A COMPARISON BETWEEN TANDEM AND OVOIDS AND INTERSTITIAL GYNECOLOGIC TEMPLATE BRACHYTHERAPY DOSIMETRY USING A HYPOTHETICAL COMPUTER MODEL

PII S0360-3016(01)02691-8 Int. J. Radiation Oncology Biol. Phys., Vol. 52, No. 2, pp. 538 543, 2002 Copyright 2002 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/02/$ see front matter PHYSICS CONTRIBUTION A COMPARISON BETWEEN TANDEM AND OVOIDS AND INTERSTITIAL GYNECOLOGIC TEMPLATE BRACHYTHERAPY DOSIMETRY USING A HYPOTHETICAL COMPUTER MODEL I-CHOW J. HSU, M.D.,* JOYCELYN SPEIGHT, M.D., PH.D.,* JENNY HAI, PH.D., ERIC VIGNEAULT, M.D., THEODORE PHILLIPS, M.D.,* AND JEAN POULIOT, PH.D.* *Department of Radiation Oncology, University of California at San Francisco, San Francisco, CA; Department of Radiation Oncology, Stanford University, Palo Alto, CA; Department of Radiation Oncology, Centre Hospitalier Universitaire de Quebec (CHUQ) Hotel-Dieu of Quebec, Quebec City, Quebec, Canada Purpose: To evaluate the dose distribution within the clinical target volume between two gynecologic brachytherapy systems the tandem and ovoids and the Syed-Neblett gynecologic template using a hypothetical computer model. Methods and Materials: Source positions of an intracavitary system (tandem and ovoids) and an interstitial system (GYN template) were digitized into the Nucletron Brachytherapy Planning System. The GYN template is composed of a 13-catheter implant (12 catheters plus a tandem) based on the Syed-Neblett gynecologic template. For the tandem and ovoids, the dwell times of all sources were evenly weighted to produce a pear-shaped isodose distribution. For the GYN template, the dwell times were determined using volume optimization. The prescribed dose was then normalized to point A in the intracavitary system and to a selected isodose line in the interstitial system. The treated volume in the two systems was kept approximately the same, and a cumulative dose volume histogram of the treated volume was then generated with the Nucletron Brachytherapy Planning System to use for comparison. To evaluate the dose to a hypothetical target, in this case the cervix, a 2-cm-long, 3-cm-diameter cylinder centered along the tandem was digitized as the clinical target volume. The location of this hypothetical cervix was based on the optimal application of the brachytherapy system. A visual comparison of clinical target coverage by the treated volume on three different orthogonal planes through the treated volume was performed. The percentage dose volume histograms of the target were generated for comparison. Multiple midline points were also placed at 5-mm intervals away from the tandem in the plane of the cervix to simulate the location of potential bladder and rectal dose points. Doses to these normal structures were calculated for comparison. Results: Although both systems covered the hypothetical cervix adequately, the interstitial system had a better coverage of the region lateral to the cervix. Smaller volumes of the vagina and uterine fundus received the full dose from the interstitial implant. The cumulative dose volume histograms revealed larger high-dose regions within the treatment volume for the intracavitary system. The volumes receiving >180% of the prescription dose were 31 cc and 17 cc for the intracavitary system and interstitial system, respectively. The isodose lines showed that most of this difference results from the high-dose region around the tandem. The percentage dose volume histograms showed that a larger percentage of cervix received a higher dose in the intracavitary system. Fifty-two percent of the target volume received 200% or higher of the prescription dose with tandem and ovoids, compared with only 20% with the template system. Analysis of dose points outside of the 100% isodose lines showed a slightly more rapid dose drop-off with the interstitial system compared to the intracavitary system. Point doses at 20, 25, and 30 mm from the tandem in the interstitial system were 100%, 69%, and 51% of prescribed dose, and from the intracavitary system were 101%, 76%, and 58%, respectively. Conclusions: Our dosimetric analysis revealed a better coverage in the parametrial regions, but underdosage of the central cervical region, for the interstitial system. On the other hand, because of the increased distance of source to dose point, there is a more rapid dose drop-off outside the treated volume with the interstitial system, which has the potential to improve tissue sparing. Based on this analysis, we caution against using a radiotherapy system with a homogenous central dose distribution when treating cervical cancer with an intact uterus. We recommend differential loading of the implant catheters with the majority of dose delivered from the tandem when using an interstitial GYN template with remote afterloader. 2002 Elsevier Science Inc. Brachytherapy, Interstitial implant, GYN template. Reprint requests to: I-Chow Hsu, M.D., Department of Radiation Oncology, 1600 Divisadero Street, Suite H1031, San Francisco, CA 94143-1708. Tel: (415) 353-7175; Fax: (415) 353-8980; E-mail: hsu@radonc17.ucsf.edu Received Mar 1, 2001, and in revised form Jul 17, 2001. Accepted for publication Jul 23, 2001. 538

