Stage II colon cancer: chemo and circulating tumor DNA

A 58-year-old professor underwent surgery for an adenocarcinoma of the sigmoid colon. The pathology report shows a pT3 tumor, 18 lymph nodes examined and none involved, no lymphovascular or perineural invasion, with preserved mismatch repair proteins. It is a stage II without high-risk factors. At the first oncology visit, he arrives with a ready question: a colleague had a blood test that looks for tumor DNA and, since it came back negative, did not need chemotherapy. He wants to know whether he should have the same test and whether he needs chemotherapy. The scene is fictional and was written for teaching.

This text uses the case as a scenario for clinical reasoning for students and health professionals. The question comes from Dr. Alexei's list for the oncology series: who, in stage II colon cancer, benefits from adjuvant chemotherapy, and what does circulating tumor DNA add to that decision? The reference studies are QUASAR, the pooled analyses of fluoropyrimidine and the DYNAMIC study, which tested decision-making guided by circulating tumor DNA. The text does not replace your country's guideline or your service's protocol. If you play the case of the day on NEXO, you will recognize the challenge: deciding when the average benefit is small and the individual risk is different for each patient.

Key points

• In stage II, surgery alone cures about 80 out of every 100 patients, which limits the possible gain from chemotherapy.
• The absolute benefit of fluoropyrimidine in 5-year survival was between 1 and 5 points in the large studies and pooled analyses.
• Clinical and pathological high-risk factors, such as pT4, obstruction, perforation and fewer than 12 lymph nodes, guide the traditional indication.
• Tumors with mismatch repair deficiency have a better prognosis and do not gain from fluoropyrimidine alone.
• Positive circulating tumor DNA after surgery identifies a very high risk of recurrence.
• DYNAMIC showed that decision-making guided by circulating tumor DNA reduced the use of chemotherapy without worsening recurrence-free survival at 2 years.

Why this is a reasoning scenario

In stage III colon cancer, the benefit of adjuvant chemotherapy is well established. In stage II, the discussion has continued for decades. Most patients are already cured by surgery, and chemotherapy can only help the minority who still have microscopic disease. The problem is that, until recently, there was no way to know who that minority was. The decision relied on factors in the pathology report that increase the average risk, but that do not show each person's residual disease.

The professor in the scenario has a low-risk tumor by the classic criteria. At the same time, he arrives with information from a colleague and with expectations about a new test. The consultation requires separating three questions: what his risk is, how much chemotherapy would reduce that risk and what the blood test would add. Each question has a different answer, and mixing them leads to bad decisions.

The post on why the multidisciplinary team differs in colon and rectal cancer shows who takes part in this decision. Here the focus is narrower: the patient with stage II who has already been operated on.

The mistake the scenario wants to avoid

The most common mistake is to indicate chemotherapy because the patient has cancer, without estimating the absolute gain. Another mistake is to order a new test without knowing what to do with each result. A third mistake is to ignore mismatch repair, which changes the prognosis and the usefulness of fluoropyrimidine.

The size of the benefit in stage II

A review on the topic recalls that surgery alone cures about 80 out of every 100 patients with stage II colon cancer (Rebuzzi and colleagues, 2020). This starting point limits what any adjuvant treatment can add. If recurrence happens in about 20 out of every 100, and chemotherapy reduces part of that risk, the absolute gain will be small.

What the large studies showed

The same review summarizes the absolute gain from fluoropyrimidine in 5-year overall survival: 2 points in the IMPACT B2 analysis, 5 points in the Intergroup analysis, 4 points in QUASAR and 1 point, from 80 to 81 in every 100, in the pooled ACCENT analysis (Rebuzzi et al., 2020). All the studies showed a gain in disease-free survival, of 3 to 4 points at 5 years (Rebuzzi et al., 2020).

In QUASAR, with 3,239 patients with colorectal cancer, most in stage II, the absolute gain in 5-year overall survival was 2.9 points with chemotherapy, according to the description of the French PRODIGE 22 protocol (Karoui et al., 2015). The authors of that protocol summarize that the absolute benefit in stage II is usually estimated at between 2 and 5 points (Karoui et al., 2015).

