0022-5347/01/1665-1707/0
T
®
HE JOURNAL OF UROLOGY
Vol. 166, 17071714, November 2001
Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Printed in U.S.A.
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE
CANCER: METHOD OF TREATMENT AND RESULTS OF A
PHASE I/II TRIAL
MICHAEL D. SHERAR,* MARK R. GERTNER, CLARENCE K. K. YUE, MARTIN E. O'MALLEY,
ANTS TOI, AVIV S. GLADMAN, SEAN R. H. DAVIDSON AND JOHN TRACHTENBERG
From the Division of Medical Physics, Ontario Cancer Institute/Princess Margaret Hospital, Department of Medical Imaging, University
Health Network and Mount Sinai Hospital and Division of Urology, Departments of Surgery, Medical Biophysics and Radiation Oncology,
University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Purpose: Interstitial microwave thermal therapy is experimental treatment for prostate cancer
with the goal of curing disease, while causing fewer complications than standard treatment
options. We present a method for delivering interstitial microwave thermal therapy using
microwave radiating helical antennae inserted percutaneously under transrectal ultrasound
guidance. We report the results of a trial of this method in 25 patients in whom primary external
beam radiation therapy had previously failed. This patient group currently has limited curative
options that are associated with a high complication rate. However, these recurrent tumors often
remain localized to the prostate, and so they may be amenable to localized therapy.
Materials and Methods: Patients with proved prostatic adenocarcinoma were candidates for
treatment when prostate specific antigen (PSA) was 15 ng./ml. or less and prostate volume was 50 cc.
or less. Followup included PSA measurement, digital rectal examination, urinalysis, and documen-
tation of adverse events at 4, 8, 12 and 24 weeks. Sextant biopsy was performed at week 24. The
procedure involved the insertion of 5 antennae percutaneously through a modified brachytherapy
template. The antenna arrangement was determined based on computer simulated predictions of
temperature throughout the prostate. The prostate was dissected away from the rectum by an
injection of sterile saline to provide a thermal barrier that protected the rectum from thermal
damage. Temperatures were monitored using interstitial mapping thermistor probes that were also
inserted through the template. A minimum peripheral target temperature of 55C but less than 70C
was maintained for 15 to 20 minutes, while the urethra, rectum and hydrodissection space remained
below 42C. The urethra and rectum were actively cooled in addition to hydrodissection.
Results: Peripheral target temperatures of 55C were achieved. The urethra and rectum remained
at a safe temperature. The procedure, including setup and treatment, required approximately 2.5
hours of operating room time. At 24 weeks the PSA nadir was 0.5 ng./ml. or less in 52% of patients
and 0.51 to 4 ng./ml. was achieved in an additional 40%. The negative biopsy rate at 24 weeks was
64%, assuming that 3 patients lost to followup would have had positive results. No major complica-
tions were observed and in most cases minor complications resolved within 3 months.
Conclusions: Interstitial microwave thermal therapy for prostate cancer was developed to heat
the prostate safely to a cytotoxic temperature. Experience with 25 patients in whom external
beam radiation therapy for prostate cancer had failed indicates that the treatment is safe.
Although our series indicates that this therapy may be effective, further studies and longer
followup are required in larger patient groups to confirm the potential role of this therapy as an
option for recurrent and primary prostate cancer.
KEY WORDS: prostate; prostatic neoplasms; hyperthermia, induced; microwaves
Interstitial microwave thermal therapy is being developed
patients currently have few effective curative options. Sal-
at Ontario Cancer Institute/Princess Margaret Hospital as a
vage prostatectomy and cryotherapy are associated with a
treatment option for prostate cancer with the goal of curing
significant complication rate.17 Thermal therapy using
disease, while causing fewer complications than standard
transrectal ultrasound is also associated with a relatively
alternatives such as radiation therapy or surgery. Initial
high complication rate.8 Hormone therapy can be effective in
studies have been done in patients with recurrence in whom
controlling tumor progression but it inevitably fails due to
primary external beam radiation therapy has failed. These
hormone resistant tumor cells.
The goal of developing microwave treatment was to heat
Accepted for publication June 15, 2001.
the whole prostatic region that may have tumor cells to a
Supported by Canadian Prostate Center, Inc., CADMIT, Inc., Na-
cytotoxic temperature of 55C to 70C, while protecting critical
tional Cancer Institute of Canada with funds from Terry Fox Run
and Canadian Cancer Society, Princess Margaret Hospital Founda-
adjacent tissues, such as the rectum, bladder and urethra.
tion and Photonics Research Ontario.
