Colon and rectal cancer (CRC) combined is the third most commonly diagnosed cancer and the second deadliest in the United States. Rectal cancer has distinct environmental associations and genetic risk factors different from colon cancer. The incidence of new cases and mortality of rectal cancer has been steadily declining for the past years, although noticed to rise in recent years on younger adults (less than 50 years). The transformation of the normal rectal epithelium to a dysplastic lesion and eventually an invasive carcinoma requires an accumulation of genetic mutations either somatic (acquired) and/or germline (inherited) in approximately 10 to 15 years period. Tumor regression grade after pre-operative therapy and pathological staging are the most important prognostic indicators of rectal cancer. All newly diagnosed rectal cancer patients should be universally screened for DNA mismatch repair/microsatellite status that is present in up to 13% of all sporadic rectal cancer cases. A careful history and physical examination, including digital rectal exam, are paramount on clinical suspicion. An endoscopy examination with rigid sigmoidoscopy is required to measure the distance from the lesion to the anal verge (less than 15 cm) and for tissue biopsy for pathological confirmation of rectal cancer. Baseline computed tomography of the chest, abdomen, and pelvis may initially grossly rule out metastatic stage. Nevertheless, a combined approach by magnetic resonance imaging or transrectal ultrasound will accurately determine tumor extension and node status of the local rectal disease at diagnosis. A multidisciplinary evaluation by medical oncology, radiation oncology, and surgical oncology should discuss the best peri-operative chemo-radiotherapy route that could augment the chance of cure on high-risk patients. Oligo-metastatic, liver and lung, and local-recurrence rectal cancer patients are potentially curable with multimodality therapies. Palliative systemic therapy is reserved for non-surgical candidates to ameliorate symptoms, improve quality of life and prolong life expectancy.
The majority of colorectal cancer (CRC), including rectal cancer (RCA), is sporadic (70%) with an average age diagnosis after 50 years old. It is mostly linked to environmental factors, contrasts from a minority of patients (10%) with a true inherited pattern that carries a higher risk in patients younger than 50 years of age), and the remaining 20% is familial clustering in the absence of identifiable inherited syndrome. Approximately 5% of all CRC cancers are attributed Familial Adenomatous Polyposis (FAP), and Lynch syndrome (hereditary non-polyposis colorectal cancer [HNPCC]), but as high as 10% of unselected CRC patients will carry a high-risk mutation. Hereditary CRC syndromes are discussed in a different StatPearls chapter. Personal or family history of CRC, adenomatous polyps and polyps with villous or tubulovillous dysplasia are at high risk for synchronous or metachronous CRC primary cancer up to 3% to 5% at five years or even longer after resection requiring close screening. Inflammatory bowel disease (mainly ulcerative colitis (UC) and rarely Crohn's disease) that causes proctitis increases cancer risk with an estimated incidence 0.5% per year between 10 and 20 years after the time of UC diagnosis, and 1% per year after that with a 30% high probability by the fourth decade. Large studies have seen an increased risk of RCA associated with the history of radiation to prostate cancer (HR, 2.06; 95% CI, 1.42 to 2.99). Other diseases with a higher risk of colon cancer compared to rectal are diabetes mellitus/insulin resistance, uncontrolled acromegaly disease, and long-term immunosuppressed renal transplant. Epidemiologic studies’ results indicate strong environmental and lifestyle associations for CRC, but distinct patterns between colon and rectum potentially exist. Modest weak increased CRC risks are seen with obesity, red/processed meat, tobacco, alcohol, androgen deprivation therapy, and cholecystectomy among others, favoring colon cancer with no strong contributors to RCA. On the other hand, large population studies with variable strength evidence have found CRC protective factors such as physical activity, diet (fruits and vegetables, fiber, resistant starch, fish), vitamin supplements (folate, folic acid, pyridoxine B6, calcium, vitamin D, magnesium), garlic and coffee, drugs (aspirin, non-steroidal anti-inflammatory drugs [NSAIDs]), hormonal replacement therapy in postmenopausal women, statins, bisphosphonate, and angiotensin inhibitors, remaining true for Rca. Interestingly, a randomized controlled clinical trial found that 600 mg of aspirin given to patients with Lynch syndrome had a protective effect against colorectal adenomas and cancer with substantially reduced cancer incidence after 55.7 months with an HR of 0.56 (95% CI; 0.32, 0.99; p = 0.05).
CRC affects approximately 135,439 estimated new patients with a corresponding 39, 910 (30%) due to RCA in the United States every year. Among all cancers sites, it is the second leading cause of death in the United States, with an estimated 50,260 deaths, when colon and rectum are combined because a large number of deaths from RCA are misclassified as the colon. An 18% of RCA have a young onset (less than 50 years) with more advanced stage and poorer prognosis. The CRC incidence rate has been declining 3% per year since 2004 while increasing by 2% per year among screened young adults. Young adults increase is driven by left-sided colon cancer and rectal cancer (3.9% per year). CRC incidence varies worldwide with higher rates in developed countries than in developing countries. Low socioeconomic status has an increased risk of CRC; strongest in the rectum and weakest in the right colon, associated with poor risk behavior and access to medical care. White American lifetime average incidence of CRC is 5% but is higher in men than in woman and African American than non-Hispanic whites. From 1975 to 2014, there has been a 51% decrease in the mortality of CRC in the United States, attributed to early detection and improvement of treatment modalities. National Cancer Institute estimates that 65% of all treated patients for CRC will be alive at 5-years.
The transformation of normal rectal or colonic epithelium to a precancerous lesion (adenomas) and ultimately to invasive carcinoma requires an accumulation of genetic mutations either somatic (acquired) and/or germline (inherited). The theory of colorectal carcinogenesis features a clonal mutation evolution that gives a cell survival-immortality advantage and allows for the development of more mutations providing for cancer hallmarks such as proliferation, invasion, metastasis, and others. Clinical evidence has shown that CRC frequently arises from adenomatous polyps that typically acquire dysplastic changes in a 10 to 15 years period prior to developing invasive carcinoma and early detection-removal of polyps will reduce the incidence of CRC. Alternative new evidence has demonstrated that hamartomatous and serrated polyps could lead to CRC.