A comparison of dosimetry in two gynecological brachytherapy systems I-C. J. HSU et al. 539 INTRODUCTION The intracavitary system using the tandem and ovoids is an integral part of radiation therapy for treatment of gynecologic malignancies. The classic Fletcher-Suit intracavitary applicator system includes an intrauterine tandem and intravaginal ovoids. Using the standard loading, this system produces a pear-shaped high-dose region centered around the cervix. This elegant brachytherapy system allows a very high dose to be delivered to the cervix while sparing the bladder and bowel, located in its proximity. The flexibility of this system allows it to be tailored to a variety of different types of patient anatomy. Because of this remarkable brachytherapy system, even large cervical tumors can be controlled using radiotherapy alone. Interstitial brachytherapy has the potential for delivery of brachytherapy treatment to tumor located away from an accessible anatomic cavity. A variety of template systems have been developed for the treatment of pelvic malignancies (1, 2). These template systems provide guidance for the insertion of implant catheters and security once the catheters are inserted. Using these templates, implant catheters can be evenly placed in the clinical target volume with or without additional imagery guidance (3 11). In the treatment of cervix cancer, the Syed-Neblett interstitial gynecologic template system has been used for patients with advanced-stage cervical cancer. The interstitial system is applied most commonly in an attempt to increase the dose to the region outside the standard pear-shaped isodose distribution, such as the parametrial, paravaginal, or paraurethral regions. In some series in the literature, encouraging results have been reported using the interstitial system (2, 5, 6, 11 15). However, most of these series also reported a high incidence of complications. Some authors have postulated that the high rate of complication is related to the high dose to rectum and bladder resulting from loading the tandem and the central 6 catheters around the vaginal obturator. This has led some to recommend unloading the central 6 catheters around the vaginal obturator when the tandem is loaded. Others have proposed differential manual loading of the catheters around the vaginal obturator to minimize the high-dose volumes. With the development of computer-controlled remote afterloading brachytherapy systems, such as the Nucletron Microseletron remote afterloading unit, the iridium source position and dwell times can be controlled using a computer-controlled stepping motor. These systems allow flexible variations in the loading of the implant catheters. Furthermore, computerized geometric optimization algorithms can be used to produce a more homogeneous dose distribution within the treated volume and to improve coverage of the clinical target volume (16). With these new capabilities, a better understanding of the dosimetry of these brachytherapy systems is needed. In this study, we will compare the treatment plans produced by the interstitial gynecologic template and the traditional intracavitary system, using a computer model. The applicator and source positions used in this study are based on a hypothetical patient with ideal anatomy. We will use this system to illustrate the difference in dosimetry between the two systems. This comparison will illustrate the potential shortfalls and possible advantages of each system. We hope a better understanding of the potential advantages of each system might lead to better treatment planning and optimization protocols in the future. METHODS AND MATERIALS Source positions of an intracavitary system (tandem and ovoids) and an interstitial system (GYN template) were digitized into the Nucletron Brachytherapy Planning System version 11.4. The GYN template is composed of a 13- catheter implant (12 catheters plus a tandem) based on the Syed-Neblett