Dj. Grossklaus et al., Ratio of free-to-total prostate specific antigen correlates with tumor volume in patients with increased prostate specific antigen, J UROL, 165(2), 2001, pp. 455-458
Purpose: We evaluated the relationship between the ratio of free-to-total p
rostate specific antigen (PSA) and prostate pathology, including grade, sta
ge and tumor volume, among patients with prostate cancer who underwent radi
cal prostatectomy.
Materials and Methods: We prospectively analyzed 54 consecutive patients wi
th prostate cancer who underwent radical prostatectomy and in whom frozen s
erum was available for assessment of free-to-total PSA ratio. Pathological
review was done with whole mount sections, and total tumor volume was deter
mined by planimetry. Comparison between free-to-total PSA ratio and patholo
gical parameters was performed using the Pearson correlation coefficient.
Results: Among the 54 patients mean total and free-to-total PSA ratio were
5.81 and 14.2 ng./ml., respectively, and free-to-total PSA ratio directly c
orrelated with prostate volume (p = 0.037), and inversely correlated with G
leason score (p = 0.012) and extracapsular disease (p = 0.0074). Furthermor
e, there was a significant relationship between free-to-total PSA ratio and
pathological stage pT2a/b in 39 cases versus pT3a/b in 15 (p = 0.005). Ove
rall, there was no correlation between free-to-total PSA ratio and tumor vo
lume. However, among 37 patients with an increased PSA, defined as greater
than 4.0 ng./ml., a significant inverse relationship between free-to-total
PSA ratio and tumor volume was identified (p = 0.01). Among this subset the
re was only a weak correlation with prostate volume (p = 0.049).
Conclusions: Our findings suggest that free-to-total PSA ratio may be predi
ctive of tumor biology among those patients with a total PSA of greater tha
n 4 ng./ml. as evidenced by good correlation with tumor grade and volume. T
his finding appears to be independent of prostate volume. These preliminary
results suggest the need for additional studies among patients with an inc
reased PSA designed to evaluate the potential role of free-to-total PSA rat
io in combination with traditional clinical variables in the prediction of
prostate cancer pathology.