There are three major molecular pathways linked to CRC. These are chromosomal instability, mismatch repair, and hypermethylation. The chromosomal instability pathway is a gain of mutations unbalancing oncogene and tumor suppressors equilibrium as seen with mutations in the adenomatous polyposis coli (APC), a hallmark of FAP. Cells with deficiency of DNA mismatch repair (dMMR), commonly MLH1 or MSH2, accumulate errors within the genome that further will be repeated causing high levels of microsatellite instability (MSI-H), a hallmark of Lynch syndrome. CpG hypermethylation of DNA could activate or silence the expression of certain genes, BRAF and MLH1 respectively. Sporadic oncogenes somatic mutations (RAS, SRC, MYC) have been implicated in CRC, being RAS the most clinical relevance. On the other hand, tumor suppressors genes require bi-allelic loss (“2-hit model”) and are described in loss of APC 5q21 gene (80% sporadic), TP53 17p gene (50% to 70% sporadic), and DCC/SMAD2-4 18q gene (73% sporadic). Specific MMR gene mutations could occur in hMSH2, hMLH1, hPMS1 and hPMS2, hMSH6, and hMLH3; each one of them can interact with MLH1 and approximately found in 15% of all sporadic CRC causing a Lynch-like syndrome with MSI-H calling for universal testing. MUTYH defects have a recessive inheritance pattern at a time requiring bi-allelic second hit or in conjunction with APC gene mutation. Cyclooxygenase (COX-2) and peroxisome proliferator–activating receptor (PPAR) genes have been implicated in CRC tumorigenesis currently under investigation for chemo-protection.
The majority of all Rca are adenocarcinomas (90%) and others not frequently seen are adenosquamous, spindle, squamous and undifferentiated. RCA adenocarcinoma can be further differentiated in cribriform comedo-type, medullary, micropapillary, serrated, mucinous and signet-ring cell. Adenocarcinomas are categorized by the percentage of gland formation into well (greater than 95%), moderately (greater than 50%) and poorly (less than 49%) differentiated, but further divided in 2-tier low grade (well-moderate)/high grade (poor) with prognostic significance. Mucinous or signet ring cells pathological description denotes that more than half of the stained cells posses that particular characteristic. Differential clinicopathological diagnoses include neuroendocrine, hamartomas, mesenchymal, and lymphomas. Cytokeratin 20 (CK20) and caudal-type homeobox 2 (CDX2) immunohistochemistry (IHC) can accurately identify CRC adenocarcinoma origin, except medullary carcinoma with MSI-H expressing other markers such calretinin, CK7, SABT2, and CDH17.
The tumor, node, metastasis (TNM) staging system of the American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) 2017 is the preferred staging system for CRC. Of note, peritoneal carcinomatosis (M1c) and nodal micrometastases (cluster greater than 0.2 mm) are included as poor prognostic factors. Other prognostic indicators that the classification continues to exclude are extramural deposits, lymphovascular invasion (LVI), perineural invasion (PNI), poor histological grade, diagnostic serum carcinoembryonic antigen (CEA), MSI-H and RAS/BRAF mutation status. The most important prognostic indicator is pathological stage at presentation supported by data from the SEER observed overall survival (OS) at 5-year rates for rectum in Stage I includes 74%; Stage IIA, 64%; Stage IIB, 51%; Stage IIC, 32%; Stage IIIA, 74%; Stage IIIB, 45%; Stage IIIC, 33%; and Stage IV, 6%. Complete, intermediate, and poor tumor regression grade (TRG) at resection after neoadjuvant CRT in the German Rectal Cancer study group showed improved disease-free survival (DFS) with 90%, 74%, and 63% at 10 years, respectively. Surgical pathology specimens with neoadjuvant treatment effects are reported with “upstage and TRG” to more accurately reflect outcomes.
Complete mesorectal resection (R0) with all negative circumferential resection margins (CRM) is an important goal to avoid local-distant recurrence (38% CRM+ versus 10% CRM-) and furthermore increase survival. Lymph node involvement is the strongest outcome predictor and the total number resected directly influence prognosis on Stage II (node-negative) and Stage III (node-positive) disease. Current guidelines by the American Society of Clinical Oncology (ASCO), National Comprehensive Cancer Network (NCCN), and European Society for Medical Oncology (ESMO) recommend a minimum of 12 nodes surgical resection, although the numbers of lymph nodes are significantly less on rectal cancer treated with neoadjuvant therapy related to a higher TRG. Lymph node ratio from the INT-0089 trial showed a 5-year OS of 80% with less than 0.05 LNR versus 50% greater than 0.4 ratios depending on the total of node retrieved.
The NCCN and the College of American pathologists among other associations recommend universal MMR/MSI status (15 to 20% sporadic CRC), BRAF p.V600 E (less than 10% sporadic CRC) and RAS mutational (12 to 75% sporadic CRC) testing for prognostic and predictive of chemotherapy efficacy. hMLH1 and hMSH2 IHC provides a high sensitive (92.3%), and specific (100%) method, and positive (96.7%) / negative (100%) predictive value method for screening for MMR/MSI status. MSI-H is synonymous with dMMR, and MSI-stable will refer to proficient MMR state. The prevalence of MSI-H in CRC is about 15% in stage II, 8% in stage III, and 4% to 5% in stage IV. BRAF gene mutation testing is indicated after a negative MLH1 IHC indicating MLH1 gene down-regulation through somatic methylation. Recently, 6 independent classification systems coalesced into 4 consensus molecular subtypes (CMSs) with distinguishing features: CMS1 (MSI-immune, 14%), hypermutated burden, dMMR, microsatellite unstable and strong immune activation; CMS2 (canonical, 37%), high chromosomal instability, epithelial, marked WNT and MYC signaling activation; CMS3 (metabolic, 13%), epithelial and evident metabolic dysregulation, KRAS mutation; and CMS4 (mesenchymal, 23%), CpG hyper-methylation, prominent transforming growth factor-beta activation, stromal invasion, and angiogenesis. CMS classification had a prognostic value, CMS1 good, CMS4 poor and CMS2/3 intermediate.
