Vous êtes sur la page 1sur 7

Int. J. Radiation Oncology Biol. Phys., Vol. 58, No. 4, pp.

1056 –1062, 2004


Copyright © 2004 Elsevier Inc.
Printed in the USA. All rights reserved
0360-3016/04/$–see front matter

doi:10.1016/j.ijrobp.2003.08.021

CLINICAL INVESTIGATION Prostate

EFFECT OF CIGARETTE SMOKING ON BIOCHEMICAL OUTCOME AFTER


PERMANENT PROSTATE BRACHYTHERAPY

GREGORY S. MERRICK, M.D.,*† WAYNE M. BUTLER, PH.D.,*† KENT E. WALLNER, M.D.,‡


ROBERT W. GALBREATH, PH.D.,*† JONATHAN H. LIEF, PH.D.,*† AND EDWARD ADAMOVICH, M.D.§
*Schiffler Cancer Center and §Department of Pathology, Wheeling Hospital; †Wheeling Jesuit University, Wheeling, WV; ‡Puget
Sound Health Care System, Department of Veterans Affairs, Seattle, WA

Purpose: Recent studies have suggested that cigarette smoking may be associated with an increased risk of death
from prostate cancer. In this study, we evaluated the effect of cigarette smoking on the presentation and
biochemical outcome after permanent prostate brachytherapy for prostate cancer.
Methods and Materials: A total of 582 patients underwent brachytherapy with generous periprostatic margins
using either 103Pd or 125I with or without supplemental external beam radiotherapy between April 1995 and
September 2000. Of the 582 patients, 178 (30.6%) had never smoked, 306 (52.6%) were former smokers, and 98
(16.8%) were current smokers. The median patient age was 67.9 years, and the median follow-up was 54.5
months. No patient was lost to follow-up. No patient underwent routine seminal vesicle biopsy or pathologic
lymph node staging. The clinical, treatment, and dosimetric parameters evaluated included tobacco status, age,
clinical stage, Gleason score, pretreatment prostate-specific antigen level, risk group, percentage of positive
biopsies, ultrasound volume, isotope used, planning volume, hormonal status, use of external beam radiotherapy,
and postimplant dosimetry (percentage of target volume receiving 100%, 150%, and 200% of the prescribed dose
and percentage of prescribed dose covering 90% of the target volume). Biochemical outcome was determined
using the American Society for Therapeutic Radiology and Oncology consensus definition.
Results: No differences in the clinical, treatment, or dosimetric parameters were identified, except that current
smokers were statistically younger than those who had never smoked or former smokers (65.9 vs. 67.8 vs. 68.3
years, respectively, p ⴝ 0.016). Specifically, no relationship was discerned between tobacco history and risk
group, supplemental external beam radiotherapy, choice of isotope, or use of hormonal therapy. The overall
biochemical freedom from progression survival rate at 7 years was 96.2%, 95.6%, and 91.6% for patients who
had never smoked, former smokers, and current smokers, respectively (p ⴝ 0.126). When stratified by risk group
and hormonal status, tobacco consumption did not predict outcome, although a trend for poorer biochemical
progression-free survival was noted in current smokers. The median prostate-specific antigen level for hormone-
naive and hormonally manipulated disease-free patients was <0.1 ng/mL. In multivariate Cox regression
analysis, Gleason score, pretreatment prostate-specific antigen level, risk group, and hormonal status were
predictors of biochemical outcome.
Conclusion: In this prostate brachytherapy cohort, tobacco consumption did not predict for risk group strati-
fication or treatment approach. Although no statistically significant difference was found in biochemical
progression-free survival, a trend for poorer biochemical outcome was demonstrated in current smokers.
© 2004 Elsevier Inc.

Prostate, Brachytherapy, Tobacco, Smoking, Survival.

