Androgen suppression, primarily castration, has
been the key objective of treatment of metastatic
Surgical castration, achieved by the use of bilateral orchiectomy,
produces a short-term symptomatic and objective tumor response in 70%
to 80% of patients. Medical castration, by the use of leuteinizing
hormone-releasing hormone (LHRH) agonists, produces an almost
equivalent effect. However, use of medical or surgical castration
eliminates only 90% to 95% of the daily testosterone production. The
remainder is produced in the adrenal glands.
In the 1980s, Labrie hypothesized that counteracting adrenal
androgens would further inhibit tumor growth. In response to this
hypothesis, a number of antiandrogen agents were identified and used
in combination with medical or surgical castration to obtain maximum
In a Southwest Oncology Group (SWOG) trial (INT 0036), Crawford and
colleagues reported a 24% improvement in survival for patients
receiving an antiandrogen (flutamide, Eulexin) and an LHRH agonist
(leuprolide, Lupron), compared with medical castration with
leuprolide alone. This early observation initiated interest and
further research into the efficacy of maximum androgen blockade in
advanced prostate cancer.
Interestingly, the current Intergroup report (INT 0105), unlike the
first, concludes that there is no clinically significant difference
between surgical castration alone or with an antiandrogen
(flutamide). The relative risk point estimate of 0.91 in the
current report favors orchiectomy plus flutamide, but the 90%
confidence limit includes 1.0, ranging from 0.81 to 1.01. Further, in
a parallel report on quality of life measures, orchiectomy plus
flutamide was associated with several reduced quality scores.
Given these new results, it is important to review the strength of
the evidence provided by the recently reported SWOG study and then
interpret the results in conjunction with other trials addressing the
same clinical questions.
The INT 0105 study was a well-designed, randomized, controlled trial
with special precautions to ensure blinded allocation. In terms of
baseline characteristics, the two arms of the trial differed only
with respect to bone pain: A higher proportion of patients in the
placebo group had bone pain. To the extent that bone pain reflects
disease severity, control patients (orchiectomy only) would be
expected to do more poorly.
The study was double blinded, and the ascertainment of outcomes was
performed relatively equally. Follow-up was complete with the
exception of three patients. Once disease progression occurred,
however, patients were allowed to cross over to treatment with
orchiectomy plus flutamide.
If delayed orchiectomy plus flutamide actually improves survival,
then control patients would be expected to have a later survival
advantage. This would possibly minimize differences between the
treatment arms. The treatments administered after disease progression
were not captured or reported, making it difficult to determine the
magnitude of crossover exposure.
Even though the relative risk in this study is not significant, the
wide confidence interval suggests that differences as large as 20%
between the two groups, or as small as no difference at all, are
possible. The median survival difference was 3.6 months, favoring the
orchiectomy plus flutamide arm, although again this was not
Life Expectancy a Better Measure?
It is important to note that the measures of outcome in this study
(median survival and relative risk) were taken at individual points
on the survival curve. Some argue that life expectancy, which was not
reported in this study, may provide a better measure of overall
differences in survival.
Using the Declining Exponential Approximation of Life Expectancy and
the reported mean survival for each treatment arm, life expectancy
for the orchiectomy plus flutamide arm is estimated to be 48.3 months
and for the orchiectomy alone arm, 43.1 months. Consequently, the
point estimate for life expectancy gained could be in the range of
This gain in life expectancy is within the same range of benefit
achieved by performing a prostatectomy on a 65-year-old man with
localized prostate cancer and is similar to the benefit with other
Quality of Life
The second publication from the recent SWOG trial focused on quality
of life differences between the two treatment arms at 1, 3, and 6
months. For this separate report, only half of the patients were
included. The sampling of patients was not clearly described.
The two arms in this subgroup of the study population differed at
baseline with respect to severity of disease. For those assigned to
receive orchiec-tomy plus flutamide, 82% had extensive disease vs 74%
assigned to orchiectomy alone.
To the extent that both disease severity and early toxicity from
treatment affect the patients reporting of quality of life, the
measures are likely to be lower in the intervention group. This, in
fact, was the case in this study.
Diarrhea and worse emotional functioning were statistically more
likely in the orchiectomy plus flutamide arm. At 3 months, the
absolute risk difference for diarrhea was 5.9% (8.6% vs 2.7%). Viewed
in a different manner, for every 17 patients treated with flutamide,
one will experience diarrhea. In practice, the diarrhea is easily
reversed by reducing the dose or stopping the drug, but, in the
trial, discontinuation of the drug was the only available option.