We also required that treatment should be delivered in a
* Financial interest and/or other relationship with Canadian Pros-
technically straightforward and reproducible manner with a
tate Center, Inc. and CADMIT, Inc.
total time of less than 2.5 hours in the operating room.
Financial interest and/or other relationship with Amgen, Astra-
Zeneca and Canadian Prostate Center, Inc.
Experience with tissue mimicking phantoms indicated that a
1707
1708
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
temperature of 55C would be achieved reproducibly in a
images acquired at 0.5 cm. intervals from prostate base to
prostatic volume by delivering microwave energy via percu-
apex were used to define the relative positions of the urethra,
taneously placed interstitial antennae.9
rectum and prostate. Temperature distribution in the pros-
The main challenge of thermal and other local therapy is
tate and surrounding tissue was calculated by applying the
directing treatment toward the posterior capsule of the pros-
finite element method to the Pennes linear bioheat equa-
tate, while preserving the rectum directly posterior. We have
tion.11 The thermal dose in the prostate and surrounding
previously reported a method to address this challenge,
tissues was calculated based on the concept of equivalent
called hydrodissection, in which the prostate is dissected
minutes at 43C for predicting the amount of thermal damage
away from the rectum by injecting sterile saline into the
accumulated at any time during treatment.12
space between these 2 structures.10 This maneuver increases
Antenna insertion. Five 5Fr Desilets-Hoffman introducers
the distance between the prostate and rectum, and provides
(Cook Canada, Inc., Stouffville, Ontario, Canada) used as
a thermal barrier that is maintained throughout treatment
antenna insertion sheaths were advanced into the prostate
by saline drip. We combined hydrodissection, and active cool-
under transrectal ultrasound guidance. Figure 2 shows the
ing of the urethra and rectum with percutaneously inserted
planned antenna array arrangement, which was based on the
helical microwave antennae to develop a technical protocol
4-antenna array used in tissue equivalent phantom experi-
for thermal therapy of recurrent prostate cancer. We describe
ments, as previously reported.9 In our study 5 antennae were
the treatment planning and delivery method, and the re-
used to increase control over spatial heating patterns. The
sults of applying this new treatment in 25 patients with
additional central posterior antenna enables heating of the
locally confined, recurrent or persistent prostate cancer
lateral and posterior prostatic aspects without damaging the
treated between May 1998 and May 2000 at Princess
periurethral zone or rectum. We used a Panther 2002 ADI
Margaret Hospital.
ultrasound console (B-K Medical Systems, Inc., North Bil-
lerica, Massachusetts) coupled to a model 8558 S/T biplanar
MATERIALS AND METHODS
transrectal probe (B-K Medical Systems, Inc.). This system
was combined with a model RTP-6000 Precision Stabilizer
Microwave apparatus. Prostate tissues were heated by 5,
(Radiation Therapy Products, Seattle, Washington) support-
17 gauge helical tip microwave radiating antennae (fig. 1, A).
ing a modified brachytherapy insertion template (Radiation
The power deposition or heating pattern of the antenna has
Therapy Products). A virtual template grid superimposed on
a peanut shape with a maximum diameter of the 50% power
the prostatic ultrasound image allowed us to select appropri-
deposition contour of 0.75 cm. (fig. 1, B). Microwave energy
ate template holes to guide the introducers to the desired
was delivered at a frequency of 915 MHz. by a model 500
prostate locations. As each introducer was advanced through
Precision Hyperthermia System (BSD Medical Corp., Salt
the template into the perineum, its arrival at the desired
Lake City, Utah). The power delivered to each antenna was
axial prostate location was verified on transverse ultrasound.
adjustable up to a maximum of 25 W.
The introducers were then advanced to the prostatic base
Treatment planning. A computer simulation program was
under sagittal ultrasound guidance and the antennae were
developed to plan microwave thermal treatments. The pro-
fed through the introducers. The introducers were retracted,
gram has previously been described in detail.9 Briefly, the
leaving the polytetrafluoroethylene coated antennae in direct
geometry of the prostate and associated tissue structures
contact with tissue. The antennae were retracted 0.5 cm.
were determined from transrectal ultrasound. Transverse
from the base under sagittal ultrasound guidance. Figure 3
shows a representative planned antenna arrangement based
on patient transrectal ultrasound.
Normal tissue cooling. Active cooling systems were used to
ensure that heating did not extend beyond the prostate to the
urethra and rectum, eliminating potential damage to these
structures.
Urethra: A custom modified Foley catheter was inserted in
the urethra during treatment. The catheter had a closed end
to contain a water circulation channel. Cool water main-
tained at approximately 10C was propelled through this
channel by a model 7550 60 Masterflex circulation pump
(Cole Parmer Instrument Co., Chicago, Illinois).