INTRODUCTION Daniell (4) reported that cigarette smokers were more


likely to present with metastatic prostate cancer than non-
Although recent studies have suggested that environmental smokers, with a 5-year cancer-specific mortality rate of 88%
factors are likely responsible for the promotion of prostate vs. 63% in favor of nonsmokers. In addition, a review of the
cancer from a latent state to a clinically evident disease, the three largest prospective prostate cancer mortality studies
potential causative relationship between tobacco use and demonstrated a modest association between current ciga-
prostate cancer remains unclear (1). Despite the lack of a rette smoking and early death from prostate cancer (5).
clear causal relationship (2), multiple studies have reported Recently, Roberts and colleagues (1) reported that smokers
a correlation between cigarette smoking and aggressive ⬍55 years old who underwent radical prostatectomy had a
prostate cancer and/or prostate cancer-related death (1– 6). greater risk of extracapsular extension or Gleason score 7 or

Reprint requests to: Gregory S. Merrick, M.D., Schiffler Cancer SchifflerCancerCenter@wheelinghospital.com


Center, Wheeling Hospital, 1 Medical Park, Wheeling, WV Received Jun 6, 2003, and in revised form Aug 15, 2003.
26003-6300. Tel: (304) 243-3490; Fax: (304) 243-5047; E-mail: Accepted for publication Aug 19, 2003.

1056
Cigarette smoking and prostate cancer ● G. S. MERRICK et al. 1057

Table 1. Clinical treatment and dosimetric values for patient population stratified by tobacco smoking status

Never smoked Former smoker Current smoker


(n ⫽ 178) (n ⫽ 306) (n ⫽ 98) Overall (n ⫽ 582)

Variable Median Mean ⫾ SD Median Mean ⫾ SD Median Mean ⫾ SD P* Median Mean ⫾ SD

Age at implant (y) 67.8 67.3 ⫾ 6.6 68.3 67.2 ⫾ 6.9 65.9 65.1 ⫾ 6.9 †
0.016 ‡
67.9 66.9 ⫾ 6.9
Follow-up (mo) 55.6 58.5 ⫾ 17.8 53.9 55.6 ⫾ 18.1 52.4 54.3 ⫾ 21.3 0.130 54.5 54.3 ⫾ 18.6
Pretreatment PSA
(ng/ml) 7.1 8.7 ⫾ 5.2 7.4 9.3 ⫾ 6.6 7.6 9.6 ⫾ 6.9 0.442 7.4 9.1 ⫾ 6.3
Gleason score 7.0 6.6 ⫾ 1.0 7.0 6.6 ⫾ 1.0 6.5 6.4 ⫾ 0.9 0.357 7.0 6.6 ⫾ 1.0
Clinical stage§ T2a 3.7 ⫾ 0.8 T2a 3.7 ⫾ 0.8 T2a 3.7 ⫾ 0.8 0.977 T2a 3.7 ⫾ 0.8
Prostate volume
(cm3) 36.3 37.3 ⫾ 10.0 35.0 35.7 ⫾ 9.6 34.9 34.6 ⫾ 9.8 0.066 35.4 36.0 ⫾ 9.8
Planning volume
(cm3) 63.1 62.2 ⫾ 13.9 59.8 60.5 ⫾ 12.5 59.2 60.2 ⫾ 13.2 0.389 60.6 60.9 ⫾ 13.0
V100 95.3 93.5 ⫾ 5.7 95.5 94.1 ⫾ 5.4 94.9 93.7 ⫾ 4.2 0.750 95.1 93.8 ⫾ 5.3
V150 54.7 55.3 ⫾ 11.1 52.6 54.8 ⫾ 12.9 50.4 52.1 ⫾ 9.6 0.382 53.0 54.5 ⫾ 11.9
V200 26.2 26.9 ⫾ 7.4 25.8 27.4 ⫾ 9.6 23.2 24.9 ⫾ 6.0 0.310 25.8 26.8 ⫾ 8.5
D90 (% mPD) 111 111 ⫾ 11 109 112 ⫾ 14 107 108 ⫾ 9.0 0.290 109 111 ⫾ 12
% positive biopsies 33.3 41.2 ⫾ 24.9 33.3 40.7 ⫾ 24.7 33.3 42.5 ⫾ 26.4 0.825 33.3 41.1 ⫾ 25.0
Hormone use (mo) 0.0 2.5 ⫾ 3.9 0.0 3.0 ⫾ 6.5 0.0 2.8 ⫾ 5.9 0.694 0.0 2.8 ⫾ 5.7
Risk group (n)
Low 70 (39.3) 111 (36.3) 38 (38.8) 0.936㛳 219 (37.6)
Intermediate 71 (39.9) 122 (39.9) 38 (38.8) 231 (39.7)
High 37 (20.8) 73 (23.9) 22 (22.4) 132 (22.7)
Isotope (n)
125
I 72 (59.6) 119 (61.1) 42 (57.1) 0.777㛳 233 (60.0)
103
Pd 106 (40.4) 187 (38.9) 56 (42.9) 349 (40.0)
Hormones (n)
None 91 (53.5) 168 (58.9) 51 (57.3) 0.830㛳 310 (57.0)
ⱕ6 mo 61 (35.9) 90 (31.6) 28 (31.5) 179 (32.9)
⬎6 mo 18 (10.6) 27 (9.5) 10 (11.2) 55 (10.1)
EBRT (n) 97 (54.5) 176 (57.5) 49 (50.0) 0.413㛳 322 (55.3)

