Minimum 18-week chemotherapy improves survival in locally advanced rectal cancer after neoadjuvant radiotherapy: a post hoc analysis of the STELLAR trial.

BACKGROUND

The survival benefit of adjuvant chemotherapy after chemoradiotherapy in locally advanced rectal cancer (LARC) remains unproven, whereas total neoadjuvant therapy (TNT) incorporating preoperative chemotherapy has demonstrated improved outcomes. However, the total chemotherapy duration delivered across neoadjuvant and adjuvant phases varies substantially in clinical practice. We investigated the impact of total chemotherapy duration in the STELLAR trial.

METHODS

This post hoc analysis was based on the phase III randomized trial, comparing short-course radiotherapy followed by four cycles of chemotherapy (TNT) with long-course chemoradiotherapy (CRT) in LARC patients. Five hundred thirty-nine patients with available chemotherapy duration data were included, with a median follow-up of 68.1 months. Patients were categorized: group 1 (no chemotherapy, n = 121), group 2 (3 to 12 weeks, n = 113), group 3 (15 weeks, n = 30), and group 4 (≥ 18 weeks, n = 275). Disease-free survival (DFS), overall survival (OS), distant metastasis (DM), and locoregional recurrence (LRR) were assessed using time-dependent Cox regression.

RESULTS

Group 4 achieved the highest 5-year OS (82.1%) and DFS (66.0%) rates. Compared with groups 1 and 2, group 4 demonstrated significantly improved OS (adj. P ≤ 0.001) and improved DFS versus group 1 (HR 0.621, 95% CI 0.443-0.870, adj. P = 0.017). In the TNT cohort, group 4 was associated with significantly improved OS and DFS compared with group 2 (adj. P < 0.01), but not with group 3. Additionally, group 4 showed a significantly lower risk of LRR than group 3. In the CRT cohort, group 4 was associated with improved OS compared with group 1 (adj. P = 0.005); however, this association was not retained in surgical patients. No significant differences in DFS, DM, or LRR were observed across groups in the CRT cohort.

CONCLUSIONS

In TNT cohort, minimum 18 weeks of chemotherapy was associated with improved OS and DFS compared to 3 to 12 weeks. The observed OS benefit of minimum 18 weeks versus no chemotherapy in the CRT cohort was not retained among surgical patients. These findings suggest caution in shortening chemotherapy duration, particularly in high-risk patients treated with TNT, and warrant confirmation in prospective TNT-specific trials.

TRIAL REGISTRATION: The STELLAR trial was registered at ClinicalTrials.gov (identifier: NCT02533271); however, this post hoc analysis was retrospectively conducted.