The association between timing in adjuvant chemotherapy administration and overall survival for women with breast cancer within the National Comprehensive Cancer Network (NCCN).

Introduction: Population based studies (eg, Hershman et al. BCRT 2006 and Lohrisch et al. JCO 2006) showed poorer survival associated with long delays in adjuvant chemotherapy (CTX) initiation following definitive surgery (DS) for women with breast cancer (BC). Delays in CTX following diagnosis (DX) have not been evaluated. The ASCO/NCCN quality measures (QMs) recommend CTX <=120 days after DX for patients with stage II/III ER/PR negative disease. We sought to examine the impact of delayed CTX on survival overall and stratified by disease-specific prognostic factors. Methods: 4,608 women with stage I-III HER2 negative breast cancer diagnosed between 2000 and 2006 at 8 NCCN centers were identified using the NCCN outcomes database. Patients with T3/4 disease or who received neoadjuvant therapy were excluded. The association between CTX timing and OS was evaluated using multivariate Cox models adjusted for CTX type, age, race, BMI, residential distance, insurance, SES, comorbidity, ER/PR, LVI, grade, T stage, and N stage. The impact of CTX timing was evaluated using a >90-day (d) DS-to-CTX threshold, based on poor outcomes observed in prior studies, and a >120d DX-to-CTX threshold, based on the ASCO/NCCN QMs. Results: Median follow-up was 7.2 years and OS at 7 years was 89%. Overall, 401 (8.7%) patients received CTX >120d after DX and 113 (2.4%) patients received CTX >90d after DS. The DX-to-CTX interval was more strongly correlated with the DX-to-DS (r = 0.74) interval than DS-to-CTX (r = 0.54) interval. A >90d DS-to-CTX interval was significantly associated with poorer survival (HR: 1.65, 95% CI 1.04-2.60, p = 0.03) in adjusted analyses. Shorter DS-to-CTX thresholds of >60d (n = 636, HR: 1.13, 95% CI: 0.89-1.43, p = 0.319) or >75d (n = 273, HR: 1.05, 95% CI 0.74-1.49, p = 0.76) were not associated with OS. The association between a >120d DX-to-CTX interval and OS was not statistically significant (HR: 1.32, 95% CI 0.99-1.76, p = 0.06). Patients who received CTX >135d(n = 231, HR1.25, 95% CI: 0.87-1.81, p = 0.22)or>150d(n = 128, HR1.15, 95% CI: 0.59-2.24, p = 0.69) after DX did not display an increased risk of death. Excluding pathological staging factors from the model had no effect on the results. In subgroup analyses stratified by ER/PR, LVI, grade, T stage or N stage, a >120d delay in CTX did not display significant associations with OS. Among ER/PR negative patients, the association between a >120d delay and OS was borderline non-significant after adjusting the p-value for multiple hypothesis testing using the false discovery rate method (HR: 1.80, 95% CI: 1.16-2.79, p = 0.09). Conclusion: Consistent with previous studies, CTX delays of >90 days following surgery were associated with poorer survival. OS was not significantly compromised in patients with DX-to-CTX intervals >120 days although this analysis may have limited power to detect small effects. More variation in the DX-to-CTX interval was attributed to pre-surgery time which may explain the differences observed between the DX-to-CTX and DS-to-CTX intervals. Among patients with ER/PR negative disease, a non-significant association between OS and a >120 day DX-to -CTX interval was observed that warrants further examination. [ABSTRACT FROM AUTHOR]