FIG. 1. Helical radiating tip microwave antenna with 3 cm. active
element 1.5 mm. in diameter and part of semirigid copper feed line
(A). Helical tip and feed line are polytetrafluoroethylene coated.
Proximal end of copper feed line attaches to section of flexible
braided coaxial cable, which connects to microwave generator. Power
deposition contour plot of helical antenna measured by infrared
thermographic technique9 with antenna position superimposed, and
limits of active helical tip (horizontal lines) and 20% to 90% contours
in 10% steps shown as percent normalized to point of maximum or
FIG. 2. Transverse prostatic aspect with planned antennae and
100% heating (B).
thermometry probe sites.
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
1709
assessed. Active cooling of the urethra, rectum and hydrodis-
section space was verified. Power to the antennae was then
turned on. Prostatic thermometry probes were allowed to
reach a target of 55C, after which heating progressed for a
further 15 minutes. Power delivered to each antenna was
manually adjusted in real time based on temperature read-
ings to maintain 55C at the prostatic boundary. Urethral,
rectal and hydrodissection space temperature was monitored
during heating to ensure that it did not exceed 42C. After
prostatic tissues had been maintained at or above 55C for 15
minutes the power to all antennae ceased. Temperature data
were obtained until all tissue had cooled to below 40C. Treat-
ment was terminated, and the antennae, thermometry probe
catheters and cooling devices were removed. The perineal
puncture wounds were dressed.
Patient recruitment. Patients were enrolled in our study if
they met certain criteria, including age 18 years or older,
written informed consent provided, histological proof of a
history of adenocarcinoma of the prostate of 12 months or
longer after definitive external beam radiotherapy, stage T1
or T2 disease confined to the prostate without evidence of
regional and/or distant disease on recent abdominal and pel-
vic computerized tomography, recent cystoscopy as clinically
FIG. 3. Planned antenna placement (1 to 5) superimposed on pa-
warranted, prostate volume 50 cc or less on transrectal ul-
tient transverse ultrasound showing virtual template grid positions
trasound, serum prostate specific antigen (PSA) 15 ng./ml. or
spaced at 1 cm. In practice template holes are spaced at 0.5 cm.
less and patient ability to comply with study requirements.
Antenna sites are no greater than 1 cm. from prostatic capsule and
Excluded from study were those who were currently receiv-
no greater than 2 cm. apart. Anterior antenna positions (1 and 5) are
at least 1 cm. from anterior prostatic margin to avoid heating blad-
ing any treatment for prostatic carcinoma, had previously
der.
received chemotherapy for prostatic carcinoma, presented
with an extensive transurethral prostatic resection defect or
in whom radiation therapy had caused extensive cystitis
Rectum: We used 2 safety mechanisms to protect the rec-
and/or proctitis. In addition, we excluded patients who had a
tum from thermal damage. For hydrodissection10 a 5.5Fr
metallic implant close to the pelvis, any condition, history of
Desilets-Hoffman introducer placed under transrectal ultra-
illness or surgery that in the opinion of the urologist may
sound guidance formed an access channel to the virtual space
have confounded study results or caused additional patient
between the prostate and rectum. We injected 20 to 30 cc
risk, life expectancy less than 5 years, a history of noncom-
sterile normal saline through this introducer to separate the
pliance with medical therapy, unwillingness or inability to
prostate from the rectum. Enlargement of this space was
complete self-administered questionnaires, or participated in
monitored on transverse and sagittal ultrasound to ensure
a clinical study or investigational treatment within the pre-
that all prostatic and rectal surfaces were separated by at
vious 90 days. Before the procedure a medical history was
least 1 cm. A steady drip of room temperature saline through
obtained and a physical examination, including a digital
the introducer throughout treatment maintained these 2
rectal examination and urinalysis, was performed. A Fleet
structures separate from each other and provided a thermal
enema was self-administered the evening before the proce-
blanket between them. Tissue temperature in this area was
dure.
controlled by adjusting the drip rate. The other safety mech-
Biopsy before and after thermal therapy. Informed consent
anism used to protect the rectum was a rectal cooling unit
was obtained. The patient was placed in the left lateral
inserted into the rectum after needle insertion and removal
decubitus position. After digital rectal examination an endo-
of the transrectal ultrasound probe. A model 7554 80 circu-
cavitary probe was inserted into the rectum. The peripheral
lation pump (Cole Parmer Instrument Co.) propelled room
zone of the prostate was examined for nodules and the pros-
temperature saline through this unit throughout treatment.
tate was measured in 3 planes to determine volume. Under
On-line thermometry. Six Bowman-type thermistor probes
direct ultrasound guidance sextant biopsy was performed
controlled by the BSD 500 system were inserted into the
using an 18 gauge needle. When a peripheral nodule was
prostate and surrounding critical normal structures to mon-
identified, additional biopsies were obtained from the nodule.
itor tissue temperature in these regions in real time (fig. 2).