gynecologic template #3 made by Alpha Omega (#712). In the intracavitary system, the dwell positions of the ovoids are separated by 4 cm with a 15-degree tilt from the vertical. The dwell positions of the tandem are located in a plane that bisects the plane of the dwell positions of the ovoids. To simplify the comparison, a straight tandem with 5 cm of active length was used for each brachytherapy system. The dwell positions were activated every 5 mm along each channel. In the interstitial system, there were 13 active dwell positions in the tandem and 108 active dwell positions in the peripheral catheters. In the intracavitary system, there were five active dwells in each ovoid and 13 active dwells in the tandem. The dose calculations were performed without accounting for the internal shielding in the ovoids. The dwell times of all sources were evenly weighted to produce a pear-shaped isodose distribution. No additional manipulation of the dwell time was done to change the shape of the pear-shaped isodose. For the template, the dwell times were determined using geometric volume optimization (17, 18). The result of the geometric optimization improved the overall coverage of the implant and decreased dose heterogeneity in the treated volume (16). The prescribed dose was then normalized to the average point A dose in the intracavitary system. In the interstitial system, the prescription dose was normalized to the selected isodose surface to cover the planning target volume (19). Incidentally, the actual volume covered by this selected isodose surface was very close to the treated volume created by the intracavitary system. Because this simplifies the percent dose volume histogram (DVH) comparison, the prescription dose was normalized to the isodose line created by the interstitial system. The dimensions of each treated volume were measured and recorded for comparison. The isodose lines were also plotted for visual comparison of the treatment volume. The cumulative DVH of the treated area was generated using the Nucletron Brachytherapy Planning System and used for comparison. To evaluate the dose to a hypothetical target, in this case the cervix, a 2-cm-long, 3-cm-diameter cylinder centered along the tandem was digitized as the target volume. The location of this hypothetical cervix was based on an optimal application of the brachytherapy system. The percent dose

540 I. J. Radiation Oncology Biology Physics Volume 52, Number 2, 2002 Table 1. Dimension of the prescription isodoses measured at the central axis Numbers of dwell positions Dwell times Height Width Thickness Tandem and ovoids 23 Uniform loading 7.3 cm 6.0 cm 3.8 cm GYN template 121 Geometrically optimized 6.0 cm 6.4 cm 3.9 cm volume histogram to the cervix was then calculated for each system. Multiple midline points were also plotted at 20, 25, and 30 mm away from the tandem in the inferior plane of the hypothetical cervix to simulate locations of potential bladder and rectal dose points. Doses to these points were then calculated and plotted for comparison. RESULTS Treated volume comparison The dimensions of the 100% isodose volume were 7.3 cm 6.0 cm 3.8 cm for the tandem and ovoids and 6.0 cm 6.4 cm 3.9 cm for the GYN template (Table 1). Visual comparison of the isodose volumes showed the cervix to be well covered by both brachytherapy systems (Fig. 1). However, the paracervical and parametrial extensions are better covered with the interstitial system. The opposite is true in the area around the fundus of the uterus. In this comparison, the upper vaginal area was better spared by the interstitial system. DVH comparison A comparison of the cumulative dose volume histograms revealed a significantly larger high-dose volume in the intracavitary system. The cumulative dose volume histograms showed that the volumes receiving 180% of prescription dose were 31 cc and 17 cc for the intracavitary system and interstitial system, respectively. Visual inspection of the isodose curves showed that most of this difference is due to the high-dose volume around the tandem. The percentage dose volume histogram provided additional evidence supporting this observation. A larger percentage of cervix received a higher dose in the intracavitary system (Fig. 2). Fifty-two percent of the target volume received 200% or more of the prescription dose with tandem and ovoids, whereas with the Fig. 1. Isodose plots of the intracavitary and interstitial brachytherapy systems. The isodose lines of each system were aligned by their origin and superimposed on each other for comparison.

