P

IccSswIlsg

mrepattci

MM

Int. J. Radiation Oncology Biol. Phys., Vol. 63, No. 3, pp. 934–939, 2005Copyright © 2005 Elsevier Inc.

Printed in the USA. All rights reserved0360-3016/05/$–see front matter

doi:10.1016/j.ijrobp.2005.07.963

HYSICS CONTRIBUTION

CERVICAL BRACHYTHERAPY UTILIZING RING APPLICATOR:COMPARISON OF STANDARD AND CONFORMAL LOADING

SUSAN BROOKS, F.R.A.N.Z.C.R., PETER BOWNES, M.SC., GERRY LOWE, M.SC.,LYNDA BRYANT, D.C.R.T., AND PETER J. HOSKIN, M.D., F.R.C.P., F.R.C.R.

Mount Vernon Cancer Center, Northwood, Middlesex, United Kingdom

Purpose: Afterloading high-dose-rate brachytherapy (HDR) treatment of cervical cancer with cross-sectionalimaging and three-dimensional (3D) reconstruction offers opportunities for individualized conformal treatmentplanning rather than fixed point-A dosimetry.Methods and Materials: Between June 2003 and September 2004, 15 patients with FIGO Stage 1B-4A cervicalcarcinoma, median age 56 years, were treated with radical external-beam radiotherapy to pelvis, includingparaortic nodes if positive on staging investigations. Fourteen patients received concurrent cisplatin chemother-apy. All patients received HDR brachytherapy administered by intrauterine tube and ring applicator. Clinicaltarget volume (CTV) and organs at risk (OAR)—rectum, bladder, and small bowel—were outlined from postinsertionCT planning scans. Planning target volume (PTV) was derived by use of 2-mm to 3-mm 3D expansion. A standardplan was produced that delivered 6 Gy to point A, and a second plan delivered 6 Gy to PTV. Constraints were definedfor the OAR: bladder, 6 Gy; rectum, 5 Gy; and small bowel, 5 Gy. Dosimetric comparison was performed by use ofthe Baltas conformal index (COIN).Results: Mean COIN values were 0.39 for conformal plans and 0.33 for standard plans (p � 0.001); mean D95values were 4.79 Gy and 4.50 Gy, respectively.Conclusion: The majority of patients achieved a plan closer to ideal for coverage of PTV, with minimization ofradiation received by normal tissues for conformal loading measured by COIN compared with fixed point-Aprescription that used the cervical ring applicator. © 2005 Elsevier Inc.

Cervix neoplasm, Brachytherapy, Radiotherapy, Conformal, Radiotherapy dosage.

tmlo

aavAtTddlpv

P

F

0

INTRODUCTION

ntracavitary brachytherapy is well recognized as an integralomponent of definitive radiotherapy treatment of cervicalancer and is recommended by the American Brachytherapyociety on the basis of patterns of care studies that havehown a reduction in recurrence and complication rateshen compared with external-beam radiotherapy alone (1).

ntracavitary brachytherapy enables high-dose radiation de-ivery to the central tumor volume with relative sparing ofurrounding normal tissues in particular dose-limiting or-ans such as rectum, bladder, and small bowel.High-dose-rate (HDR) brachytherapy enables short treat-ent times that can be outpatient based, with associated

eduction in exposure to caregivers and visitors. It alsomploys smaller diameter applicators, with increased dis-lacement and reduced dose to surrounding normal tissuesnd reduced applicator movement. The ability to optimizehe treatment-dose distribution by the variation of dwellimes and adjustment of source position allows a greaterontrol of the dose distribution and potentially less morbid-ty (2–5). The use of applicators compatible with computed

Reprint requests to: Peter J. Hoskin, M.D., F.R.C.P., F.R.C.R.,ount Vernon Cancer Centre, Rickmansworth Road, Northwood,

iddlesex, HA6 2RN, UK. Tel: (�44) 01923-826111; Fax: (�44)

934

omography (CT) and magnetic resonance imaging (MRI)ay further improve the optimization of treatment by al-

owing improvement in tumor and normal-tissue delineationn 3-dimensional (3D) imaging at the point of planning (6).Adjustment of the source loading within the applicators

llows improvement of the dose distribution and individu-lized source loading to conform to the residual tumorolume that has been delineated on 3D CT planning images.dequate dose to the tumor is then possible while the dose

o normal tissues is maintained within defined constraints.o evaluate this approach relative to standard point-A doseefinition, we have compared the dose distribution of stan-ard loading to achieve a point-A dose with a conformaloading of the intrauterine tube and ring applicator foratients treated with definitive radical radiotherapy for cer-ical cancer at Mount Vernon Hospital.