Postoperative care and followup. After treatment was com-
Three probes were placed within the peripheral prostate to
plete a Foley catheter was inserted and the patient was
base using 3, 6Fr flexible implant catheters (Cook Canada,
hospitalized overnight. Patients were then discharged home
Inc.) as thermometry probe channels. The remaining 3
on oral analgesia. Table 1 shows followup assessments.
probes were inserted into 16 gauge catheters placed in the
urethra, rectum and hydrodissection space, respectively. The
urethral catheter was permanently attached to the custom
RESULTS
Foley catheter, the rectal probe catheter was attached to the
Figure 4 shows temperature and thermal dose patterns
rectal cooling unit and the hydrodissection catheter was ad-
calculated using the treatment planning program. In anten-
vanced through the hydrodissection introducer to the pros-
nae 1 and 5, 2 and 4, and 3 power delivery was simulated as
tatic base. Thermal scans were done from the distal ends of
12 W. for 5 minutes followed by 5 W. for 15 minutes, 15 W. for
all catheters proximal at 2-minute intervals throughout
5 minutes followed by 5 W. for 15 minutes and 5 W. for 5
treatment and posttreatment cooling. The probes were
minutes followed by 3 W. for 15 minutes, respectively. The
mapped at 0.5 cm. intervals for at least 6 cm. to ensure that
delivery of high power at the start of treatment was designed
data were obtained along the length of the prostate and in
to increase prostatic temperature rapidly, while the 15 min-
surrounding normal tissue.
utes of lower power delivery maintained the prostate at a
Treatment delivery. Before applying power to the antennae
steady target temperature of approximately 55C at the pe-
the temperatures measured by all thermometry probes were
riphery. Calculations show that this target was achieved at
1710
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
TABLE 1. Followup procedures after treatment
Procedure
3 Wks.
4 Wks.
8 Wks.
12 Wks.
24 Wks.
Catheter removal
Yes
--
--
--
--
Trial voiding
Yes
--
--
--
--
Adverse events
Yes
Yes
Yes
Yes
Yes
Vital signs
Yes
Yes
Yes
Yes
Yes
Urine culture and stain
Yes
When clinically warranted
When clinically warranted
--
--
Digital rectal examination
--
Yes
Yes
Yes
Yes
Physical examination
--
Yes
--
--
--
PSA
--
Yes
Yes
Yes
Yes
Urinalysis
--
Yes
Yes
When hematuria noted
When hematuria noted
Hematology complete blood count
--
--
Yes
--
--
Transrectal ultrasound guided biopsy
--
--
--
--
Yes
FIG. 4. Transverse ultrasound at prostatic apex (A) midplane (B) and base (C). Temperature contours (colored lines) calculated at end of
20 minutes of heating are superimposed with prostatic outline (black line), urethra (black square), rectum (thin green line) and antenna
positions (1 to 5). In planes distal to antennae tips (C and F) antennae positions are not visualized. Cumulative thermal dose contours of
equivalent minutes at 43C calculated at end of 20 minutes are shown at prostatic apex (D) mid plane (E) and base (F).
the apex and mid plane, while the temperature in the lateral
regions at the base were between 50C and 55C (fig. 4, A to C).
Sparing the periurethral tissue and anterior zone of the
prostate was predicted by the low temperatures calculated
for these tissues in each plane. Figure 4, D to F shows the
corresponding contour diagrams of the thermal dose achieved
by the end of treatment. Greater than 105 equivalent min-
utes per plane at 43C were delivered to the target tissue. As
with the temperature plots, sparing the periurethral tissue
and prostatic anterior zone was predicted by the low thermal
dose accumulation predicted in these regions.
B-ultrasound was performed at the end of the insertion
procedure (fig. 5). The antenna insertion sheaths and 3 tem-
perature probe catheters on the lateral and posterior pros-
tatic margins were clearly visualized. Figure 5 shows that
antennae and temperature probes were placed according to
plan (figs. 2 and 3).
Figure 6 shows mid plane sagittal ultrasound before and
after hydrodissection. The prostate was separated from the
rectum by an additional 1.2 cm. by this procedure. Separa-
tion was maintained by sterile saline drip throughout treat-
ment. Typically 100 to 200 cc saline were delivered.
