Background: Neuroendocrine liver metastases (NELM) are typically associated with high recurrence rates following surgical resection. Conditional disease-free survival (CDFS) estimates may be more clinically relevant compared to actuarial survival estimates.
Methods: CDFS was assessed using a multi-institutional cohort of patients. Cox proportional hazards models were used to evaluate factors associated with disease-free survival (DFS). Three-year CDFS (CDFS3) estimates at "x" year after surgery were calculated as CDFS3 = DFS(x + 3)/DFS(x).
Results: A total of 521 patients met the inclusion criteria. While actuarial 3-year DFS gradually decreased from 49% at 1 year to 39% at 5 years, CDFS3 increased over time. CDFS3 at 5 years was estimated as 89% vs actuarial 8-year DFS of 39% (P < .001). The probability of remaining disease-free at 5 years after resection increased as patients remained disease-free. For example, the probability of being disease-free for an additional 3 years was 66.3% and 88.8% for patients who lived 2 and 5 years, respectively. Overall, CDFS3 in each subgroup increased postoperatively as years elapsed, however, the impact of each prognostic factor on CDFS3 changed over time.
Conclusion: CDFS of patients who underwent resection of NELM exponentially improved as patients survived additional years without recurrence. CDFS provides more accurate prognostic measures compared with traditional DFS measures.
Keywords: conditional survival; disease-free survival; neuroendocrine liver metastasis.
© 2019 Wiley Periodicals, Inc.