METHODS AND MATERIALS

atientsThe study population included 15 consecutive patients with

IGO Stage 1B-4A cervical carcinoma who were treated with

1923-844167; E-mail: peterhoskin@nhs.netReceived July 4, 2005. Accepted for publication July 17, 2005.

ra(npcatSIpwtm

T

ppvawpttpc

tUad(tw

apam

T

wfp2pte

owaboCqweVeiac(tatIccp

tt

Tivthprrtagcm

S

P

L

E

C

I

I

B

p

935Cervical brachytherapy utilizing ring applicator ● S. BROOKS et al.

adical external-beam radiotherapy to pelvis between June 2003nd September 2004. The median age of the patients was 56 yearsrange, 30–87 years). Thirteen patients had squamous cell carci-oma, and 2 patients had adenocarcinoma. They were staged by ahysical examination under anesthetic that included biopsy andystoscopy, an MRI of the pelvis, and either a CT scan of chest andbdomen or an abdominal ultrasound and chest X-ray. Two pa-ients underwent a lymph node dissection. Four patients had FIGOtage IB cancer, 8 patients had Stage IIB, 2 patients had StageIIB, and 1 patient had Stage IVA. Nine patients had enlargedelvic lymph nodes on imaging. In 1 of these patients, metastasesere confirmed pathologically. Two patients had enlarged paraor-

ic lymph nodes on staging investigations; in 1 of these patients,etastases were confirmed pathologically (Table 1).

reatmentAll patients were treated with external-beam radiotherapy to the

elvis before brachytherapy. Paraortic nodes were also included ifositive on staging investigations. Thus, 13 patients received pel-ic radiotherapy alone to a dose of 50 Gy in 25 fractions by use of3-field or 4-field plan prescribed to the isocenter. The 2 patientsith enlarged paraortic nodes received 45 Gy in 25 fractions to theelvis and paraortic nodes with anteroposterior fields, prescribed tohe midplane, followed by a boost of 5 Gy in 3 fractions, deliveredo the pelvis alone. Staging scans were used to help determinelanning volumes. Fourteen patients received concurrent cisplatinhemotherapy (40 mg/m2 weekly) (Table 1).

All patients received subsequent HDR brachytherapy adminis-ered via a Gammamed-Plus (Varian Medical Systems, Crawley,K) intrauterine tube and cervical ring applicator set that are CT

nd MRI compatible. All treatments were completed within 42ays. The tube and cervical ring applicator, with either a 90°straight) or a 60° angulation, was inserted under general anes-hetic by use of a cervical sleeve to enable subsequent insertions

Table 1. Patient and treatment characteristics

tage IB 4IIA 0IIB 8IIIA 0IIIB 2IVA 1

athology Squamous cell carcinoma 13Adenocarcinoma 2

ymph nodes (positive) Pelvic only 9PAN only 1Pelvic � PAN 1

BRT Pelvis alone 13Pelvis � PAN 2

hemotherapy (cisplatin) Yes 14No 1

U tube angle 60° 1190° 4

U tube length 4 cm 66 cm 78 cm 2

uild-up cap Yes 12No 3

Abbreviations: EBRT � external-beam radiotherapy; PAN �araortic lymph nodes.

ithout anesthetic. Thirteen patients had a 2-mm-thick ring spacer n

dded, and all patients had nonradiopaque packing placed to dis-lace the rectum. Planning CT scans were performed immediatelyfter applicator insertion, when applicator and anatomic move-ents are minimized.

reatment planningThe clinical target volume (CTV) and the organs at risk (OAR),

hich included rectum, bladder, and small bowel, were outlinedrom postinsertion CT planning scans. These volumes were inde-endently defined by 2 clinicians. The OAR were volumed up tocm beyond the superior and inferior limits of the applicators. Thelanning target volume (PTV) was then derived by use of a 2-mmo 3-mm 3D expansion on the Brachyvision planning system, withxpansion limited by the OAR.