FIG. 5. Transverse ultrasound reveals antenna insertion sheaths
Temperature profiles were measured just before power was
(1 to 5) and right-to-left thermometry probe catheters (a to c). Re-
discontinued at approximately 18 minutes of heating (fig. 7).
verberations are visible as repetitive bright bands trailing upward
behind some sheaths.
Similar temperature profiles were achieved at all 3 periph-
eral prostatic probe sites. The target temperature of 55C was
achieved along most of the length of each temperature scan,
although at the apical end the temperature was below the
scan limits. The low temperature maintained in the urethra,
target. However, due to prostatic curvature the temperature
hydrodissection space and rectum indicates that these tis-
probes were located outside of prostatic tissue at the axial
sues were protected. The maximum temperature reached
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
1711
FIG. 6. Mid plane sagittal view of prostatic-rectal space before (A) and after (B) hydrodissection. Posterior prostatic aspect was initially
separated from rectum (horizontal lines) with eventual separation by as much as 1.2 cm. (vertical lines) by clearly visible saline bolus.
Custom Foley catheter and posterior prostatic insertion catheters are visible.
FIG. 7. Temperature profiles measured just before power ceased during treatment. Prostatic temperature probes demonstrated safe
temperatures inferior to apex (dashed line), indicating that this region was spared thermal damage during treatment. Zero point indicates
prostatic base.
was 31.0C, 34.5C and 31.2C in the urethra, hydrodissection
Efficacy. At 6 months 13 (52%), 10 (40%) and 2 (8%) of the
space and rectum, respectively.
25 men treated with microwave thermal therapy achieved a
Figure 8, A shows the temperatures at the 2 cm. position of
PSA nadir of 0.5 or less, 0.51 to 4 and greater than 4 ng./ml.,
the 3 prostate tissue temperature probes as a function of
respectively. Biopsy results available in 22 of the 25 cases
treatment time. This position corresponded approximately to
were negative in 16 and positive in 6. If we assume that the
the mid plane of the prostate. The target temperature was
3 patients lost to followup would also have had positive
achieved in less than 4 minutes and maintained at that level
biopsy findings, results were positive in 36% and negative in
until about 18 minutes into treatment, when the power was
64%.
turned off. The somewhat oscillatory nature of the tempera-
Complications. Table 4 lists complications that developed
ture patterns was due to manual changes in power in re-
after thermal therapy. Most patients tolerated the procedure
sponse to changing temperatures, as measured by the map-
well and none required hospitalization for more than 1 night.
ping thermometry probes. For comparison figure 8, B shows
Complications were categorized as short-term and long-term
equivalent temperatures as a function of time that were
major, irritative, obstructive, incontinence and pain. No ma-
recorded at the warmest location per probe. The patterns are
jor complications were observed. The rate of most complica-
only slightly higher in temperature but similar to previous
tions decreased significantly in the initial 3 months after
values (fig. 8, A).
treatment. Notably no patient had complete incontinence as
Patient and tumor characteristics. Table 2 shows the char-
a result of this procedure.
acteristics of the patient population, including co-morbid dis-
ease, race and previous therapy. Median patient age was 68.6
years (range 55 to 78). Table 3 lists the characteristics of
DISCUSSION
patient disease before initial external beam radiation ther-
The development of curative therapy for recurrent prostate
apy and before thermal therapy. Immediately before thermal
cancer remains a major challenge. Approximately a third to
therapy median PSA was 6.69 ng./ml., including greater than
half of patients with disease recurrence have locally persis-
10 ng./ml. in 6 (24%) and 10 ng./ml. or less in 19 (76%).
tent but no micrometastatic disease, as implied by PSA meas-
Patients were hospitalized after therapy for 1 night.
urement after salvage prostatectomy.1, 2 However, fibrosis
1712
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
TABLE 3. Distribution of tumor stage and Gleason grade in 25
patients before initial radiation therapy and immediately before
microwave thermal therapy
No. Radiation
No. Microwave
Stage:
T1
4
7
T2
15
16
T3
6
2
Grade:
57
19
14
810
5
7
Unknown
1
4
No patients had grade 3/4 disease.
TABLE 4. Patients with complications of microwave thermal
therapy at any time within and after 3 months after therapy
No. Up to 3 Mos. (%)
No. After 3 Mos. (%)
Irritative:
11 (44)
7 (28)
Frequency
5 (20)
3 (12)
Urgency
4 (16)
3 (12)
Infection
3 (12)
1 (4)
Dysuria
1 (4)
0
Bladder spasm
1 (4)
0
Nocturia
4 (16)
2 (8)
Obstructive:
8 (32)
3 (12)
Retention
4 (16)
2 (8)
Hesitancy
1 (4)
1 (4)
Poor stream
3 (12)
1 (4)
Incontinence:
3 (12)
6 (24)
Stress
2 (8)
5 (20)
Urge
1 (4)
1 (4)
Perineal pain
3 (12)
1 (4)
No patients had major complications.
