Purpose: We evaluated the hypothesis that telomere DNA content (TC) in prostate tumor tissue is associated with time to prostate cancer recurrence.
Materials and methods: The cohort was comprised of 77 men who underwent prostatectomy between 1982 and 1995. Slot blot assay was used to measure TC in DNA extracted from paraffin embedded tumor and nearby, histologically normal prostate (NHN) tissues. Multivariate Cox proportional hazards analysis was done to relate TC, patient age at diagnosis, Gleason sum and pelvic node involvement to time of prostate cancer recurrence. Regression analysis was done to relate TC in paired tumor and NHN tissues. Nonparametric Kruskal-Wallis analysis was done to relate TC in tumor and NHN tissues with 72-month disease-free survival.
Results: TC was a predictor of time to prostate cancer recurrence when controlling for age at diagnosis, Gleason sum and pelvic node involvement (RH = 5.02, 95% CI 1.40 to 17.96, p = 0.0132). TC in tumor tissue was associated with TC in NHN tissue (R = 0.601, p <0.0001). Median TC in tumor and NHN tissues from men in whom cancer recurred within 6 years was approximately half that in men who remained disease-free (p = 0.012 and 0.024, respectively).
Conclusions: Decreased TC in prostate tissues obtained by radical prostatectomy predicts prostate cancer recurrence independent of age at diagnosis, Gleason sum and pelvic node involvement. TC in tumor tissue is also associated with TC in NHN prostate tissue. Thus, mechanisms known to generate genomic instability are operative in fields of cells beyond the tumor margins prior to histological changes.