Abbreviations: PSA ⫽ prostate-specific antigen; V100/150/200 ⫽ percentage of target volume receiving 100%, 150%, 200% of prescribed
dose; D90 ⫽ percentage of prescribed dose covering 90% of target volume; MPD ⫽ minimal peripheral dose; EBRT ⫽ external beam
radiotherapy
Data in parentheses are percentages.
* Determined using analysis of variance.

Significantly younger than the never and former smokers.

Statistically significant.
§
Mean clinical stage determined by scaling T1a–T3c from 1 to 9.

Determined using chi-square.

greater histologic features and extraprostatic extension than tatic margins using either 103Pd or 125I with or without
did nonsmokers, with former smokers possessing an inter- supplemental EBRT between April 1995 and September
mediate risk. 2000 for clinical Stage T1b-T3a (1997 American Joint
After permanent prostate brachytherapy, most studies Committee on Cancer) prostate cancer. The mean and me-
have reported long-term biochemical outcomes that com- dian follow-up was 54.3 ⫾ 18.6 months and 54.5 months,
pare favorably with radical prostatectomy and external respectively (range, 30 –97 months). No patient was lost to
beam radiotherapy (EBRT), with low rates of urinary in- follow-up. Patients underwent clinical staging by medical
continence and rectal injury and acceptable rates of sexual history, physical examination, including digital rectal exam-
dysfunction (7, 8). Cigarette smoking has not been evalu- ination, and serum prostate-specific antigen (PSA) measure-
ated as a risk factor for biochemical progression after ment. Bone scans, CT of the pelvis, and prostatic acid
brachytherapy. Accordingly, we evaluated the effect of cig- phosphatase levels were obtained as clinically indicated. No
arette smoking on the presentation and biochemical out- patient underwent seminal vesicle biopsy or pathologic
come after permanent prostate brachytherapy. lymph node staging.
Because of well-documented inaccuracies in Gleason
grading, all cases originating from outside institutions were
METHODS AND MATERIALS
reviewed by a single pathologist (E.A.) with a special in-
A total of 582 consecutive patients at the Schiffler Cancer terest in urologic malignancies (9). The calculation algo-
Center underwent brachytherapy with generous peripros- rithms and seed parameters used in preplanning and post-
1058 I. J. Radiation Oncology ● Biology ● Physics Volume 58, Number 4, 2004

Table 2. Most recent PSA level in patients free of biochemical progression, stratified by risk and months of hormone use and
smoking status

Never smoked Former smoker Current smoker


(n ⫽ 172*) (n ⫽ 293*) (n ⫽ 90*) Overall (n ⫽ 555*)