Most all CRC will present either by diagnostic colonoscopy for suspicious signs and symptoms (80%), asymptomatic, routine screening (11%), or incidental finding at an acute abdomen emergent admission (7%). Patients diagnosed with routine cancer screening are frequently earlier stage compared to advanced disease of incidental surgical findings. Diagnostic colonoscopy’s triggers are blood per rectum (37%), abdominal pain (34%) and anemia (23%). Most common indications of emergent surgery are an obstruction (57%), peritonitis (25%) and perforation (18%).
Symptoms according to tumor location on clinical presentation of rectosigmoid are more frequently associated with a change in bowel habits (diminish stool caliber), bright red blood per rectum (hematochezia), pain (tenesmus), leakage diarrhea (mucus discharge) and constipation (obstruction). Late symptomatic presentation of metastatic disease at diagnosis will depend on the affected organ according to the dissemination route; more frequently lung via rectal veins draining to inferior vena cava rather into portal veins to the liver, to supraclavicular adenopathy via lymphatic or to neighbor structures by contiguous invasion. Physical examination should explore signs of ascites, hepatomegaly, and lymphadenopathy, and must extend to a digital rectal exam for fixed mass. A thorough family history is of great relevance to identify familial clusters and inherited patterns that would change surveillance and therapy of a high-risk patient.
Initial evaluation may involve barium enema or computed tomography (CT) colonography, but ultimately an endoscopy is required for tissue biopsy. Flexible sigmoidoscopy is no replacement for a complete diagnostic colonoscopy, still, a screening modality that reduces CRC mortality. The Federal Drug Administration (FDA) has approved PILLCAM 2 for those non-obstructed patients with incomplete colonoscopy, and not for routine screening. CRC screening modalities and recommendations for average and high-risk patients are discussed in a different StatPearls article. Routine laboratory workups with complete blood count (CBC), iron panel, basic metabolic panel, liver function test, and coagulation tests are not diagnostic but often useful for management. CEA greater than five ng/mL has a poor prognostic value when present, but lacks diagnostic sensitivity 46% (95% CI 0.45 to 0.47) and has limited specificity 89% (95% CI 0.88 to 0.92). Pre-operative CEA is indicated on all newly diagnosed CRC, normalization after surgical resection expected and serial assays monitor on follow-ups.
Baseline CT of the chest, abdomen, and pelvis with intravenous (IV) and oral contrast is the preferred cost-effective staging imaging study prior to surgical resection. CT abdomen and pelvis provides a moderate specificity for initial assessment of accurately staging T (50%) and N (73%) but rather provide immediate high screening sensitivity for distant metastasis (87%). CT chest remains controversial, as 9% will show indeterminate lesions, of that 11% represent metastatic lesions. MRI and CT triple phase imaging have improved the detection of liver metastasis. Positron emission tomography (PET) is not routinely indicated on the preoperative staging of CRC. Biopsy of a suspicious metastatic site should be performed to confirm the diagnosis.
Rigid sigmoidoscopy measures the distal extension of the tumor from the anal verge and is further divided to low (less than 5 cm), middle (less than 10 cm) or high (less than 15 cm) rectal cancer. Loco staging of Rca will require optimal imaging by transrectal ultrasonography (TRUS) and pelvic magnetic resonance imaging (MRI) to accurately determine T, N and predicting CRM optimal candidates for upfront surgery, radiation therapy or CRT. TRUS accuracy for T ranges from 80% to 95% and N from 70% to 75%, however, carries limitation with CRM on posterior tumors. MRI accuracy ranges for T from 81% to 92%, N from 69% to 84% and CRM 57% to 90%. In clinical practice, the information obtained with TRUS and MRI is often complementary and center-dependent directed. Although TRUS may be comparable to MRI for initial stating, it has shown significant limitations evaluating pre-operative treatment tumor response (re-stage) whereas MRI has greater post-treatment anatomical accuracy with the newer acquisition methods (sensitivity 88%) allowing a more precise surgical plan.
An accurate Rca staging will determine the most appropriate treatment route. Endoscopic resection (ER) is reserved for selected candidates with favorable-risk, and early stage (cT1N0M0) found in a completely excised rectal polyp. Upfront Rca surgical resection is appropriate for lesions that do not invade the muscularis propria and with negatives lymphatic nodes (cT2N0M0) on appropriate surgical candidates. Neoadjuvant combined CRT is recommended for locally advanced resectable colon cancer (cT3-4N0-2M0). Adjuvant therapy is strongly suggested for all pathological T3 and/or N positive tumors. Surgery in conjunction with peri-chemotherapy continues to provide a curative option on oligo-metastatic lung and liver disease CRC patients. Palliative systemic chemotherapy is offered to non-surgical candidates with unresectable locally advanced disease or high metastatic burden to improve quality of life and prolongs life expectancy. RCA local-recurrence may achieve cure with further multimodality therapy on selected cases. Colon cancer treatment modalities are discussed in a different StatPearls article.
Local excisions by endoscopy should achieve complete tumor resection. Endoscopic resections are offered to patients that will agree with close surveillance, open to accept surgical resection if T2 disease and/or are categorized unfit for surgical resection. Endoscopic resection procedures are to be offered by an experienced physician at a center of excellence. Invasion of muscularis propria (T2), poor histological grade, LVI, PNI, at the stalk or flat/depressed sessile polyp are considered ER Rca high-risk features and needs further surgical resection. RCA ER is indicated for T0-1, less than 3 cm diameter, mobile tumor, feasible resections margins by TRUS and no distant metastatic radiographic evidence plus the absence of high-risk features previously mentioned. ESMO recently recommended against lesions with submucosal invasion higher than 1000 micrometers due to a 12% nodal involvement rate. RCA lesions can be resected either by transanal excision (TAE) or transanal endoscopic surgery (TES) further divided in TE microsurgery, TE operation and TA minimally invasive surgery. Regardless of the procedure, all lesions should be tattooed to allow site identification on further steps. The DFS and OS at ten years of T1, RCA with favorable-risk features, local excisions are 92% and 98%, respectively, with 6% local recurrence seen within three years. Patients with high-risk features after endoscopic resections are strongly recommended to undergo immediate surgical resection rather than salvage approach based on a lower recurrence rate (8% versus 37%), higher metastatic rate (10% versus 23%), and prolong 10-year mOS (89% versus 72%). Adjuvant CRT has only shown to reduce local recurrence rate (7% versus 14%) although no mOS benefit.
