Stratifying prostate cancer patients by relative lymph node involvement: population- and modeling-based study

Cancer Med. 2016 Aug;5(8):1850-5. doi: 10.1002/cam4.776. Epub 2016 May 26.

Abstract

It is estimated that about 10% of new prostate cancer (PCa) cases are lymph node-positive (LN+). We have previously discussed the role of the inflection point (IP) of an inverse Gompertzian survival curve as a surrogate for disease incurability. In this study, we aimed to stratify curability of different patient cohorts with pathologically positive lymph nodes through modeling survival curves by different percentages of LN involvement (%LN+) postoperatively and calculating associated IPs. From the Surveillance, Epidemiology, and End Results (SEER) database, we selected LN+ PCa patients undergoing radical prostatectomy. Modeling of relative survival curves using inverse Gompertzian kinetics for increasing value of maximal %LN+ involvement allowed stratification of cohort into groups with <10%, 10-40%, and greater or equal to 40% of LN+ out of all LNs sampled. Data were retrieved for 5903 patients. For the entire cohort, relative survival was 96%, 87%, and 76% at 5, 10, and 15 years, respectively. For %LN +, <10% the IP was about 27 years postoperatively. Patients with 10-40% LN+ had an IP at about 10 years; for those with more than 40% LN+, the IP was 7 years. A 10-year relative survival decreases from 97% for <10% LN+ to 71% for more than 40% LN+. While better therapies for LN + PCa are badly needed, this patient cadre is not homogenous and should be stratified by %LN+ in future clinical trials.

Keywords: Health services research; lymph node-positive; mathematical oncology; relative survival.

MeSH terms

  • Aged
  • Humans
  • Kaplan-Meier Estimate
  • Lymphatic Metastasis
  • Male
  • Middle Aged
  • Models, Statistical
  • Neoplasm Grading
  • Prostatectomy
  • Prostatic Neoplasms / mortality
  • Prostatic Neoplasms / pathology*
  • Prostatic Neoplasms / surgery
  • SEER Program
  • United States / epidemiology