|
Immediate Versus Delayed Radiation
Therapy For Patients At High Risk For Failure After Radical
Prostatectomy - A comparison of outcome for two options in this
management dilemma (SWOG Trial 8794)
(May 2008)
In their review of this
issue, Kibel and Nelson in Prostate Cancer and Prostatic Diseases
(2007)10, succinctly cited the current state of affairs:
"...recent randomized
trials have demonstrated a biochemical advantage to adjuvant radiation
therapy, but it remains to be seen if this will translate to an
improvement in survival end points or if salvage radiation would be just
as effective." A plethora of articles address and confirm this analysis,
with nearly all, however, stopping the comparison at the point of PSA
relapse. As the results of the SWOG trial (below) document, following
PSA relapse after primary therapy much of the long and unpredictable
course of prostate cancer still remains to be revealed.
SWOG Trial 8794 studied
425 men with pT3 NO MO disease, reported by Ian Thompson et al. in JAMA
November 15, 2006, provides illuminating data by virtue of a uniquely
long follow-up of 10.6 years (interquartile range, 9.2-12.7 years). The
study found no significant difference in overall survival
(including death from any cause) between immediate adjuvant radiation
with 60-64 Gy (a bit less than current doses of 64-70 Gy; median dose 65
Gy) and delayed therapy in patients with locally advanced prostate
cancer. The median overall survival for the immediate therapy group was
14.7 years and for delayed, 13.8 years (P=.16). As expected, the median
PSA relapse-free survival (relapse point set at PSA >0.4 ng/mL) was an
impressive 10.3 years for immediate vs. 3.1 years for delayed therapy.
The primary study
endpoint was "metastases-free survival, defined as time to first
occurrence of metastatic disease or death due to any cause." These two
end points were chosen because of their special clinical importance. For
the immediate therapy group the median metastatic-free estimate was 14.7
years vs. 13.2 years for the delayed XRT group, just short of
significance at P=.06. This lengthy interval to metastases for the
delayed group surprised the study planners, who at the outset had "the
assumption that the primary end point, metastases free survival, would
be 6 years if radiation were delayed and that immediate treatment would
decrease the metastases-free survival hazard by one third."
Seventy of the 211 men
on the observation arm ultimately "crossed over" and received radiation,
and for the 65 men whose PSA values were know at the time of initiation
of XRT, 55.4% started treatment at the point of PSA relapse (>0.4 ng/mL),
and 41.5% at objective recurrence. The protocol schema did not require
an undetectable PSA post surgery for men to be study eligible. In the
adjuvant group 65% had a post surgery PSA of less than 0.2 ng/mL, hardly
different from the 68% in the observation group with the same PSA value.
The crossover to XRT from observation status took place at between 3
days and 9.7 years after randomization (median 2 years, interquartile
range 11 months - 4.5 years).
The protocol did
not specify the PSA value at which hormonal therapy should be initiated,
but "among the observation group, 21% had received hormone therapy by 5
years, compared with 10% among those of the radiotherapy group."
One take away point
that emerges from this study focuses on the endpoint of PSA relapse
ubiquitously used in so many comparison studies. The SWOG study raises
questions about the reliability of time-to-PSA failure as a relevant
surrogate for treatment effectiveness when the important endpoints of
metastases free survival and life duration are more clinically
meaningful concerns. A questioning of the significance of time-to-PSA
relapse has also arisen in chemotherapy trials where the
time-to-objective disease recurrence seems to be a more appropriate
comparison point between regimens.
The SWOG 8794 data
established that the early radiation therapy clearly delayed the
initiation of hormonal intervention therapy, which conventionally
utilizes Lupron with its known toxicity and degradation of quality of
life. However, the development of substantially less toxic alternatives
to LHRH agonists such as (say) dutasteride/bicalutamide may
significantly replace this classic Lupron paradigm and erode the
importance of this delay of hormone intervention.
A second observation
from SWOG 8794 - which again was unexpected by the authors - led to
their conclusion that "The pattern of treatment failure in high-risk
patients is predominantly local with a surprisingly low incidence
of metastatic failure" (Swanson, JCO, June 2007). The data showed that
at a median follow-up of 10.2 years for the patients in both arms who
had a post-surgical PSA of <0.2ng/mL reductions in
PSA failure, local failure, and distant failure for the adjuvant group
vs. observation were 72% v. 42%, 20% v. 7%, and 12% v. 4%, respectively.
For the two study groups in which the post-surgical PSA values were
between >0.2 and 10 ng/mL the 10-year risk reduction
figures in the three categories were 80% v. 73%, 25% v. 9%, and 16% v.
12%, again respectively. This led to the authors to conclude that
"further improvement in reducing local treatment failure is likely to
have the greatest impact on outcome in high-risk patients after
prostatectomy."
Several generalizations
emerge from the data of SWOG 8794: 1), the interval of time to PSA
failure for prostate cancer therapy is an unreliable predictor of
metastases-free and overall survival; 2) the predominant treatment
failure after radical prostatectomy is local; and 3), if prolongation
of freedom from PSA failure is a goal, then in surgically treated
men with advanced prostate cancer a high priority should be given to
treatment of the post surgical prostatic bed with immediate external
beam radiotherapy or [I will add] radiation administered at a low PSA
level (see below).
[SWOG 8794 was not
designed to explore the finer point of defining a potential window of
opportunity for the serviceable application of radiotherapy following
surgery for men at high risk for recurrence, i.e. XRT given in the very
low ranges of PSA after relapse, (say) at less than 1 ng/mL or at least
between 1 and 2 ng/mL. For an analysis of this issue see the articles in
the PCa Commentary of November 2007, July 2007, and especially September
2004 indexed under "Adjuvant and Salvage Treatment."]
«
Back to Article List |
