INTRODUCTION — Approximately 106,180 new cases of colon cancer are diagnosed annually in the United States . Surgical resection is the only curative treatment for locoregional disease. Outcome is most closely related to the extent of disease at presentation (table 1 and figure 1) . (See "Clinical presentation, diagnosis, and staging of colorectal cancer", section on 'TNM staging system'.)
For patients who have undergone potentially curative resection, disease recurrence is thought to arise from clinically occult micrometastases that are present at the time of surgery. The goal of postoperative (adjuvant) therapy is to eradicate these micrometastases, thereby increasing the cure rate.
The benefits of adjuvant chemotherapy have been most clearly demonstrated in stage III (node-positive (table 1)) disease (an approximately 30 percent relative reduction in the risk of disease recurrence and a 22 to 32 percent relative reduction in mortality), and it has become a standard approach in this setting. The benefits of adjuvant chemotherapy in stage II (node-negative) disease are less certain, and the use of chemotherapy in this group is variable. Decision-making is influenced by tumoral deficiency of DNA mismatch repair (MMR) enzyme status/high levels of microsatellite instability (dMMR/MSI-H), which is associated with better prognosis overall. All patients with stage II colon should be tested for dMMR (either by microsatellite instability [MSI] testing or immunohistochemistry [IHC] for expression of MMR proteins) to aid in chemotherapy decision-making. (See 'Testing for Lynch syndrome' below.)
This topic review will cover adjuvant therapy for patients with resected stage II colon cancer. Trials demonstrating the benefit of adjuvant therapy for patients with stage III (node-positive) colon cancer, optimal timing of adjuvant therapy, use of adjuvant regional liver-directed therapy and adjuvant radiation therapy, adjuvant therapy for colon cancer in older adult patients, benefits of adjunctive therapy with diet, exercise, aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs), vitamin D, and coffee consumption in patients with early stage colorectal cancer, surgical management and prognosis of colon cancer, chemotherapy after resection of colorectal cancer liver metastases, and recommendations for post-treatment follow-up are discussed separately. (See "Adjuvant therapy for resected stage III (node-positive) colon cancer" and "Adjuvant therapy for resected colon cancer in older adult patients" and "Adjunctive therapy for patients with resected early stage colorectal cancer: Diet, exercise, NSAIDs, and vitamin D" and "Overview of the management of primary colon cancer" and "Potentially resectable colorectal cancer liver metastases: Integration of surgery and chemotherapy" and "Post-treatment surveillance after colorectal cancer treatment".)
IMPACT OF DNA MISMATCH REPAIR — Approximately 5 to 15 percent of colorectal cancers (CRCs) have sporadic or inherited (Lynch syndrome) deficiency of a DNA mismatch repair (MMR) protein, most commonly MutL homolog 1 (MLH1) or MutS homolog 2 (MSH2). A high level of microsatellite instability (MSI-H) is the biologic footprint of deficient MMR (dMMR). (See 'Testing for Lynch syndrome' below and "Molecular genetics of colorectal cancer", section on 'Mismatch repair genes' and "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis".)
MMR status influences decision-making for adjuvant therapy in two ways:
●MMR status is an important prognostic factor in stage II disease. Patients with dMMR have an approximately 50 percent lower risk of recurrence and a relatively higher survival rate after surgery alone compared with tumors that are proficient in MMR [3,4]. (See "Pathology and prognostic determinants of colorectal cancer", section on 'Mismatch repair deficiency'.)
●The majority of published data also suggest that MMR status is an important predictor of benefit from fluoropyrimidine-based adjuvant chemotherapy. Most studies support the view that dMMR tumors are resistant to fluoropyrimidine chemotherapy but retain sensitivity to oxaliplatin. (See 'Patients with deficient MMR' below.)
ADJUVANT CHEMOTHERAPY — Our general approach to adjuvant chemotherapy in patients with resected stage II disease is outlined in the algorithm (algorithm 1), and is informed by the DNA mismatch repair (MMR) status as well as the presence or absence of high-risk clinicopathologic features.
Patients with proficient MMR
Indications for adjuvant therapy — For patients with stage IIA colon cancers that are proficient in mismatch repair (pMMR) and who are at low risk for recurrence based on clinicopathologic features, we suggest observation rather than adjuvant chemotherapy. For patients with a pT4 primary tumor or two or more other high-risk features (fewer than 12 sampled lymph nodes in the surgical specimen, perineural or lymphovascular invasion, poorly differentiated or undifferentiated histology, clinical intestinal obstruction, tumor perforation, or grade BD3 tumor budding), we suggest adjuvant chemotherapy. For patients with one high-risk clinicopathologic feature other than a T4 primary, we discuss the risks and benefits of adjuvant chemotherapy and individualize decision-making.
Is there benefit for fluoropyrimidine-based chemotherapy?
Low-risk patients with stage II disease — For most patients without specific high-risk features, we suggest observation rather than adjuvant therapy. Studies in patients with both high- and low-risk disease (most of whom had pMMR disease) have not shown conclusive evidence of benefit, and if present, the absolute benefit is likely to be small and offset by chemotherapy side effects.
The benefits of fluoropyrimidine-based adjuvant systemic therapy have been clearly shown for patients with stage III disease. Multiple clinical trials have tested the value of such therapy in resected stage II colon cancer, showing at best a small disease-free survival (DFS) and a potentially very small overall survival (OS) benefit from adjuvant therapy, which is no greater than 5 percent at five years. Adjuvant chemotherapy cannot be considered a standard approach to all patients with stage II colon cancer.
●Chemotherapy benefit in pMMR tumors – The available data addressing the benefit of adjuvant chemotherapy in patients with pMMR tumors are mixed:
•In a pooled analysis of data from two fluoropyrimidine adjuvant therapy trials, a treatment benefit was not found in patients with pMMR stage II disease (hazard ratio [HR] for DFS 0.84, 95% CI 0.57-1.24) . (See 'Patients with deficient MMR' below.)
•Other series have reported modest benefits for adjuvant chemotherapy in patients with pMMR [6-9].
•A pooled analysis of data from two studies [5,8,10] specifically addressing the impact of adjuvant fluoropyrimidine chemotherapy in patients with stage II colon cancer and pMMR tumors concluded that OS was significantly improved (1093 patients in two studies, HR 0.31, 95% CI 0.18-0.52, translating to 242 fewer deaths per 1000 (95% CI, 298 fewer to 160 fewer), but DFS was not significantly different (1162 patients, HR 0.56, 95% CI 0.25-1.26, translating to 171 fewer recurrences per 1000 [95% CI 323 fewer to 81 more]) . (See 'Higher-risk stage II disease' below.)
●Trials in all risk groups – Four large trials that specifically address the benefit of fluoropyrimidine-based chemotherapy in populations consisting either entirely or predominantly of stage II disease have been undertaken, all of which fail to show unequivocal OS benefit for adjuvant chemotherapy, although a DFS benefit has been consistently demonstrated [12-15].
The largest of these, the QUASAR trial, randomly assigned 3238 patients with an "uncertain indication for adjuvant therapy" (91 percent node-negative) to fluorouracil (FU) plus leucovorin (LV) with or without levamisole, or observation following resection of colon (71 percent) or rectal cancer [13,16]. Adjuvant chemotherapy was associated with a significant 22 percent lower risk of disease recurrence and an 18 percent reduction in death, which translated into a 3 to 4 percent absolute benefit in five-year OS . Among patients with stage II colon cancer (n = 2146), there was only a trend towards better OS (HR 0.86, 95% CI 0.54-1.19, five-year survival 83.9 versus 81.5 percent).