Rca has strong recommendations for neoadjuvant therapy for stage II (T3 or T4 node-negative) and III (node-positive), although the best regimen has not been established. The recommended timing for surgery after neoadjuvant therapy varies by guidelines NCCN (5 to 12 weeks) and ESMO (4 to 12 weeks). Based on clinical trials, CRT does not increase surgical complication between recommended surgical times.
The landmark phase III German Rectal Cancer Study Group (CAO/ARO/AIO-94) randomly assigned patients with clinical stage T3 or T4 or node-positive disease to receive either preoperative or postoperative CRT (RT: 5040 cGy delivered in fractions of 180 cGy per day, 5 days per week, and FU: 120-hour continuous IV infusion at a dose of 1000 mg/m2 per day during the first and fifth weeks). After surgery, four 5-day additional cycles of FU (500 mg/m2) were given. Preoperative chemoradiotherapy, as compared with postoperative chemoradiotherapy, reduced local relapse (6% versus 13%), toxicity (27% versus 40%) but did not improve distant metastasis (29%) or overall survival (59%) at the 10-year, follow up. Secondary analysis showed worse outcome on pathological node positivity (N2) for postoperative treatment and better prognosis with higher tumor regression scores with pre-operative treatment. The National Surgical Adjuvant Breast and Bowel Project (NSABP) R-03 trial compared neoadjuvant versus adjuvant CRT (FU/LV /RT) in the treatment of locally advanced rectal carcinoma but did not reach accrual and closed early. NSABP showed an improving trend for DFS and OS (75% versus 66%, p = 0.065) in the neoadjuvant arm. The Korean trial failed to show benefit with CRT by using capecitabine as a radiosensitizer.
The benefit of adding chemotherapy to radiation preoperative followed by either by three cycles of adjuvant chemotherapy or surveillance was address by the European Organization for Research and Treatment of Cancer (EORTC) 22921. CRT consisted of 45 Gy to the posterior pelvis in 25 fractions of 1·8 Gy over five weeks and each course of FU (350 mg/m(2) per day) and LV (20 mg/m(2) per day intravenous bolus. The primary endpoint of survival did not differ at ten years follow-up of neoadjuvant RT (49%) and CRT (50.7%), moreover, adjuvant chemotherapy (51.8%) did not differ from observation (48.4%). Three trials led by the Italian Study group, Dutch Colorectal PROCTOR/SCRIPT and UK Chronicle trial showed no survival benefit of post-surgical adjuvant chemotherapy after pre-surgical neoadjuvant therapy, further supported by a combined meta-analysis.
FU/LV is the standard accepted chemotherapy, but capecitabine has become an acceptable alternative surrogate. Addition of oxaliplatin has no significant impact outcomes but increased toxicity corroborated by different designed trials (STAR-01 trial, ACCORD 12/0405 PRODIGE 2 trial, PETACC-6 trial, NSABP R-04 trial) except for the German CAO/ARO/AIO-04 trial with FOLFOX6-RT that modestly improved pathological complete response (17% versus 13%) and 3-year DFS (76% versus 71%) compared to FU/LV-RT, while toxicity remained equal. Irinotecan, bevacizumab, cetuximab, and panitumumab have shown no benefit in neoadjuvant treatment.
The ultimate goal in invasive Rca is the complete resection of the tumor and removal of the lymphovascular spread by a goal of the minimum negative proximal margin of 5 cm, distal of 2 cm and radial greater than 1 mm. A secondary aim is the restoration of the bowel continuity, either by one-stage primary anastomosis or two-stage approach temporary diversion. Open transabdominal Rca surgery is the preferred surgical modality, but the laparoscopic approach is recommended on selected patients with experienced surgeons, robotic surgery does not confer an advantage in RCA. Potential palliative surgical procedures for unresectable RCA tumors include resection with primary anastomosis, diverting colostomy, bypass procedure, and endoluminal stent fulguration endocavitary radiation.
RCA surgical technique depends mainly on the sphincter-sparing procedure by low anterior resection (LAR) or not sparing by abdominal perineal resection (APR) according to the preoperative staging which determines if distal margins are achievable by LAR. All APR/LAR must include total mesorectal excision (TME) to ensure margins and lymph node retrieval. LAR continues to be the gold-standard technique and is preferred if potential negative margins less than 1 cm with LAR, low-lying Rca, salvage procedure, and poor anorectal function. Multivisceral resection is reserved only for locally advanced rectal cancer involving adjacent organs or bones (T4). Meta-analysis of randomized trials (including COLOR trial, CLASSIC trial, COREAN trial, ACOSOG Z6051 and AlaCart) of laparoscopic Rca resection had 13% rate of incomplete TME compared to open surgery 10%, but similar LN retrieval and margins goal, pending long-term outcomes but early reports showed OS and DFS. Rca with positive margins is associated with an HR of 16.8 for local recurrence, and 2.35 decreased 5-year OS even with adjuvant therapy. The adequate node sampling number beyond the field for RCA is not clear specifically after neoadjuvant CRT. Expected local recurrence either by LAR or APR with TME ranges from 4% to 7%.
Surgery alone can be curative for patients with stage I disease. For stage II (T3-T4) and III diseases (node-positive), the 5-year survival rate can range from 30% to 60% with surgical resection alone thus adjuvant therapy is recommended. Neoadjuvant therapy is the preferred treatment route for stage II and III RCA, nonetheless clinically some patients opt upfront surgical resection requiring postoperative therapy. Adjuvant therapy for stage II and III RCA should be offered to all patients regardless of the surgical procedure outcomes according to ASCO and NCCN guideline. ESMO recommends only be an offer to patients with high-risk recurrence features (positive margins, perforation, invasion of adjacent organs, incomplete TME, tumor deposits and extra-capsular nodal deposits). NCCN guidelines recommend post-operative chemotherapy (FOLFOX or CAPOX) sequenced before and/or after combined chemoradiotherapy (fluoropyrimidine as radio-sensitizer) within 4 to 6 weeks of surgical resection depending on patient health recovery for up to 6 months length or 4 months when preoperative CRT was given. ASCO guidelines recommend full weight-based chemotherapy dosing. The optimal sequencing of peri-operative chemotherapy and radiotherapy remains investigational.
