Pii: s0360-3016(99)00351-x

Int. J. Radiation Oncology Biol. Phys., Vol. 46, No. 1, pp. 221–230, 2000 Copyright 2000 Elsevier Science Inc.
PII S0360-3016(99)00351-X
THE AMERICAN BRACHYTHERAPY SOCIETY RECOMMENDATIONS FOR
PERMANENT PROSTATE BRACHYTHERAPY POSTIMPLANT
DOSIMETRIC ANALYSIS
SUBIR NAG, M.D.,*† WILLIAM BICE, PH.D.,*‡ KEITH DEWYNGAERT, PH.D.,*§ BRADLEY PRESTIDGE, M.D.,*࿣ RICHARD STOCK, M.D.,*¶ AND YAN YU, PH.D.*# *Clinical Research Committee, The American Brachytherapy Society, Oak Brook, IL; †Ohio State University, Columbus, OH; ‡University of Texas Health Science Center,San Antonio, TX; §New York University, New York, NY; ࿣Cancer Therapy and Research Center, San Antonio, TX; ¶Mt. Sinai Hospital, New York, NY; and #University of Rochester, Rochester, NY Purpose: The purpose of this report is to establish guidelines for postimplant dosimetric analysis of permanent
prostate brachytherapy.
Methods: Members of the American Brachytherapy Society (ABS) with expertise in prostate dosimetry evalu-
ation performed a literature review and supplemented with their clinical experience formulated guidelines for
performing and analyzing postimplant dosimetry of permanent prostate brachytherapy.
Results: The ABS recommends that postimplant dosimetry should be performed on all patients undergoing
permanent prostate brachytherapy for optimal patient care. At present, computed tomography (CT)-based
dosimetry is recommended, based on availability cost and the ability to image the prostate as well as the seeds.
Additional plane radiographs should be obtained to verify the seed count. Until the ideal postoperative interval
for CT scanning has been determined, each center should perform dosimetric evaluation of prostate implants at
a consistent postoperative interval. This interval should be reported. Isodose displays should be obtained at 50%,
80%, 90%, 100%, 150%, and 200% of the prescription dose and displayed on multiple cross-sectional images of
the prostate. A dose-volume histogram (DVH) of the prostate should be performed and the D
(dose to 90% of
the prostate gland) reported by all centers. Additionally, the D
the fractional V
(i.e., the percentage of prostate volume receiving 80%, 90%, 100%, 150%, and 200% of the prescribed dose,
respectively), the rectal, and urethral doses should be reported and ultimately correlated with clinical outcome
in the research environment. On-line real-time dosimetry, the effects of dose heterogeneity, and the effects of
tissue heterogeneity need further investigation.
Conclusion: It is essential that postimplant dosimetry should be performed on all patients undergoing permanent
prostate brachytherapy. Guidelines were established for the performance and analysis of such dosimetry.
2000 Elsevier Science Inc.
Prostate neoplasm, Brachytherapy, Dosimetry, Guidelines, 125Iodine, 103Palladium.
INTRODUCTION
for, and role of postimplant dosimetry following permanentradioactive seed implantation is occasionally questioned.
Postimplant dosimetric analysis is standard practice follow- This prompted the ABS to organize a panel with expertise in ing temporary brachytherapy procedures. Its role following the field of implant evaluation to perform a literature review permanent implants is less well established. Previous sur- and to share their experience and knowledge to develop veys have shown wide variation in dosimetric methods (1, guidelines for the performance and analysis of postimplant 2); however, there are no established clinical standards or guidelines for performing prostate dosimetry. The recently Because the treatment plan and the actual implant have issued American Brachytherapy Society (ABS) guidelines already been completed at the time of postimplant analysis, for prostate brachytherapy recognized the need for such the rationale for its use needs elucidation. The first issue guidelines (3). Although an increasing number of prostate arises from the fact that brachytherapy is an imperfect brachytherapy procedures are performed annually, the need modality, and certainly, the permanent ultrasound-guided Reprint requests to: Subir Nag, M.D., Chief of Brachytherapy, Brachytherapy Society, and thank Drs. Patrick Swift, Mack Roach, The Ohio State University, 300 West 10th Avenue, Columbus, OH Frank Waterman, David Beyer, and Michael Zelefsky for their 43210. Tel: (614) 293-8415; Fax: (614) 293-4044; E-mail: Presented at the annual meeting of the American Society for Acknowledgments—The authors wish to express their gratitude to Therapeutic Radiology and Oncology, San Antonio, TX, Novem- Mr. David Carpenter for editorial assistance. The authors acknowl- edge the support of the Board of Directors of the American Accepted for publication 6 August 1999.
I. J. Radiation Oncology ● Biology ● Physics prostate implant technique is no exception. The dose distri- Table 1. Relative advantages and disadvantages of each imaging butions following implantation are not the same as those modality for performing post implant dosimetric analysis* planned prior to the implant (4–10). Because the dose dis- tributions differ, it is important to document the actual dosethat the prostate and normal adjacent tissues will receive over the life of the implant. This can only be determined if a postimplant dosimetric assessment is performed.
The information obtained is essential for optimal patient care. Significant underdosing of the prostate, which can lead to treatment failure, can be potentially rectified using sup- plemental external beam irradiation or additional seed im- * Grading scale: ϩϩ, ϩ, 0, Ϫ and ϪϪ, where ϩϩ is the plantation (11). In patients who experience a biopsy-proven highest ranking and ϪϪ is the lowest.
local failure, the knowledge of the original dose distribution † MRI (magnetic resonance imaging) performed with a body may prove useful when considering salvage therapy, whether in the form of external beam irradiation reimplan- CT ϭ computed tomography; TRUS ϭ transrectal ultrasound.
tation or radical prostatectomy (12–14). This information isalso important in determining the cause of potential com- as their spatial relationship from the cross-sectional images.
plications and appropriate patient management While the advantages of this type of dosimetry over plane Postimplant dosimetry is invaluable for those physicians film are overwhelming, each of the above imaging modal- who are just starting to perform permanent seed implants.
ities has limitations. Table 1 summarizes their relative ad- Dosimetry results can help physicians assess and modify vantages and disadvantages. This has prompted efforts to their implant technique. This is essential because there is a combine imaging methods by spatially coregistering (i.e., learning curve involved in performing prostate brachyther- fusing) the information from two or more imaging modal- apy (15, 16). In addition, experienced physicians can use dosimetry to further refine and perfect the procedure (17).