FIG. 8. Temperature profile throughout treatment for 3 prostatic
thermometry probes at mid plane (A) and at warmest points per
target volume. Antenna arrangements were based on the
probe (B). Maximum temperatures were below 70C. Power ceased at
approximately 18 minutes and tissue cooled to safe level before
ability to achieve cytotoxic temperatures to the prostatic
treatment was completed.
capsule in a reasonable treatment time of approximately 20
to 25 minutes. However, applying heat conduction to deliver
required thermal doses to the prostatic margins results in
TABLE 2. Characteristics of the 25 patients who underwent
the possibility of heat spreading beyond the prostatic cap-
microwave thermal therapy
sule. The close proximity of the rectum to the prostate would
No. (%)
normally preclude effective treatment to the posterior mar-
Co-morbid disease:
gins due to the requirement of no heat damage to the rectum.
Cardiac
8 (32)
The hydrodissection method enables aggressive treatment to
Hypertension
10 (40)
be delivered to the posterior prostatic margins, while protect-
Diabetes
3 (12)
ing the rectum because large temperature gradients can be
Other:
11 (44)
Arthritis
3
created. As a safety requirement, rectal, urethral and hydro-
Cholesterolemia
1
dissection space temperature did not exceed 42C. If this
Bleeding ulcer
2
situation were to have occurred, power delivery would have
Sleep apnea
1
been immediately stopped to enable these tissues to cool. The
Diverticulosis
2
Cholelithiasis
2
temperature at the posterior prostatic margins was greater
Race:
than 60C, while the hydrodissection space was maximum of
White
24 (96)
just below 35C and the rectum was maintained at just below
Black
1 (4)
32C (fig. 7).
No patients had stroke and none was Hispanic or Asian.
Although data on the temperature probes acquired during
treatment allow a high level of safety, they do not indicate
temperature throughout the whole target tissue. Our as-
and scarring after primary radiation therapy result in a high
sumption of cell death throughout the target was based on
complication rate when salvage prostatectomy is attempt-
the assumption that temperature within the target volume
ed.3, 4 Our series shows that interstitial microwave thermal
was not less than that measured at the prostatic periphery.
therapy can be delivered safely and in a technically feasible
Results of the treatment planning simulation support these
manner.
hypotheses. However, our current treatment planning does
The insertion of temperature probes and antennae is sim-
not consider all individual parameters, such as blood flow,
ilar to the insertion of needles in a prostatic brachytherapy
tissue electrical and thermal properties, and changes in these
procedure. Standard brachytherapy stabilizing and insertion
parameters during treatment. Measuring the thermal dose
equipment were used, except the holes in the template nor-
or thermal damage throughout the target volume during
mally used for inserting brachytherapy seeds were enlarged
treatment would be a preferred indicator of treatment effi-
by the manufacturer to accommodate the microwave anten-
cacy. Magnetic resonance imaging (MRI) during treatment
nae. The antenna sheaths and probe catheters were clearly
has been investigated for this purpose at our institution
visualized on transrectal ultrasound during setup (fig. 5).
based on T2 measurements and magnetic resonance ther-
This interstitial microwave heating method relies on the
mometry.13 Furthermore, we have previously reported that
conduction of heat from close to the antennae to the whole
calculating the thermal dose based on MRI temperature
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
1713
mapping throughout the heated tissues correlates well with
ity of the rectum to the prostate limits the therapeutic dose
posttreatment histology in a rabbit brain model.14 Although
delivered to the posterior prostatic aspect.
these MRI techniques are promising methods for on-line
Our data show that 8% of patients achieved a PSA nadir
3-dimensional temperature monitoring, the practical prob-
that was greater than 4 ng./ml. 6 months after treatment and
lems of performing interstitial microwave thermal therapy
there was a 36% positive biopsy rate at 6 months, assuming
inside the bore of a conventional magnet are considerable.
that 3 patients lost to followup would have had positive
Applying MRI temperature mapping techniques on open con-
results. These treatment failures may have occurred because
cept magnets would provide a practical solution for on-line
sublethal thermal doses were achieved in certain prostatic
monitoring.
