High frequency of allelic losses in high-grade prostate cancer is associated with biochemical progression after radical prostatectomy

• Published in: Urologic oncology 2005-03, Vol.23 (2), p.87-92
• Main Authors: Valeri, Antoine, Fromont, Gaelle, Sakr, Wael, Azzouzi, Rahmène, Dey, Jyotirmoy, Chantrel-Groussard, Karine, Latil, Alain, Berthon, Philippe, Cussenot, Olivier, Pontes, J. Edson, Cher, Michael L.
• Format: Article
• Language: English
• Subjects: Aged; Biological and medical sciences; Case-Control Studies; Disease Progression; Female; Genetics; Humans; LOH; Loss of Heterozygosity; Male; Medical sciences; Microsatellite Repeats; Middle Aged; Neoplasm Recurrence, Local; Nephrology. Urinary tract diseases; Prognosis; Prostate cancer; Prostate-Specific Antigen - blood; Prostatectomy; Prostatic Neoplasms - genetics; Prostatic Neoplasms - pathology; Prostatic Neoplasms - surgery; Treatment Outcome; Tumors; Tumors of the urinary system; Urinary tract. Prostate gland
• ISSN: 1078-1439; 1873-2496
• DOI: 10.1016/j.urolonc.2004.08.015

Loss of heterozygosity (LOH) is the most consistent genetic change in prostate cancer (CaP). We aimed, to correlate specific LOH and the overall LOH frequency, to disease progression after radical prostatectomy (RP) in high-grade CaP. Between January 1990 through December 1998, 126 patients who underwent RP (cT1-T2), Gleason 8–10, were pT3, or pN1, or SM(+) (surgical margins). Nine were lost of follow-up, 39/117 (33%) had no biochemical progression (mean follow-up: 45 months). After exclusion for preoperative PSA >50 ng/mL, a case-control study was designed by matching 26 of these cases with 26 similar patients without biochemical progression (criteria: pT, pN, year of surgery). Using microsatellite markers, LOH were assessed on six chromosomal regions (7q31, 8p22, 12p13, 13q14, 16q23.2, 18q21). No prognostic value was associated with LOH at any one specific locus. However, the overall LOH frequency (five classes, cutoff of 60%), was significantly higher if progression ( P = 0.02; P = 0.03) in SM(+) patients, and was near statistical significance ( P = 0.08; P = 0.07) for the overall case-control population. In multivariate analysis (overall population), the overall LOH rate ≥60% was independently associated with progression [ P = 0.035; Odds Ratio (OR) = 5.54]. An overall LOH rate ≥60% predicted poor outcome in 85% of SM(+) patients and 69% of the whole population. Our results suggest that the overall rate of LOH at chromosomal “hot spots” is more likely to be predictive of recurrence than the presence of LOH at any one particular locus. Moreover, the identification of a threshold of LOH could help in predicting patients with poor outcome who may be candidates for local or systemic adjuvant therapies.