Stage II and III RCA have a high 35% and 65% local recurrence rate, respectively. Radiation therapy (RT) proved to reduce 5-year local recurrence compared to observation alone (17 vs. 28%) but did not improve survival (58% versus 59%). Early survival evidence of combined chemotherapy fluorouracil (FU) and RT (40 Gy) showed lower recurrence rate compared to surgery alone (33% versus 55%) although not different from RT alone was seen in the Gastrointestinal Tumor Study Group (GITSG) protocol. A significant superiority of CRT (FU plus semustine and RT) over RT (50 Gy) was seen until the North Central Cancer Treatment Group (NCCTG) trial with a reduced recurrence of RCA by 34% (P = 0.0016; 95% CI, 12% to 50%) and the overall death rate by 29% (P = 0.025; 95% CI, 7% to 45%). Of note, both trials FU-based CRT regimens are considered inferior to current guidelines. Leucovorin (LV) benefit to FU failed to improve DFS and OS in the United States Intergroup trial 0114 (INT-0114). Capecitabine (825 mg/m2 daily, five days per week during RT) was non-inferior to FU reported by the National Surgical Adjuvant Breast and Bowel Project 04 trial.
Chemotherapy alone by oxaliplatin (FOLFOX) or irinotecan (FOLFIRI) postoperative regimens either receiving prior neoadjuvant or adjuvant CRT showed to be safe but unfortunately added no benefit in the INT-E3201 trial. In the ADORE open-label, multicenter, phase 2 trial Rca stage II and III patients were randomly assigned to receive adjuvant chemotherapy either by four cycles of FU/LV (FU: 380 mg/m(2) and LV: 20 mg/m(2) on days 1 through 5, every 4 weeks) or 8 cycles of FOLFOX (O: 85 mg/m(2), LV: 200 mg/m(2), and FU: bolus 400 mg/m(2) on day 1, and FU: infusion 2400 mg/m(2) for 46 hours, every 2 weeks) after preoperative CRT. 3-year DFS was 72% (95% CI 64% to 78%) in the FOLFOX group and 63% (55% to 70%) in the FU/LV group (hazard ratio 0·657, 95% CI 0·434-0·994; p=0·047).
Based on these results, the preferred adjuvant chemotherapy for RCA is an oxaliplatin-containing regimen (FOLFOX, CAPOX) administered for two months, followed by a fluoropyrimidine-based CRT and completed by two months of additional chemotherapy. Alternative four months of chemotherapy followed by two months of CRT is acceptable. Fluoropyrimidine radiation sensitizer current regimens include weekly bolus FU (the Mayo and Roswell Park regimen), short-term infusional FU (the de Gramont regimen), and single-agent capecitabine explained in systemic therapy section. Standard radiation dose is 45 to 50 Gy in 25 to 28 fractions to the pelvis with a tumor bed boost with a 2 cm margin of 5.4 to 9 Gy in 3 to 5 fractions. At this point in time, both irinotecan and targeted-based therapies have no indication in the adjuvant therapy of Rca outside of investigational trials. There are two web-based validated tools to calculate the relative risk of disease recurrence and mortality based on clinicopathological features and potential adjuvant benefit (an ACCENT by the Mayo Clinic and Adjuvant! Online). Furthermore, gene profiling by Oncotype DX Colon, ColDX or ColoPrint are currently available in clinical practice to assist in risk assessment but has no valid predictive value, thus not recommended by clinical practice guidelines.
More than half of all CRC patients will develop metastasis and the majority to the liver (80% to 90%). The prognosis for advanced non-resectable and metastatic CRC patients with best supportive care is poor by an mOS 5 to 6 months, except for a subset of patients with oligo-metastatic hepatic or pulmonary patient that are potentially curable with peri-operative chemotherapy. The goals of systemic therapy for advanced non-resectable and metastatic CRC of therapy is palliation of symptoms, improvement of the quality of life and prolong survival. Systemic therapy for cancers of the colon and rectum are addressed jointly by clinical research and practice. In the 1990s, FU/LV monotherapy was the standard first-line therapy with an approximately mOS 12 months. It was not until the early 2000s that the addition of oxaliplatin and irinotecan to the backbone of FU/LV resulted in an improvement of mOS to nearly 24 months. The introduction of biologic agents monoclonal antibodies (MAbs) targeting vascular endothelial growth factors (VGFRs) and epidermal growth factors receptor (EGFR) further enhanced the efficacy of systemic medical therapy with mOS reaching 36 months and when exposed to multiple lines of therapy an impressive 20% 5-year OS even at metastatic stage.
Current availability of nine different antineoplastic class and more than a dozen drug options (Fluoropyrimidines [FU], Capecitabine [CAP], S-1, tegafur plus uracil [UFT]), Irinotecan, Oxaliplatin, anti-EGFRs (Cetuximab, Panitumumab), anti-VEGFRs (Bevacizumab, Ramucirumab), recombinant fusion protein (Aflibercept), tyrosine kinase inhibitors (Regorafenib), antimetabolites (TAS-102, Lonsurf), immunotherapy (Nivolumab, Pembrolizumab) for mCRC has resulted in an abundance of therapeutic possibilities with different combinations and sequences, not yet established. The most appropriate treatment regimen is conceivably the one that generates the highest overall response rate (ORR), greatest impact to metastases for surgical conversion and/or ultimately offers the longest PFS and OS, balanced with a favorable toxicity profile. Predictive biomarker (RAS/BRAF type and MMR/MSI status), primary location (tumor and metastasis), patient (performance status and comorbidities) and therapy goals (palliate or conversion) can help guide treatment decisions for specific patient subpopulations.