regions. Correlating the thermal dose achieved by microwave
We have investigated other methods of monitoring thermal
therapy with treatment outcome would aid in the develop-
therapy in real time, including imaging changes in ultra-
ment of improved methods of treatment planning and deliv-
sound attenuation during heating and monitoring blood flow
ery. In our study the temperature measured along the length
changes using functional computerized tomography.1518
of the prostate at the 3 peripheral locations was used to
There are 2 major challenges of using ultrasound for moni-
control manually the heat delivered via each antenna. Com-
toring thermal therapy for prostatic cancer. Artifacts in the
puter simulations were used to estimate the thermal dose
ultrasound image from needles restricts the visibility of the
achieved throughout the prostate as a result of the heating
target and normal tissues. In addition, pressure caused by
protocols. Thus, they were used to judge treatment efficacy
the ultrasound probe may move the rectal wall closer to the
after the fact. A further goal of these computer simulations is
prostate during treatment. The therapy method presented
to prescribe a heating plan on an individual basis that would
requires a separate small rectal cooling device and conse-
lead to a higher likelihood of treatment success.
quently no ultrasound is performed during treatment.
As on-line image monitoring techniques mature, we antic-
CONCLUSIONS
ipate that the problem of determining 3-dimensional tissue
The results of our study indicate that interstitial micro-
temperature and thermal damage during treatment may be
wave thermal therapy with hydrodissection is safe in pa-
solved in a practical way. Before it is achieved, interstitial
tients with recurrent or persistent localized prostate cancer
temperature mapping, as presented, is adequate to ensure
after failed external beam radiation therapy. Early followup
treatment safety. A higher likelihood of effective treatment
biopsy and PSA data indicate that treatment may be effective
can be achieved using pretreatment planning as a means of
in a significant percent of patients. This method was de-
determining antenna sites that maximize the probability
signed to avoid complications by protecting the urethra and
that temperatures throughout the whole target volume
rectum with active cooling mechanisms. Temperatures meas-
would be cytotoxic.
ured during treatment indicate that cytotoxic heating was
Interstitial microwave thermal therapy with hydrodissec-
achieved out to the prostatic capsule and critical normal
tion resulted in a negative biopsy in 64% of patients 6 months
tissues were protected. Due to the variable aggressiveness of
after therapy. Although these early data are promising, we
prostatic cancer among patients and the potential for long-
expect that subsequent biopsy failures may decrease the
term asymptomatic survival with local disease developing a
long-term success rate. Longer followup in larger groups is
treatment option with a low side effect profile is extremely
necessary to determine the durability of the response to
desirable. Longer term followup of PSA and biopsy results is
microwave thermal therapy. Pisters et al reported that 93%
required in larger groups of patients to confirm the potential
of patients who had received previous radiation therapy had
of interstitial microwave thermal therapy as a treatment
a negative sextant biopsy 6 months after cryotherapy when
option for recurrent, persistent and primary prostate cancer.
they had been treated with a double freeze-thaw cycle.5 The
equivalent result for a single freeze-thaw cycle was 71%.
In their cryotherapy series Pisters et al noted a short-term
REFERENCES
and long-term incontinence rate of 42% and 28%, respective-
1. Rogers, E., Ohori, M., Kassabian, V. S. et al: Salvage radical
ly.5 When a written questionnaire was administered, the
prostatectomy: outcome measured by serum prostate specific
incontinence rate increased to 73%. In contrast, microwave
antigen levels. J Urol, 153: 104, 1995
thermal therapy resulted in a 12% and 24% incontinence
2. Lerner, S. E., Blute, M. L. and Zincke, H.: Critical evaluation of
rate, respectively. However, there are several possible con-
salvage surgery for radio-recurrent/resistant prostate cancer.
J Urol, 154: 1103, 1995
founders in this comparison. The 42% and 28% incontinence
3. Neerhut, G. J., Wheeler, T., Cantini, M. et al: Salvage radical
rates after cryotherapy were measured by retrospective chart
prostatectomy for radiorecurrent adenocarcinoma of the pros-
review and telephone survey, while in our series the rate was
tate. J Urol, 140: 544, 1988
measured only by retrospective chart review. The cryother-
4. Link, P. and Freiha, F. S.: Radical prostatectomy after definitive
apy cohort included patients who had undergone previous
radiation therapy for prostate cancer. Urology, 37: 189, 1991
hormonal and/or chemotherapy in addition to radiation ther-
5. Pisters, L. L., von Eschenbach, A. C., Scott, S. M. et al: The
apy in group 2,5 whereas chemotherapy precluded enroll-
efficacy and complications of salvage cryotherapy of the pros-
ment in our study. Our study participants had received pre-
tate. J Urol, 157: 921, 1997
vious radiation therapy only, equivalent to group 1 in the
6. Perrotte, P., Litwin, M. S., McGuire, E. J. et al: Quality of life
after salvage cryotherapy: the impact of treatment parame-
study of Pisters et al. Complications in the cryotherapy study
ters. J Urol, 162: 398, 1999
were not provided according to these 2 groups.