More important, to more fully assess the overall impact of a
treatment on patients overall health-related quality of life,
one should assess the short-term side effects as well as the
potential long-term clinical benefits of a treatment. By limiting
quality of life assessment to the first 6 months of treatment, this
study focuses on short-term side effects and fails to account for the
potential long-term clinical and quality of life benefits of flutamide.
In summary, the evidence from both the clinical trial and quality of
life reports of the recently reported SWOG study (INT 0105) suggests
that the survival differences between the orchiectomy plus flutamide
and orchiectomy alone groups are small. Furthermore, the evidence
suggests that quality of life may be worse in the first 6 months of
treatment in those receiving orchiectomy plus flutamide. In addition,
the overall clinical significance of the differences is unclear when
compared to the potential of longer term quality of life benefits
stemming from extended time to progression.
These results need to be considered alongside those reported in other
studies of maximum androgen blockade. A recent systematic review with
metaanalysis has integrated the SWOG trial into an overview
summary. This literature-based review by Bennett et al included
only trials that used flutamide with medical or surgical castration
in the maximum androgen blockade arm. Nine studies, involving 4,128
patients with advanced prostate cancer, were reviewed.
In this metaanalyis, overall survival was significantly better in
those receiving maximum androgen blockade than in those undergoing
castration alone. The relative risk was .90, an estimate that is
similar to estimates included in previous metaanalyses based on
smaller numbers of patients treated with flutamide as part of a
maximum androgen blockade regimen as well as to the results of the
recent SWOG report.[4,9,10] The 95% confidence interval was between
0.79 and 1.00.
If the relative risk point estimate of .90 in this overview is indeed
correct, then one must determine if this difference is likely to be
clinically significant. As stated previously, this point estimate is
associated with expected improvements in survival that, on average,
are comparable to other currently employed cancer interventions.
In 1999, fewer than one-fifth of metastatic prostate cancer patients
are expected to undergo a surgical castration procedure. Thus, the
role of flutamide in the treatment of advanced prostate cancer seems
less clear as an adjunct therapy to surgical castration than to
medical castration with an LHRH agonist, which is the more common
castration therapy in the current era.
In addition, the most recent meta-analysis, as well as the
preliminary results of the updated Prostate Cancer Trialists
Collaborative Group, from Oxford University, in 1997, suggests that,
overall, there appears to be a small survival difference that favors
maximum androgen blockade. In any case, more trials (and even more
metaanalyses) are unlikely to further refine the relative risk point
estimate associated with maximum androgen blockade.
1. Resnick MI, Grayhack JT: Treatment of stage IV carcinoma of the
prostate. Urol Clin North Am 2:141-161, 1975.
2. Labrie F, Dupont A, Belanger A, et al: New hormonal therapy in
prostatic carcinoma: Combined treatment with an LHRH agonist and an
antiandrogen. Clin Invest Med 5:267-275, 1982.
3. Crawford ED, Eisenberger MA, McLeod DG, et al: A controlled trial
of leuprolide with and without flutamide in prostatic carcinoma. N
Engl J Med 321:419-424, 1989.
4. Eisenberger MA, Blumenstein BA, Crawford ED, et al: Bilateral
orchiectomy with or without flutamide for metastatic prostate cancer.
N Engl J Med 339:1036-1042, 1998.
5. Wright JC, Weinstein MC: Gains in life expectancy from medical
interventions-standardizing data on outcomes. N Engl J Med
6. Sox Jr. HC, Blatt MA, Higgins MC, et al: Medical Decision Making,
pp 186-191. Boston, Butterworths, 1998.
7. Moinpour CM, Savage MJ, Troxel A, et al: Quality of life in
advanced prostate cancer: Results of a randomized therapeutic trial.
J Natl Cancer Inst 90:1537-1544, 1998.
8. Bennett CL, Tosteson TD, Schmitt BP, et al: Maximum
androgen-blockade with medical or surgical castration in advanced
prostate cancer: A meta-analysis of 9 published randomized controlled
trials and 4,128 patients using flutamide. Prostate Cancer and
Prostatic Diseases 2(1):4-8, 1999.
9. Caubet JF, Torteson TD, Dong EW, et al: Maximum androgen blockade
in advanced prostate cancer: A meta-analysis of published randomized
controlled trials using nonsteroidal antiandrogens. Urology 49:71-78, 1997.
10. Prostate Cancer Trialists Collaborative Group: Maximum
androgen blockade in advanced prostate cancer: An overview of 22
randomised trials with 3283 deaths in 5710 patients. Lancet