A plan was produced that delivered a standard prescription dosef 6 Gy to point A, and a second plan delivered 6 Gy to the PTV,ith constraints defined for the OAR: bladder, 6 Gy; rectum, 5 Gy;

nd small bowel, 5 Gy. A dosimetric comparison was performedy use of the Baltas conformal index (COIN) (7), which is an indexf treatment quality that incorporates anatomic information fromT directly input into an algorithm. This measurement of implantuality considers conformity of the reference isodose to the PTV,hich is a ratio expression of the V100 (C1 � PTVref/PTV),

xcess radiation to normal tissues outside PTV (C2 � PTVref/

ref), and volume of critical structures that receive radiation inxcess of defined tolerance (C3 � �Ncs[1�VCSref/VCS]). Thedeal situation is a reference isodose outline identical to the PTVnd no dose to surrounding normal tissues outside the PTV be-ause of the rapid dose falloff. Thus, the ideal conformal indexCOIN � C1�C2�C3) is 1. In practice, such an ideal conformalreatment rarely occurs, and, thus, those with a value closest to 1re considered closest to the ideal. Other values of interest includehe dose delivered to 95% of the target volume (D95) and to thenternational Commission on Radiation Units (ICRU) point A,omparison of maximal dose to a contiguous 2cc volume of aritical normal tissue (D2cc), and the ICRU rectal and bladderoints.Statistical analysis was applied to document the significance of

he difference between the outcome means by use of 2-sided pairedtests.

RESULTS

Details of the dosimetric comparisons are shown in Table 2.he parameters C1, C2, and C3 together with the conformal

ndex comparison are shown in Table 3. The mean D95

alues are 4.79 Gy for the conformal plans and 4.50 Gy forhe standard plans. The mean point-A dose was significantlyigher than the D95 value for both conformal and standardlans at 5.77 Gy (p � 0.002) and 5.99 Gy (p � 0.03),espectively. The mean C1 calculation or conformity of theeference isodose to the PTV was significantly greater forhe conformal plan at 0.82 compared with the standard plant 0.77 (p � 0.05). Figure 1 and Figure 2 show a set of plansenerated for one patient and is an example of the improvedonformity of the 6 Gy isodose to the PTV for the confor-al compared with the standard arrangement.The mean C2 value, which accounts for the volume of

ormal tissue outside the PTV that receives 100% of the

dm

tt0ri0wp

tt0bb(gd

st0

tTfd

c

sgmbm

iVccltTdwsamdetwds

c3mCvihrB

ission

936 I. J. Radiation Oncology ● Biology ● Physics Volume 63, Number 3, 2005

ose, is 0.48 for the standard plans and 0.50 for the confor-al plans (p � 0.23).The mean rectal C3 value, which reflects the volume of

he rectum that receives dose in excess of the definedolerance level of 5 Gy, is 0.99 for the standard plans and.99 for the conformal plans (p � 0.90). The mean D2cc

ectum value for conformal plans is 4.02 Gy and is signif-cantly less than the ICRU rectal-point dose at 5.48 Gy (p �.0002). A similar comparison is found for standard plansith the mean D2cc value at 3.74 Gy and the ICRU rectal-oint dose at 4.52 Gy (p � 0.009).The mean bladder C3 value, which reflects the volume of

he bladder that receives dose in excess of the definedolerance level of 6 Gy is 0.97 for the standard plans and.98 for the conformal plans (p � 0.14). The mean D2cc

ladder value for conformal plans is greater than the ICRUladder-point dose at 5.48 Gy and 4.60 Gy, respectivelyp � 0.006). The mean D2cc value was 6.05 Gy and, thus,reater than the 4.75 Gy ICRU bladder-point dose for stan-ard plans (p � 0.01).

The mean bowel C3 value, which reflects the volume ofmall bowel that receives dose in excess of 5 Gy, is 0.99 forhe standard plans and 0.99 for the conformal plans (p �.84).Fourteen of the 15 patients had a COIN calculation closer

o 1 for the conformal plan compared with the standard plan.he mean COIN calculation for standard plans is 0.33 and

or the conformal plans is 0.39, reflecting a significantifference of 18% (p � 0.001).

DISCUSSION

The goal of radical radiotherapy is to achieve the highestytotoxic dose to the tumor while minimizing the dose to

Table 2. Comparison of dose

Mean values Standard (range) Gy

Point-A dose 5.99 (5.83–6.19)D95 4.50 (1.87–7.76)ICRU rectum 4.52 (2.64–5.79)D2cc rectum 3.74 (1.82–6.98)ICRU bladder 4.75 (1.74–7.28)D2cc bladder 6.05 (3.35–8.07)

Abbreviation: ICRU � International Comm

Table 3. Analysis of dos

Mean values Standard (range) Gy

C1 0.77 (0.45–1.0)C2 0.48 (0.19–0.81)C3 rectum 0.99 (0.91–1.0)C3 bladder 0.97 (0.92–1.0)C3 small bowel 0.99 (0.98–1.0)COIN 0.33 (0.18–0.46)