For fit candidates, the standard first-line treatment approach is to start with a FOLFOX, CAPOX or FOLFIRI regimen plus anti-EGFR (Cetuximab or Panitumumab) when the tumor is RAS/BRAF wild-type and left-sided or anti-VGFR (Bevacizumab) when the tumor is RAS/BRAF mutated or right-sided. At progression of disease or unacceptable toxicity switch to the opposite second-line regimen with either continuation of prior anti-VGFR or new biological agent regardless of the tumor location. ASCO, ESMO, and NCCN recommend comprehensive testing of KRAS, NRAS exons 2 (codons 12 and 13), 3 (codons 59 and 61), 4 (codons 117 and 146) and BRAF V600E for anti-EGFR candidates based the results of the Panitumumab Randomized Trial in Combination with Chemotherapy for Metastatic Colorectal Cancer to Determine Efficacy (PRIME) trial and other major meta-analyses. Biological agents additions to chemotherapy backbone have significantly improved response rate, DFS, and OS in the first and second line setting. Cetuximab and Panitumumab have comparable efficacy in first-line, and second-line addition and even single-agent salvage therapy, but may share cross-resistance limiting use after progression of anti-EGFRs. Bevacizumab can be continued beyond progression of first-line to second-line therapy (e.g., RAS/BRAF-mutated) but not in conjunction with a new anti-EGFRs candidate (e.g., RAS/BRAF wild-type and left sided). For patients who progressed anti-VEGF bevacizumab may consider aflibercept or ramucirumab addition to backbone chemotherapy regimen.
Sequential single agents remains a valid option for mCRC patients compared to combination treatments demonstrated in the United Kingdom Medical Research Council's FOCUS (Fluorouracil, Oxaliplatin, CPT-11: Use and Sequencing) and the Dutch Colorectal Group CAIRO (CApecitabine, IRinotecan, Oxaliplatin) trials, although biological agents were not used in both trials, and most of all patients did not receive all three drugs. Multiple phase III trials have confirmed similar ORR, PFS, and OS between FOLFOX and FOLFIRI regimens, even with added anti-VEGFs or anti-EGFRs. In clinical practice, the choice between FOLFOX and FOLFIRI is based on treatment toxicity profile. CAPOX combination is a statistically non-inferior substitute to FOLFOX regimen in PFS and OS with tolerable toxicity profile; conversely, XELIRI had shown higher intolerable rates of gastrointestinal toxicities. Maintenance therapy with low dose capecitabine-bevacizumab after the first-line FOLFOX-bevacizumab good response is an accepted treatment option based on CAIRO3 trial results. Complete discontinuation of chemotherapy may be detrimental and remains controversial, but intermittent therapy has not resulted in significant reduction in OS but rather, a better quality of life.
FOLFOXIRI with or without bevacizumab is reserved for conversion approach in potential resectable liver/lung metastasis on selected excellent performance status patients. Patients not fit for a triplet or doublet regimen, consensus-based guidelines recommend FU/LV (IV) or capecitabine PO monotherapy. For patients after FOLFOX, CAPOX, FOLFIRI, anti-VEGFs and anti-EGFRs agent progression that remains eligible for further chemotherapy may consider regorafenib or Lonsurf (TAS-102) salvage therapy, or immunotherapy (in MSI-H/dMMR tumors). Novel targeted agents may be available through clinical trials at different lines of therapies failure.
Recommended treatment monitoring consists of close observation for signs/symptoms of adverse reaction (before each treatment cycle and as needed), frequent checking on blood work parameters (before each treatment cycle and as indicated), CEA levels serial assays (every 1 to 3 months) and radiographic evaluation (every 2 to 3 months) by a multidisciplinary team. Patients should be frequently screened for somatic symptoms and psychosocial distress.
Selected Landmark Clinical Trials
Fluorouracil-Leucovorin (FU/LV): An updated-meta analysis compared to FU-alone resulted on higher overall response rates (ORR 21% versus 11%) and 1-year OS (47% versus 37%) favoring FU/LV. The 2 most commonly used regimens in the United States include the Mayo regimen (bolus FU: 425 mg/m2 and LV: 20 mg/m2 on days 1 to 5 every 4 to 5 weeks) and the Roswell Park regimen (Bolus FU: 500 mg/m2 and LV: 500 mg/m2 administered weekly for 6 out of 8 weeks). Further improved by the de Gramont regimen (LV: 200 mg/m2 as a 2-hour infusion followed by bolus FU: 400 mg/m2 and 22-hour infusion FU: 600 mg/m2 for 2 consecutive days every 2 weeks) with significantly better ORR (33% vs 14%), PFS (7 vs 5 months) and OS (15.5 vs 14.2 months) with significantly less grade 3/4 toxicities (23.9% versus 11.1%) compared to the Mayo regimen.
Capecitabine (CAP): A phase III clinical trial by Hoff PM et al prospectively randomly assigned patients to oral CAP (1250 mg/m2 twice daily for 14 days every 21 days) or FU/LV (the Mayo regimen) resulting in non-inferior ORR (24.8% versus 15.5%), time to progression (TTP 4.3 versus 4.7 months) and mOS (12.5 versus 13.3 months), respectively. CAP, common side effects of this drug included diarrhea (overlapping toxicities with irinotecan), hyperbilirubinemia and hand-foot syndrome. CAP has never been directly compared with infusional FU/LV (de Gramont regimen), CAPOX versus infusional FOLFOX has shown to be of similar efficacy in the treatment of advanced CRC, as shown by the phase TREE-1 trial and AIO trial. In the United States clinical practice, dose-reducing CAP by about 20% (1000 mg/m2) alone or in combination regimens does not appear to decrease the treatment efficacy, but it greatly improves the side effect profile of the treatment. Other alternative oral fluoropyrimidines not approved in the United States include S1- derived three different agents tegafur-uracil, gimeracil and oteracil, and raltitrexed.
