Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer
Phase observationalN = 570
- Population
- 570 tumour specimens from patients with resected stage II (312) or stage III (258) colon cancer; 95 (16.7%) high-frequency MSI; 287 patients received no adjuvant therapy and 283 received adjuvant chemotherapy.
- Design
- Retrospective biomarker analysis of tumour specimens from patients enrolled in five earlier randomised phase 3 trials of fluorouracil-based adjuvant chemotherapy versus no adjuvant treatment (three with leucovorin: NCIC-CTG C.03, NCCTG 874651 and FFCD; two with levamisole: NCCTG 784852 and Gastrointestinal Intergroup 0035). Microsatellite instability assessed with mononucleotide and dinucleotide markers; samples with instability at 30% or more of loci were high-frequency MSI, and low-frequency MSI was grouped with microsatellite-stable tumours. Primary outcomes overall and disease-free survival, censored at eight years. Not a new randomised comparison: the treatment comparison within each trial was randomised, the MSI classification was not.
- Primary endpoint
- OS, DFS
- Arms
- No adjuvant therapy (control, n = 287): Surgery without adjuvant chemotherapy (control arms of the source trials)
- Fluorouracil-based adjuvant chemotherapy (experimental, n = 283): Fluorouracil with leucovorin or levamisole (source-trial regimens)
- People
- Ribic CM
OSHR 0.72 (95% CI 0.53–0.99)Δ 60-month OS−17.3 pointsΔ 60-month OS+7.1 points
Show the numbers
| Endpoint | Analysis | Fluorouracil-based adjuvant chemotherapy | No adjuvant therapy | Effect |
|---|---|---|---|---|
| OSPatients with microsatellite-stable or low-frequency MSI tumours; multivariate analysis adjusted for stage and grade | exploratoryRibic 2003 | — | — | HR 0.72 (95% CI 0.53–0.99)P=0.04Chemotherapy versus no chemotherapy. Retrospective analysis of specimens from randomised trials. The abstract states that the benefit of treatment differed significantly according to MSI status (P=0.01 for the interaction) and that there was no benefit of adjuvant chemotherapy in the group with high-frequency MSI. |
| OSPatients with high-frequency MSI tumours (95: 42 without and 53 with adjuvant chemotherapy) | exploratoryRibic 2003 | 70.7% at 5 y (59.2–84.5%) | 88% at 5 y (78.7–98.4%) | P=0.07Univariate comparison from Table 3 of the full text (P=0.07 is the Table 3 comparison of the two rates); the confidence intervals are the 95% intervals printed there. Small subgroup (95 patients); the multivariate hazard ratio for death with chemotherapy in this group is 2.14 (95% CI 0.83 to 5.49; P=0.11) in Table 4 of the full text. |
| OSPatients with microsatellite-stable or low-frequency MSI tumours (475: 245 without and 230 with adjuvant chemotherapy) | exploratoryRibic 2003 | 75.5% at 5 y (70.1–81.4%) | 68.4% at 5 y (62.7–74.6%) | P=0.02Univariate comparison from Table 3 of the full text; the confidence intervals are the 95% intervals printed there. |
| Endpoint | Analysis | Fluorouracil-based adjuvant chemotherapy | No adjuvant therapy | Effect |
|---|---|---|---|---|
| OSPatients with microsatellite-stable or low-frequency MSI tumours; multivariate analysis adjusted for stage and grade | exploratoryRibic 2003 | — | — | HR 0.72 (95% CI 0.53–0.99)P=0.04Chemotherapy versus no chemotherapy. Retrospective analysis of specimens from randomised trials. The abstract states that the benefit of treatment differed significantly according to MSI status (P=0.01 for the interaction) and that there was no benefit of adjuvant chemotherapy in the group with high-frequency MSI. |
| OSPatients with high-frequency MSI tumours (95: 42 without and 53 with adjuvant chemotherapy) | exploratoryRibic 2003 | 70.7% at 5 y (59.2–84.5%) | 88% at 5 y (78.7–98.4%) | P=0.07Univariate comparison from Table 3 of the full text (P=0.07 is the Table 3 comparison of the two rates); the confidence intervals are the 95% intervals printed there. Small subgroup (95 patients); the multivariate hazard ratio for death with chemotherapy in this group is 2.14 (95% CI 0.83 to 5.49; P=0.11) in Table 4 of the full text. |
| OSPatients with microsatellite-stable or low-frequency MSI tumours (475: 245 without and 230 with adjuvant chemotherapy) | exploratoryRibic 2003 | 75.5% at 5 y (70.1–81.4%) | 68.4% at 5 y (62.7–74.6%) | P=0.02Univariate comparison from Table 3 of the full text; the confidence intervals are the 95% intervals printed there. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| OS | 0.72 | 0.53–0.99 |