Variable Median Mean ⫾ SD Median Mean ⫾ SD Median Mean ⫾ SD Median Mean ⫾ SD

Low risk
No hormones ⬍0.1 0.18 ⫾ 0.37 ⬍0.1 0.11 ⫾ 0.20 0.06 0.20 ⫾ 0.36 ⬍0.1 0.15 ⫾ 0.29
Hormones ⱕ6 mo ⬍0.1 0.12 ⫾ 0.22 ⬍0.1 0.08 ⫾ 0.11 ⬍0.1 0.09 ⫾ 0.07 ⬍0.1 0.10 ⫾ 0.16
Hormones ⬎6 mo 0.20 0.17 ⫾ 0.15 ⬍0.1 0.01 ⫾ 0.00 ⬍0.1 0.16 ⫾ 0.30 ⬍0.1 0.11 ⫾ 0.14
Intermediate risk
No hormones ⬍0.1 0.08 ⫾ 0.13 ⬍0.1 0.12 ⫾ 0.20 ⬍0.1 0.13 ⫾ 0.29 ⬍0.1 0.11 ⫾ 0.20
Hormones ⱕ6 mo ⬍0.1 0.08 ⫾ 0.09 ⬍0.1 0.05 ⫾ 0.09 ⬍0.1 0.04 ⫾ 0.08 ⬍0.1 0.05 ⫾ 0.09
Hormones ⬎6 mo ⬍0.1 0.01 ⫾ 0.00 ⬍0.1 0.01 ⫾ 0.01 ⬍0.1 0.01 ⫾ 0.00 ⬍0.1 0.01 ⫾ 0.00
High risk
No hormones ⬍0.1 0.06 ⫾ 0.16 ⬍0.1 0.10 ⫾ 0.22 0.10 0.11 ⫾ 0.20 ⬍0.1 0.09 ⫾ 0.19
Hormones ⱕ6 mo ⬍0.1 0.02 ⫾ 0.03 ⬍0.1 0.04 ⫾ 0.09 0.30† 0.30 ⫾ 0.00 ⬍0.1 0.05 ⫾ 0.09
Hormones ⬎6 mo ⬍0.1 0.05 ⫾ 0.13 ⬍0.1 0.02 ⫾ 0.04 0.10 0.03 ⫾ 0.04 ⬍0.1 0.03 ⫾ 0.06
All risks
No hormones ⬍0.1 0.11 ⫾ 0.25 ⬍0.1 0.11 ⫾ 0.20 ⬍0.1 0.16 ⫾ 0.31 ⬍0.1 0.12 ⫾ 0.24
Hormones ⱕ6 mo ⬍0.1 0.09 ⫾ 0.17 ⬍0.1 0.06 ⫾ 0.10 ⬍0.1 0.07 ⫾ 0.09 ⬍0.1 0.07 ⫾ 0.13
Hormones ⬎6 mo ⬍0.1 0.06 ⫾ 0.10 ⬍0.1 0.02 ⫾ 0.04 ⬍0.1 0.02 ⫾ 0.03 ⬍0.1 0.03 ⫾ 0.07

Abbreviation: PSA ⫽ prostate-specific antigen.


Main effects: Tobacco use, p ⫽ 0.401; risk p ⫽ 0.364; hormones, p ⫽ 0.080.
* Not including patients with failure (overall failures, n ⫽ 27).

n ⫽ 1.

