CT or MRI to assess distant metastases, and biopsy to rule out other rectal tumors such as squamous cell carcinoma, melanoma, sarcoma, or lymphoma). Low anterior resection (LAR) is suitable for tumors located in the middle and upper third of the rectum, whereas an abdominoperineal resection (APR) may be necessary for low-lying rectal cancer. T stage and grade, lymphovascular involvement, lymph node metastases, and achievement of a negative radial margin are important prognosticators and predictive of local and distant recurrences. Total mesorectal excision (TME) in conjunction with LAR or APR has been advocated as the optimal surgical procedure for rectal cancer. A sharp (rather than blunt or avulsive) dissection is performed to remove the entire rectum. This technique has been shown to achieve a higher negative radial margin rate than blunt dissection (93% vs. 80%). Laparoscopic procedures are safe in the hands of experienced surgeons.
40. T1–T2 N0 rectal cancer. Transrectal ultrasound staged rectal lesions confined to the submucosa may be treated with full thickness local resection. Regional lymph node involvement occurs in 10% to 15% of these patients. T1 tumors invading the deepest part of the submucosa (SM3) carry a significantly higher risk of lymph node involvement than more superficial lesions (SM1 or SM2). Lesions invading the muscularis propria (T2) have a higher incidence of lymph node involvement (12% to 22%). If pathologically confirmed, patients require additional chemoradiotherapy. Preoperative radiation may be indicated for low-lying rectal lesions in an attempt to convert an APR to a potential sphincter-preserving LAR.
41. Locally advanced T3–T4 or N1 rectal cancer. Neoadjuvant chemoradiation is a standard treatment for locally advanced rectal cancer. 5-FU–based infusional chemoradiation is associated with lower local recurrence rate (6% vs. 13%), reduced acute and long-term toxicities, improved compliance, and promotes higher sphincter preservation, but achieved similar 5-year OS compared with postoperative therapy (N Engl J Med2004;351:1731). The degree of downstaging achieved with neoadjuvant therapy also provides prognostic information. Resection is generally done 6 to 10 weeks after chemoradiation, and the decision to administer adjuvant chemotherapy is based on the initial clinical staging.

 IV. COMPLICATIONS

1. Cancer-related. Bowel obstruction, hemorrhage, abdominal pain, perforation, fistula formation, peritonitis, anemia, and malnutrition. Extensive liver metastases hepatomegaly, liver failure, or jaundice.
2. Treatment-related
3. Surgical and radiation complications. Common bowel dysfunction after surgery includes stool frequency, episodic frequency, pressure sensation, urgency, nocturnal movements, and soilage. Anastomotic stricture, leak, ulceration, bleeding obstruction, infection, bladder, and sexual dysfunction may occur.
4. Chemotherapy and targeted therapy
5. Fluoropyrimidines -5-FU or capecitabine may cause myelosuppression, mucositis, diarrhea, excessive lacrimation, skin discoloration, dehydration, palmar–plantar erythrodysesthesia (PPE), and rare cardiotoxicity. Patients with dihydropyrimidine dehydrogenase (DPD) deficiency may have severe myelosuppression, ataxia, and diarrhea.
6. Oxaliplatin may cause cold-induced peripheral neuropathy, laryngodysesthesias, myelosuppression, nausea, and fatigue.
7. Irinotecan causes acute and delayed diarrhea, nausea, vomiting, myelosuppression, and alopecia. Patients with the UGT1A1 7/7 polymorphism may require irinotecan dose reductions owing to increased risk for grade 4 to 5 neutropenia.
8. Bevacizumab can cause common side effects such as epistaxis, hypertension, and proteinuria. Rarer side effects include arterial thrombotic events (ATEs) such as myocardial infarction or stroke, perforation, wound dehiscence, and reversible leukoencephalopathy, among others.

 i. Should any of these issues occur while on bevacizumab therapy, dose holding or discontinuation may be warranted. Brain metastasis must be treated prior to considering bevacizumab. Grade 3 venous thromboembolic events may be resumed WITH CAUTION (after a 2-week holding period) after therapeutic dose anticoagulation therapy is provided and the patient has no hemorrhagic complications. In patients with arterial thrombus (TIA, MI, CVA), bevacizumab should not be administered for at least 6 months and should then be reevaluated. Reasons for permanent discontinuation include: Gastrointestinal perforation, surgery (at least 28 days prior to and after elective surgery) and wound-healing complications, hemorrhage, hypertensive crisis, nephrotic syndrome, non-GI fistula formation, reversible posterior leukoencephalopathy syndrome (RPLS), severe/life-threatening thrombosis. Hold when 24-hour urine protein ≥2 grams.

