S. Vijayakumar et al., LOCALIZED PROSTATE-CANCER - USE OF SERIAL PROSTATE-SPECIFIC ANTIGEN MEASUREMENTS DURING RADIATION-THERAPY - AN UPDATE, Onkologie, 17(6), 1994, pp. 578-585
Background: Serum prostate-specific antigen (PSA) has proven to be the
most useful tumor marker for prostate cancer. We previously reported
the decline rate of PSA levels in patients with cancer of the prostate
during radiotherapy; however, on the basis of a small number of patie
nts, no definite conclusions were drawn. Material and Methods: In this
report, we update our findings from a prospective study in which chan
ges in PSA levels during radiotherapy (RT) were determined. Pre-RT and
weekly PSA levels during 7-8 weeks of RT were measured in 76 stage A-
C patients with prostate cancer. Results: In a majority (78%) of patie
nts, the PSA levels declined. The mean rate of PSA decline was 1.3%/da
y or about 9%/week during RT. 13 patients received hormonal therapy pr
ior to RT; their mean rate of decline (1.8%/day) was not significantly
different from the average rate in 63 patients who had no hormonal tr
eatment (1.2%/day). 11 patients had one or more digital rectal examina
tions (DRE) during the RT course; the mean rate of PSA decline (0.5%/d
ay) was significantly less than the mean rate of decline in 65 patient
s who had no DRE (1.5%/day; p=0.033); however, the median rates of dec
line between these groups were not statistically different. There was
a significant (p=0.0004) transient increase in the PSA values 24 h aft
er the first RT fraction. No significant association was found between
rate of decline and age, stage, grade, or race in the group of patien
ts with no hormonal therapy. Tumor volume was calculated from prostate
volume and initial PSA level prior to RT; a highly significant correl
ation was found between tumor volume and the rate of decline (p<0.001)
. In a multivariate model which included age, stage, grade, race and t
umor volume, the latter remained significantly associated (p<0.001) wi
th the rate of PSA decline, A similar univariate and multivariate anal
ysis showed that PSA density also was associated with the rate of PSA
decline during RT (p<0.001). Conclusion: The rate of PSA decline durin
g RT is associated with the calculated tumor volume; PSA density is al
so similarly correlated with the rate of PSA decline. Our findings can
be interpreted to mean that prostate cancer cells may be more radiose
nsitive than non-malignant prostate cells. Whether the rate of PSA dec
line during RT can predict subsequent outcome is not known at this tim
e.