A comparison of dosimetry in two gynecological brachytherapy systems I-C. J. HSU et al. 541 Fig. 2. Dose volume histogram of hypothetical cervix. template system, only 20% of the target volume received 200% or more of the prescription dose. Dose drop-off comparison Analysis of the dose received by the hypothetical bladder and rectal points showed a slightly more rapid dose drop-off with the interstitial system compared with the intracavitary system. Points at 15, 20, and 25 mm from the tandem in the interstitial system were receiving 100%, 69%, and 51% of the prescription dose, and in the intracavitary system they received 101%, 76%, and 58%, respectively (Fig. 3). DISCUSSION The interstitial GYN template may be a more flexible system that allows for better coverage of the cervix in patients with limited anatomy. Although this may be true, our comparison of the dosimetry also revealed some potential shortfalls of the interstitial system. In our comparison of the treated volumes, we have demonstrated the potential for the interstitial system to cover parametrial and paracervical tissues beyond the high-dose volume created by the intracavitary system. Without a significant increase in the thickness of the treated volume, the interstitial implant was able to deliver dose to regions more lateral to the cervix than the intracavitary system could. However, our study was limited by using a comparison of only one source-loading combination. The treated volume of the intracavitary system can be easily modified by increasing the size of ovoids or by altering the loading of the applicator (20). However, ultimately the patient s anatomy and the vaginal mucosal tolerance limit the volume that can be treated with the intracavitary system. Similarly, we did not fully exploit the potential of the interstitial system in this study. Coverage of the uterine fundus can be improved by advancing the central 6 catheters. Activation of additional dwell positions can treat more paravaginal and paraurethral tissue. The interstitial implant is a flexible system, but there are also additional risks for morbidity to bowel, because more catheters and active dwell positions are placed inside of the pelvis. The goal of this study is not to compare all the potential applications of the

542 I. J. Radiation Oncology Biology Physics Volume 52, Number 2, 2002 Fig. 3. Hypothetical dose points were placed along the central axis of the cervix to simulate bladder and rectal points. Dose measured at these points showed different rates of dose drop-off between the two brachytherapy systems. two systems, but to illustrate the intrinsic differences and possible pitfalls of each system. Our calculations showed that there is poor coverage at the fundus of the hypothetical uterus using our geometrically optimized interstitial implant. This is partially because of the positioning of catheters, and it is a potential limitation of the interstitial system. If the catheters are inserted too deeply or are loaded mistakenly, there is a greater risk of bowel damage. Some innovative techniques have been described, including transrectal ultrasound, fluoroscopy, computerized tomography (CT), and laparoscopy guidance to place the catheters (3, 7, 9, 10). These techniques may decrease the risk of mechanical trauma to bowel, but a CT anatomy-based brachytherapy planning system is still needed to minimize the dose delivered to the bowel or bladder. We used a straight tandem in our analysis, but in reality the uterus is almost never directly superior to the vaginal canal, and a curved tandem is almost always used. This creates additional problems for the GYN template system, because most catheters are straight and rigid. The insertion and loading of the central 6 catheters around the vaginal obturator in a Syed-Neblett template must be done with great care, because they are most likely to be located near small bowel or the sigmoid colon. A simple solution may be to eliminate the central 6 catheters and deliver most of the central dose via the intrauterine tandem. In the DVH comparison, the high-dose volume was found around the tandem in the intracavitary system. Using the DVH of the cervix, we showed that this high-dose volume represents a significantly larger volume of the cervix receiving a much higher dose. Although dose heterogeneity in the target area is generally avoided in external beam radiotherapy planning, this ultra-high-dose region in our plan might be beneficial in the treatment of cervical cancer. The safety of this concentric high-dose region in the uterus is well documented in the literature. Tandem- and ovoid-type applicators have been the backbone of successful gynecologic radiotherapy (21 23). Furthermore, the necrotic and potentially radioresistant region of cervical cancer is