Abbreviation: COIN � conformal index.

urrounding normal tissues and particularly to critical or-ans at risk. In cervical cancer, brachytherapy is the idealethod to achieve this goal, and the development of image-

ased conformal techniques can optimize the use of thisodality, as demonstrated in this study.The mean D95 results indicate that the dose to the 95%

sodose is greater for the conformal treatment. The mean100, and therefore the C1, value is also greater for the

onformal treatment, which indicates the reference isodoseoverage of the PTV is greater compared with standardoading. The mean C2 results reflect that the dose to normalissues outside the PTV is less for the conformal treatment.he individual C3 results are mixed, which reflects that theose to the bladder is lower for the conformal loading,hereas the mean rectal and small bowel C3 results are the

ame for both conformal and standard plans. These resultsre not unexpected, as the plans were designed to achieve asuch dose as possible to the PTV while staying within the

ose limits defined for the individual OAR. Overall, how-ver, the COIN calculation performed for the conformalreatment was 18% higher than that of standard loading,hich was statistically significant (p � 0.001); this resultemonstrates an advantage for the conformal approach overtandard loading to achieve a point-A dose.

The COIN calculation offers a useful tool for numericallyomparing conformity between plans. With the advent ofD treatment planning, we can calculate an index of treat-ent quality that incorporates anatomic information fromT directly input into the conformal-index algorithm. Pre-ious attempts to assess the quality of a brachytherapymplant have been criticized because of their dependence onomogeneity of dose and the subjectivity of defining doseange (7). The COIN used in this exercise, described byaltas et al. (7), does not rely on homogeneity and defined

standard or conformal plan

formal (range) Gy P values (95% CI)

5.77 (4.37–7.17) 0.41 (�0.33–0.76)4.79 (3.05–6.70) 0.22 (�0.75–0.19)5.48 (3.61–8.68) 0.02 (�1.78–0.16)4.02 (2.08–4.93) 0.23 (�0.84–0.27)4.60 (1.86–6.76) 0.52 (�0.34–0.65)5.48 (3.58–6.46) 0.06 (�0.04–1.17)

on Radiation Units.

using conformal index

rmal (range) Gy P values (95% CI)

2 (0.65–0.98) 0.05 (�0.11–�0.0003)0 (0.30–0.74) 0.23 (�0.11–�0.01)9 (0.97–1.0) 0.90 (�0.01–0.01)8 (0.94–1.0) 0.14 (�0.03–0.004)9 (0.95–1.0) 0.84 (�0.006–0.007)9 (0.29–0.66) 0.001 (�0.09–�0.03)

s using

Con

imetry

Confo

0.80.50.90.90.90.3

pu

oelcc

(ieci

scpicTlmmsIIbssci

cMdl

tbtatotsvtblhcptfdrb“v(udrd

Ft

937Cervical brachytherapy utilizing ring applicator ● S. BROOKS et al.

oints, isodoses, and dose rates and is, therefore, moreseful in clinical practice.A limitation of the COIN calculation is that the 3 critical

rgans and the PTV are equally weighted. Thus, as anxample, the same COIN result may be associated with aow dose to the critical structures (high C3) but poor PTVoverage (low C1) compared with a plan that has high

ig. 1. Example of the standard (a) and conformal (b) computedomography plans generated in the transaxial plane.

ritical-structure doses (low C3) with good PTV coverage r

high C1). Therefore, each situation must be assessed bydentification of the individual parameters that make upach overall COIN calculation. A weighting factor to ac-ount for the individual parameters could be incorporated tomprove the utility of COIN.

This index uses a chosen dose constraint for each criticaltructure and calculates the volume of the organ that re-eives that dose or a greater dose. It does not utilize doseoints such as the ICRU 38 rectal and bladder points. Thedeal situation is to have 0 volume receive more than theonstraining dose, which would result in a C3 value of 1.he use of 3D imaging for planning and subsequent calcu-

ation of maximal dose to specific volumes of critical nor-al tissues derived from dose–volume histograms (DVH) isore accurate than dose points (8–12). The results of this

tudy are mixed when the D2cc values are compared with theCRU reference points. The rectal dose was higher for theCRU rectal point for both conformal and standard plans,ut the bladder ICRU point underestimated the dose, as isuggested in previous reports. The point-A dose in thiseries was higher compared with the D95 results for bothonformal and standard plans, which again supports find-ngs of previous studies.