This trial included patients with rectal cancer (29 percent of the total enrolled), the median number of lymph nodes examined was only six, some patients received radiation therapy (14 percent in each group), and a small number of patients in both groups received portal venous, rather than systemic, infusions of chemotherapy (six percent in the chemotherapy group and five percent in the observation group).
●Meta-analyses in all risk groups – Several meta-analyses have evaluated the benefits of adjuvant chemotherapy in patients with stage II disease, all of which have concluded that if there is a survival benefit from adjuvant chemotherapy in this group, that it is no greater than 5 percent at five years [3,17-21]. As examples:
•An Intergroup analysis included pooled individual data from 3302 patients with stage II or III colon cancer who were enrolled in seven randomized trials comparing LV- or levamisole-modulated FU versus surgery alone . In multivariate analysis adjusted for T-stage, histologic grade, and nodal status, adjuvant chemotherapy was associated with a significant 30 percent relative reduction in the risk of recurrence and a 26 percent relative reduction in the risk of death. Treatment effects were consistent across all subsets, with similar relative risk reductions for stage II and III disease. For patients with stage II disease, this benefit translated into a significant improvement in five-year DFS favoring chemotherapy (76 versus 72 percent), but the difference in OS was not statistically significant (81 versus 76 percent, an absolute benefit of 5 percent).
A follow-up analysis of long-term outcomes of adjuvant therapy was conducted by the Adjuvant Colon Cancer End Points (ACCENT) collaboration. This analysis included the individual patient data from nine trials, seven of which were in the Intergroup analysis. Among the nearly 6900 patients with stage II cancer, adjuvant FU-based chemotherapy was associated with a 5 percent absolute improvement in survival at eight years (72 versus 66.8 percent, p = 0.026) .
•By contrast, a year 2015 meta-analysis of adjuvant chemotherapy for stage II colon cancer (37 trials, 5 randomized the remained observational, 15,559 patients) concluded that the overall estimated five-year DFS for patients treated with adjuvant chemotherapy was lower, 79.3 (95% CI 75.6-83.1) percent compared with 81.4 (95% CI 75.4-87.4) percent for patients who did not receive adjuvant therapy . OS results were not reported.
Largely based upon these data, updated year 2021 American Society of Clinical Oncology (ASCO) guideline on adjuvant therapy for stage II colon cancer reiterated its position from 2004  that a significant subgroup of patients with stage II colon cancer are not expected to benefit from adjuvant chemotherapy, and recommended against routine use for all patients with stage II disease .
●Lower-risk disease – The only data addressing the benefit of adjuvant chemotherapy for low-risk patients with stage II disease without definable high-risk factors come from retrospective reports, and the results are mixed, with some suggesting benefit (mostly in terms of DFS) [23-25], and others, no benefit, and even a detrimental impact on outcomes [26,27]. Based upon these limited data, the ASCO expert consensus group recommended against adjuvant chemotherapy for individuals with low-risk stage II disease .
Higher-risk stage II disease — As discussed above, the magnitude of the absolute OS benefit from adjuvant therapy seems to be very small for most patients with stage II disease. However, most trials have shown at least a DFS benefit for adjuvant chemotherapy in patients with stage II disease. Furthermore, as indicated by the Intergroup meta-analysis, the relative improvement in outcomes with adjuvant fluoropyrimidine-based chemotherapy (as measured by the HR or odds ratio [OR]) appears to be independent of stage and similar in stage II and III disease. (See 'Low-risk patients with stage II disease' above.)
This has prompted efforts to utilize clinicopathologic and molecular features to select groups of patients with higher-risk stage II disease who have a greater risk of recurrence and might derive a greater absolute degree of benefit from adjuvant chemotherapy.
Definitions of expert groups — Not surprisingly, definitions of "high-risk" early colon cancer from expert groups are variable; we use the ASCO definitions :
●In its 2000 consensus guidelines, the College of American Pathologists recognized only T4 disease, nodal status, lymphovascular invasion (LVI), a high preoperative carcinoembryonic antigen (CEA) level, and positive surgical margins to represent category I prognostic factors (ie, factors definitively proven to be of prognostic import based on evidence from multiple statistically robust published trials) . These guidelines have not been updated since 2000. (See "Pathology and prognostic determinants of colorectal cancer".)
●The definitions of "high-risk" stage II disease from ASCO , the National Comprehensive Cancer Network (NCCN) , and the European Society for Medical Oncology (ESMO)  are available (table 2). An important point is that, while all three guidelines consider poorly differentiated histology to represent an adverse feature, this is only true for tumors that are not microsatellite unstable (MSI-H). (See 'Prevalence of MMR enzyme deficiency' below.)
Clinicopathologic features — Clinicopathologic features that have been associated with a worse prognosis in patients with stage II disease are a T4 primary (table 1) [31,32]; high-grade/poorly differentiated histology (including signet ring, mucinous, and undifferentiated tumors) [19,33]; LVI [32,34-36]; perineural invasion (PNI) [32,37-40]; clinical bowel obstruction or tumor perforation [41-44]; close, indeterminate, or positive margins; inadequately sampled lymph nodes [45,46]; a high preoperative serum CEA level [32,47-53]; high levels of tumor budding ; and occult nodal micrometastases, as detected by molecular or immunohistochemical methods. (See "Pathology and prognostic determinants of colorectal cancer", section on 'Preoperative serum CEA' and "Pathology and prognostic determinants of colorectal cancer", section on 'Nodal micrometastases' and "Pathology and prognostic determinants of colorectal cancer", section on 'Tumor border and tumor budding'.)
The magnitude of risk conferred by these factors in patients with stage II disease, particularly the independent contribution of obstruction and perforation, is difficult to reliably estimate from the literature:
●Some of the best data on clinicopathologic prognostic factors and outcomes of stage II disease come from an analysis of 1738 patients treated on Cancer and Leukemia Group B (CALGB) 9581, a phase III trial of adjuvant edrecolomab versus observation after resection of stage II adenocarcinoma . The study was negative for benefit from edrecolomab, and the two arms were combined for an analysis of prognostic features in surgically treated patients. In multivariate analysis, the following factors were significant independent predictors of disease-specific OS: race, sex, age, poorly differentiated histology, vascular/lymphatic invasion, T4 depth of tumor invasion, and PNI; the number of nodes examined (<12 versus 12 or more) and age (<70 versus 70 years or older) were both of borderline statistical significance (p = 0.06).
●The contribution of clinicopathologic factors to outcomes was addressed in a study of 448 patients from Memorial Sloan Kettering who had undergone curative resection of stage II colon cancer without postoperative chemotherapy . In multivariate analysis, only three factors independently influenced prognosis: T4 primary, preoperative CEA >5 ng/mL, and LVI or PNI. Five-year disease-specific survival rates for patients with zero, one, or two or more of these poor prognostic factors were 95, 85, and 57 percent, respectively.
An association between a greater number of poor prognostic factors (especially PNI and LVI) and poorer survival has been observed by others [40,56,57]. As an example, in the International Duration Evaluation of Adjuvant Chemotherapy (IDEA) Collaborative, the five-year DFS was 74.8 percent in patients with stage II disease and two or more high-risk factors, compared with 87.3 percent among those with only one risk factor .
Molecular factors, including circulating tumor DNA — Other factors that may be associated with high risk include BRAF mutation status, gene expression profiling, and circulating tumor DNA; none of these factors are yet used widely or considered a standard approach to management of stage II disease.
●Circulating tumor DNA – In studies of localized colon cancer, high levels of postoperative circulating tumor DNA (ctDNA) reliably indicate minimal residual disease after curative-intent surgery and predict recurrence with a good sensitivity and extremely high specificity [58-66]. In fact, ctDNA testing outperforms the current standard assessment of clinicopathologic features in terms of prognostic value. In one study, patients with stage II colon cancer who were not treated with chemotherapy and who had ctDNA detected after curative-intent surgery were 18-fold more likely to have a recurrence as compared with those without detectable ctDNA (79 versus 9.8 percent) .