Irinotecan (IRI): In a landmark phase II clinical trial, patients with FU-refractory mCRC were randomly assigned either single-agent IRI (300mg/m2 every 3 weeks) or BSC, showed a significant 1-year OS (36% versus 14%) with improved quality of life (performance status, weight loss, and pain-free), respectively. Following this, 3 key trials were conducted to test the role of IRI versus FU/LV in the front-line setting. A 3-arm trial compared three treatment regimens: the Roswell Park regimen, IRI plus FU/LV weekly bolus regimen (IFL or Saltz regimen); and IRI alone.
The results of the trial significantly favored IFL with an ORR (39% versus 21%) and mOS (14.8 versus 12.6 months). In Europe, 3 pivotal phase III trials compared the FU/LV regimens to FOLFIRI regimens (the Douillard regimen-IRI: 180 mg/m2 on day 1, FU: 400 mg/m2 bolus followed by 600 mg/m2 over 22 hours, both on days 1 and 2, and LV: 200 mg/m2 on days 1 and 2). The Douillard FOLFIRI regimen trial demonstrated a significant ORR (49% versus 31%), TTP (6.7 versus 4.4 months) and extended mOS (17.4 versus 14.1 months). Higher side effects of irinotecan-group were diarrhea, myelosuppression, and alopecia. Of note, IRI combinations require an adequate biliary function for its active glucuronide metabolite, SN-38, be excreted. Approximately 10% of US patients are homozygous for the UGT1A1*28 allele polymorphism increasing SN-38 bioavailability and therefore recommended starting at a lower dose of irinotecan.
Oxaliplatin (OXA): OXA has very limited activity in CRC as a single agent thus not recommended unless synergist with fluoropyrimidines. In phase III trial, FU/LV (de Gramont regimen) was compared with or without OXA (OXA: 85 mg/m2 on day one over two hours, FOLFOX4 regimen) as first-line therapy for patients with mCRC. FOLFOX4 regimen had significantly higher ORR (51% versus 22%), PFS (9 versus six months) but comparable mOS (16.2 versus 14.7 months) with more grade 3/4 neutropenia, diarrhea, and neurotoxicity. Because no overall survival benefit was achieved in these first-line trials, the FDA did not approve OXA for CRC until years later for second-line therapy that showed prolonged PFS and increased ORR compared with FU/LV for patients who experienced disease progression while receiving first-line IRI-regimens. The most important side effect and dose-limiting toxicity of OXA is neurotoxicity. It may present as an acute and reversible, cold-triggered sensory neuropathy, or a chronic dose-limiting cumulative sensory neurotoxicity.
Irinotecan versus Oxaliplatin-Based Regimens: After encouraging results of previous trials conducted in the United States and Europe using OXA and IRI, the North Central Cancer Treatment Group (NCCTG)/Intergroup trial N9741performed a pivotal and practice-changing trial comparing FOLFOX4, the standard combination IFL and new IROX combination (OXA:85 mg/m plus IRI 200 mg/m, both on day 1 every 3 weeks). The N9741 results demonstrated the superiority of FOLFOX compared with IFL and IROX as first-line therapy for mCRC with ORR (45% versus 31% versus 36%), PFS (8.7 versus 6.9 versus 6.7 months) and mOS (19.5 versus 15 versus 17.3 months), respectively. The toxicity profile likewise favored FOLFOX, except for neurotoxicity. FOLFOX emerged as new standard first-line therapy with rapid and widespread adaptation in the United States.
Bevacizumab (BEV): In randomized placebo-controlled phase III trial, IFL plus BEV (5 mg/kg every two weeks) was assessed in first-line therapy for mCRC. For the first time, the addition of an anti-VEGF inhibitor was validated as an efficacious antineoplastic treatment by significantly improving ORR (45% versus 35%), PFS (10.6 versus 6.2 months), and mOS (20.3 versus 15.6 months). This trial was the first phase III validation of an antiangiogenic agent as an effective treatment option in a human malignancy. Subsequently, FOLFOX-BEV also showed to improve mOS in first-line (TREE-2 trial 23.7 versus 18.2 months) and second-line (ECOG 3200 trial 12.9 versus 10.8 months) settings. The AVEX phase III trial selected patients age 70 or older, who were not deemed to be candidates for OXA or IRI-based chemotherapy first-line regimens, to receive CAP regimen alone or with BEV (7.5 mg/kg intravenously on day 1 given every three weeks). The study showed a significantly longer mPFS (9.1 versus 5.1 months) and a remarkable improvement in mOS (20.7 versus 16.8 months) for CAP-BEV than with CAP alone, respectively. Prolonged VEGF inhibition with BEV beyond the first-line progression was evaluated by the European TML (ML18147) phase III trial, which randomly assigned at progression within three months to either continue second-line with or without BEV. The mOS, primary endpoint of the study, favored BEV-chemotherapy based continuation 11.2 months versus 9.8 months chemotherapy alone. This effect was confirmed in PFS (5.7 versus 4.1 months) but not in ORR (4% versus 5%). The toxicity profile observed with BEV consists of hypertension, bleeding, gastrointestinal perforations, impaired wound healing and arterial-venous thrombotic events.
Aflibercept (AFL): AFL was evaluated in the second-line setting among patients who had all progressed OXA-based with or without BEV first-line chemotherapy by the placebo-controlled phase III VELOUR trial. Patients randomly selected to receive FOLFIRI with AFL (4 mg/kg IV every two weeks) had a significant improvement in ORR (19.8% versus 11.1%), PFS (6.9 versus 4.7 months and mOS (13.5 versus 12.1 months).
The aflibercept arm toxicity profile had an intensified rate of grade 3/4 diarrhea, mucositis, neutropenia, infection, and fatigue plus hypertension, proteinuria, hemorrhage, and arterial-venous thromboembolic events, usually seen in FOLFIRI-BEV second-line.
Ramucirumab (RAM): In a double-blind placebo-controlled phase III RAISE trial, the addition FOLFIRI-RAM (8 mg/IV every 2 weeks) as second-line therapy for patients progressing with a FOLFOX-BEV improved PFS (5.7 versus 4.5 months) and mOS (13.3 versus 11.7 months). Grade 3/4 neutropenia, hypertension, diarrhea and fatigue was worse on the combination arm.
