EvidenceHealthCancer
Adjuvant Chemotherapy After Surgery for Colon Cancer (Stage III)
- Study type:
- Set of randomised clinical trials comparing different adjuvant chemotherapy regimens after surgery; early trials include surgery‑alone vs surgery+chemo, later trials compare chemo vs chemo (no modern surgery‑alone arm).
- Topic:
- Adjuvant chemotherapy after curative surgery for stage III colon cancer
- Added:
- 2026-02-05
- Status:
- published
Strength of evidence
Moderate – multiple randomised trials with hard endpoints exist, but modern regimens are only tested against older chemo (not against surgery alone); primary reliance on disease‑free survival and incomplete harm reporting limit the strength of conclusions.
Source
Moertel CG et al. Adjuvant therapy with 5‑fluorouracil and levamisole for resected colon carcinoma. J Clin Oncol. 1991;9:286–296. PMID: 1794023. André T et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer (MOSAIC). N Engl J Med. 2004;350:2343–2351. PMID: 15175436. André T et al. Improved overall survival with oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment in stage II or III colon cancer: 10‑year follow‑up of MOSAIC. J Clin Oncol. 2015;33:3733–3740. PMID: 26527776.
Several randomised trials have tested whether adding chemotherapy after curative surgery for colon cancer (especially stage III with lymph‑node involvement) reduces recurrences and improves survival. Early trials compared surgery alone with 1 year of 5‑fluorouracil (5‑FU) plus levamisole and found relative reductions in recurrence and mortality, although absolute survival gains were modest and toxicity was substantial. Later studies replaced levamisole by leucovorin (5‑FU/LV), and finally added oxaliplatin (FOLFOX), showing additional absolute 10‑year survival benefits in stage III patients (about +8 percentage points vs 5‑FU/LV), but at the cost of more neuropathy and other side‑effects. From a WorthDay perspective, adjuvant chemotherapy for resected stage III colon cancer rests on moderate‑quality evidence: it probably produces a real but limited survival advantage for some patients, yet the absolute effect size is small and a full, transparent accounting of long‑term harm vs benefit is still missing.
Key findings
- Early adjuvant trials comparing surgery alone vs surgery + 5‑FU/levamisole in Dukes C / stage III colon cancer reported relative reductions in recurrence (~31–41%) and overall mortality (~13–33%) with chemotherapy, but absolute 5‑year survival gains were only in the range of a few to perhaps 10 percentage points.
- Long‑term toxicity and quality‑of‑life data for 5‑FU/levamisole were poorly documented; the regimen is clearly not harmless.
- Schema‑comparison trials (e.g. adjCCA‑01) found that 5‑FU/leucovorin produced better disease‑free and overall survival than 5‑FU/levamisole in stage III patients (e.g. median OS ~88.9 vs 78.6 months; 117 vs 144 deaths), but did not include a surgery‑only arm.
- The MOSAIC trial randomised 2 246 stage II–III patients to 6 months of 5‑FU/LV (LV5FU2) or FOLFOX4 (5‑FU/LV + oxaliplatin). FOLFOX improved 3‑year disease‑free survival by about 5–6 absolute percentage points (relative risk reduction ~20%).
- Long‑term MOSAIC follow‑up showed a 10‑year overall survival of ~71.7% with FOLFOX vs 67.1% with 5‑FU/LV (absolute gain ~4.6 percentage points). In stage III patients the absolute 10‑year OS gain was ~8.1 percentage points (~67.1% vs 59.0%). No clear survival benefit was seen in stage II.
- FOLFOX caused substantially more acute and chronic toxicity, especially cumulative peripheral neuropathy; some patients experienced long‑lasting or permanent sensory deficits.
- Modern regimens like FOLFOX are not compared directly against contemporary surgery‑only care; their incremental value vs no chemo is inferred from older studies rather than measured in a dedicated randomised trial.
- Overall, the evidence suggests a real but modest survival benefit for some stage III patients, counterbalanced by non‑trivial treatment‑related harm, leading to a WorthDay evidence rating of moderate rather than high.
Background
- Colon cancer is frequently diagnosed in stages II and III. After radical surgical resection (R0), many patients have no visible disease, yet some later relapse due to microscopic residual tumour. Adjuvant chemotherapy aims to eradicate these cells and improve long‑term survival. In the 1980s, randomised trials introduced 5‑fluorouracil (5‑FU) plus levamisole as adjuvant therapy in stage III patients. Over time, levamisole was replaced by leucovorin (5‑FU/LV) and oxaliplatin was added to form FOLFOX, which became a guideline standard. The key question is: to what extent do these regimens actually extend life, and at what cost, compared with surgery alone or earlier chemo schemes?