Prospective trials have been designed to address the clinical utility of detecting minimal residual disease after potentially curative surgery and whether such testing can refine the selection of patients at highest risk to receive chemotherapy. This issue was directly addressed in the randomized phase II DYNAMIC trial, in which 455 patients with stage II colon cancer were randomly assigned in a 1:2 ratio after surgery to standard risk-factor-guided chemotherapy (at the discretion of the treating clinician) or to ctDNA-based management in which patients who had a positive ctDNA test at week 4 or 7 received chemotherapy and ctDNA-negative individuals did not . Notably, the ctDNA assay was personalized for each patient based on probes that were developed on the basis of specific mutations identified in the tumor tissue.
At a median follow-up of 37 months, a lower percentage of patients in the ctDNA-guided management group received chemotherapy (15 versus 28 percent), and even within the high-risk clinical group (defined as proficient mismatch repair along with any clinicopathologic risk factor, 40 percent of those enrolled) the likelihood of receiving chemotherapy was 2.14 times as high in the standard management group. Despite these differences, two-year recurrence-free survival was similar in both groups (93.5 versus 92.4 percent, absolute difference 1.1 percent, 95% CI -4.1 to 6.2 [noninferiority margin -8.5]).
However, in the ctDNA-negative patients not treated with chemotherapy, the three-year recurrence rate was threefold higher among those with clinical high-risk tumors (ie, high grade, lymphovascular invasion, <12 nodes examined, obstruction or perforation) than among those with low-risk (three-year recurrence-free survival 86.4 versus 96.7 percent, HR for recurrence 3.04, 95% CI 1.26-7.34). Similarly, among ctDNA-negative patients, three-year recurrence-free survival was higher for those with T3 as compared with T4 tumors (94.2 versus 81.3 percent, HR for recurrence 2.60, 95% CI 1.01-6.71).
Notably, more patients in the ctDNA-guided chemotherapy group completed all planned cycles of chemotherapy (84 versus 78 percent). Furthermore, there were significant discrepancies in the type of adjuvant chemotherapy administered to those in the standard management arm (10 percent oxaliplatin-based, 90 percent fluoropyrimidine monotherapy) versus the ctDNA-guided management group (62 percent oxaliplatin-based). Furthermore, among the ctDNA-positive patients who were treated with adjuvant chemotherapy, three-year recurrence-free survival was 92.6 percent among those who received an oxaliplatin-containing regimen compared with 76 percent with single-agent fluoropyrimidine. While these results are suggestive that oxaliplatin-containing regimens are superior for this ctDNA positive subset, this conclusion is not definitive as the chemotherapy regimen was not a randomized assignment and only 45 ctDNA-positive patients received chemotherapy.
The DYNAMIC results confirm that a ctDNA-guided strategy of chemotherapy selection can achieve similar outcomes in a population of patients with stage II colon cancer as the currently used clinical features, with lower rates of chemotherapy exposure. However, given the nonrandomized nature of the chemotherapy strategy for ctDNA-positive patients, and the lack of chemotherapy in the ctDNA-negative patients, and concerns for higher recurrence rates in those with high-risk features who did not receive chemotherapy, these results offer individual patients prognostic information but do not provide estimates of chemotherapy treatment effect to help guide decision making about adjuvant chemotherapy.
Until additional data are available, we agree with the 2021 ASCO expert panel that there is insufficient evidence of a predictive value for chemotherapy to include ctDNA in the list of high-risk features for stage II colon cancer to warrant consideration of adjuvant chemotherapy .
●BRAF – Emerging data support the view that BRAF V600E mutations are a negative prognostic factor among patients with pMMR stage II colon cancer [68-71]. As examples:
•In an analysis of data from the three adjuvant chemotherapy trials conducted in patients with stage II or III colon cancer, BRAF mutation was not prognostic for relapse-free survival (RFS) in any subset, but it was prognostic for OS in patients with MSI-low and microsatellite stable (MSS) tumors (HR for death 2.2, 95% CI 1.4-3.4) .
•In a large population-based study of patients with all stages of colon cancer, BRAF V600E mutation was seen in 5 percent of MSS tumors (40 of 803), and 52 percent (43 of 83) of MSI-H tumors. Among patients with all stages of disease, MSI-H tumors were associated with an excellent five-year survival, whether the V600E mutation was present or absent (76 and 75 percent, respectively), while for MSS tumors, the presence of a V600E mutation significantly worsened five-year survival (17 versus 60 percent) . Among patients with stage II MSS tumors, the risk of dying was significantly higher in those with a BRAF mutation (4 of 17 [24 percent] versus 47 of 889 [5.3 percent], HR for death 4.88, 95% CI 1.73-13.76).
•In a combined analysis of 2299 patients enrolled on two National Surgical Adjuvant Breast and Bowel Project trials testing the value of adjuvant chemotherapy in patients with stage II or III colon cancer, BRAF mutations were associated with poor OS, but there was a significant interaction between DNA mismatch repair (MMR) status and BRAF mutations . The five-year survival rate was highest in patients with deficient MMR (dMMR) and BRAF wildtype tumors (90 percent), and worst in those with pMMR and BRAF-mutated tumors (69 percent), while it was intermediate in those with pMMR, BRAF wildtype tumors (82 percent) or dMMR BRAF-mutant tumors (84 percent).
While these results are intriguing, in our view, using BRAF V600E mutation status to select patients with pMMR stage II colon cancer for adjuvant therapy requires further study. The prognostic impact on non-V600 BRAF mutations is just beginning to be studied. (See "Pathology and prognostic determinants of colorectal cancer", section on 'RAS and BRAF'.)
●Gene expression profiling – There have been multiple efforts to identify molecular signatures that provide an accurate and personalized assessment of the risk of relapse (ie, prognostic signatures) and the benefit of chemotherapy (predictive signatures). Several of these gene expression assays are available for clinical use in the United States for patients with stage II colon cancer (table 3). All of these tests have been shown to have prognostic value in independent patient series, though the design and sample numbers of the validation analyses have varied . None of the five commercially available assays, including the 12-gene recurrence score (RS), has specific US Food and Drug Administration approval for use in stage II colon cancer. Furthermore, none of these assays has been able to predict which patients with stage II disease could benefit from adjuvant therapy.
The best documented and validated tool is the 12-gene RS (Oncotype-DX Colon Cancer Assay). The 12 genes included in this assay were selected based on their independent association with recurrence in each of four adjuvant therapy trials . Patients with an RS <30 had a three-year recurrence rate of 8 percent, patients with an RS of 31 to 40 had a recurrence risk of 11 percent, and patients with RS >40 had a recurrence rate of 25 percent at three years. The ability of this 12-gene assay to predict recurrence rate was independently validated in an analysis of data from the prospective QUASAR trial , in a separate analysis of data from the CALGB 9581 trial , and in the SUNRISE study . In the CALGB 9581 analysis, the RS appeared to be most discerning for patients with T3N0 tumors that were pMMR, in whom the prespecified low- and high-RS groups had average five-year recurrence risks of 13 and 21 percent, respectively. Forty-four percent of patients with T3N0 pMMR tumors fell into the low-RS category and, thus, may have been spared from receiving chemotherapy. (See 'Prevalence of MMR enzyme deficiency' below.)
On the other hand, the added prognostic value provided by the 12-gene RS and three other colon cancer RS assays based upon microarray gene expression was called into question by an analysis of data from the PETACC-3 trial of adjuvant FU and LV with or without irinotecan . All four risk scores had a statistically significant association with both OS and RFS, but each could only marginally improve the prognostic value of an RFS and OS model with the known factors of T-stage, N-stage, and MMR status.