1. Cetuximab/Panitumumab can cause acneform rash/folliculitis, trichomegaly, paronychial inflammation, xerosis, diarrhea, fatigue, hypomagnesemia, and rare pulmonary toxicities.

 i. Management of rash will vary for each patient. Moisturizers and sunscreen may reduce severity. Folliculitis usually occurs within 1 to 3 weeks. If it is restricted to facial area only or there is mild-to-severe reaction, topical antibiotics, such as clindamycin 2%, may be used. Topical steroids such as hydrocortisone 1% may also improve symptoms. Oral therapy with minocycline may be added to the topical regimen. Referral to a dermatologist may be necessary if symptoms worsen or are severe at baseline.

1. FOLLOW-UP. After curative resection and completion of therapy, a history, physical examination, and CEA determination is recommended every 3 months for the first 2 years, then every 4 to 6 months until the 5th year. Colonoscopy is recommended 1 year after diagnosis or 1 year from surgical resection, and repeated in 1 year if abnormal; otherwise, repeated in 2 to 3 years. For patients without prior full colonoscopy (i.e., due to obstructing lesion or emergent surgery), colonoscopy in 6 months is recommended.

 VI. EPIDEMIOLOGY AND SCREENING

74. CRC ranks fourth in frequency among men and women, with approximately 140,000 new cases per year. It is the second leading cause of cancer deaths in the United States. In regard to mortality rates, it ranks second in men and women. Median age of diagnosis is 69, and the median age at death is 74. Patients with average risk (age ≥50, no prior history or adenoma, no Inflammatory bowel disease (IBD)) should have a colonoscopy done at age 50 (90% of cases occur after age 50). If no polyps are seen, colonoscopy should be repeated in 10 years. If polyps are seen, polypectomy should be done. Alternatively, fecal occult blood tests (FOBT) annually with flexible sigmoidoscopy every 5 years or double-contrast barium enema every 5 years could be performed.
75. If there is a family history of colon cancer in a first-degree relative (parent, sibling, child), risk increases about twofold. If an adenoma is diagnosed under age 60, screening should begin at age 40, or 10 years earlier than the youngest cancer in the family. According to the American Cancer Society and the American College of Gastroenterology, screening for African Americans should begin at age 45.

 VII. RISK FACTORS. The lifetime risks for developing CRC are as follows:

1. General population 5%.
2. Personal history 15% to 20%.
3. Inflammatory bowel disease 15% to 40%.
4. Hereditary nonpolyposis colorectal cancer (HNPCC) 70% to 80%—or Lynch syndrome. HNPCC is characterized by early onset of colon cancer and adenomas but not polyposis. HNPCC is autosomal dominant with 80% penetrance and is due to a mutation in MMR genes MLH1, MSH2, MSH6, PMS1, PMS2, and MSH3. Extracolonic manifestations include endometrial, ovarian, gastric, urogenital, bile duct, and sebaceous gland (Muir–Torre) cancer.
5. Familial adenomatous polyposis (FAP) >95%.FAP is characterized by the presence of thousands of polyps with an autosomal dominant inheritance with high penetrance and a prevalence of 1 in 8,000. The APC suppressor gene is located on chromosome 5q21, which is important in cell adhesion, signal transduction, and transcriptional activation in its interaction with beta catenin and the Wnt pathway. Cyclin D1 and c-myc are downstream targets. Cancer occurs at a median age of 39, and >90% will develop adenomas by age 30. Extracolonic manifestations include cholangiocarcinoma, duodenal carcinoma, gastric cancer, desmoid tumors, osteomas, thyroid cancer, and brain tumors.
6. Other risk factors include age >50, high red meat intake and low fiber diets, Peutz–Jeghers syndrome, juvenile polyposis, immunosuppression, smoking, alcohol, and others.

VIII. PATHOGENESIS. There are three major pathways to CRC: Chromosomal instability (CIN) pathway, mutator-phenotype/DNA MMR pathway, and hypermethylation phenotype hyperplastic/serrated polyp pathway (CIMP+).

1. Chromosomal instability or suppressor pathway accounts for 85% of sporadic CRC, with an adenoma as the precursor lesion. FAP coli is the prototype for this model characterized by the loss of APC gene. A potential subtype of this model is characterized by the silencing of methylation of the DNA repair enzyme methylguanine DNA methyltransferase with serrated polyps as precursor lesions.
2. Microsatellite instability or mutator pathway accounts for 15% of sporadic CRC. HNPCC is the prototype for this model characterized by a loss of MMR genes (MLH1, MSH2, MSH6, etc).
3. Hypermethylation phenotype (CIMP1). Hyper- or hypomethylation can silence the expression of certain genes including MMR enzymes. A defect in the CpG islands causes particularly high-frequency methylation, therefore promoting silencing of enzymes such as MLH1.