Historically, the earliest method was plain film dosime- Finally, the data that is provided by postimplant dosim- try. The geometric reconstruction of source locations from etry could be used in future outcome analysis. It would two projection radiographs has been used for many years to allow comparison of treatment results from various institu- perform pre- and postimplant dosimetry. Techniques are tions and could be used as a quality assurance tool in available for using films with a common axis and for those prospective multi-institutional clinical trials.
taken with a stereo shift (19, 20). Methods of correcting forfilm skew—films not perpendicular to the axis of the x-raysource—have been developed (21). The errors inherent in these methods have been studied (22–24). Unfortunately, Members of the ABS with expertise in prostate dosimetry two-film techniques may be used reliably only when the evaluation performed a literature review, and supplemented operator is able to match each individual source from one with their clinical experience, formulated guidelines for film with its corresponding image on the other film. Because performing and analyzing postimplant dosimetry for perma- of the number seeds and their irregular spacing, this is nent prostate brachytherapy. The areas of consensus and extraordinarily difficult in permanent prostate brachyther- controversy were noted. Specific dosimetric recommenda- tions (for use in a community setting as well as in the To address this problem, three film techniques have been research environment) and directions for future investiga- developed (25–27). These methods dramatically improve tions were made. This report was reviewed and approved by the accuracy of seed localization in permanent prostate implants, achieving true localization rates on the order of90% (28). Radiographs or fluoroscopy can be performed inthe operating room during or immediately after the implant with equipment that is readily available.
Available modalities for obtaining postimplant dosimetry The principal disadvantage of plane film techniques is Methods of performing postimplant evaluation of pros- that they cannot be used to visualize the target (prostate tate implants can be best categorized by the modality used gland) and critical structures (rectum, urethra, bladder).
to generate the images. The brachytherapist is no longer While the dose distribution may be computed and displayed limited to using plane films. Initially developed for com- in axial planes, or even viewed as a three-dimensional (3D) puted tomography (CT) (18), prostate brachytherapy dosim- object, there is no information about the spatial relationship etry based upon axial images has been applied to magnetic of this distribution to the prostate or adjacent structures.
resonance imaging (MRI), and transrectal ultrasound While other deficiencies exist with this method (organ mo- (TRUS). The term cross-sectional image set will be used to tion between films, for instance), it is this shortcoming that refer generically to all three. Each modality allows, at least led to the development of cross-sectional image set dosim- to some degree, localization of structures and seeds, as well etry such as CT-based dosimetry.
Prostate postimplant dosimetry ● S. NAG et al. CT-based dosimetry was first applied to prostate implants implantation of the gland (14). The seed locations are very by Roy et al. at Memorial Sloan-Kettering Cancer Center in difficult to discern, and their disruption of the ultrasound New York (18). The advantage of being able to visualize signal makes delineation of the prostatic borders more dif- sources in relation to the target became immediately appar- ficult than with the preimplant ultrasound. Because of this, ent, particularly with regard to low energy isotopes such as it seems unlikely that ultrasound can be used as a single 125I and 103Pd, for which the dose distribution is highly modality for postimplant dosimetric analysis, unless ultra- dependent on precisely locating the seed positions. In this sound technology is improved. The patient discomfort as- technique, abutting slices taken through the gland were sociated with this examination particularly after surgery displayed and digitized into the treatment planning system.
adds to the disadvantages of this option unless the dosimetry Because the sources often appeared on more than one slice, is performed on-line, during surgery.
a seed location reduction method (seed sorting) based upon Nevertheless, there are some features of TRUS that make it appealing. Ultrasound examination is relative easy and The basic methods of CT-based implant dosimetry have inexpensive. The possibility of using the same imaging changed little. Various authors have published techniques modality that was used to perform the preplan and the that adjust the slice spacing or the distance between slices, implant procedure to generate the post plan is enticing.
and the task of seed sorting has been automated (28).
TRUS potentially offers the only practical option for per- Properly performed, the accuracy of seed location is on par forming on-line dosimetric analysis during the procedure, with, if not superior to, three-film techniques (28, 29).
allowing the brachytherapist to adjust the dose distribution Limitations of this technique include the required a priori by adding seeds in regions where the dose is inadequate.
knowledge of the number of seeds in the image set at the Like MRI, longitudinal imaging is also possible.
beginning of the sorting process. This information can be Because each imaging modality offers it own advantages, garnered from a single plane film, usually taken in the some authors have combined imaging techniques to opti- anterior-posterior direction, or less reliably, from documen- mize the information available for the postimplant analysis tation detailing the number of sources implanted within the (31, 33, 34). Combining two or more modalities usually patient. Additionally, some inherent uncertainty is intro- involves using a modality that optimizes source localization duced when the location of the seeds in the axial (i.e., and another that best delineates the prostatic and critical cranio-caudal) direction is determined. This is because axial volume sampling limits the resolution in this direction to the Coregistration, sometimes called fusion, relies upon de- width of each individual slice. Soft tissue contrast with CT termining a transformation matrix that converts data from is often poor, making it difficult to reliably contour the one image set to the other. Image information can then be borders of the prostate, especially at the base and the apex overlaid to calculate and display information from both sets.