Tumor definition and surrounding normal-tissue volumesan be difficult to determine by use of CT planning scans.RI has been proposed to achieve a more accurate tumor

efinition, and the use of MRI in the planning of gyneco-ogical malignancies may give greater certainty (13–18).

Concerns have been raised about the potential for damageo normal tissue from high dose rates with the use of HDRrachytherapy (19–24). Optimization of the dose distribu-ion is one of the techniques available to HDR brachyther-py that will reduce the likelihood of significant normal-issue toxicity. Early work by Houdek et al. (19) assessedptimization potential of computer-controlled HDR brachy-herapy by simulation of the dwell times and positions of theources in the tube and ring applicators for cervical andaginal treatments. They found that the optimized distribu-ion of physical doses was superior to that of standardrachytherapy, particularly with regard to a significantlyower dose in the rectum. Similarly, Noyes et al. (20) reportigher doses at bladder and rectum for nonoptimized HDRervical brachytherapy, and Decker et al. (21), who com-ared LDR cervical brachytherapy to optimized and nonop-imized HDR tandem and ovoid dosimetry in 2 patients alsoound that nonoptimized HDR dosimetry resulted in higheroses at selected points, including dose to bladder andectum. Furthermore, over repeated fractions of HDRrachytherapy, the prescription points move and result inmultiple point A’s,” which can be overcome by an indi-idualized dose plan based on CTV (12). Cetingoz et al.22) theoretically evaluated the optimization possibilities ofterine and vaginal brachytherapy by assessing the doseistribution variations in rectum, bladder, mean point-Beference points, and volume parameters while keeping theose to point A standard. By increasing the dwell-time

atios of intrauterine and vaginal sources, they observed a

dr5iirtd(icTc

aflfmam

ss

l plan

938 I. J. Radiation Oncology ● Biology ● Physics Volume 63, Number 3, 2005

ecrease in dose to critical structures, particularly to theectum, by 23% at 2.5-mm step positions and by 28% at-mm step positions and also a decrease in the referencesodose volume by 14% and 17%, respectively. A compar-son of cervical cancer treatment outcomes in patients whoeceived individualized brachytherapy dosimetry (105 pa-ients) and historical series in which patients received noosimetry (58 patients) was performed by Faroudi et al.23). They reported that individualized dosimetry resultedn the delivery of lower brachytherapy dose with decreasedomplications but no decrease in local control or survival.hus, a number of studies confirm that important changes

Fig. 2. Example of the standard and conforma

an occur in reference points and volumes if dwell times p

REFEREN

Brachytherapy Society recommendations for high-dose-rate

nd source positions are manipulated, which may be re-ected in clinically important outcomes. Our study providesurther support with the finding that the COIN calculation isore ideal with regards to isodose coverage of the tumor

nd lower dose to critical organs, for conformal or opti-ized plans.

CONCLUSION

Improved coverage of PTV and reduced dose to criticaltructures reflected in an improved COIN has been demon-trated for a conformal plan compared with a fixed point-A

s generated in the frontal and sagittal planes.

rescription that used the cervical ring applicator.

CES

1. Nag S, Erickson B, Thomadsen B, et al. The American

brachytherapy for carcinoma of the cervix. Int J Radiat Oncol Biol Phys 2000;48:201–211.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

939Cervical brachytherapy utilizing ring applicator ● S. BROOKS et al.

2. Petereit DG, Sarkaria JN, Potter DM, et al. High-dose-rate versuslow-dose-rate brachytherapy in the treatment of cervical cancer:Analysis of tumor recurrence—the University of Wisconsin ex-perience. Int J Radiat Oncol Biol Phys 1999;45:1267–1274.

3. Arterbery VE. High-dose rate brachytherapy for carcinoma ofthe cervix. Curr Opin Oncol 1993;5:1005–1009.

4. Nakano T, Kato S, Ohno T, et al. Long-term results ofhigh-dose rate intracavitary brachytherapy for squamous cellcarcinoma of the uterine cervix. Cancer 2005;103:92–101.

5. Potter R, Knocke TH, Fellner C, et al. Definitive radiotherapybased on HDR brachytherapy with iridium 192 in uterinecervix carcinoma: Report on the Vienna University Hospitalfindings (1993–1997) compared to the preceding period in thecontext of ICRU 38 recommendations. Cancer Radiother2000;4:159–172.