An Austrian phase III study with 5-fluorouracil and leucovorin in stage II showed no significant difference in death from colon cancer: 13.1 in every 100 in the chemotherapy group versus 15.3 in every 100 in the control group, with a hazard ratio of 0.83 and a confidence interval crossing 1 (Schippinger et al., 2007). Isolated studies, with few events, rarely have the power to show a gain of this order.

What registry data add

In a Korean national analysis with 7,880 patients operated on for stage II colon cancer, adjuvant chemotherapy was associated with better survival both in the high-risk group and in the low-risk group, also in patients over 70 years old (Kim et al., 2018). Adding oxaliplatin showed no difference in survival in this cohort, with a hazard ratio of 1.36 and an interval of 0.91 to 2.03 (Kim et al., 2018). Registry data suffer from selection bias, because those who receive chemotherapy tend to be healthier, and for this reason they do not replace randomized trials.

The practical conclusion is that, for a patient at average risk, chemotherapy offers few points of gain. For the professor in the scenario, with below-average risk, the expected gain is even smaller. This does not mean that chemotherapy should never be offered, but rather that the conversation needs to start from the absolute number.

The high-risk factors

The traditional indication for chemotherapy in stage II rests on factors that increase the risk of recurrence. The 2020 review lists the main ones (Rebuzzi et al., 2020):

• pT4 tumor, which invades the visceral peritoneum or neighboring organs.
• Obstruction or perforation at presentation.
• Lymphovascular invasion.
• Perineural invasion.
• Fewer than 12 lymph nodes examined in the specimen.
• Poorly differentiated tumor, when mismatch repair is preserved.

The Korean analysis used a similar list, including positive margin, and showed that the risk grew with the number of factors present in those who did not receive chemotherapy (Kim et al., 2018). In this cohort, 27.9 in every 100 patients had at least one high-risk factor (Kim et al., 2018).

Why the factors are not enough

The factors in the pathology report indicate increased risk in the group, but they do not say whether that patient has residual disease. A Dutch cost-effectiveness study recalls that recurrence in patients without a pT4 tumor remains considerable, about 12.5 in every 100, and that some high-risk patients cured by surgery receive unnecessary treatment (Kramer et al., 2024). This imprecision in both directions is the reason for the search for better markers.

The lymph node count deserves special attention. Fewer than 12 lymph nodes may mean that an involved lymph node was not examined, and the patient would in fact be in stage III. In that case, the risk factor reflects the quality of the specimen and of the pathology exam, not only the biology of the tumor.

Mismatch repair and tumor side

Mismatch repair deficiency appears in about 15 out of every 100 stage II colon tumors and is about 4 times more frequent in the right colon than in the left (Rebuzzi et al., 2020). Without treatment, survival of stage II patients with repair deficiency was 90 out of every 100, versus 66 out of every 100 with preserved repair, in an analysis cited by the review (Rebuzzi et al., 2020). The same review reports that, in this group, fluoropyrimidine alone was associated with worse 5-year overall survival, with a hazard ratio of 2.95 (Rebuzzi et al., 2020).

When high risk coexists with deficient repair

In a Chinese series of 262 patients with stage II and repair deficiency, 68.3 out of every 100 had at least one high-risk factor (Liao et al., 2024). Three-year disease-free survival was 92.6 out of every 100 with surgery alone and 88.1 out of every 100 with adjuvant chemotherapy, with no significant difference (Liao et al., 2024). The authors found no benefit from oxaliplatin chemotherapy even in the subgroups with pT4 or fewer than 12 lymph nodes, and they identified high preoperative CEA and pT4 as independent risk factors for recurrence (Liao et al., 2024).

A Norwegian cohort of 544 patients showed that repair deficiency was associated with a trend toward better disease-free survival in stage II, but with worse survival in stage III, with a hazard ratio of 4.17 (Hestetun et al., 2022). The prognostic value of the marker depends on the stage, and this reinforces the need to interpret the result within context.