Defining this transformation matrix requires at least three The ability of MRI to visualize soft tissue anatomy makes data points, although the most successful coregistration it an enticing choice as an imaging modality for prostate methods use a much larger number of data points. Examples brachytherapy dosimetry. Several authors have used MRI in in prostate brachytherapy include using marker seeds (10), this regard (31, 32). The MRI set is not restricted to axial the urethral surface (33), and multiple seed locations (35).
acquisition, a particularly useful attribute for delineating the There are pitfalls associated with coregistration of two glandular borders at the troublesome apex and base. Critical image sets. There can be changes in the patient position structures such as the urethra and the neurovascular bundle relative to the coordinates used to generate the transforma- tion matrix, or changes in the relative positions between the There are many problems associated with MRI dosimetry coordinates themselves. For instance, using a urethra dis- of the prostate. In addition to the same seed sorting prob- tended by the presence of a catheter in one image set to lems inherent to any cross-sectional image set, visualization align an image set that had been generated without a cath- of the seeds themselves is difficult. Because there is no eter would likely produce errors. For the same reason, signal from them, they image as low signal areas, making extreme care must be exercised when aligning image sets them difficult to distinguish from vessels, calcifications, and based upon source locations from images produced at two other structures with no signal. This is particularly difficult widely different times after the implant. A similar argument at the periphery and just outside the gland. While some can be made against coregistering TRUS images taken success has been achieved by choosing an imaging se- before implant with the CT images taken after implant, quence using bone windows with a narrow bandwidth (thus unless sufficient time has passed for the postimplant edema enhancing the artifact from the seeds), MRI does not image sources as well as CT does. The acquisition process is Distortion can also be a problem. A simple transforma- certainly slower than CT, possibly contributing to motion tion that results in scaling translation and rotation cannot artifacts. Distortion of the image set may also be a problem correct for a distorted data set. Fortunately, over the dis- tances of concern in prostate brachytherapy, and with the Although no reports have been published on the use of equipment that is currently available, distortion of any sin- TRUS for postimplant dosimetry, it has been used for re- gle data set is usually minimal. Coregistration techniques I. J. Radiation Oncology ● Biology ● Physics that ignore distortion have thus far proved adequate for thickness and spacing are commonly reported in the litera- permanent prostate brachytherapy, because the distortion is ture) (6, 10, 17, 18, 42–44).
minimal due to the small distances in prostate brachyther- A catheter placed in the bladder and filled with contrast can be used to localize the urethra and internal bladder wall.
A simple example of coregistration is the overlay of However, the use of a catheter should be weighed against isodose curves generated from plane film dosimetry on axial the discomfort and potential morbidity of this procedure CT images. Alignment is performed visually, sometimes (especially if the CT scan is not performed in the immediate aided by the placement of a gold marker seed placed at the postimplant period when the patient already has an indwell- apex of the gland. Transverse slices in the plane film coor- dinate system are generated by the planning system and then CT images are acquired using normal body-CT settings.
overlaid on the appropriate CT slice. Although this practice If hardcopy films are to be used for digitization of seeds and is common, it is fraught with uncertainties, and therefore is prostate, an optimal window setting must be chosen that of marginal value in permanent prostate brachytherapy. The balances the ability to resolve seeds with the ability to rapid changes in dose within relatively short distances make delineate the prostate and adjacent structures of interest. The it necessary to be as accurate as possible in determining the geometry of the implant, and therefore the dosimetry, is transformation matrix. This level of accuracy can be derived directly from the CT images themselves. In some achieved only with methods of determination that are quan- CT scans, the images may contain distortions (such as unequal x and y scaling), and it is important that means of Roberson, Narayana, and colleagues have used marker identifying and accounting for such scaling variations be in seeds, as well as the urethral and rectal surfaces, to coreg- ister the preimplant ultrasound and the postimplant CT scan The TG-43 formalism is recommended for both the pre- (10, 33). A similar technique in which the urethra and and postimplant dosimetry (45–48). Due to the difficulties bladder base are visually aligned to coregister postimplant in using CT scans to determine seed orientation, the use of CT and MRI image sets has recently been used by Amdur et a point source approximation with anisotropy constant is al. (36). A more rigorous method of coregistering image recommended (49). Calculations should be performed using sets based on the available source locations in each data set a matrix with resolution limited to 2 mm or less (50) in an has been outlined by Dubois et al. (35). This method has effort to minimize the effects of the large dose gradients been used to coregister postimplant CT and MRI data sets inherent in a brachytherapy procedure.
and to fuse postimplant CT data sets to ultrasound image The target is defined as the prostate (without margin) on sets acquired to plan a second salvage implant (37).
the individual CT images. Care should be taken to distin-guish the prostate from the peri-prostatic tissue. Severalstudies have noted discrepancies in volume of prostate, as determined by TRUS, MRI, and CT, reflecting the difficul- At the present time, CT-based evaluation of the prostate ties in differentiating the prostate from the periprostatic implant appears to best satisfy the requirements of seed musculature and venous plexus using CT (23, 33, 44, 51).
localization target and normal structure delineation and Normal structures of interest that can be defined by using seed-target registration. It is also readily available. Due to CT include the urethra and the rectum (17, 18, 52, 53). For possible seed migration or embolization (38–41), the num- the urethra, the entire prostatic urethra should be defined.
ber of seeds implanted may not be the same as the number This can be done through use of a central lumen point of seeds present in the prostate at the time of the postimplant identified on each slice, or by contouring the urethral wall scan. Therefore, a better approximation of the number of (52, 54). Catheterization is an accurate method for localiza- seeds may be obtained by using plane radiographs. The tion of the urethra within the prostate. If, however, the recommended technique for performing CT-based dosime- urethra cannot be visualized, an alternative is to identify, as the urethral dose point, the geometric center of the prostate The region to be imaged by CT should include the pros- as imaged on successive CT slices (55). It must be recog- tate, all the seeds within and around the prostate, and any nized that doing this gives only an estimate of the urethral critical structures for which the dose is to be reported. To dose, and is valid only if peripheral seed-loading configu- accomplish this, it is suggested that at minimum, a 2-cm margin be added to the superior and inferior extent of the For purposes of dosimetry, only the anterior rectal wall, prostate. A reduced field of view that completely encom- and not the entire rectum, is considered the structure of passes the volumes and structures of interest, but offers a interest. As with the urethra, several different methods may finer resolution in the plane of the implant, should be used.