operative dosimetry were those recommended by the hormone-releasing hormone agonist and an antiandrogen.
American Association of Physicists in Medicine Task The mean and median duration of hormonal manipulation in
Group No. 43 (10). The monotherapeutic brachytherapy this subgroup was 7.0 ⫾ 7.2 and 4 months, respectively
dose was 125 Gy minimal peripheral dose for 103Pd (78 (range, 1–36 months). For all 582 patients, the mean and
patients) and 145 Gy for 125I (182 patients). The brachy- median duration of hormonal therapy was 2.8 ⫾ 5.7 and
therapy boost dose was 90 Gy for 103Pd (271 patients) and 0.0 months, respectively.
110 Gy for 125I (51 patients). Patients were monitored by physical examination, includ-
Our patient selection, preplanning, and intraoperative ing digital rectal examination and PSA determination at
philosophy and dosimetric evaluation method have been 3– 6-month intervals. The end point of the analysis was
previously described (11–16). The brachytherapy planning biochemical progression-free survival as defined by the
target volume consisted of a 3– 8-mm enlargement of the American Society of Therapeutic Radiology and Oncology
prostate gland on each ultrasound slice, with a resultant consensus definition (17). Although in this series, all pa-
planning volume approximately 1.75 times the ultrasound- tients with treatment failure presented with PSA progres-
determined prostate volume (13, 14). The minimal periph- sion, abnormal digital rectal examination findings or the
eral dose was prescribed to the planning target volume with development of distant metastases in the absence of PSA
margin. At the time of implantation, the prostate gland, progression also would have been scored as a treatment
periprostatic region, and base of the seminal vesicles were failure. No patient underwent routine postimplant prostate
implanted. On average, approximately 40% of the seeds biopsy.
were placed in periprostatic locations (11). The brachyther- The clinical, treatment, and dosimetric parameters eval-
apy procedure was performed with preloaded 18-gauge nee- uated for biochemical progression-free survival included
dles using transverse and sagittal ultrasonography, along cigarette consumption (never, former, or current), age, clin-
with fluoroscopy. ical T stage, Gleason score, pretreatment PSA level, isotope,
Of the 582 patients, 322 received supplemental EBRT hormonal status, supplemental EBRT, risk group (low risk,
before brachytherapy. In general, EBRT consisted of 45 Gy Gleason score 6 or less, PSA level ⱕ10 ng/mL, clinical
to the prostate, periprostatic region, seminal vesicles, and Stage T1c-T2a; intermediate risk, Gleason score 7 or greater
first echelon lymph nodes in 1.8-Gy fractions using 15– or PSA ⬍10 ng/mL or clinical Stage T2b or greater; and
18-MV photons delivered by a multifield technique with high risk, two or three of the intermediate-risk criteria),
shielding of the posterior half of the rectum via the lateral percentage of positive biopsies, prostate volume, percentage
portals. Of the 582 patients, 234 (40.2%) received hormonal of the target volume receiving 100%, 150%, and 200% of
manipulation for either cytoreduction or multiple poor prog- the prescribed dose, and the percentage of the prescribed
nosticators. Hormonal therapy consisted of a luteinizing dose covering 90% of the target volume.
Cigarette smoking and prostate cancer ● G. S. MERRICK et al. 1059

Fig. 1. Clinical and biochemical outcomes for 582 study patients.

Comparisons of the continuous variables across the three among the three levels of smoking. Univariate Cox regression
levels of smoking were made using analysis of variance. The analysis was used to determine the univariate predictors of
three levels of risk were compared across the three levels of biochemical disease-free survival, and all variables with p
smoking using two-way analysis of variance. A Tukey posthoc ⱕ0.10 were included as covariates in a multivariate Cox re-
analysis was made when significance was found to cross the gression model to identify independent predictors of biochem-
levels of a given factor. Associations between categorical vari- ical disease-free survival. For all tests, p ⱕ0.05 was considered
ables and levels of smoking were tested with Pearson’s chi- to be statistically significant. Statistical analysis was performed
square test. Survival curves were analyzed using the Kaplan- with Statistical Package for Social Sciences, version 11.0,
Meier method, and the log–rank test was used for comparison software (SPSS, Chicago, IL).
1060 I. J. Radiation Oncology ● Biology ● Physics Volume 58, Number 4, 2004

Fig. 2. Seven-year overall biochemical progression-free (bNED) Fig. 3. Seven-year biochemical progression-free (bNED) survival
survival stratified by tobacco status. Symbols indicate follow-up for hormone-naı̈ve, low-risk patients stratified by tobacco status.
time for censored patients. Open circles indicate never smoked; Symbols indicate follow-up time for censored patients. Open cir-
open triangles, former smokers; plus signs, current smokers. cles indicate never smoked; open triangles, former smokers; plus
signs, current smokers.