Tumor side

A 2017 Chinese analysis combined tumor side with mismatch repair in stage II (Yang et al., 2017). Right colon tumors had lower survival, and the only subgroup that appeared to benefit from chemotherapy was that of right colon tumors with preserved repair (Yang et al., 2017). These are retrospective data, useful for generating hypotheses, but they show how great the heterogeneity of stage II is.

For the professor in the scenario, the sigmoid tumor with preserved repair and no risk factors places the patient in the group where the benefit of chemotherapy is most uncertain. The conversation should start with his favorable prognosis, not with the list of treatments.

Circulating tumor DNA as a marker of residual disease

After surgery, DNA fragments carrying the tumor's mutations may remain in the blood if tumor cells are left. Detection of this circulating tumor DNA is interpreted as a sign of molecular residual disease. Unlike the factors in the pathology report, which estimate average risk, the test looks for the disease itself.

The study that showed the signal

In an Australian prospective study of 230 patients operated on for stage II colon cancer, circulating tumor DNA was detected after surgery in 14 of 178 patients not treated with chemotherapy, or 7.9 out of every 100 (Tie et al., 2016). Of these, 11, or 79 out of every 100, had recurrence over a median follow-up of 27 months, versus 9.8 out of every 100 among patients with a negative test, with a hazard ratio of 18 (Tie et al., 2016). The Dutch MEDOCC-CrEATE protocol notes that the prognostic value of the test was much greater than that of pT4 in this study, with a hazard ratio of 14 versus 2.6 in the univariate analysis (Schraa et al., 2020).

A 2026 review on the topic reports that, on the Japanese CIRCULATE platform, postoperative positivity was associated with a hazard ratio of 11.99 for disease-free survival (Yang et al., 2026). The authors highlight that this association holds after adjustment for known clinical and pathological factors (Yang et al., 2026).

The difference between prognosis and prediction

A prognostic marker tells who is at higher risk. A predictive marker tells who benefits from a treatment. Circulating tumor DNA is a strong prognostic marker, but that does not prove that chemotherapy works better in those with a positive test. This distinction is central to interpreting DYNAMIC and to talking with the professor in the scenario.

DYNAMIC: deciding by circulating tumor DNA

DYNAMIC randomized 455 patients with stage II colon cancer, in a 2 to 1 ratio, between a decision guided by circulating tumor DNA and a standard decision based on clinical and pathological factors (Tie et al., 2022). In the guided group, a positive test in the fourth or seventh week after surgery led to chemotherapy with oxaliplatin or with a fluoropyrimidine alone, and those with a negative test received no treatment (Tie et al., 2022).

What the study showed

Adjuvant chemotherapy was used in 15 out of every 100 patients in the guided group and in 28 out of every 100 in the standard group (Tie et al., 2022). Recurrence-free survival at 2 years was 93.5 out of every 100 in the guided group and 92.4 out of every 100 in the standard group, which met the non-inferiority criterion, with a margin of 8.5 points (Tie et al., 2022). At 3 years, recurrence-free survival was 86.4 out of every 100 among treated patients with a positive test and 92.5 out of every 100 among untreated patients with a negative test (Tie et al., 2022).

Outcomes in DYNAMIC	Decision guided by circulating tumor DNA	Standard decision
Randomized patients	302	153
Use of adjuvant chemotherapy	15 out of every 100	28 out of every 100
Recurrence-free survival at 2 years	93.5 out of every 100	92.4 out of every 100
Recurrence-free survival at 3 years	91.7 out of every 100	92.4 out of every 100

Source of the table numbers: Tie et al., 2022 (Tie et al., 2022).

How to read the result

DYNAMIC was designed to show that the guided strategy was not worse, not to show that it was better (Tie et al., 2022). The non-inferiority margin of 8.5 points is wide, and recurrence-free survival at 3 years was 91.7 out of every 100 in the guided group versus 92.4 out of every 100 in the standard group, with a hazard ratio of 0.96 (Tie et al., 2022). The main gain was sparing chemotherapy for about half of the patients who would have received it under the standard decision, with no detectable loss of disease control.

In the guided group, recurrence or death occurred in 15 of 246 patients with a negative test and in 8 of 45 with a positive test (Tie et al., 2022). A negative test therefore does not guarantee a cure: about 7 out of every 100 negative patients had a recurrence at 3 years (Tie et al., 2022).