be used to define the rectal wall. These include the use of This will reduce the error associated with seed localization single points located along the anterior wall of the rectum, contouring the outer anterior rectal wall for use with surface Contiguous axial slices are recommended to reduce the dosimetry or contouring the anterior rectal wall as a volume chance of missing seeds between scans. The slice thickness excluding the lumen (53, 56, 57). As most commercial and spacing should be no greater than 5 mm (3-mm slice planning systems are unable to define the dose to surfaces, Prostate postimplant dosimetry ● S. NAG et al. Table 2. Effect of timing on CT-based dosimetric evaluation based on interseed spacing in an effort to eliminate theuncertainty introduced by CT-based prostate margin delin- eation. It was proposed that this alternative would serve as a more accurate estimation for target volume changes. How- ever, the volumes estimated by this method were in reason-able agreement with those determined by contouring the The optimal time to evaluate permanent prostate implant dosimetry is controversial, and may differ by isotope (be- cause of the difference in half-lives). Time-averagedweighting factors (58) and computer modeling (60–62) sug-gest that 103Pd and 125I implants would best be evaluated contouring the anterior rectal wall as a volume represents a after about 2 and 4 weeks, respectively. However, the do- simetric compromise introduced by performing evaluation It is possible for an individual seed to appear on multiple of 103Pd implants at 1 month was demonstrated to be quite CT slices. Although the frequency of this event may be small (58). For various practical and logistical reasons, reduced by using the larger (5 mm) slice spacing, it is not many brachytherapists prefer to rely on early scanning (2, 3, eliminated. Therefore, a seed-sorting computer program is 6, 18, 38, 59, 63). Many patients come from great distances needed to eliminate this duplication or redundancy, and to and may be unwilling to make return trips just for postop- yield a final seed count consistent with the presumed num- erative imaging studies. Additionally, early feedback can be ber of seeds within the volume. As previously stated, plane used to compensate an underdosed prostate (by reimplant- radiographs are recommended in conjunction with CT- ing or adding external beam) and to improve the implanta- based dosimetry to aid seed sorting routines that require tion technique. With early dosimetry, Willins and Wallner prior knowledge of the number of seeds to be identified estimated that coverage of the gland by at least 80% of the from the larger set of seeds localized on CT. The “z” or target isodose line was adequate (6, 63). These consider- cranio-caudal coordinate of seeds identified on multiple CT ations may outweigh the 10% underestimation of prostate slices may be better defined by averaging over the CT coverage that can be produced by early dosimetry (58).
coordinates of the different slices on which that seed has Based on these considerations it can be stated that: 1. There is controversy and lack of consensus regarding the ideal time to obtain postoperative dosimetry. The clinical The degree of volumetric enlargement of the prostate significance of obtaining dosimetry at different time in- induced by the multiple needle punctures associated with this procedure has been described. It is presumed that the 2. The most practical postoperative time interval for scan- etiology of this volume increase is the trauma-associated fluid accumulation and bleeding within the gland. Although 3. The most reproducible dosimetric results will be ob- the percent of volume increase has been reported as ranging tained if the scan is performed 1 month postimplant, from 0 to 96%, mean values range from approximately 20% although this may not be practical in all patients.
to 50% (5, 31, 33, 58–60) (Table 2). The broad range of 4. Until the ideal postoperative interval for scanning has values is most likely related to a number of factors, which been determined, each center should perform dosimetric might include biological variation between patients, as well evaluation of prostate implants at a consistent postoper- as differences in experience and technique among brachy- ative interval. This interval should be stated in the do- therapists. There seems, however, to be better agreement on simetry report. It should be kept in mind that dosimetry the rate of resolution of this edema, with reported half-lives obtained from CT scan in the immediate postimplant period will underestimate prostate coverage by about The magnitude dynamics and resolution of edema may 10%, compared to dosimetry obtained from CT scan have obvious implications for the timing at which the dose- volume relationship is described. There are few reports ofchanges in CT-based dosimetry in a serial fashion over time Dosimetric evaluation and reporting postimplant. In the first, Prestidge and colleagues (58) re- Evaluation of postimplant dosimetry is typically carried ported a mean maximum volume increase of 19%. This out in three separate steps: (a) examination of isodose resulted in a 10% underestimate of prostate coverage by the distribution, (b) generation of the dose-volume histogram prescribed dose on postoperative day 1 relative to day 180.
(DVH), and (c) determination of dose uniformity and dose Waterman et al. (5) found a mean volume increase of conformity indices. These three aspects of dosimetric eval- 52% on day 1 relative to preimplant, which resulted in a uation provide complementary information for assessing the mean decrease of approximately 10% in calculated dose coverage. In this report, edema was initially calculated A two-dimensional isodose distribution should be gener- I. J. Radiation Oncology ● Biology ● Physics Fig. 1. (a) Cumulative DVH of dose in percent of D versus volume in percent of the target volume. (b) Differential versus fractional volume in arbitrary scale.The full width at half maximum (FWHM) ated on multiple slices throughout the prostate and in other areas of concern. Outline of the prostate and any adjacent 100%, 90%, and 80% of the prostate, respectively).
critical structures as determined by tomographic imaging should be superimposed on the isodose distribution. Such tional volume of the prostate that receives 200%, 150%, isodose plots offer the most direct assessment of dose cov- 100%, 90%, and 80% of the prescribed dose, respective- erage, because the location of any underdosage in the pros- tate can be evaluated based on supplemental clinical judg- 3. The total volume of the prostate (in cc) obtained from ment. It is recommended that at least the following set of isodose lines be generated as a percentage of the prescrip- 4. The number of days between implantation and the date tion dose: 200%, 150%, 100%, 90%, 80%, and 50%.
of the imaging study used for dosimetric reconstruction.
Generation of the DVH of the prostate is recommended.