Anti-EGFR Monoclonal Antibodies
Cetuximab (CET): CET monotherapy (400 mg/m2 followed by a weekly infusion of 250 mg/m2) for patients who had experienced disease progression on prior FU, OXA and IRI-based therapy had an improved ORR (40% versus 11%) and mOS (6.1 versus 4.6 months) compared to BSC. The EPIC and BOND trials proved PFS and ORR superiority of IRI-CET over IRI-alone on patients with prior IRI failure, without significantly improving survival. A large multicenter randomized phase III CRYSTAL trial, FOLFIRI with or without cetuximab in treatment naïve CRC patients, confirmed that selected KRAS wild-type patients have a significantly higher ORR, PFS, and mOS (23.5 versus 20 months). The benefit of FOLFOX-CET remains uncertain with three trials (OPUS, CALBG 80203, CALBG 80405) showing modest significant ORR and PFS improvements but no mOS benefit, and three other trials (MRC COIN, NORDIC VII, New EPOC) with no resulted benefit. CET arm had higher-grade diarrhea, magnesium-wasting syndrome, infusion reaction and ocular-skin toxicity, last correlated to response rate.
Panitumumab (PAN): Single-agent PAN (6 mg/kg every two weeks) was compared to BSC in a large international phase III trial in chemotherapy-refractory mCRC patients. PAN had an ORR (37% similar to CET) and modest prolonged mPFS (8 versus 7.3 weeks), but mOS was not increased likely because crossed over from BSC to the PAN arm. The ASPECCT phase III trial compared head-to-head PAN vs. CET monotherapies on chemotherapy-refractory patients and showed non-inferior mOS (10 months) with similar expected toxicity profile. The FOLFOX4-PAN benefit in first-line setting was seen in phase III PRIME trial with mPFS improvement (9.6 versus eight months) and trend for mOS (23.0 versus 19.7 months) not significant at 55 weeks follow up.
VEGF Inhibitors versus Anti-EGFR Monoclonal Antibodies
The FIRE-3 trial recruited treatment naïve mCRC KRAS exon2 wild-type patients and randomly assigned to receive FOLFIRI+CET or FOLFIRI+BEV. The primary endpoint of the trial, objective response rate analyzed by intention-to-treat analysis, was not reached (CET 62% versus. BEV 58%, p = 0.18). Although no difference in PFS was noted (CET 10 versus BEV 10.3 months), the mOS was significantly longer in the FOLFIRI+CET arm (CET 28.7 versus BEV 25.0 months; HR, 0.77; p = 0.017). An updated analysis, which accounted for additional mutations in KRAS exon 3 and 4, as well as NRAS mutations exon 2 and 3, demonstrated a longer mOS survival of 33.1 months for FOLFIRI+CET. In contrast, the U.S. Intergroup study, CALGB/South-west Oncology Group (SWOG) 80405 trial that compared FOLFOX or FOLFIRI (“dealer’s choice) with CET compared with BEV as first-line therapy had no difference in mOS (30 versus 29.9 months, respectively), not even whit expanded RAS-mutated analysis. A preliminary retrospective analysis suggests that patients treated with CET for KRAS wild-type left-sided tumors had an mOS that was significantly higher than right-sided counterpart by 33.3 versus 19.4 months, respectively. Combination therapy with VEGF inhibitors and anti-EGFRs monoclonal antibody have failed and even showed detrimental outcomes in multiple trials (BOND-2, PACCE, CAIRO-2), thus not recommended.
Salvage Therapy for Refractory Disease
Regorafenib (REG): REG (160 mg by mouth daily for 3 weeks of 4 weeks cycle) efficacy was investigated in a placebo-controlled, multicenter international, randomized, phase III CORRECT trial after progression of multiple therapies. REG compared with placebo had ORR (41% versus 15%), PFS (1.9 versus 1.7 months) and modest mOS benefit (6.4 versus 5.0 months) but significant (p = 0.0052). The observed toxicity profile with regorafenib was hand-foot skin reaction, fatigue, hypertension, diarrhea, and rash, with a 1.6% fatal hepatic failure. REG benefit was later confirmed by the CONCUR trial.
Trifluridine-tipiracil (TAS-102): The international, double-blind, placebo-controlled randomized-controlled, phase III RECOURSE study of patients with refractory showed that TAS-102 (35 mg/m2 by mouth twice daily on days 1 through 5, and 8 to 12 of each 28-day cycle) improved ORR (44% versus 16%), DFS and mOS (7.1 versus 5.3 months); The most adverse event was neutropenia (38%), febrile neutropenia (4%) and 1 treatment-related death.
Immune Checkpoint Inhibitors
Pembrolizumab (PEM): In a phase II study, PEM (10 mg/kg every 14 days) was given to 11 heavily pretreated patients with dMMR mCRC with a remarkable ORR of 71% and PFS rate of 67%. An expanded cohort of 54 patients, PEM showed OOR of 50% and an 89% disease control rate with a durable response for more than a year, not seen in pMMR CRC.
Nivolumab (NIV): In CheckMate-142, NIV (3 mg/kg every 2 weeks) was given to 74 dMMR CRC patients resulted in 31% objective response, 69% disease control and median duration response was not reached at 12 months.
Utilize CRC screening guidelines to guide and avoid repeating other screening modalities after a high-quality colonoscopy for ten years or with less than ten years life expectancy.
Do not use epidermal growth factor targeted-therapy against colorectal cancer unless the patient has a tumor biomarker (RAS/BRAF or MMR/MSI) that predicts effective response, advanced non-surgical stage, and satisfactory performance status.
Offer early palliative care for all CRC patients with uncontrolled physical symptoms or psychosocial stress even during oncological disease treatment, and encourage early conversations of the end of life of those with the metastatic stage.
All CRC patients should be supported for survivorship care after completion of treatment. A discussion on lifestyle changes including to follow a healthy diet, obtain and maintain ideal body weight, establish an active exercise routine, minimize alcohol consumption, and quit smoking, should be encouraged.
Establish a route for prevention and detection of new or recurrent CRC to minimize psychological stress, offer genetic risk assessment when appropriate and manage long-term effects of cancer treatment.