often centrally located around the os, and therefore this centrally located high-dose region is justified, if not required. One potential down side of the high-dose sleeve around the tandem is in a situation where the uterine canal is eccentrically located. This is possible when the cervix is destroyed by the tumor or when the tumor or fibroid displaces the uterine tandem. If the canal is significantly displaced, the high-dose region can fall on an undesirable area. For example, the tandem may be pushed away from the tumor and closer to the bladder or the rectum. In these suboptimal situations, if imaging information from ultrasound, CT, or MR is available, the dose delivered by tandem and ovoids can be modified to compensate. The incidence and significance of this type of suboptimal geometry is unknown. However, based on the well-documented safety and efficacy records of the tandem and ovoid system, we caution against using a radiotherapy system with a homogenous central dose distribution when treating cervical cancer with an intact uterus. This may seem ironic at first, because most of the modern brachytherapy and radiotherapeutic optimization routines strive to decrease dose heterogeneity within the treated volume. However, we believe this is one occasion where we may be better off with the older system, or may at least learn from the features of the older system, before developing a better, more sophisticated optimization program. In our analysis of dose to the hypothetical bladder and rectal points, we found that the interstitial implant system produced a steeper dose falloff outside the implant volume. This illustrates the benefit of inverse square law and improved depth dose with distance. The difference between the dose falloff in the two systems is a reflection of the relationship between the source, the prescription point, and the normal structure. In the interstitial system, the sources were closer to the bladder or the rectal points outside of the prescription isodose surface than in the intracavitary system. Because the dose falloff is steeper closer to the source, the dose to bladder and rectum was lower for the interstitial system. By having the active source closer to the border of the target volume, the interstitial implant takes full advantage of the steeper dose dropoff near the source, because of the inverse square law. In addition to the inverse square law, internal shielding in the ovoids can significantly decrease the dose to bladder and rectum. We did not include the effect of shielding in our calculations, because there are a variety of shielded ovoids available in the market. The quality and effects of these shields must be individually calibrated and recorded in the brachytherapy planning system to produce accurate dosimetric information. As with any type of radiotherapy technique, the objective of

A comparison of dosimetry in two gynecological brachytherapy systems I-C. J. HSU et al. 543 a treatment system is to cover the intended treatment volume with high dose and minimize the dose to the nearby normal structures. Although the interstitial system may be less limited by the anatomy, the underlying requirement of the interstitial system is that implant catheters have to be placed accurately in the target volume. This is not always straightforward, because the catheters also have to be placed away from normal structures, such as the bladder, rectum, and ureters. Furthermore, because of the bony anatomy of the pelvis, the implant catheters often need to be placed in a nonparallel fashion. Some innovative techniques have been described, including transrectal ultrasound, fluoroscopy, CT, and laparoscopy guidance. However, these techniques still require significant amounts of equipment, technical skill, and effort. Alternatively, some would compensate for the limitation of the intracavitary system by adding a parametrial boost using external beam radiotherapy or CT-based conformal radiotherapy. At our institution, the indication for interstitial implant is based on the relationship between the tumor extension and the size of the vaginal vault, based on physical examination and CT/MR imaging studies. In the era of chemoradiotherapy, better response and smaller residual tumors are often seen at the time of brachytherapy application (24 26). This could mean fewer patients needing interstitial brachytherapy. However, for those patients in whom interstitial implant is indicated, application of hard-won clinical wisdom, along with improvement in treatment planning and technique, might decrease complications and improve local control. REFERENCES 1. Martinez A, Cox RS, Edmundson GK. A multiple-site perineal applicator (MUPIT) for treatment of prostatic, anorectal, and gynecologic malignancies. Int J Radiat Oncol Biol Phys 1984;10(2):297 305. 