The most common format is the cumulative DVH, whichshows the percent volume (or total volume) of the prostate All of the above volumetric parameters are obtainable from that receives greater than or equal to a given dose. A less a single compilation of the DVH. Of these, only D commonly used representation, the differential dose-volume been shown to correlate with PSA-based clinical outcome histogram (DDVH), displays the relative volume of the (11) and should be reported by all. However, in the research prostate that receives a given dose (Fig. 1). The full width at environment, a complete set of dosimetric data should be half maximum (FWHM) of the DDVH is a measure of the collected to facilitate future clinical correlation with respect uniformity of the dose distribution. It is generated on the to local control radiation toxicity and for comparison of DDVH by taking the peak volume value, dividing by two, results between various institutions.
and drawing a horizontal line on the graph. The dose where A number of dose conformity quantifiers exist in the the line first hits the rising curve is subtracted from the dose literature for prostate brachytherapy (64, 69, 70). Of these, represented by the last intersection of the line and the falling the target volume ratio (TVR) is traditionally defined as the curve, giving the FWHM. A larger value implies a wider ratio of the reference dose volume to the target volume. The range of doses or a less uniform dose distribution. A smaller concept of TVR is similar to the historical matched periph- value thus reflects a more uniform dose distribution. The eral dose and has the same limitation of not addressing the geometrical relationship between the target volume and the Typically, the DDVH peaks at a dose that is higher than reference dose volume: a geometrical miss will not be the prescription dose. The spread of the peak is a useful reflected in the TVR value. It is possible to perform an indicator of dose homogeneity (18, 64, 65). A smaller implant with a TVR Ͼ 1.0 (seemingly good), where very spread indicates greater dose uniformity. It is recommended little dose was actually delivered to the prostate (bad) if that a grid size of 2 mm or smaller be used to ensure most of the sources were outside the prostate. A modified adequate resolution of the reported parameters in the dosi- TVR (TVR2), as described by Bice and Prestidge (43), metric calculation (17, 66–68).
takes into account the volume and the location encompassed It is recommended that the following be reported to allow by the reference isodose surface. TVR2 is defined as the adequate evaluation of postimplant dosimetry and to allow reference dose volume divided by the volume of the target that receives the reference dose or greater. While this quan- Prostate postimplant dosimetry ● S. NAG et al. tifier still has some flaws, TVR2 has the advantage of Ͻ 140 Gy, compared to 92% for those with a including the spatial relationship between the target and the dose, but is dependent upon who and how the target is Treatment-related morbidity has also been correlated drawn. Because the clinical target volume in prostate with postimplant dosimetry findings. Wallner et al. (52) brachytherapy is not yet fully understood, this dose confor- analyzed 45 patients treated with 125I implantation who had mity parameter was not found to be a useful enough param- CT-based dosimetry performed 2–4 h after implantation and eter to receive a strong endorsement or a recommendation related these findings to urinary and rectal morbidity. He from the panel. While it may be of value in assessing future found that in patients who developed RTOG grade 0–1 clinical outcomes, it is not required of the community urinary morbidity, an average of 10 mm of urethra was irradiated to doses Ͼ 400 Gy (pre-TG43) compared to 20 Calculation and reporting of dose to the prostatic urethra mm for patients experiencing Grade 2–3 morbidity (p ϭ are important components of dosimetric evaluation. Dose to 0.07). He concluded that both the dose and length of urethra the urethra may be represented in a number of ways. If the irradiated were related to urinary morbidity. Similarly, urethra is adequately visualized (e.g., by catheterization) in when examining rectal morbidity, he found that in patients postimplant imaging, a DVH or dose-surface histogram developing RTOG Grade 1–2 rectal morbidity an average of (DSH) can be generated in addition to point dose calculation 17 mm2 of rectal wall was irradiated to doses Ͼ 100 Gy, at the center of the urethra on each axial slice. Less reliably, compared to 11 mm2 for patients experiencing no rectal the geometric center of the prostate may be used as a surrogate for the location of the urethra, particularly for the Desai et al. (54) analyzed acute urinary morbidity in 117 peripheral loaded implants (55). The urethral dose through- patients treated with 125I implants by correlating urinary out the prostate should be examined on multiple sections.
symptoms as measured by the international prostate symp- The length of urethra receiving Ͼ 200% of the prescribed tom score with findings from CT-based dosimetry per- dose should be recorded to allow correlation with urethral formed 1 month after implantation. She found that the highest symptom score in each patient correlated with the Similarly, dose to the anterior rectal wall is an important following dose descriptions of the prostate: D component of postimplant evaluation. Rectal dose may be represented in a DSH or DVH within an annulus that approximates the anterior rectal wall (56). Alternatively, for doses delivered to 5 cm2 of urethra, as measured by DSH.
simplicity, point dose sampling along the anterior rectal The conclusion of this analysis was that attempts at reduc- wall may be used. Again, the rectal dose should be recorded ing urethral doses can translate into reduced urinary symp- toms, and that trials of prostate dose escalation may be It is recognized that detailed DVH analysis may be very limited by the acute urinary symptoms (54).
labor intensive and may not be supported by all treatmentplanning systems at present. Therefore, it may not be prac- DISCUSSION
tical to report all of the above dosimetric parameters in thecommunity setting. However, it is recommended that at a Postimplant dosimetry of the prostate is a constantly minimum, postimplant dosimetry be performed and the D evolving dynamic field. The above recommendations rep- reported at all centers, and that all the other parameters be resent the current consensus opinion of the ABS. Because of additionally obtained in a research environment.
the current paucity of published data, there are areas ofcontroversy that cannot be resolved. For example, the ideal time for obtaining the postimplant dosimetry or an exact The data collected from dosimetric analysis is relevant in dose/volume recommendation to the urethra or rectum can- that it has been shown to correlate with treatment outcomes.
not currently be identified. The panel identified a number of Historically, measures of implant quality of retropubic pros- other parameters that should be considered for further de- tate brachytherapy have been related to disease control velopment and refinement of the dosimetric process.
