FAMILIAL PROSTATE CANCER: A DIFFERENT DISEASE?
PATRICK A. KUPELIAN, ERIC A. KLEIN, JOHN S. WITTE, VARANT A. KUPELIAN AND JOHN H. SUH

Fron The Departments of Radiation,, Oncology and Urology, Cleveland Clinic Foundation, and Department of Epidemiology and
Biostatistics, Case Western Reserve University, Cleveland, Ohio

J Urol 158:2197-2201, December 1997.

ABSTRACT
Purpose: We analyzed the outcome after radical prostatectomy of patients with familial
prostate cancer versus patients with sporadic prostate cancer.
Materials and Methods: The study included 720 patients with prostate carcinoma who were
treated with prostatectomy between 1987 and 1996. Patients were excluded from the study if
they had received adjuvant or neoadjuvant treatment, or had no available pretreatment prostatic
specific antigen (PSA) level, no available biopsy Gleason score, incomplete pathological informa-
tion or no available followup PSA levels. The analysis was performed on 529 cases. Patients were
considered to have a positive family history for prostate cancer when the index patient confirmed
the diagnosis of prostate cancer in a first degree relative (brother or father). The outcomes of
interest were biochemical relapse-free survival, local failure and distant metastases. Proper
tional hazards were used to analyze the effect of family history and confounding variables (that
is age, stage, biopsy Gleason score, initial PSA levels, surgical specimen Gleason score, extra-
capsular extension, lymph node metastasis, seminal vesicle invasion and surgical margin in-
volvement) on treatment outcome.

Results: Median followup was 30 months. Of all cases 12% had a positive family history.
Younger age was the only factor associated with positive family history, with 18% of pa
tients younger than 65 years having a positive family history versus 6% of older patients
(chi-square p <0.001). The 5-year biochemical relapse-free survival rate for the entire group was
64%. The 5-year biochemical relapse-free survival rates for patients with negative family history
versus positive history were 66% and 46%, respectively (p = 0.001). A multivariate time-to-
failure analysis using the proportional hazards model was performed based on family history, age
(less than 65 versus 65 to 69 versus 70 or greater, initial PSA (10 or less versus greater than 10),
biopsy Gleason score (6 or less versus 7 or greater), clinical T stage (T1-TBA versus TBB-C),
prostatectomy specimen Gleason score (6 or less versus 7 or greater), extracapsular extension,
seminal vesicle involvement, surgical margin involvement and lymph node involvement. After
adjusting for the potential confounding factors, positive family history remained strongly asso-
ciated with biochemical failure. The clinical failure rate for the entire group was 14%. The 5-year
local failure rate was 7%, with positive surgical margins being the only independent predictor of
local failure. The 5-year distant metastasis rate was 8%, with family history and initial PSA
levels being independent predictors of distant relapse.

Conclusions: Our study suggests that patients with a familial prostate cancer have a higher
likelihood of biochemical failure after radical prostatectomy than patients with sporadic cancer.
This effect is independent of pretreatment or pathological factors. Our results suggest that the
higher failure rates associated with familial prostate cancer are mainly secondary to higher
distant relapse rates, and that familial prostate cancer may be more biologically aggressive than
sporadic cancers.

J Urol 158:2197-2201, December 1997.