Conditional Probability of Survival After Neoadjuvant Chemoradiation and Proctectomy for Rectal Cancer: What Matters and When

BACKGROUND: Disease-free survival estimated from the time of surgery does not account for the changing likelihood of survival based on time already accrued. Conditional disease-free survival is defined as the probability of remaining disease free after reaching a specific time point without recurrence. OBJECTIVE: The purpose of this study was to evaluate conditional disease-free survival for patients with rectal cancer who were treated by proctectomy after neoadjuvant chemoradiation. DESIGN: Demographics, tumor characteristics, and tumor regression scores were assessed. Three-year conditional disease-free survival was estimated at x year after surgery based on the formula cDFS(3) = DFS(x+3)/DFS(x), where DFS is disease-free survival and cDFS is conditional disease-free survival. Analyses were performed using Cox proportional hazards models. SETTING: The study was conducted at a single tertiary referral center. PATIENTS: A total of 545 patients with rectal cancer who were treated by neoadjuvant chemoradiation and curative intent surgery between 1992 and 2012 were included. MAIN OUTCOME MEASURES: Disease-free survival and conditional disease-free survival were measured. RESULTS: The median patient age was 57.5 years, and 28.4% were women. Median follow-up was 5.9 years. Disease-free survival at 1, 3, and 5 years was 89%, 71%, and 63%. The probability of remaining disease free for an additional 3 years for patients disease free at 1, 3, and 5 years was 75%, 83%, and 82%. Tumor regression, pathologic stage, margin status, differentiation, and procedure (low anterior versus abdominoperineal resection) were associated with disease-free survival on multivariable analysis (p < 0.05), but their relevance varied over time. R1 resection and differentiation were initially significant but not at 5 years. In contrast, tumor regression after neoadjuvant chemoradiation had a long-lasting impact on survival (at 5 y, conditional disease-free survival for an additional 3 y: 91%, 85%, 76%, and 71% for regression scores 0, 1, 2, and 3; p = 0.002). LIMITATIONS: This was a retrospective study over 20 years, with evolution in adjuvant therapies during this time. CONCLUSIONS: Conditional disease-free survival estimates improved over time, confirming that most patients will see a recurrence within the first few years. The impact of specific prognostic factors evolves variably over time. This information is useful to patients and providers and can help guide counseling and surveillance. See Video Abstract at .