Currently, dosimetric analysis is performed after the im- Stock et al. analyzed the results of CT-based postimplant plant has been completed. This does not provide a mecha- dosimetry (using TG43 guidelines) performed 1 month after nism for correction if suboptimal dose distribution is ob- implantation in 134 patients treated with 125I implants for tained. Ideally, one should strive for on-line real-time T1 to T2 prostate cancer over a 6-year period. This study intraoperative dosimetry to allow for adjustment in seed correlated dosimetric findings with PSA control and nega- placement to achieve the intended dose. Current ultrasound technology must be improved to localize the individual seed 100–119.9 Gy, 120–139.9 Gy, 140–159.9 Gy, and Ն 160 position within the prostate, and isodose calculations must Gy were associated with improved freedom from PSA fail- be rapidly performed on-line and updated as subsequent ure rates of 53%, 82%, 80%, 95%, and 89%, respectively seeds are implanted. Correlation of the resultant implant (p ϭ 0.02) at 4 years. A dose cutoff point was found at 140 dose distribution to the clinical outcome has yet to be Gy, with PSA control rates of 68% for those patients re- I. J. Radiation Oncology ● Biology ● Physics The dose distribution in a prostate implant is very inho- complex to be practical for the practicing community radi- mogeneous. The degree of dose heterogeneity varies from ation oncologist, and may be more relevant for the larger implant to implant. The tumor control probability (TCP) brachytherapy centers planning to compare their outcome depends on the degree of heterogeneity in addition to the results. This differentiation has been mentioned in the rel- prescribed dose (74). For example, in two implants with the same D , the dose may be much higher (or lower) in some regions of one than in similar regions of the other. The implant with the more heterogeneous dose may have agreater TCP, because parts of the tumor will receive a dose The ABS recommends that postimplant dosimetry should be that is much higher than the prescribed dose. The therapeu- performed on all patients undergoing permanent prostate tic advantage and tradeoff of dose heterogeneity are not yet brachytherapy for optimal patient care. At present, CT-based adequately documented for the purpose of clinical correla- dosimetry is recommended based on availability, cost and the ability to image the prostate as well as the seeds. Additional Another factor to be considered is the presence of large plane radiographs should be obtained to verify the seed count.
prostate calcifications that can affect the dose delivered. The Until the ideal postoperative interval for CT scanning has been higher atomic number of calcium (z ϭ 20) compared to that determined, each center should perform dosimetric evaluation of prostatic tissue (z ϭ 7.6) leads to a greater absorbed dose of prostate implants at a consistent postoperative interval. This increased attenuation and increased dose deposition at the interval should be reported. Isodose displays should be ob- calcium/soft tissue interface. Interseed effects may also tained at 50%, 80%, 90%, 100%, 150%, and 200% of the adversely affect the dose distribution, because the seeds, prescription dose and displayed on multiple cross-sectional being denser than tissue, will absorb some of the radiation images of the prostate. A DVH of the prostate should be from other seeds (75, 76). The actual effect of these heter- ogeneities on the dose distribution needs further investiga- rectal, and urethral doses should be reported and ultimately Finally, these recommendations are intended to be advi- correlated with clinical outcome at larger centers. On-line, sory in nature; the responsibility for the medical decisions real-time dosimetry, the effects of dose heterogeneity, and the ultimately rests with the treating physician who has to effects of tissue heterogeneity need further investigation. These consider the cost-benefit ratio. We also recognize that some recommendations should be a practical guide for performing of the recommendations given in this report may be too postimplant dosimetry for permanent prostate brachytherapy.
REFERENCES
1. Prete JJ, Prestidge BR, Bice WS, et al. A survey of physics ment deviations from pre-planned positions in ultrasound and dosimetry practice of permanent prostate brachytherapy in guided prostate implants. Radiother Oncol 1994;32:268–270.
the United States. Int J Radiat Oncol Biol Phys 1998;40:1001– 9. Narayana V, Roberson PL, Winfield RJ, et al. Optimal place- ment of radioisotopes for permanent prostate implants. Radi- 2. Nag S, Baird M, Blasko J, et al. American Brachytherapy ology 1996;199:457–460.
Society (ABS) survey of current clinical practice for perma- 10. Roberson PL, Narayana V, McShan DL, et al. Source place- nent brachytherapy of prostate cancer. J Brachyther Int 1997; ment error for permanent implant of the prostate. Med Phys 3. Nag S, Beyer D, Friedland J, et al. American Brachytherapy 11. Stock RG, Stone N, Tabert A, et al. A dose-response study for Society (ABS) recommendations for transperineal permanent I-125 prostate implants. Int J Radiat Oncol Biol Phys 1998; brachytherapy of prostate cancer. Int J Radiat Oncol Biol 12. Lerner SE, Blute ML, Zincke H. Critical evaluation of salvage surgery for radiorecurrent/resistant prostate cancer. J Urol 4. Yu Y, Waterman FM, Suntharalingam N, et al. Limitations of the minimum peripheral dose as a parameter for dose speci- 13. Stock RG, Stone NN. Salvage brachytherapy for recurrent fication in permanent 125I prostate implants. Int J Radiat prostate cancer following radiation therapy. In: Zelefsky, Oncol Biol Phys 1996;34:717–725.
Greco C, editors. Radiotherapy of prostate cancer. United 5. Waterman FM, Yue N, Corn BW, et al. Edema associated Kingdom: Harwood Academic. In press.
with I-125 or Pd-103 prostate brachytherapy and its impact on 14. Grado GL, Collins JM, Kriegshauser JS, et al. Salvage brachy- post-implant dosimetry: An analysis based on serial CT ac- therapy for localized prostate cancer after radiotherapy failure.
quisition. Int J Radiat Oncol Biol Phys 1998;41:1069–1077.
Urology 1999;53:2–10.
6. Willins J, Wallner K. CT-based dosimetry for transperineal 15. Prestidge BR, Bice WS, Prete JJ, et al. A dose-volume anal- I-125 prostate brachytherapy. Int J Radiat Oncol Biol Phys ysis of permanent transperineal prostate brachytherapy. (Ab- str.) Int J Radiat Oncol Biol Phys 1997;39:289.
7. Bice WS, Prestidge BR, Grimm PD, et al. Centralized multi- 16. Sharkey J, Chovnick SD, Behar RJ, et al. Outpatient ultra- institutional post-implant analysis for interstitial prostate sound-guided palladium-103 brachytherapy for localized ad- brachytherapy. Int J Radiat Oncol Biol Phys 1998;41:921– enocarcinoma of the prostate: A preliminary report of 434 patients. Urology 1998;51:796–803.