6. Nag S, Cardenes H, Chang S, et al. Proposed guidelines forimage-based intracavitary brachytherapy for cervical carci-noma: Report from Image-Guided Brachytherapy WorkingGroup. Int J Radiat Oncol Biol Phys 2004;60:1160–1172.

7. Baltas D, Kolotas C, Geramani K, et al. A conformal index(COIN) to evaluate implant quality and dose specification inbrachytherapy. Int J Radiat Oncol Biol Phys 1998;40:515–524.

8. Wachter-Gerstner N, Wachter S, Reinstadler E, et al. Bladderand rectum dose defined from MRI based treatment planningfor cervix cancer brachytherapy: Comparison of dose-volumehistograms for organ contours and organ wall, comparisonwith ICRU rectum and bladder reference point. RadiotherOncol 2003;68:269–276.

9. Kim RY, Pareek P. Radiography-based treatment planningcompared with computed tomography (CT)-based treatmentplanning for intracavitary brachytherapy in cancer of the cer-vix: Analysis of dose-volume histograms. Brachytherapy2003;2:200–206.

0. Kapp KS, Stuecklschweiger GF, Kapp DS, et al. Dosimetry ofintracavitary placements for uterine and cervical carcinoma:Results of orthogonal film, TLD, and CT-assisted techniques.Radiother Oncol 1992;24:137–146.

1. Fellner C, Potter R, Knocke TH, et al. Comparison of radi-ography- and computed tomography-based treatment planningin cervix cancer in brachytherapy with specific attention tosome quality assurance aspects. Radiother Oncol 2001;58:53–62.

2. Datta N, Basu R, Rajasekar D, et al. Problems and uncertain-

ties with multiple point A’s during multiple high-dose-rate

intracavitary brachytherapy in carcinoma of the cervix. ClinOncol 2004;16:129–137.

3. Roach M 3rd, Faillace-Akazawa P, Malfatti C, et al. Prostatevolumes defined by magnetic resonance imaging and comput-erized tomographic scans for three-dimensional conformalradiotherapy. Int J Radiat Oncol Biol Phys 1996;35:1011–1018.

4. Milosevic M, Voruganti S, Blend R, et al. Magnetic resonanceimaging (MRI) for localization of the prostatic apex: Compar-ison to computed tomography (CT) and urethrography. Ra-diother Oncol 1998;47:277–284.

5. Debois M, Oyen R, Maes F, et al. The contribution of mag-netic resonance imaging to the three-dimensional treatmentplanning of localized prostate cancer. Int J Radiat Oncol BiolPhys 1999;45:857–865.

6. Tardivon AA, Kinkel K, Lartigau E, et al. MR imaging duringintracavitary brachytherapy of vaginal and cervical cancer:Preliminary results. Radiographics 1996;16:1363–1370.

7. Schoeppel SL, LaVigne ML, Martel MK, et al. Three-dimen-sional treatment planning of intracavitary gynecologic im-plants: Analysis of ten cases and implications for dose spec-ification. Int J Radiat Oncol Biol Phys 1994;28:277–283.

8. Wachter-Gerstner N, Wachter S, Reinstadler E, et al. Theimpact of sectional imaging on dose escalation in endocavitaryHDR-brachytherapy of cervical cancer: Results of a prospec-tive comparative trial. Radiother Oncol 2003;68:51–59.

9. Houdek PV, Schwade JG, Abitbol AA, et al. Optimization ofhigh dose-rate cervix brachytherapy, Part I: Dose distribution.Int J Radiat Oncol Biol Phys 1991;21:1621.

0. Noyes WR, Peters NE, Thomadsen BR, et al. Impact of“optimized” treatment planning for tandem and ring, andtandem and ovoids, using high dose rate brachytherapy forcervical cancer. Int J Radiat Oncol Biol Phys 1995;31:79–86.

1. Decker WE, Erickson B, Albano K, et al. Comparison oftraditional low-dose-rate to optimized and nonoptimized high-dose-rate tandem and ovoid dosimetry. Int J Radiat Oncol BiolPhys 2001;50:561–567.

2. Cetingoz R, Ataman OU, Tuncel N, et al. Optimization in highdose rate brachytherapy for utero-vaginal applications. Ra-diother Oncol 2001;58:31–36.

3. Foroudi F, Bull A, Gebski V. Radiation therapy for cervixcarcinoma: Benefits of individualized dosimetry. Clin Oncol2002;14:43–49.

4. Lanciano R. Optimizing radiation parameters for cervical can-

cer. Semin Radiat Oncol 2000;10:36–43.