The 12-gene RS was created as a prognostic marker. While it may have the ability to better estimate a patient's risk of recurrence, whether such increased risk can be overcome by adjuvant chemotherapy is not clear. The original investigators did generate a separate "treatment score" that appeared to be predictive of FU benefit in stage II disease from the initial four-study patient sample . However, the predictive value of this treatment score could not be validated in the QUASAR sample .
Consistent with the recommendations of most expert groups, we do not utilize the 12-gene RS or any other gene expression profiling assay for clinical decision making in stage II colon cancer. A December 2012 technical brief on gene expression profiling for predicting outcomes in stage II colon cancer prepared for the United States Agency for Healthcare Research and Quality concluded that the clinical utility of all of the gene expression assays, including the 12-gene RS, was uncertain, and that information was lacking about the extent to which assay results do or do not classify patients into distinct groups with clinical relevance. ASCO guidelines do not address use of this assay in stage II disease . NCCN guidelines state that there are insufficient data to recommend the use of multigene assays, such as the Oncotype DX assay, to determine adjuvant therapy . Updated ESMO guidelines suggest that gene expression signatures are not recommended for routine practice, but could be used to complement clinicopathologic information in intermediate-risk scenarios (ie, T3N0 disease with pMMR), although their role in predicting chemotherapy benefit is uncertain .
Does chemotherapy benefit high-risk patients? — Despite the adverse influence of these features on prognosis, few high-quality studies have addressed the benefit of chemotherapy in higher-risk subsets of stage II disease, and the results are disparate, with some suggesting no benefit [19,26,78-83], and others, benefit for certain subgroups [26,39,84-86]. Two systematic reviews are available:
●A systematic review and meta-analysis of 23 cohort studies and one randomized trial (the MOSAIC trial of oxaliplatin versus fluoropyrimidines alone in patients with stage III or high-risk stage II colon cancer (see 'Should an oxaliplatin-containing regimen be used?' below)) concluded that adjuvant chemotherapy improved OS in high-risk stage II colon cancer (HR for death 0.64, 95% CI 0.51-0.80) . However, in subgroup analysis, benefit was limited to those with localized intestinal perforation, obstruction, pT4 lesions, or <12 sampled lymph nodes; patients with LVI, PNI, and poorly differentiated histology did not gain a significant OS benefit from adjuvant chemotherapy.
●ASCO conducted a meta-analysis in preparation for its updated guidelines for adjuvant therapy in stage II colon cancer . The following conclusions were drawn:
•T4 primary - Patients with a resected stage T4N0 (IIB) colon cancer actually have worse outcomes than do those with stage IIIA disease (figure 1).
In the ASCO analysis, across six retrospective studies [23,26,56,83,84,87] totaling 18,517 patients with T4 disease, there was an OS advantage (HR 0.64, 95% CI 0.56-0.75) favoring adjuvant chemotherapy versus surgery alone. This translated into 139 fewer deaths per 1000 patients (95% CI 174 fewer to 93 fewer). Across two studies [26,84] totaling 7711 patients, there was a consistent RFS advantage (HR 0.70, 95% CI 0.63-0.77). There was no assessment for risk of bias or adjustment for confounding factors.
•Fewer than 12 sampled nodes – Across four studies (6800 patients), there was a significant positive effect of adjuvant chemotherapy on OS (HR 0.67, 95% CI 0.57-0.77), which translated into 126 fewer deaths per 1000 (95% CI 170 fewer to 85 fewer). There was also a consistent benefit of adjuvant chemotherapy on DFS/RFS (HR 0.71, 95% CI 0.61-0.82). However, the quality of evidence was judged to be low.
•Obstruction and other risk factors – Across four studies of patients with intestinal obstruction (911 patients), there was a significant benefit for adjuvant chemotherapy for OS (HR 0.57, 95% CI 0.38-0.85) and DFS/RFS (HR 0.63, 95% CI 0.44-0.89). By contrast, the impact of adjuvant therapy in the setting of other high-risk factors (tumor perforation, LVI, PNI, high tumor budding or poorly/undifferentiated histology) was less certain, with wide confidence intervals overlapping with 1.0. However, there were significant limitations to the evidence base for all of these high-risk factors, mostly because of a high degree of heterogeneity.
•Multiple risk factors – At least three studies have addressed the impact of multiple risk factors on benefit from chemotherapy, with disparate results:
-In one study, a significant benefit from chemotherapy was found for T4 tumors in combination with LVI, a smaller number of sampled lymph nodes, and poorly differentiated histology .
-A second report found that with the exception of T4 disease, patients with two or more risk factors did not benefit from adjuvant chemotherapy .
-Others note the adverse influence of the combination of LVI and PNI, which has been associated with particularly poor outcomes .
Should an oxaliplatin-containing regimen be used? — There is insufficient evidence to routinely recommend the addition of oxaliplatin to fluoropyrimidine-based chemotherapy for any patients with stage II colon cancer.
We suggest a fluoropyrimidine-based regimen (LV-modulated FU [FU/LV] or capecitabine alone) rather than an oxaliplatin-based regimen for most patients with stage II disease who elect chemotherapy. However, for patients who have high-risk stage II disease and either a T4 primary or multiple high-risk factors, we discuss the potential benefits and risks of oxaliplatin, as seen in the MOSAIC trial (table 4). Recommendations for patients with dMMR stage II disease are presented below.
Benefits in stage II disease — Compared with stage III (node-positive) colon cancer, the benefits of oxaliplatin in stage II disease are uncertain. (See "Adjuvant therapy for resected stage III (node-positive) colon cancer", section on 'Oxaliplatin-based therapy'.)
The following data inform this issue:
●MOSAIC trial – The MOSAIC trial compared six months of adjuvant FU/LV versus FOLFOX (oxaliplatin plus short-term infusional FU and LV (table 5)) in patients with resected stage II (40 percent) or III (60 percent) colon cancer . (See "Adjuvant therapy for resected stage III (node-positive) colon cancer".)
At the initial analysis (minimum six years of follow-up in all patients), the addition of oxaliplatin resulted in a small increase in six-year OS (79 versus 76 percent; HR for death 0.84, p = 0.046), but the survival benefit was limited to patients with stage III disease (73 versus 69 percent, p = 0.023). In the 569 patients with high-risk stage II cancer, five-year DFS (defined as time from randomization to recurrence or death from any cause) was slightly but not significantly higher with FOLFOX as was six-year OS, and time to tumor recurrence (defined as the time from randomization to recurrence of the same cancer, but second cancers and death were not counted) was significantly longer with FOLFOX . The interaction test between treatment and risk in high-risk stage II patients for TTR was not significant (p = 0.235). These benefits are outlined in the table (table 4).
In 2015, updated data show an increasing absolute survival benefit for oxaliplatin in stage III disease with time (67 versus 59 percent, p = 0.043), whereas FOLFOX continues not to show a survival benefit over FU/LV for patients with stage II cancer (78 versus 79 percent, p = 0.98) . In an exploratory analysis, there was a suggestion of benefit in patients with high-risk stage II disease; however, because of the small number of patients, the 7 percent absolute improvement in DFS (82 versus 77 percent) and the 2 percent absolute improvement in survival (85 versus 83.3 percent) with FOLFOX as compared with FU/LV in this group were not statistically significant.
Largely based on these data, the updated ASCO expert consensus group for adjuvant chemotherapy in stage II colon cancer concluded that oxaliplatin-containing chemotherapy could be considered an option for some patients, and that shared decision-making should follow a discussion of efficacy results of the MOSAIC trial and potential for harms (table 4) .
