Long-term survival among men with conservatively treated localized prostate cancer

JAMA. 1995 Aug;274(8):626-31.


Objective: To determine age-specific, all-cause mortality, disease-specific mortality, and life expectancy for men aged 65 to 75 years who are treated only with immediate or delayed hormonal therapy for newly diagnosed, clinically localized prostate cancer.

Design: A population-based, retrospective cohort study.

Setting: Patient records were abstracted from 37 acute care hospitals and two Veterans Affairs medical centers in Connecticut. Original pathology slides were sent to a referee pathologist who was blinded to case outcomes.

Subjects: All men identified by the Connecticut Tumor Registry with clinically localized prostate cancer diagnosed in 1971 to 1976 who were aged 65 to 75 years at the time of diagnosis and were untreated or treated with immediate or delayed hormonal therapy.

Main outcome measures: Parametric proportional hazards models incorporating tumor histologic findings, comorbidity, and age at the time of diagnosis to compare cohort survival with that of men in the general population.

Results: After a mean follow-up of 15.5 years, the age-adjusted survival for men with Gleason score 2 to 4 tumors was not significantly different from that of the general population. Maximum estimated lost life expectancy for men with Gleason score 5 to 7 tumors was 4 to 5 years and for men with Gleason score 8 to 10 tumors was 6 to 8 years. Tumor histologic findings and patient comorbidities were powerful independent predictors of survival.

Conclusions: Compared with the general population, men aged 65 to 75 years with conservatively treated low-grade prostate cancer incur no loss of life expectancy. Men with higher-grade tumors (Gleason scores 5 to 10) experience a progressively increasing loss of life expectancy. Case series reports of survival/mortality experienced by men with clinically localized prostate cancer that fail to control for age, tumor histologic features, and comorbidities risk significant bias.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Aged
  • Cohort Studies
  • Humans
  • Life Expectancy*
  • Male
  • Neoplasm Staging
  • Neoplasms, Hormone-Dependent / mortality*
  • Neoplasms, Hormone-Dependent / pathology
  • Neoplasms, Hormone-Dependent / therapy
  • Proportional Hazards Models
  • Prostatic Neoplasms / mortality*
  • Prostatic Neoplasms / pathology
  • Prostatic Neoplasms / therapy
  • Retrospective Studies
  • Severity of Illness Index
  • Survival Analysis