RESULTS
manipulated patients were stratified by risk group, no sta-
Table 1 summarizes the mean clinical, treatment, and tistically significant difference in biochemical outcome was
dosimetric variables of the patients stratified by tobacco demonstrated among the three tobacco cohorts.
status. The mean and median follow-up for the entire cohort In multivariate Cox regression analysis, pretreatment
was 54.3 ⫾ 18.6 and 54.5 months, respectively. The three PSA level, Gleason score, hormonal status, and risk group
patient cohorts were well-matched, except that current predicted for biochemical failure (Table 3). Additional clin-
smokers were statistically younger (p ⫽ 0.016), with a trend ical, treatment, and dosimetric parameters, including to-
for smaller prostate glands. Tobacco status did not predict bacco status, failed to predict for biochemical outcome.
for pretreatment PSA level, Gleason score, clinical stage,
risk group, percentage of positive biopsies, or the addition
DISCUSSION
of supplemental EBRT or hormonal therapy.
Table 2 displays the most recent posttreatment PSA level In the current brachytherapy study, tobacco consumption
in patients free of biochemical progression for the three did not predict a more aggressive presentation, as deter-
tobacco cohorts stratified by risk group. In the biochemi-
cally disease-free patients, neither tobacco status nor risk
group predicted for the absolute posttreatment PSA level,
and a trend for lower posttreatment PSA levels was noted in
patients who had received hormonal therapy (p ⫽ 0.080).
Figure 1 diagrams the clinical and biochemical outcomes
for all 582 patients. Tobacco did not influence the overall
prostate cancer death rate; however, current and former
smokers were more likely to die of other causes than those
who had never smoked (p ⫽ 0.006).
Figure 2 displays the freedom from biochemical progres-
sion for all three tobacco cohorts. A trend was noted for
poorer outcome in current smokers vs. those who had never
smoked or former smokers (p ⫽ 0.126). Patients who had
never smoked and former smokers had nearly equivalent
biochemical outcomes (96.2% vs. 95.6%, respectively).
Figures 3–5 illustrate the freedom from biochemical pro-
gression in hormone-naive low-risk, intermediate-risk, and
high-risk patients. In the low-risk and high-risk, but not Fig. 4. Seven-year biochemical progression-free (bNED) survival
for hormone-naı̈ve, intermediate-risk patients stratified by tobacco
intermediate-risk groups, those who had never smoked and status. Symbols indicate follow-up time for censored patients.
former smokers had nearly identical outcomes, with a trend Open circles indicate never smoked; open triangles, former smok-
for poorer outcomes in current smokers. When hormonally ers; plus signs, current smokers.
Cigarette smoking and prostate cancer ● G. S. MERRICK et al. 1061

gression was evident in current smokers compared with


those who had never smoked or were former smokers, the
trend did not reach statistical significance in univariate
analysis. In addition, tobacco consumption was not associ-
ated with an increased death rate from prostate cancer.
However, current and former smokers were more likely to
die of other causes compared with nonsmokers (Fig. 1).
However, because a trend for increased biochemical failure
was related to tobacco consumption, it is possible that with
longer follow-up of more patients, tobacco consumption
would result in an increased incidence of prostate cancer-
related death. As such, our current results are not necessar-
ily contradictory to the previously published literature. A
review (5) of the three largest prospective prostate cancer
mortality studies demonstrated a modest association be-
tween current cigarette smoking and early death from pros-
tate cancer, and Daniell (4) reported that cigarette smokers
Fig. 5. Seven-year biochemical progression-free (bNED) survival
were more likely to present with metastatic prostate cancer
for hormone-naı̈ve, high-risk patients stratified by tobacco status.
Symbols indicate follow-up time for censored patients. Open cir- with a decreased 5-year cancer-specific mortality rate.
cles indicate never smoked; open triangles, former smokers; plus Using patient-administered quality-of-life instruments, to-
signs, current smokers. bacco consumption has been correlated with deleterious effects
on brachytherapy-related late urinary and bowel function (18,
19). Using the Expanded Prostate Cancer Index, tobacco con-
mined by pretreatment PSA level, Gleason score, clinical sumption was the strongest predictor of late urinary function,
stage, percentage of positive biopsies, or risk group. Al- with statistically significant differences in all urinary-specific
though a trend for poorer biochemical freedom from pro- domains, including function, incontinence, irritation/obstruc-