●NSABP C-07 – The NSABP trial C-07 randomly assigned 2246 patients with stage II (29 percent) or III (71 percent) colon cancer to bolus FU/LV or weekly oxaliplatin plus bolus FU/LV (FLOX) . The addition of oxaliplatin resulted in a significant improvement in four-year DFS (73 versus 67 percent, HR for recurrence 0.80, 95% CI 0.69-0.93). In the stage-based subgroup analysis, patients with stage II disease had neither DFS benefit (HR 0.94, p = 0.67) nor OS benefit (HR 1.04, p = 0.84) from FLOX over FU/LV .
●ACCENT database analysis – These conclusions were also supported by a later analysis derived from the 12,233 patients enrolled in the five randomized trials that comprised the ACCENT database . Among patients with stage II disease, the addition of oxaliplatin significantly reduced the risk of recurrence within the first 14 months post-treatment but not thereafter, and there was no OS benefit. The majority of patients in these analyses would have had pMMR tumors, although the trials were carried out prior to routine testing for MMR status. The contribution of oxaliplatin to reversing FU resistance in patients with dMMR tumors is addressed below. (See 'High-risk tumors and benefit of oxaliplatin' below.)
Duration of chemotherapy — If a fluoropyrimidine alone is administered for adjuvant therapy of resected colon cancer, six months of therapy is considered standard adjuvant chemotherapy, based on the trials reported above, which utilized six months of therapy. (See 'Is there benefit for fluoropyrimidine-based chemotherapy?' above.)
For high-risk patients with stage II disease who opt to receive oxaliplatin-containing chemotherapy, the optimal duration (three versus six months) remains uncertain. In higher-risk stage III colon cancer, three months of oxaliplatin-containing chemotherapy is considered adequate for those with earlier substages of disease (T1-3N1) based on a pooled analysis of data from the IDEA collaboration, which shows a very small and likely clinically nonmeaningful difference in DFS and OS with longer therapy, but a nearly threefold higher risk of treatment-related peripheral neuropathy. (See "Adjuvant therapy for resected stage III (node-positive) colon cancer", section on 'Duration of therapy'.)
For stage II disease, based on a preliminary report of data from the IDEA collaboration, the absolute benefit of three additional months of an oxaliplatin-based regimen is greater than seen in stage III disease (3.3 percent in DFS), but it is also small and must be counterbalanced by the higher risk of toxicity with six months of treatment.
In our view, both three and six months of chemotherapy are reasonable options, and the decision must be individualized. Although the IDEA trials did not randomly assign the chemotherapy regimen, if three months of therapy is chosen, we would suggest capecitabine plus oxaliplatin (CAPOX) rather than FOLFOX, whereas if six months of therapy is chosen, we suggest FOLFOX. If significant neuropathy develops after three months of therapy, we discontinue oxaliplatin and complete the remainder of the six-month course with a fluoropyrimidine alone.
The duration of oxaliplatin-containing chemotherapy for high-risk stage II disease was addressed in a set of six parallel noninferiority trials comparing three versus six months of oxaliplatin therapy for stage II or III colon cancer, four of which included high-risk stage II disease (table 6). The regimen (FOLFOX or CAPOX) was preselected and not randomized. A pooled analysis of all six trials suggests that three months of therapy was adequate for earlier substages of stage III colon cancer (ie, T1-3N1) . (See "Adjuvant therapy for resected stage III (node-positive) colon cancer", section on 'Duration of therapy'.)
The following data are available for high-risk stage II disease (defined as T4, poorly differentiated histology, LVI or PNI, inadequate nodal harvest, obstructed, or perforated) from the IDEA collaboration:
●Published results in stage II disease are available only from the TOSCA trial [95,96]. Of the 3759 enrolled patients in this trial, approximately one-third had high-risk stage II disease. At a median follow-up of 62 months, noninferiority could not be shown for three as compared with six months of oxaliplatin-based chemotherapy . In the per-protocol analysis, the HR for RFS was 1.41 (95% CI 1.05-1.89), with an absolute difference in five-year RFS of 5.9 percent favoring six months of therapy. In a later preplanned subgroup analysis, among patients treated with CAPOX (38 percent of the total), five-year RFS was similar between three and six months of treatment (five-year RFS 84.6 versus 85.4 percent, difference 0.76 percent in favor of the six-month arm, 95% CI -6.28-7.8 percent), while for patients receiving FOLFOX, the difference was more pronounced (five-year RFS 81.1 versus 89.6 percent, difference 8.56 percent in favor of the longer duration of treatment, 95% CI 3.45-13.67 percent) . Despite these findings, the test for an interaction between duration and regimen was not statistically significant. Grade 3 or worse neurotoxicity rates were higher with longer therapy (8.4 versus 1.3 percent).
●A pooled analysis of all four randomized trials investigating the duration of adjuvant oxaliplatin in stage II disease from the IDEA collaboration included 3273 randomized patients; 1254 (38 percent) had FOLFOX and 2019 (62 percent) had CAPOX . In the entire population, five-year DFS was 80.7 versus 83.9 percent for three versus six months of therapy, respectively (HR 1.17, 95% CI 1.05-1.31, absolute difference -3.2 percent), and noninferiority for three months of adjuvant treatment was not shown. The risk of grade 2 to 4 neurotoxicity was 13 versus 36 percent (HR 0.36, 95% CI 0.31-0.42). In an unplanned subset analysis, three months of CAPOX was noninferior to six months, while three months of FOLFOX was inferior to six months (table 7).
●Data are also available from the phase III ACHIEVE-II trial, in which 525 Asian patients with high-risk stage II colon cancer who were randomly assigned to three versus six months of either CAPOX (84 percent) or modified FOLFOX6 . In the entire cohort, the three-year DFS was not significantly worse with three months of therapy (88.2 versus 87.9 percent, HR 1.12, 95% CI 0.65-1.96), although the confidence intervals were very wide, but the risk of grade 2 or worse sensory neuropathy was significantly lower (16 versus 43 percent).
The ASCO expert panel on adjuvant therapy for stage II colon cancer concluded that adjuvant oxaliplatin-based chemotherapy can be offered for a duration of three or six months, after a discussion with the patient of the potential benefits and harms associated with shorter versus longer treatment duration (table 7) .
Web-based tools to aid decision-making — Several web-based tools had been developed to assist the clinician in calculating the relative risk of disease recurrence and mortality, based on clinicopathologic features, and the relative benefit to be gained with adjuvant chemotherapy for stage II or III colon cancer. However, none of them (Adjuvant! Online, Numeracy, the ACCENT database calculator) is available online any longer.
For individual patients, a postoperative nomogram has been developed that permits a prediction of the risk of a colon cancer recurrence based upon clinicopathologic factors and whether adjuvant chemotherapy was administered or not . The nomogram, which was revised in 2019 , and has been independently validated [100-102], is available online. It is one of two prognostic tools endorsed for use in patients with colon cancer by the American Joint Committee on Cancer, meeting all quality criteria . The other approved tool is the ACCENT database calculator, which is limited to stage III disease, but it is no longer available online.
While programs such as these can assist in estimating the potential benefit of adjuvant therapy, the decision to pursue treatment requires careful consideration of all aspects of the individual patient by the clinician, including those that are not contained in any of the models (eg, perforation). A major problem is that molecular prognostic factors such as dMMR are not included in this model. A proposed revision of the nomogram for determining the likelihood of postoperative recurrence in resected stage I to III colon cancer that includes molecular (MMR status) and unique clinicopathologic features, such as the presence of tumor-infiltrating lymphocytes, has been published . Validation in independent datasets is needed.
In our opinion, tools such as these are best used as a rough guide to explain the degree of absolute benefit that may be achieved by receipt of adjuvant chemotherapy.