8. Dawson JE, Wu T, Roy T, et al. Dose effects of seed place- 17. Stone NN, Stock RG, DeWyngaert JK, et al. Prostate brachy- Prostate postimplant dosimetry ● S. NAG et al. therapy: Improvements in prostate volume measurements and 38. Nag S, Scaperoth DD, Badalament R, et al. Transperineal dose distribution using interactive ultrasound guided implan- palladium-103 prostate brachytherapy: Analysis of morbidity tation and three-dimensional dosimetry. Radiat Oncol Invest and seed migration Urology 1995;45:87–92.
39. Tapen EM, Blasko JC, Grimm PD, et al. Reduction of radio- 18. Roy JN, Wallner KE, Harrington PJ, et al. A CT-based eval- active seed embolization to the lung following prostate uation method for permanent implants: Application to pros- brachytherapy. Int J Radiat Oncol Biol Phys 1998;42:1063– tate. Int J Radiat Oncol Biol Phys 1993;26:163–169.
19. Altschuler MD, Findlay PA, Epperson RD. Rapid accurate 40. Steinfeld AD, Donahue BR, Plaine L. Pulmonary emboliza- three-dimensional location of multiple seeds in implant radio- tion of iodine-125 seeds following prostate implantation.
therapy treatment planning. Phys Med Biol 1983;28:1305– 41. Nag S, Vivekanandam S, Martinez-Monge R. Pulmonary em- 20. Siddon RL, Chin LM, Two-film brachytherapy reconstruction bolization of permanently implanted radioactive palladium- algorithm. Med Phys 1985;12:77–83.
103 seeds for carcinoma of the prostate. Int J Radiat Oncol 21. Metz C, Fencil LE. Determination of three dimensional struc- Biol Phys 1997;39:667–670.
ture in biplane radiography with prior knowledge of the rela- 42. Stock RG, Stone NN, Wesson MF, et al. A modified technique tionship between the two views: Theory. Med Phys 1989;16: transperineal prostate implantation. Int J Radiat Oncol Biol 22. Hughes HA. Accuaracy of foreign body localization from tube Phys 1995;32:219–225.
shift radiographs. Br J Radiol 1956;29:116–118.
43. Bice WS, Prestidge BR. A review of postimplant quality 23. Rosen II, Khan KM, Lane RG, et al. The effect of geometric assessment in permanent transperineal interstitial prostate errors in the reconstruction of iridium-192 seed implants. Med brachytherapy. J Brachyther Int 1997;13:297–313.
Phys 1982;9:220–223.
44. Badiozamani KR, Wallner K, Cavanagh W, et al. Compara- 24. Fitzgerald LT, Mauderli W. Analysis of errors in three-dimen- bility of CT-based and TRUS-based prostate volumes. Int J sional reconstruction of radium implants from stereo radio- Radiat Oncol Biol Phys 1999;43:375–378.
graphs. Radiology 1975;115:455–458.
45. Nath R, Anderson LL, Luxton G, et al. Dosimetry of intersti- 25. Amols HI. Rosen II. A three-film technique for reconstruction tial brachytherapy sources: Recommendations of the AAPM of radioactive seed implants Med Phys 1981;8:210–214.
Radiation Committee Task Group No. 43. Med Phys 1995;22: 26. Biggs PJ, Kelley DM. Geometric reconstruction of seed im- plants using a three-film technique. Med Phys 1983;10:701– 46. Luse RW, Blasko J, Grimm P. A method for implementing the American Association of Physicists in Medicine Task 27. Altschuler MD, Kassaee A. Automatic matching of corre- Group-43 dosimetry recommendations for 125I transperineal sponding seed images of three simulator radigraphs to allow prostate seed implants on commercial treatment planning sys- 3-D triangulation of implanted seeds. Phys Med Biol 1997;42: tems. Int J Radiat Oncol Biol Phys 1997;37:737–741.
47. Bice WS, Prestidge BR, Prete JJ, et al. Clinical impact of 28. Brinkman DH. Kline RW. Automated seed localization from implementing the recommendations of AAPM Task Group 43 CT data sets of the prostate. Med Phys 1998;25:1667–1672.
permanent prostate brachytherapy using 125I. Int J Radiat 29. Bice WS, Dubois DF, Prete JJ, et al. Source localization from Oncol Biol Phys 1998;40:1237–1241.
axial image sets by iterative relaxation of the nearest neighbor 48. Williamson J, Coursey B, DeWerd L, et al. Dosimetric pre- criterion. Med Phys Submitted.
requisites for routine clinical use of new low energy photon 30. Dubois DF, Prestidge BR, Hotchkiss LA, et al. Intraobserver interstitial brachytherapy sources. Med Phys 1998;25:2269– and interobserver variability of MR imaging and CT-derived prostate volumes after transperineal interstitial permanent 49. Yu Y, Anderson LL, Li Z, et al. Permanent prostate seed prostate brachytherapy. Radiology 1998;207:785–789.
implant brachytherapy: Report of the American Association of 31. Moerland MA, Wijrdeman HK, Beersma R, et al. Evaluation Physicists in Medicine Task Group No. 64. Med Phys. In of permanent I-125 prostate implants using radiography and magnetic resonance imaging. Int J Radiat Oncol Biol Phys 50. Anderson LL. Brachytherapy planning and evaluation. En- docurie Hyperther Oncol 1991;7:139–146.
32. Dubois DF, Prestidge BR, Hotchkiss LA, et al. Source local- 51. Roach M, Faillace-Akazawa P, Malfatti C, et al. Prostate ization following permanent transperineal prostate interstitial volumes defined by magnetic resonance imaging and comput- brachytherapy using magnetic resonance imaging. Int J Radiat erized tomographic scans for three-dimensional conformal Oncol Biol Phys 1997;39:1037–1041.
radiotherapy. Int J Radiat Oncol Biol Phys 1996;35:1011– 33. Narayana V, Roberson PL, Pu AT, et al. Impact of differences in ultrasound and computed tomography volumes on treat- 52. Wallner K, Roy J, Harrison L. Dosimetry guidelines to min- ment planning of permanent prostate implants. Int J Radiat imize urethral and rectal morbidity following transperineal Oncol Biol Phys 1997;37:1181–1185.