Methods
- The early Intergroup trials randomised patients with completely resected Dukes C / stage III colon cancer to either surgery alone or surgery plus one year of 5‑FU/levamisole. Endpoints were relapse‑free survival and overall survival, with no blinding; patients and clinicians knew who received chemotherapy. Later schema‑comparison trials, such as adjCCA‑01, randomised resected stage III patients to 5‑FU/LV vs 5‑FU/levamisole, with disease‑free survival (DFS) and overall survival (OS) as endpoints. The MOSAIC trial enrolled 2 246 patients with resected stage II or III colon cancer and randomised them to 6 months of LV5FU2 (5‑FU/LV) or FOLFOX4 (5‑FU/LV + oxaliplatin). The primary endpoint was 3‑year DFS, with OS as a key secondary endpoint; follow‑up extended to 10 years. All regimens were delivered in open‑label fashion, and modern trials do not include a surgery‑only arm.
Results
- In the early 5‑FU/levamisole trials, chemotherapy reduced recurrence and death risk in relative terms compared with surgery alone, but absolute survival gains were modest. For example, if about 60 of 100 surgery‑only patients were alive at 5 years, adjuvant treatment might raise this to roughly 65–70 of 100, not 90 of 100. Schema‑comparison work showed that 5‑FU/LV yielded better DFS and OS than 5‑FU/levamisole, with roughly 10 months longer median survival and fewer deaths, establishing 5‑FU/LV as a superior backbone. MOSAIC further demonstrated that adding oxaliplatin improved DFS and, on longer follow‑up, OS: overall 10‑year survival rose from ~67.1% with 5‑FU/LV to ~71.7% with FOLFOX (absolute gain ~4.6 percentage points), and in stage III from ~59.0% to ~67.1% (absolute gain ~8.1 percentage points). No clear OS advantage was seen in stage II. However, FOLFOX substantially increased acute toxicity and chronic neuropathy, with a subset of patients experiencing persistent sensory deficits and functional limitations.
Interpretation
- From a conventional oncology perspective, these findings justify adjuvant chemotherapy as standard care for stage III colon cancer: multiple randomised trials demonstrate statistically significant improvements in survival, and FOLFOX appears more effective than older regimens. From the WorthDay perspective, the evidence is judged as moderate rather than high. Randomisation and hard endpoints are strengths, but the absence of modern surgery‑only controls, heavy reliance on DFS as a surrogate, and incomplete long‑term harm assessment restrict the strength of causal claims. The absolute survival gains – a few extra long‑term survivors per hundred treated – are meaningful but far from curative, and must be balanced against the risk of lasting toxicity. The fair summary is that adjuvant chemotherapy probably offers a small to moderate survival advantage for some stage III patients, at the cost of significant harm, and should be presented as such rather than as an unquestioned life‑saving measure for everyone.
Limitations
- • No modern surgery‑only comparison for current regimens: 5‑FU/levamisole was compared with surgery alone, but 5‑FU/LV and FOLFOX were tested only against earlier chemo, not against no chemo, so the incremental value of FOLFOX vs surgery alone is extrapolated rather than directly measured.
- • Reliance on surrogate endpoints: many key trials use disease‑free survival (DFS) as the primary endpoint; DFS depends on imaging and definitions of relapse and does not automatically guarantee longer or better life.
- • Incomplete reporting of long‑term harm: neuropathy and other chronic toxicities are acknowledged but not fully quantified; there is no integrated net‑benefit analysis combining survival gains and serious harms.
- • Relative risk framing can exaggerate perceived benefit: emphasis on relative risk reductions (e.g. 20% less relapse) can make modest absolute gains (a few extra survivors per 100) appear larger than they really are.
- • Trial populations may not reflect real‑world patients: older, frailer or multi‑morbid individuals are often under‑represented, so benefits and harms in everyday practice may differ from trial results.
- • Open‑label design (no blinding): clinicians and patients knew which regimen was used, which can influence reporting of side‑effects and ancillary care, potentially biasing comparisons.
WorthDay perspective
- For WorthDay, S0037 is a clear example of moderate‑quality evidence in a high‑stakes medical decision. Users facing adjuvant chemotherapy after surgery for stage III colon cancer should understand that randomised trials suggest a modest increase in long‑term survival probability, especially with FOLFOX, but the benefit is measured in single‑digit percentage points and comes with real risks of acute toxicity and chronic neuropathy. The Science Lab entry can help users formulate concrete questions for their oncologists, such as: "Out of 100 people like me, how many more are alive at 10 years with this regimen?" and "What are the realistic chances of long‑term nerve damage or other serious side‑effects?"
- In the app, this topic sits primarily in Health but strongly affects Productivity & Balance (ability to work, energy management) and Social (family decisions, support networks). WorthDay Elements could include: preparing question lists before consultations, logging how treatment discussions influence daily energy and mood, scheduling time with trusted people to review options, and consciously recording the values that guide such life‑and‑death choices. The key message is that chemotherapy for stage III colon cancer is neither a guaranteed life‑saver nor pure poison; it is a trade‑off under uncertainty, and users should be supported in making informed, value‑aligned decisions.
Permalink: https://www.worth-day.com/science-compass/s0037-adjuvant-chemotherapy-after-surgery-for-colon-cancer-stage
Related studies
Science meets practice