2. Ampuero F, Doss L, Khan M, et al. The Syed-Neblett interstitial template in locally advanced gynecological malignancies. Int J Radiat Oncol Biol Phys 1983;9(12):1897 1903. 3. Erickson B, Albano K, Gillin M. CT-guided interstitial implantation of gynecologic malignancies. Int J Radiat Oncol Biol Phys 1996;36(3):699 709. 4. Aristizabal S, Surwit E, Hevezi J, et al. Treatment of advanced cancer of the cervix with transperineal interstitial irradiation. Int J Radiat Oncol Biol Phys 1983;9(7):1013 1017. 5. Martinez A, Herstein P, Portnuff J. Interstitial therapy of perineal and gynecological malignancies. Int J Radiat Oncol Biol Phys 1983;9(3):409 416. 6. Martinez A, Edmundson G, Cox R, et al. Combination of external beam irradiation and multiple-site perineal applicator (MUPIT) for treatment of locally advanced or recurrent prostatic, anorectal, gynecologic malignancies. Int J Radiat Oncol Biol Phys 1985;11(2):391 398. 7. Stock RG, Chan K, Terk M, et al. A new technique for performing Syed-Neblett template interstitial implants for gynecologic malignancies using transrectal-ultrasound guidance. Int J Radiat Oncol Biol Phys 1997;37(4):819 825. 8. Nag S, Martínez-Monge R, Ellis R, et al. The use of fluoroscopy to guide needle placement in interstitial gynecological brachytherapy. Int J Radiat Oncol Biol Phys 1998;40(2):415 420. 9. Eisbruch A, Johnston CM, Martel MK, et al. Customized gynecologic interstitial implants: CT-based planning, dose evaluation, and optimization aided by laparotomy. Int J Radiat Oncol Biol Phys 1998;40(5):1087 1093. 10. Recio FO, Piver MS, Hempling RE, et al. Laparoscopicassisted application of interstitial brachytherapy for locally advanced cervical carcinoma: Results of a pilot study. Int J Radiat Oncol Biol Phys 1998;40(2):p.411 4. 11. Demanes JD, Rodriguez RR, Bendre DD, et al. High dose rate transperineal interstitial brachytherapy for cervical cancer: High pelvic control and low complication rates. Int J Radiat Oncol Biol Phys 1999;45(1):105 112. 12. Gaddis O, Morrow P, Klement V, et al. Treatment of cervical carcinoma employing a template for transperineal interstitial Ir192 brachytherapy. Int J Radiat Oncol Biol Phys 1983;9(6): 819 827. 13. Aristizabal S, Woolfitt B, Valencia A, et al. Interstitial parametrial implants in carcinoma of the cervix stage II-B. Int J Radiat Oncol Biol Phys 1987;13:445 450. 14. Hughes-Davies L, Silver B, Kapp DS. Parametrial interstitial brachytherapy for advanced or recurrent pelvic malignancy: The Harvard/Stanford experience. Gynecol Oncol 1995;58(1):24 27. 15. Charra C, Roy P, Coquared R, et al. Outcome of treatment of upper third vaginal recurrences of cervical and endometrial carcinomas with interstitial brachytherapy. Int J Radiat Oncol Biol Phys 1998;40(2):421 426. 16. Kolkman-Deurloo IKK, Visser AG, Niël CGJH, et al. Optimization of interstitial volume implants. Radiother Oncol 1994;31:229 239. 17. Edmundson GK. Volume optimisation: An American viewpoint. In: Mould RF, et al., editors. Brachytherapy from radium to optimization. Wageningen, The Netherlands: Nucletron International; 1994. p. 314 318. 18. Van Der Laarse R, Prins TPE. Introduction to HDR brachytherapy optimisation. In: Mould RF, et al., editors. Brachytherapy from radium to optimization. Wageningen, The Netherlands: Nucletron International; 1994. p. 314 318. 19. (ICRU Report 58). Dose and volume specification for reporting interstitial therapy. Bethesda, MD: International Commission on Radiation Units and Measurements; 1997. 20. Kim R, Caranto J, Pareek P, et al. Dynamics of pear-shaped dimensions and volume of intracavitary brachytherapy in cancer of the cervix: A desirable pear shape in the era of threedimensional treatment planning. Int J Radiat Oncol Biol Phys 1997;37(5):1193 1199. 21. Eifel PJ. Intracavitary brachytherapy in the treatment of gynecologic neoplasms. J Surg Oncol 1997;66(2):141 147. 22. Perez CA, Grigsby PW, Castro-Vita H, et al. Carcinoma of the uterine cervix. I. Impact of prolongation of overall treatment time and timing of brachytherapy on outcome of radiation therapy. Int J Radiat Oncol Biol Phys 1995;32(5):1275 1288. 23. Logsdon MD, Eifel PJ. Low dose rate brachytherapy in the treatment of cervical carcinoma. Hematol Oncol Clin North Am 1999;13(3):577 584. 24. Morris M, Eifel PJ, Lu J, et al. Pelvic radiation with concurrent chemotherapy compared with pelvic and para-aortic radiation for high-risk cervical cancer. N Engl J Med 1999; 340(15):1137 1143. 25. Rose PG, Bundy BN, Watkins EB, et al. Concurrent cisplatinbased radiotherapy and chemotherapy for locally advanced cervical cancer [see comments]. N Engl J Med 1999;340(15): 1144 1153. 26. Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy for bulky stage IB cervical carcinoma. N Engl J Med 1999;340(15):1154 1161.