 IX. ANAL CANCER

1. Presentation. Bleeding, pain, constipation, tenesmus, diarrhea, discharge, and pruritus. Often, the symptoms are ascribed to other benign conditions such as hemorrhoids, fistula-in-ano, fissure, or anal condylomata. A careful history on risk factors also needs to be obtained.

 Physical examination findings include a firm, indurated, or exophytic anal mass or inguinal lymphadenopathy. A careful digital examination and anoscope is necessary to evaluate the extent of the tumor. Women with anal cancer should also have a thorough gynecological examination. Diagnosis is made with incisional biopsy of the mass and any inguinal lymphadenopathy.

 Pathology usually reveals squamous cell carcinoma or cloacogenic carcinoma. Adenocarcinomas involving the anus should be treated similar to rectal cancer. Rare cases of melanomas or neuroendocrine cancers may occur. Workup should include chest radiograph and CT or MRI of the abdomen and pelvis. HIV testing should be considered in high-risk individuals in addition to baseline laboratory tests.

1. Staging. Staging is based on the TNM system.

 T_is, carcinoma in situ

 T1, tumor is 2 cm or smaller

 T2, tumor is between 2 cm and 5 cm

 T3, tumor larger than 5 cm

 T4, tumor of any size that invades adjacent organs such as the vagina, urethra, or bladder

 N0, no regional lymph nodes involved

 N1, metastases in unilateral internal iliac or inguinal lymph node

 N3, metastases in perirectal and one inguinal lymph node and/or bilateral internal iliac or inguinal lymph nodes

 M0/1, no or (+) distant metastases present

 Stage I, T1, N0, M0

 Stage II, T2, 3, N0, M0

 Stage IIIA, T1–T3, N1, M0 or T4, N0, M0

 Stage IIIB, T4, N1, M0 or any T, N2–N3, M0

 Stage IV, any T, any N, M1

1. Prognosis. Prognosis is based on staging. T1 and T2 tumors have a 5-year survival of more than 80%, whereas T3 and T4 tumors have 5-year survivals of less than 50%. Inguinal lymphadenopathy and male sex also are related to a poorer prognosis. Tumors in the anal margin have a more favorable prognosis than do those in the canal.
2. Treatment. Standard therapy includes the Nigro protocol chemoradiation with mitomycin C and 5-FU. Higher DFS (73% vs. 51%) and lower colostomy rates (9% vs. 22%) are associated with chemoradiation compared with radiation alone. Five-year survival varied between 64% and 83% with combined-modality therapy. Patients with HIV/AIDS and anal cancer may be treated with the same Nigro protocol, but caution should be applied as they may not tolerate full doses of chemotherapy. Patients with T1 lesions may be considered for local excision alone or with chemoradiation.
3. Epidemiology and risk factors
4. Epidemiology. Anal cancer accounts for about 1.6% of all digestive system cancers in the United States. It is more common in men than in women. Its incidence generally increases with age, with peak incidence in the sixth and seventh decades of life. The incidence is increasing in men younger than 40 years.
5. Risk factors. Human papillomavirus (HPV) infection- HPV 16 and 18, women with HPV-related cervical cancer, smoking, HIV infection, and anal receptive sex.

SUGGESTED READINGS

Andre T, Boni C, Mounedji-Boudiaf L, et al.; Multicenter International Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer (MOSAIC) Investigators. Oxaliplatin, fluorouracil, and leucovorin as adjuvant therapy for colon cancer. N Engl J Med 2004;350:2343–2351.

Benson AB, Schrag D, Somerfield MR, et al. American Society of Clinical Oncology recommendations on adjuvant chemothaerapy for Stage II colon cancer. J Clin Oncol 2004;22:3408–3419.

Nishihara R. Long-term colorectal cancer incidence and mortality after lower endoscopy. N Engl J Medicine 2013;369:1095–1105.

Ryan DP, Compton CC, Mayer RJ. Medical progress: carcinoma of the anal canal. N Engl J Med 2000;342:792–800.

Sauer R, Becker H, Hohenberger W, et al.; German Rectal Cancer Study Group. Preoperative versus postoperative chemoradiation for rectal cancer. N Engl J Med 2004;351:1731–1740.