For the professor in the scenario, the colleague's information matches this result: with a negative test, the colleague was spared chemotherapy within a strategy that has been tested. But the test was not available in every service, and the decision to order it must come with a plan for each result.

What the test does not yet answer

The 2026 review summarizes the limitations that still weigh on use of the test (Yang et al., 2026):

• False-negative results when residual disease has a very low volume.
• Differences between platforms that do or do not use the mutational profile of the operated tumor.
• Lack of consensus on the ideal time to collect the sample after surgery.
• Little evidence that intensifying treatment in those with a positive test improves the outcome.
• Interference from clonal hematopoiesis on some platforms.

The authors note that, in stage II, the proportion of patients with a positive test is low, and that recurrence can occur in those who had a negative test (Yang et al., 2026). The strategy of sparing treatment has more support than the strategy of intensifying it (Yang et al., 2026).

Cost and access

A Dutch cost-effectiveness model compared selection strategies for chemotherapy in stage II (Kramer et al., 2024). Using circulating tumor DNA testing alone was less effective than the current guideline, while combining the test with pT4 and mismatch repair reduced recurrences, but at a cost of 67,413 euros per quality-adjusted life year, above the Dutch threshold of 50,000 euros (Kramer et al., 2024). The authors calculated that the combination would become cost-effective if the test cost less than 1,500 euros (Kramer et al., 2024).

In Brazil, access to the test is limited and varies widely between public and private services. That does not change the logic of the decision, but it changes what can be offered. In services without the test, the decision remains based on the pathology report and on mismatch repair.

Studies in progress

MEDOCC-CrEATE, designed within a Dutch cohort, includes 1,320 patients with stage II and no indication for chemotherapy under the guidelines and tests whether patients with a positive test accept treatment and whether it reduces recurrence (Schraa et al., 2020). Studies of this kind must answer whether a positive test is also predictive of benefit.

How to talk with the patient

The consultation in the scenario's professorship calls for a conversation organized in steps. The sequence below summarizes one way to conduct it:

• Explain the patient's risk of recurrence based on the pathology report, in absolute numbers.
• Explain how much chemotherapy would reduce that risk, also in absolute numbers.
• Explain what the circulating tumor DNA test is, what a positive and a negative result would mean and what would be done in each case.
• Explain the adverse effects of chemotherapy, including oxaliplatin neuropathy when it is considered.
• Agree on the decision and record the patient's preferences.

The post on diagnostic uncertainty helps with conversations about small benefits and known risks.

When chemotherapy is discussed

If the patient has high-risk factors and preserved mismatch repair, guidelines usually discuss chemotherapy with a fluoropyrimidine, with or without oxaliplatin. If the tumor has mismatch repair deficiency and there is no pT4, the prognosis is favorable, and a fluoropyrimidine alone should not be used. A positive test, when available, becomes one more risk element.

Follow-up

Whatever the decision, follow-up after surgery follows a schedule of visits, CEA, CT scans and colonoscopy. A negative circulating tumor DNA test does not replace this follow-up, because recurrence can still happen (Tie et al., 2022) (Yang et al., 2026).

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code. Stage II is a good exercise because the decisive clue is in the pathology report, in mismatch repair and in what the patient wants, and because the newest test is not always the first step. The post on differential diagnosis with clues shows how to train with progressive clues.

Three exercises with cases

Change the pathology report. Say what changes if the tumor is pT4 with perforation and only 9 lymph nodes examined.

Change the repair status. Say what changes if the tumor is in the right colon, with mismatch repair deficiency and no pT4.

Change the test. Say what to do if the circulating tumor DNA test comes back positive in the seventh week after surgery.

The post on clinical decision-making is about choosing tests that change management. The limits on using the game are on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch.

Frequently asked questions

How much does chemotherapy improve survival in stage II?

In large studies and pooled analyses, the absolute gain in 5-year overall survival was between 1 and 5 points (Rebuzzi et al., 2020).

Does the circulating tumor DNA test replace the pathology report?

