TOSCA

Background

Phase III, open-label RCT. 2402 patients with resected stage III (high-risk, N2 or T4) or stage II high-risk colon cancer. Italian GONO group trial comparing 3 months (6 cycles) vs 6 months (12 cycles) of adjuvant FOLFOX4. One of the 6 trials contributing to the IDEA collaboration.

Interventions and follow up

Arm A: FOLFOX4 × 6 cycles (3 months): oxaliplatin 85mg/m² D1 + leucovorin 200mg/m² D1/D2 + 5-FU 400mg/m² bolus + 5-FU 600mg/m² CI 22h D1/D2 q2wk
Arm B: FOLFOX4 × 12 cycles (6 months)
Primary endpoint: Disease-free survival (DFS) non-inferiority
mFollow up: 60 mo

Results

DFS: 5-yr DFS 64.5% vs 65.5%, HR 1.13, 95% CI 0.95–1.34 — non-inferiority not proven (CI includes 1.20 margin)
OS: HR 1.01 — similar
Neuropathy grade ≥2 residual at 12 mo: 5% vs 21% (3 vs 6 mo)

Adverse events

Neurologic: Grade ≥3 sensory neuropathy 3% vs 11% (3 vs 6 mo); treatment discontinuation due to neuropathy 2% vs 10%
GI/hematologic: Hematologic toxicity and GI toxicity (diarrhea, mucositis) similar; overall tolerability substantially better with 3-month duration

Conclusions

TOSCA alone failed to definitively prove non-inferiority of 3 vs 6 months FOLFOX4 in stage III colon cancer — the confidence interval extended above the pre-specified non-inferiority margin of 1.20. However, TOSCA enrolled a deliberately high-risk population and used FOLFOX4, and the neuropathy reduction with 3 months was marked. Its contribution to the IDEA pooled analysis provides context for the regimen-specific (FOLFOX vs CAPOX) treatment duration recommendations.

Key Limitations

Key Limitations: TOSCA deliberately enriched for high-risk disease (N2/T4), the population where shorter duration is most problematic by IDEA analysis. The FOLFOX4 regimen (bolus 5-FU) is less common than FOLFOX6 in current practice. The non-inferiority margin of 1.20 used in TOSCA differs from the 1.12 used in IDEA, affecting interpretation. The trial failed to independently demonstrate non-inferiority; conclusions depend on the IDEA pooled analysis. The enrolled population overlaps with IDEA's high-risk stratum, which showed FOLFOX 3-month non-inferiority was not demonstrated.

Clinical Context

TOSCA is primarily cited as a component of the IDEA collaboration and illustrates the challenge of demonstrating 3-month FOLFOX non-inferiority in high-risk stage III colon cancer. Its results, combined with IDEA, support the current recommendation that 3-month FOLFOX may be considered in low-risk (T1-3/N1) stage III but 6 months should be preferred in high-risk (T4/N2) disease. CAPOX 3 months is the preferred shorter-duration regimen across risk strata per IDEA.

References

References: Sobrero AF et al, Ann Oncol 2018