7. Bales, G. T., Williams, M. J., Sinner, M. et al: Short-term out-
Comparing our data with those on patients treated with
comes after cryosurgical ablation of the prostate in men with
transrectal ultrasound is difficult because that group largely
recurrent prostate carcinoma following radiation therapy.
comprised primary treatment with only 2 of 50 involving
Urology, 46: 676, 1995
radiation failure.8 In contrast to those studies and cryother-
8. Gelet, A., Chapelon, J. Y., Bouvier, R. et al: Local control of
apy results,5 we observed no major complications, such as
prostate cancer by transrectal high intensity focused ultra-
fistula, osteitis pubis, prostatic abscess or severe pain. We
sound therapy: preliminary results. J Urol, 161: 156, 1999
propose that it was a result of the hydrodissection technique,
9. Sherar, M. D., Gladman, A. S., Davidson, S. R. H. et al: Helical
antenna arrays for interstitial microwave thermal therapy for
which allows aggressive treatment of the posterior aspect of
prostate cancer: tissue phantom testing and simulations for
the prostate with no observable risk to the rectum. It is
treatment. Phys Med Biol, 46: 1905, 2001
probable that this technique may help to improve the com-
10. Lancaster, C., Toi, A. and Trachtenberg, J.: Interstitial micro-
plication and efficacy rate of all local prostatic treatments,
wave thermoablation for localized prostate cancer. Urology,
including cryotherapy and brachytherapy, in which proxim-
53: 828, 1999
1714
INTERSTITIAL MICROWAVE THERMAL THERAPY FOR PROSTATE CANCER
11. Pennes, H. H.: Analysis of tissue and arterial blood tempera-
laser photocoagulation: comparison with an Arrhenius dam-
tures in the resting human forearm. 1948. J Appl Physiol, 85:
age model. Phys Med Biol, 45: 1115, 2000
5, 1998
12. Sapareto, S. A. and Dewey, W. C.: Thermal dose determination
EDITORIAL COMMENT
in cancer therapy. Int J Radiat Oncol Biol Phys, 10: 787, 1984
13. Chen, J. C., Moriarty, J. A., Derbyshire, J. A. et al: Prostate
This article is an evaluation of a novel technique for the treatment of
cancer: MR imaging and thermometry during microwave ther-
failed primary external beam radiotherapy for prostate cancer. The
mal ablation-initial experience. Radiology, 214: 290, 2000
technique of urethral and rectal cooling units along with hydrodissec-
14. Sherar, M. D., Moriarty, J. A., Kolios, M. C. et al: Comparison of
tion to separate the rectum and prostate appears to have prevented the
thermal damage calculated using magnetic resonance ther-
complications of a fistula. Other complications include urine inconti-
mometry, with magnetic resonance imaging post treatment
nence (24%), bladder outlet obstruction (12%) and irritative voiding
and histology, after interstitial microwave thermal therapy of
symptoms (28%). A positive biopsy rate of 24% to 36% at 24 weeks and
rabbit brain. Phys Med Biol, 45: 3563, 2000
a PSA nadir of less than 0.5 ng./ml. in 52% at 24 weeks may be falsely
15. Gertner, M. R., Worthington, A. E., Wilson, B. C. et al: Ultra-
reassuring, and longer followup is required before statements of efficacy
sound imaging of thermal therapy in in-vitro liver. Ultrasound
can be made. Although this technique has less complications than
Med Biol, 24: 1023, 1998
reported series of salvage radical prostatectomy, they are still signifi-
16. Bevan, P. D. and Sherar, M. D.: B-scan ultrasound imaging of
cant and the "true risks benefit ratios" will only be known with longer
thermal coagulation in bovine liver: log envelope slope atten-
followup. The failure rate is at least 24% at 1 year and will surely
uation mapping. Ultrasound Med Biol, 27: 379, 2001
increase with time. The short-term results may be no better than those
17. Bevan, P. D. and Sherar, M. D.: B-scan ultrasound imaging of
obtained with androgen deprivation therapy.
thermal coagulation in bovine liver: frequency shift attenua-
Daniel J. Culkin
tion mapping. Ultrasound Med Biol, 27: 809, 2001
Department of Urology
18. Purdie, T. G., Lee, T. Y., Iizuka, M. et al: Dynamic contrast
University of Oklahoma Health Sciences Center
enhanced CT measurement of blood flow during interstitial
Oklahoma City, Oklahoma