I-125 prostate brachytherapy. Int J Radiat Oncol Biol Phys 34. Narayana V, Roberson PL, Winfield RJ, et al. Impact of ultrasound and computed tomography prostate volume regis- 53. DeWyngaert JK, Wesson MF. Treatment planning consider- tration on evaluation of permanent prostate implants. Int J ations of ultrasound guided I-125 prostate implants. Int J Radiat Oncol Biol Phys 1997;39:341–346.
Radiat Oncol Biol Phys 1991;21(Suppl. 1):160.
35. Dubois DF, Bice WS, Prestidge BR, et al. CT and MRI 54. Desai J, Stock RG, Stone NN, et al. Acute urinary morbidity derived source localization error in a custom prostate phantom following iodine-125 interstitial implantation of the prostate using automated image coregistration. Med Phys. In press.
gland. Radiat Oncol Invest 1998;6:135–141.
36. Amdur R, Gladstone D, Leopold K, et al. Prostate seed im- 55. Waterman FM, Dicker AP. Determination of the urethral dose plant quality assessment using MR and CT image fusion. Int in prostate brachytherapy when the urethra is not visualized in J Radiat Oncol Biol Phys 1999;43:67–72.
the post-implant CT scan. (Abstr.) Med Phys. In press.
37. Bice WS, Freeman JE, Russell J, et al. Coregistration of 56. Lu Y, Li S, Spelbring D, et al. Dose-surface histograms as post-implant ultrasound and CT image sets in permanent pros- treatment planning tool for prostate conformal therapy. Med tate brachytherapy. Int J Radiat Oncol Biol Phys. Submitted.
Phys 1995;22:279–284.
I. J. Radiation Oncology ● Biology ● Physics 57. Zelefsky MJ, Happersett L, Leibel S, et al. The effect of 66. Anderson L. Brachytherapy planning and evaluation. Endocu- treatment positioning on normal tissue dose in patients with rie Hypertherm Oncol 1991;7:139–146.
prostate cancer treated with three-dimensional conformal ra- 67. Paul J, Philip P, Brandenburg R, et al. Histograms in brachy- diotherapy. Int J Radiat Oncol Biol Phys 1997;37:13–19.
therapy. Endocurie Hypertherm Oncol 1991;7:13–26.
58. Prestidge BR, Bice WS, Kiefer EJ, et al. Timing of computed 68. Waterman F, Strubler K. Absorbed dose determination for tomography-based postimplant assessment following perma- interstitial 125I boost therapy. Med Phys 1983;10:155–158.
nent transperineal prostate brachytherapy. Int J Radiat Oncol 69. van’t Riet A, Mak AC, Moerland MA, et al. A conformation Biol Phys 1998;40:1111–1115.
number to quantify the degree of conformality in brachyther- 59. Merrick GS, Butler WM, Dorsey AT, et al. Influence of timing apy and external beam irradiation: Application to the prostate. on the dosimetric analysis of transperineal ultrasound-guided Int J Radiat Oncol Biol Phys 1997;37:731–736.
prostatic conformal brachytherapy. Radiat Oncol Invest 1998; 70. Saw CB, Suntharalingam N. Quantitative assessment of inter- stitial implants. Int J Radiat Oncol Biol Phys 1991;20:135– 60. Yue N, Nath R, Chen Z, et al. Optimum timing for CT-based dose evaluation of I-125 and Pd-103 prostate seed implants.
71. Zelefsky MJ, Whitmore WF Jr. Long-term results of retropu- (Abstr.) Int J Radiat Oncol Biol Phys 1998;42:134.
bic permanent 125iodine implantation of the prostate for clin- 61. Yue N, Dicker AP, Corn BW, et al. A dynamic model for the ically localized prostatic cancer. J Urol 1997;158:23–29.
estimation of optimum timing of computed tomography scan 72. Nath R, Roberts K, Ng M, et al. Correlation of medical for dose evaluation of 125I or 103Pd seed implant of prostate.
dosimetry quality indicators to the local tumor control in Int J Radiat Oncol Biol Phys 1999;43:447–454.
patients with prostate cancer treated with iodine-125 intersti- 62. Moerland MA. The effect of edema on post-implant dosimetry tial implants. Med Phys 1998;25:2293–2307.
of permanent iodine-125 prostate implants: A simulation 73. Fuks Z, Leibel SA, Wallner KE, et al. The effect of local study. J Brachyther Int 1998;14:225–231.
control on metastatic dissemination in carcinoma of the pros- 63. Willins J, Wallner K. Time-dependent changes in CT-based tate: Long-term results in patients treated with 125I implanta- dosimetry of I-125 prostate brachytherapy. Radiat Oncol In- tion. Int J Radiat Oncol Biol Phys 1991;21:537–547.
vest 1998;6:157–160.
74. Ling CC, Roy J, Sahoo N, et al. Quantifying the effect of dose 64. Yu Y. A non-divergent specification of the mean treatment inhomogeneity in brachytherapy: Application to permanent dose in interstitial brachytherapy. Med Phys 1996;23:905– prostatic implant with 125I seeds. Int J Radiat Oncol Biol Phys 65. van’t Riet A, Te Loo HJ, Ypma AF, et al. Ultrasonically 75. Burns G, Raeside D. The accuracy of single-seed dose super- guided transperineal seed implantation of the prostate: Mod- position of I-125 implants. Med Phys 1989;16:627–631.
ification of the technique and qualitative assessment of im- 76. Meigooni A, Meli J, Nath R. Interseed effects on dose for 125I plants. Int J Radiat Oncol Biol Phys 1992;24:555–558.
brachytherapy implants. Med Phys 1992;19:385–390.

Source: http://www.brachiterapia.it/ita/elementi/implantdosim.pdf