No. In DYNAMIC, it was used to decide on chemotherapy, but assessment of the pathology report and of mismatch repair remains necessary (Tie et al., 2022).

Does a negative test mean cure?

No. In DYNAMIC, about 7 in every 100 patients with a negative test had a recurrence within 3 years (Tie et al., 2022).

Does a positive test guarantee that chemotherapy will work?

No. The test shows high risk, but the evidence of benefit from intensification is still limited (Yang et al., 2026).

Does a tumor with mismatch repair deficiency need chemotherapy?

In general, not with a fluoropyrimidine alone, because the prognosis is better and the fluoropyrimidine showed no gain in this group (Rebuzzi et al., 2020) (Liao et al., 2024).

Final checklist

• I reviewed the pT and pN stage and the lymph node count.
• I looked for obstruction, perforation, lymphovascular and perineural invasion.
• I confirmed the mismatch repair result.
• I estimated the recurrence risk and the absolute gain from chemotherapy.
• I defined what I would do with each result before ordering the circulating tumor DNA test.
• I explained the expected adverse effects.
• I recorded the decision and the patient's preferences.
• I agreed on the follow-up schedule.

Want to practice the reasoning with progressive clues? Play the case of the day.

References

1. Hestetun KE, Rosenlund NB, Stanisavljević L, et al. Stage-dependent prognostic shift in mismatch repair-deficient tumors: Assessing patient outcomes in stage II and III colon cancer. Front Oncol. 2022. PMID 36110945.
2. Karoui M, Rullier A, Luciani A, et al. Neoadjuvant FOLFOX 4 versus FOLFOX 4 with Cetuximab versus immediate surgery for high-risk stage II and III colon cancers: a multicentre randomised controlled phase II trial--the PRODIGE 22--ECKINOXE trial. BMC Cancer. 2015. PMID 26156156.
3. Kim MK, Won DD, Park SM, et al. Effect of Adjuvant Chemotherapy on Stage II Colon Cancer: Analysis of Korean National Data. Cancer Res Treat. 2018. PMID 29216709.
4. Kramer A, Greuter MJE, Schraa SJ, et al. Early evaluation of the effectiveness and cost-effectiveness of ctDNA-guided selection for adjuvant chemotherapy in stage II colon cancer. Ther Adv Med Oncol. 2024. PMID 39175989.
5. Liao L, Tang J, Hong Z, et al. The effects of oxaliplatin-based adjuvant chemotherapy in high-risk stage II colon cancer with mismatch repair-deficient: a retrospective study. BMC Cancer. 2024. PMID 38302968.
6. Rebuzzi SE, Pesola G, Martelli V, et al. Adjuvant Chemotherapy for Stage II Colon Cancer. Cancers (Basel). 2020. PMID 32927771.
7. Schippinger W, Samonigg H, Schaberl-Moser R, et al. A prospective randomised phase III trial of adjuvant chemotherapy with 5-fluorouracil and leucovorin in patients with stage II colon cancer. Br J Cancer. 2007. PMID 17895886.
8. Schraa SJ, van Rooijen KL, van der Kruijssen DEW, et al. Circulating tumor DNA guided adjuvant chemotherapy in stage II colon cancer (MEDOCC-CrEATE): study protocol for a trial within a cohort study. BMC Cancer. 2020. PMID 32819390.
9. Tie J, Cohen JD, Lahouel K, et al. Circulating Tumor DNA Analysis Guiding Adjuvant Therapy in Stage II Colon Cancer. N Engl J Med. 2022. PMID 35657320.
10. Tie J, Wang Y, Tomasetti C, et al. Circulating tumor DNA analysis detects minimal residual disease and predicts recurrence in patients with stage II colon cancer. Sci Transl Med. 2016. PMID 27384348.
11. Yang E, Wang X, Zhang J, et al. Postoperative circulating tumor DNA in stage II colon cancer: biological rationale, clinical evidence, and unresolved challenges. Front Oncol. 2026. PMID 42798930.
12. Yang L, He W, Yang Q, et al. Combination of primary tumor location and mismatch repair status guides adjuvant chemotherapy in stage II colon cancer. Oncotarget. 2017. PMID 29228759.