Table 3. Univariate and multivariate Cox regression analysis for biochemical failure

Univariate Cox regression Multivariate Cox regression

Variable RR p* RR p

Gleason score 2.12 ⬍0.001* 2.45 0.001*


PSA 1.06 0.001* 1.05 0.029*
Hormone use 0.079* 0.006*
None vs. ⱕ6 mo 0.26 0.029* 0.36 0.101
None vs. ⬎6 mo 0.57 0.450 0.08 0.005*
ⱕ6 mo vs. ⬎6 mo 2.20 0.388 0.22 0.158
Risk ⬍0.001* 0.028*
Low vs. intermediate 1.17 0.814 0.36 0.193
Low vs. high 7.77 ⬍0.001* 1.53 0.575
Intermediate vs. high 6.62 ⬍0.001* 4.26 0.008*
Percent positive biopsies 1.03 ⬍0.001* 0.101
Stage (T1a-T2a vs. T2b-T3c) 3.38 0.003* 0.675
EBRT (no vs. yes) 3.58 0.010* 0.787
Prostate volume 1.03 0.100
Tobacco use 0.140
Never vs. former 1.31 0.586
Never vs. current 2.69 0.067
Former vs. current 2.05 0.110
Isotope (125I vs. 103Pd) 1.66 0.231
Hypertension (no vs. yes) 0.55 0.152
Diabetes (no vs. yes) 1.20 0.808
Age 1.04 0.237
V100 0.02 0.303
V150 0.01 0.208
V200 0.01 0.348
%D90 2.92 0.653

Abbreviations: EBRT ⫽ external beam radiotherapy; PSA ⫽ prostate specific antigen; RR ⫽ relative risk.
* Only those variables with p ⬍ 0.10 in univariate Cox regression analysis included in multivariate Cox regression analysis.
1062 I. J. Radiation Oncology ● Biology ● Physics Volume 58, Number 4, 2004

tion, and bother scores (18). The International Prostate Symp- the observation that the development of benign prostatic hy-
tom Scores were also significantly greater in current smokers. perplasia is inhibited by cigarette smoking (20).
An evaluation of late bowel function using the Rectal Function The data reported here are not necessarily in contradiction
Assessment Score demonstrated that the number of preimplant with the findings of increased extraprostatic extension in smok-
bowel movements, tobacco status, and median rectal dose ers (1). Mayo and Cleveland Clinic studies have shown that the
correlated with late bowel function (19). vast majority of extraprostatic extension is limited to within 5
A potential shortcoming of our study was that the pack- mm of the prostatic capsule (21, 22). Our implant philosophy,
year history was not available. Rodriguez and colleagues which includes the use of multiple periprostatic seeds with or
(5), however, reported that “no trend in risk was observed without supplemental EBRT (11–16), results in a Day 0 CT-
with the number of cigarettes per day or with the duration of determined extracapsular treatment margin of 6.5 mm and 9.6
smoking among current smokers and no increased risk was mm at the 100% and 75% isodose line, respectively (12). As
found among former smokers.” In contrast, Giovannucci et such, generous periprostatic treatment margins might neutral-
al. (6) reported that men who had smoked ⱖ15 pack years ize the adverse effect of a greater incidence of extraprostatic
extension in smokers (23).
of cigarettes within the preceding 10 years were at a greater
risk of distant metastatic disease and fatal prostate cancer
relative to nonsmokers, with the excess risk among smokers CONCLUSION
eliminated within 10 years of stopping tobacco use. In this prostate brachytherapy cohort, tobacco consumption
Consistent with a previous report, smokers in the current did not predict for risk group stratification or treatment ap-
study were diagnosed with prostate cancer at a younger age proach. Although no statistically significant difference was
than were nonsmokers, with no relationship between Gleason found in biochemical progression-free survival, a trend for
score and tobacco consumption (1, 4). The trend for smaller poorer biochemical outcome was demonstrated in current
prostate glands in current smokers (Table 1) is consistent with smokers.