Chemotherapy toxicity — Given the relatively good prognosis of stage II disease treated in the modern era , a recommendation to administer adjuvant chemotherapy must consider the potential risks associated with treatment:
●In a phase III study, severe (grade 3 or worse) side effects developed in 14 percent of patients with stage II disease who were treated with six months of short-term infusional FU plus LV; this led to premature treatment discontinuation in 6 percent . Higher rates of diarrhea, nausea/vomiting, and myelosuppression are usually seen with bolus FU/LV regimens. (See "Adjuvant therapy for resected stage III (node-positive) colon cancer", section on 'Infusional versus bolus fluorouracil'.)
With the addition of six months of oxaliplatin, 19 percent experienced severe side effects, 12 percent of which resulted in treatment discontinuation . Grade 3 sensory neuropathy developed in 13 percent during therapy but was still present at 48 months in only 0.7 percent.
●Rates of peripheral sensory neuropathy are significantly lower in patients who elect three as compared with six months of oxaliplatin-based chemotherapy, as demonstrated in the IDEA collaborative trials. (See 'Duration of chemotherapy' above.)
●In most modern adjuvant therapy trials, rates of treatment-related death range from 0.5 to 1 percent; many of these deaths are in patients who are deficient in dihydropyridine dehydrogenase (DPD), a major metabolizing enzyme for FU. (See "Chemotherapy-associated diarrhea, constipation and intestinal perforation: pathogenesis, risk factors, and clinical presentation", section on 'Fluorouracil'.)
●Long-term side effects are less well defined. In one study, quality of life decreased after adjuvant FU/LV therapy, requiring up to one year to return to baseline . Neuropathy was still present in 10 percent of patients two years after discontinuation of oxaliplatin in the C-07 trial, despite the use of a lower cumulative oxaliplatin dose than in the MOSAIC trial of FOLFOX . (See "Overview of neurologic complications of platinum-based chemotherapy", section on 'Cumulative sensory neuropathy'.)
●In the X-ACT study, six months of treatment with capecitabine was associated with hand-foot syndrome in 60 percent (17 percent severe [grade 3 or 4]) . In addition, 46 percent developed diarrhea of any grade (11 percent severe), 36 percent had nausea or vomiting (3 percent severe), 22 percent had stomatitis (2 percent severe), and 23 percent had fatigue or asthenia (1 percent severe). Six percent of patients developed partial alopecia. (See "Adjuvant therapy for resected stage III (node-positive) colon cancer", section on 'Less fit patients or a contraindication to oxaliplatin'.)
In general, across regimens and most adverse events, females experience clinically and statistically significantly greater treatment-related toxicity from adjuvant chemotherapy, especially for severe neutropenia . Females have a lower clearance of FU, which means that dosing according to body surface area (BSA; which is the standard approach) results in higher plasma fluoropyrimidine levels than occur in males [109-111]. The mechanism underlying this lower clearance is not established but could be due to sex differences in the metabolizing enzyme dihydropyrimidine dehydrogenase (DPYD) [112-114]. (See "Dosing of anticancer agents in adults", section on 'Body surface area (BSA)-based dosing' and "Chemotherapy-associated diarrhea, constipation and intestinal perforation: pathogenesis, risk factors, and clinical presentation", section on 'Predictive markers'.)
Chemotherapy dosing in obese patients — For cancer patients with a large BSA, chemotherapy drug doses are often reduced because of concern for excess toxicity. However, there is no evidence that fully dosed patients with obesity experience greater toxicity from chemotherapy for colorectal cancer; furthermore, patients with obesity who are given reduced doses may have inferior outcomes . Although limited, the available data do not support the policy of routine dose reduction (or capping the maximal BSA to 2 m2) for patients with obesity with colon cancer. Guidelines from ASCO recommend that full weight-based cytotoxic chemotherapy doses be used to treat patients with obesity with cancer, particularly when the goal of treatment is cure . (See "Dosing of anticancer agents in adults", section on 'Overweight/obese patients'.)
Patients with deficient MMR
Indications for adjuvant therapy — For most patients with stage II colon cancer and dMMR/MSI-H tumors without high-risk features, we suggest observation alone rather than adjuvant chemotherapy. The best approach for patients with stage II MSI-H/dMMR tumors that have high-risk features (eg, T4 primary, obstructed/perforated, fewer than 12 nodes in the surgical specimen, LVI or PNI) is uncertain. Few data are available to guide the use of adjuvant therapy in this subset, and the decision to pursue chemotherapy must be individualized, based upon the number of high-risk features, the patient's overall medical condition, and age. For patients who have dMMR tumors and high-risk features who choose chemotherapy, we suggest an oxaliplatin-based regimen rather than fluoropyrimidines alone.
Prevalence of MMR enzyme deficiency — Approximately 15 to 20 percent of colorectal cancers have sporadic or inherited (Lynch syndrome) deficiency of a DNA mismatch repair (MMR) protein, most commonly MLH1 or MSH2. MSI-H is the biologic footprint of DNA dMMR. (See "Molecular genetics of colorectal cancer", section on 'Mismatch repair genes' and "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis".)
dMMR is more prevalent in stage II as compared with stage III colon cancer (21 versus 14 percent in one study) . dMMR tumors are characteristically located proximally and have a mucinous histology with tumor-infiltrating lymphocytes; they also have a better prognosis than do MSS tumors. However, this better prognosis appears to be attenuated in the presence of a BRAF V600E mutation, a finding that indicates a sporadic tumor with a hypermethylation phenotype rather than a tumor that arose in the setting of hereditary nonpolyposis colorectal cancer (HNPCC) [118,119]. (See "Pathology and prognostic determinants of colorectal cancer", section on 'Mismatch repair deficiency' and "Molecular genetics of colorectal cancer", section on 'Hypermethylation phenotype (CIMP+) pathway'.)
Testing for Lynch syndrome — We test for dMMR in all patients with stage II colon cancer. In addition to obtaining prognostic and predictive information to aid in adjuvant therapy, testing is important to screen patients for Lynch syndrome (HNPCC). Patients with Lynch syndrome have germline loss of expression of one of several MMR enzymes, resulting in a predisposition to colorectal cancer and other cancers, especially endometrial cancer. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis", section on 'Clinical features'.)
IHC for the most commonly lost mismatch repair proteins (MLH1, MSH2, MSH6, and PMS2) can be used in place of the more cumbersome MSI testing (a polymerase chain reaction assay, which takes longer and is more expensive). Getting a result within a shorter time frame is particularly important for adjuvant therapy decisions, as treatment is usually started within four to six weeks of resection.
Absence of one of these proteins on screening IHC confirms the good-prognosis dMMR phenotype but does not necessarily indicate Lynch syndrome. Although over 90 percent of Lynch syndrome-related colorectal cancers will demonstrate MSI, 15 percent of sporadic colorectal cancers also have MSI. Thus, the finding of MSI in a colorectal cancer is not specific for Lynch syndrome:
●Absence of protein expression of MLH1 and PMS2, MLH1 alone, or PMS2 alone (which is rare) may be associated with either sporadic or inherited disease. The next step is analysis for BRAF V600E mutation or analysis of methylation of the MLH1 promoter. The presence of a BRAF mutation suggests sporadic rather than inherited disease, and there is no need to refer for germline genetic testing. On the other hand, if there is methylation of the MLH1 promoter and no BRAF mutation, most likely the patient will have a sporadic tumor; however, 5 to 10 percent of these patients may harbor a germline mutation. Referral for germline testing is appropriate. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis", section on 'Indications for germline testing'.)
●On the other hand, loss of expression of MSH2 alone, MSH2 in combination with MSH6, or MSH6 alone is highly specific for a disease-causing germline defect, and referral for germline genetic testing (for the genes corresponding to the absent proteins) is appropriate. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis" and "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Cancer screening and management".)