REFERENCES
1. Roberts WW, Platz EA, Walsh PC. Association of cigarette 13. Butler WM, Merrick GS, Lief JH, et al. Comparison of seed
smoking with extraprostatic prostate cancer in young men. loading approaches in prostate brachytherapy. Med Phys
J Urol 2003;169:512–516. 2000;27:381–392.
2. Hickey K, Do K-A, Green A. Smoking and prostate cancer. 14. Merrick GS, Butler WM. Modified uniform seed loading for
Epidemiol Rev 2001;23:115–125. prostate brachytherapy: Rationale, design, and evaluation.
3. Hsing AW, McLaughlin JK, Hrubec Z, et al. Tobacco use and Tech Urol 2000;6:78–84.
prostate cancer: 26-year follow-up of US veterans. Am J 15. Merrick GS, Butler WM, Dorsey AT, et al. Potential role of
Epidemiol 1991;133:437–441. various dosimetric quality indicators in prostate brachyther-
4. Daniell HW. A worse prognosis for smokers with prostate apy. Int J Radiat Oncol Biol Phys 1999;44:717–724.
cancer. J Urol 1995;154:153–157. 16. Merrick GS, Butler WM, Dorsey AT, et al. The effect of
5. Rodriguez C, Tatham LM, Thun MJ, et al. Smoking and fatal prostate size and isotope selection on dosimetric quality fol-
prostate cancer in a large cohort of adult men. Am J Epidemiol lowing permanent seed implantation. Tech Urol 2001;7:233–
1997;145:466–475. 240.
6. Giovannucci E, Rimm EB, Ascherio A, et al. Smoking and risk 17. American Society for Therapeutic Radiology and Oncology
of total and fatal prostate cancer in United States health profes- Consensus Panel. Consensus Statement: Guidelines for PSA
sionals. Cancer Epidemiol Biomarkers Prev 1999;8:277–282. following radiation therapy. Int J Radiat Oncol Biol Phys
7. Merrick GS, Wallner KE, Butler WM. Permanent interstitial 1997;37:1035–1041.
brachtherapy in the management of carcinoma of the prostate 18. Merrick GS, Butler WM, Wallner KE, et al. Long-term uri-
gland. J Urol 2003;169:1643–1652. nary quality of life after permanent prostate brachytherapy. Int
8. Merrick GS, Wallner KE, Butler WM. Minimizing prostate J Radiat Oncol Biol Phys 2003;56:454–461.
brachtherapy-related morbidity. Urology 2003;62:786–792. 19. Merrick GS, Butler WM, Wallner KE, et al. Late rectal
9. Merrick GS, Butler WM, Farthing WH, et al. The impact of function following prostate brachytherapy. Int J Radiat Oncol
Gleason score accuracy as a criterion for prostate brachyther- Biol Phys 2003;57:42–48.
apy patient selection. J Brachyther Int 1998;14:113–121. 20. Daniell H. More stage A prostatic cancers, less surgery for
10. Nath R, Anderson LL, Luxton G, et al. Dosimetry of intersti- benign hypertrophy in smokers. J Urol 1993;149:68–72.
tial brachytherapy sources: Recommendations of the AAPM 21. Davis BJ, Pisansky TM, Wilson TM, et al. The radial distance
Radiation Therapy Committee Task Group No. 43. Med Phys of extraprostatic extension of prostate carcinoma: Implications
1995;22:209–234. for prostate brachytherapy. Cancer 1999;85:2630–2637.
11. Merrick GS, Butler WM, Dorsey AT, et al. Seed fixity in the 22. Sohayda C, Kupelian PA, Levin HS, et al. Extent of extra-
prostate/periprostatic region following brachytherapy. Int J capsular extension in localized prostate cancer. Urology 2000;
Radiat Oncol Biol Phys 2000;46:215–220. 55:382–386.
12. Merrick GS, Butler WM, Wallner KE, et al. Extracapsular 23. Wallner KE, Merrick GS, Butler WM, et al. Treatment mar-
radiation dose distribution following permanent prostate gins predict two-year PSA response after Pd-103 prostate
brachytherapy. Am J Clin Oncol 2003;26:E178–E189. brachytherapy. Submitted.