For individuals with evidence of loss of expression of an MMR protein by IHC, further evaluation is outlined in the suggested algorithm (algorithm 2).
Is there benefit from fluoropyrimidines alone? — Most (but not all [6,10,120-122]) studies support the view that adjuvant, single-agent, fluoropyrimidine-based chemotherapy is less beneficial, or even potentially harmful, for patients with MSI-H or dMMR stage II or III colon cancer compared with pMMR tumors [5,107,123-128]. The following represents the range of findings:
●In an early pooled analysis of data from two fluoropyrimidine adjuvant therapy trials. there was a significantly decreased OS in patients with stage II disease and dMMR tumors who were treated with adjuvant chemotherapy relative to surgery alone (HR for death 2.95, 95% CI 1.02-8.54) .
●In a later pooled analysis of three studies of patients with dMMR tumors [5,10,129], there was no evidence for a benefit in terms of either OS (320 patients in three studies, HR 1.03, 95% CI 0.53-2.20) or DFS (HR 0.89, 95% CI 0.45-1.74) . The quality of the evidence was very low, and there was a high degree of heterogeneity. One of these observational studies included patients with high-risk stage II disease who were treated with oxaliplatin-based adjuvant chemotherapy.
●Additional data are available from the QUASAR study of 1913 patients with stage II colon cancer randomized to FU/LV versus supportive care . Patients with dMMR tumors had a recurrence risk that was one-half that of patients with pMMR cancers (risk ratio 0.53, 95% CI 0.40-0.70), and there was no significant benefit in for adjuvant chemotherapy (OR for recurrence 0.81, 95% CI 0.29-2.22), which translated to an absolute decrease in recurrence at two years of 1.6 percent.
The relatively small number of patients with dMMR cancers and the very small number of events (total 15) makes it difficult to assess whether there is a true signal of chemotherapy effect in patients with dMMR tumors, and these results are not powerful enough to nullify the previous studies indicating a predictive capacity of dMMR in stage II colon cancer.
Resistance to FU may be due to incorporation of FU metabolites into DNA or overexpression of its effective targets, thymidylate synthetase and/or DPD in dMMR tumors [130-132].
High-risk tumors and benefit of oxaliplatin — For patients who have dMMR stage II tumors and high-risk features who choose adjuvant chemotherapy, we suggest the use of an oxaliplatin-based regimen rather than fluoropyrimidines alone.
Whether the lack of benefit from adjuvant FU in patients with dMMR can be overcome by the addition of oxaliplatin is unclear; there are no prospective trials comparing oxaliplatin versus non-oxaliplatin-containing adjuvant chemotherapy in patients with dMMR tumors. However, data from retrospective analyses are suggestive:
●While one report noted a significant benefit from adding oxaliplatin to FU and LV in patients with dMMR tumors , another suggested a lower rate of disease control with FOLFOX in patients with dMMR tumors compared with those with pMMR tumors . Neither of these studies was prospectively designed.
●In an analysis of data from NSABP C07, which tested the worth of adding oxaliplatin to a fluoropyrimidine-based regimen in patients with stage II or III disease, oxaliplatin benefit was present regardless of MMR status .
●A subgroup analysis of the MOSAIC trial was conducted with data from the 1008 patients for whom MMR status was available after nearly 10 years median follow-up. Although this subgroup analysis is underpowered and should not be considered definitive evidence, oxaliplatin was associated with a considerable improvement in mortality in patients with dMMR tumor (HR 0.42, 95% CI 0.16-1.07, p = 0.069) .
●A retrospective analysis of 433 patients with resected dMMR tumors (57 percent stage II) from several French centers also supports benefit for an oxaliplatin-containing regimen in patients with dMMR tumors . Adjuvant chemotherapy was administered to 17 percent of patients with stage II disease (n = 41, mainly those with high-risk tumors) and 69 percent of those with stage III disease (n = 129). Overall, three-year RFS was 75 percent for surgery alone, 66 percent for FU alone, and 84 percent with FOLFOX. In subgroup analysis, the benefit of FOLFOX, compared with surgery alone, was significant only for stage III disease (HR for relapse 0.41, 95% CI 0.19-0.87), while there was a trend toward better outcomes in patients with stage II disease (HR for relapse 0.13, 95% CI 0.02-1.05, p = 0.06). FU alone did not provide any advantage over surgery for either stage II or III disease.
The ASCO expert panel on adjuvant therapy for resected stage II colon cancer stated that when shared decision-making results in the choice to proceed with adjuvant chemotherapy in patients who have dMMR tumors and a pT4 primary or other high-risk features, an oxaliplatin-containing regimen is recommended .
Recommendations from expert groups — There is no consensus from expert groups as to which, if any, patients with stage II disease should receive adjuvant chemotherapy. The definition of "high-risk" stage II disease is not standardized, as reflected in discordant guidelines from expert groups (table 2), and there is a lack of agreement as to whether clinicopathologic factors should be used at all in the therapeutic decision-making process. We generally use the ASCO definitions . (See 'Definitions of expert groups' above.)
●ASCO – Updated guidelines for adjuvant therapy in stage II colon cancer from ASCO continue to highlight the importance of shared decision-making regarding the use of adjuvant therapy in stage II colon cancer . They recommend against routine use of adjuvant chemotherapy for stage II colon cancer, especially individuals at low risk of experiencing a disease recurrence, but recommend that clinicians offer adjuvant therapy to those with a pT4 primary tumor or other high-risk clinicopathologic features (fewer than 12 sampled nodes, LVI or PNI, poorly differentiated/undifferentiated histology [unless dMMR], obstruction, or perforation). The number of risk factors should be incorporated into shared decision-making. The panel endorsed an oxaliplatin-containing regimen for patients with dMMR tumors with a pT4 primary or other high-risk factors, but noted the limited evidence to support the addition of oxaliplatin to fluoropyrimidine-based chemotherapy in other settings. For higher-risk patients with pMMR, oxaliplatin could be offered for three or six months, after a discussion with the patient on the benefits and risks of harm with the proposed treatment duration. These and other specific recommendations are outlined in the table (table 8).
●ESMO – Updated year 2020 guidelines from ESMO suggest that the major prognostic parameters for stage II risk assessment are pT4 stage (including perforation), MMR/MSI status, and the number of lymph nodes sampled . However, other additional factors such as histologic subtype and grade; lymphatic, venous, or perineural invasion; lymphoid inflammatory response; involvement of resection margins; and serum CEA should all be taken into consideration for refining risk assessment.
●NCCN – Updated consensus-based guidelines from the NCCN also suggest a discussion about the potential risks versus benefits of adjuvant therapy for all patients with stage II disease, with treatment decisions being influenced by the number of lymph nodes analyzed, presence of high-risk clinicopathologic features (defined as fewer than 12 nodes in the specimen; poorly differentiated histology; colonic obstruction or localized perforation; perineural, vascular, or lymphatic invasion; close/indeterminate or positive margins), MMR status, and assessment of other comorbidities and anticipated life expectancy . We agree with this viewpoint.
ADJUVANT RADIATION THERAPY — The evidence is inconclusive as to the benefit of adding radiation therapy (RT) to chemotherapy for patients at high risk for a local recurrence. We suggest adjuvant RT for patients with tumors involving the ascending or descending colon with penetration to a fixed structure (T4b disease, (table 1) or a positive margin).
Postoperative RT is not usually considered a routine component of care for completely resected colon cancer. This is in contrast to patients with rectal cancer, in whom effective adjuvant therapy for both transmural and node-positive disease includes RT. Local recurrence is more frequent with rectal cancer due to the local anatomy and the difficulty in obtaining adequate resection margins. (See "Adjuvant therapy for resected rectal adenocarcinoma in patients not receiving neoadjuvant therapy", section on 'Introduction'.)
Selected patients with colon cancer who are at high risk for local recurrence (ie, positive resection margins, T4b disease (table 1), perforation, abscess formation) might potentially benefit from adjuvant RT. These groups have an estimated risk of local recurrence that is 30 percent or higher [135-139]. However, there is a paucity of high-quality evidence addressing the role of adjuvant RT (with or without concurrent chemotherapy) in patients with resected locally advanced colon cancer:
●In the largest retrospective series, 171 patients underwent RT for completely resected high-risk colon cancer (T4 tumors; T3N+ disease at sites other than the midsigmoid colon, midtransverse colon, and cecum; and selected patients with high-risk T3N0 disease and a <1 cm radial margin) . Concurrent chemoradiotherapy (mostly fluorouracil on days 1 to 3 during the first and last week of RT) was administered to 68 patients. The outcomes of this irradiated group were compared with those of a historical cohort of 395 patients of similar T and N stage who underwent surgery alone at the same institution.
Local failure rates in patients treated with RT for T4N0 and T4N+ disease were 7 and 28 percent, respectively; the corresponding values for patients undergoing surgery alone were 31 and 53 percent. Disease-free survival rates for T4N0 and T4N+ tumors in the irradiated group were 79 and 53 percent, compared with 63 and 38 percent, respectively, with surgery alone.
●A single randomized United States Intergroup trial, conducted in patients with high-risk T1-3/N1-2 disease, failed to show benefit for adjuvant RT but had been closed prior to achieving its accrual goal because of slow accrual. (See "Adjuvant therapy for resected stage III (node-positive) colon cancer", section on 'Adjuvant radiation therapy'.)
Despite the lack of evidence proving benefit from randomized trials, consensus-based guidelines from the National Comprehensive Cancer Network  suggest that adjuvant RT be "considered" for patients with T4 disease with penetration to a fixed structure. Others suggest that RT be considered on a case-by-case basis for positive resection margins and disease complicated by perforation or abscess formation [135,140].
We only offer RT for tumors involving fixed parts of the colon, which are the ascending or descending colon.
SPECIAL CONSIDERATIONS DURING THE COVID-19 PANDEMIC — The coronavirus disease 2019 (COVID-19) pandemic has increased the complexity of cancer care. Important issues include balancing the risk from delaying cancer treatment versus harm from COVID-19, minimizing the number of clinic and hospital visits to reduce exposure whenever possible, mitigating the negative impacts of social distancing on delivery of care, and appropriately and fairly allocating limited health care resources. Specific considerations for patients with stage II colon cancer include a discussion of the relative benefits versus risks of pursuing chemotherapy treatment; utilizing oral rather than intravenous chemotherapy, where possible, if chemotherapy is chosen; reducing the duration of adjuvant chemotherapy (ie, three versus six months of an oxaliplatin-based regimen), if appropriate; and transitioning outpatient care (eg, pump disconnection, administration of growth factors) to home whenever possible. (See 'Duration of chemotherapy' above.)
These and other recommendations for cancer care during active phases of the COVID-19 pandemic are discussed separately. (See "COVID-19: Considerations in patients with cancer".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Colorectal cancer".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Colon and rectal cancer (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Rationale and benefits of adjuvant chemotherapy
•Following potentially curative resection of colon cancer, the goal of adjuvant chemotherapy is to eradicate micrometastases and improve the likelihood of cure.
•The absolute magnitude of benefit for adjuvant chemotherapy in resected stage II disease is not as great as in stage III disease; the best evidence suggests no more than a 5 percent absolute improvement in five-year survival for most patients. Benefit may be greater in high-risk subsets. (See 'Is there benefit for fluoropyrimidine-based chemotherapy?' above.)
●Impact of deficient mismatch repair
•All patients with stage II colon cancer should have their tumors tested for deficiency in DNA mismatch repair (dMMR; as indicated by high levels of microsatellite instability [MSI-H] by molecular testing or immunohistochemistry for loss of MMR protein expression). (See 'Testing for Lynch syndrome' above.)
•Patients with dMMR have a better prognosis after surgery alone compared with tumors that are proficient in MMR (pMMR). MMR status is also an important predictor of lack of benefit from fluoropyrimidine-based adjuvant chemotherapy. (See 'Impact of DNA mismatch repair' above.)
●Indications for adjuvant chemotherapy – Our general approach to adjuvant chemotherapy in patients with resected stage II disease is outlined in the algorithm (algorithm 1), and is informed by the MMR status as well as the presence or absence of high-risk clinicopathologic features.
•Definition of high-risk clinicopathologic features – For patients with resected stage II colon cancer, high-risk clinicopathologic features include a pT4 primary, poorly differentiated/undifferentiated histology (unless dMMR), fewer than 12 nodes in the surgical specimen, lymphovascular or perineural invasion, high levels of tumor budding, and clinical obstruction or perforation).
•Patients with proficient mismatch repair
-For patients with T3N0 pMMR tumors and no high-risk clinicopathologic features, we suggest observation alone rather than adjuvant chemotherapy (Grade 2C). (See 'Low-risk patients with stage II disease' above.)
-For most patients with pMMR tumors and a T4 tumor or multiple high-risk features, we suggest adjuvant chemotherapy (Grade 2C). For patients with only one high-risk feature other than T4 disease, we discuss the risks and potential benefits of adjuvant chemotherapy and individualize decision-making. (See 'Does chemotherapy benefit high-risk patients?' above.)
-Choice of regimen
For most patients with pMMR higher-risk stage II colon cancer who do not have a T4 primary or multiple high-risk factors and who choose adjuvant chemotherapy, we suggest a fluoropyrimidine-based regimen (leucovorin-modulated fluorouracil [LV/FU] or capecitabine alone) rather than an oxaliplatin-based regimen (Grade 2C). Six months of treatment is a standard approach. (See "Treatment protocols for small and large bowel cancer" and 'Duration of chemotherapy' above.)
For patients who have a T4 primary or multiple high-risk factors, we discuss the potential benefits and risks of an oxaliplatin containing regimen, as was used in the MOSAIC trial, and individualize decision-making. (See 'Should an oxaliplatin-containing regimen be used?' above.)
If an oxaliplatin-based regimen is chosen, the optimal duration is unclear. In our view, both three and six months of chemotherapy are reasonable options, although neurotoxicity rates are clearly higher with longer treatment, and the decision must be individualized.
If three months of therapy is chosen, we suggest capecitabine plus oxaliplatin (CAPOX) rather than oxaliplatin plus short-term infusional FU and LV (FOLFOX) (Grade 2C). If six months of chemotherapy is chosen, we suggest initial FOLFOX rather than CAPOX (Grade 2C). If neuropathy develops after three months, we discontinue oxaliplatin and complete the six-month adjuvant regimen with a fluoropyrimidine alone. (See 'Duration of chemotherapy' above.)
•Patients with deficient mismatch repair
-For patients without high-risk features who have tumors with dMMR, we suggest observation alone (Grade 2C). These patients have a favorable prognosis and are not likely to derive significant benefit from adjuvant therapy. (See 'Patients with deficient MMR' above.)
-The best approach for patients with stage II dMMR tumors that have high-risk features is uncertain given the better prognosis compared with pMMR tumors. The decision to pursue chemotherapy must be individualized with shared decision-making, taking into account the number of high-risk features, the patient's overall medical condition, and age.
●Adjuvant radiation therapy – We suggest adjuvant radiation therapy (RT) for patients with tumors of the ascending or descending colon with penetration to a fixed structure (T4b disease (table 1)) or a positive margin (Grade 2C). The evidence is inconclusive as to the benefit of adding RT to chemotherapy for patients at high-risk for a local recurrence. (See 'Adjuvant radiation therapy' above.)