Radiation therapy target volume reduction in pediatric rhabdomyosarcoma

Radiation Therapy Target Volume Reduction in Pediatric

Rhabdomyosarcoma

Implications for Patterns of Disease Recurrence and Overall Survival

Bree R. Eaton, MD1; Mark W. McDonald, MD1; Sungjin Kim, MS2; Robert B. Marcus, Jr, MD1; Anna L. Sutter, MS1;

Zhengjia Chen, PhD2; and Natia Esiashvili, MD1

BACKGROUND: The use of radiation therapy (RT) ‘‘cone-down’’ boost to reduce high-dose treatment volumes according to tumor

response to induction chemotherapy in patients with pediatric rhabdomyosarcoma (RMS) may reduce treatment morbidity, yet the

impact on tumor control is unknown. METHODS: Fifty-five children, including 18 (33%) with parameningeal (PM) RMS and 37 (67%)

with non-PM RMS, who received definitive treatment with chemotherapy and RT from April 2000 through January 2010 were retro-

spectively reviewed. RESULTS: In total, 28 patients (51%) received a cone-down boost. The high-dose boost volume was reduced by a

median of 56% of the initial target volume (range, 5%-91%). The median time to initiating RT was 3 weeks for patients with PM RMS

and 16 weeks for patients with non-PM RMS (P < .001). After a median follow-up of 41 months, local failure occurred in 5 patients

(9%), including 2 patients who received a cone-down boost, and there were no marginal failures. Twelve patients (67%) with PM RMS

had intracranial tumor extension. In this subgroup, 4 patients (30%) who received a cone-down boost and had � 3 weeks between

chemotherapy and RT initiation experienced leptomeningeal failure as their first site of disease progression, and a delayed time to RT

initiation was associated with decreased survival (P ¼ .055) CONCLUSIONS: A cone-down boost allowed for significant reductions in

high-dose RT treatment volume while maintaining excellent tumor control in most patients. However, in the subset of patients with

PM RMS and intracranial tumor extension, early RT initiation and wider margin RT to cover adjacent areas at high risk for meningeal

extension may be more important for adequate disease control. Cancer 2013;119:1578-85. VC 2012 American Cancer Society.

KEYWORDS: radiation therapy, planning target volume, cone-down boost, rhabdomyosarcoma.

INTRODUCTIONRadiation therapy (RT) is a critical component of multimodality therapy for pediatric rhabdomyosarcoma (RMS). Induc-tion chemotherapy followed by concurrent chemoradiation is the current standard of care for patients with unresected dis-ease, those with microscopic or gross residual disease after surgery, and for all patients with alveolar histology based onimprovements in failure-free survival with the addition of RT in these groups.1,2 Despite the prominent role that RT playsin the management of RMS, uncertainty remains regarding the optimal timing for the initiation of RT in relation to theonset of chemotherapy and whether RT volumes may be modified in response to tumor shrinkage after inductionchemotherapy.

Early cooperative protocols by the International Rhabdomyosarcoma Study Group (IRSG) used conventional RTtechniques with various timing of RT initiation. Retrospective analyses of data from IRSG studies as well as multiple single-institution series have assessed the impact of RT timing on local control (LC). Some studies suggest that early initiation ofRT is important for LC in patients with meningeal impingement,3 whereas other investigations have demonstrated thatdelay of RT by 16 weeks or even longer may not compromise outcomes in patients with parameningeal (PM) RMS.4-6

More recently, the development of advanced RT techniques, such as diagnostic image fusion for target delineationand intensity-modulated radiation therapy (IMRT) with daily image guidance, have enabled the confident delivery ofmore conformal treatments with the possibility of reduced margins for treatment uncertainties. Several institutions havenow reported promising clinical experience with IMRT for pediatric RMS.7-10 The impetus for IMRT is to improve the

DOI: 10.1002/cncr.27934, Received: October 5, 2012; Revised: November 2, 2012; Accepted: November 6, 2012, Published online December 27, 2012 in Wiley

Online Library (wileyonlinelibrary.com)

Corresponding author: Bree R. Eaton, MD, 1365 Clifton Road NE, Building A, Suite CT 104, Atlanta, GA 30322. Fax: (404) 778-3643; [email protected]

1Department of Radiation Oncology, Winship Cancer Institute of Emory University, Atlanta Georgia; 2Department of Biostatistics and Bioinformatics, Winship Can-

cer Institute of Emory University, Atlanta Georgia

Mark McDonald’s current affiliation: Indiana University Health Proton Therapy Center, Bloomington, Indiana

Robert Marcus’ current affiliation: Gulf Region Radiation Oncology, Pensacola, Florida

1578 Cancer April 15, 2013

Original Article

conformality of the prescribed RT dose to the target vol-ume, sparing more nontarget tissues from high-doseradiation.

Even more normal tissue volume could be sparedfrom high-dose radiation by the implementation of aresponse-based radiation treatment volume that targetsthe prechemotherapy tumor volume to an intermediatedose before ‘‘coning down’’ to boost the postchemother-apy tumor volume to full dose. With this technique,delayed onset of RT allows for continued tumor responseto induction chemotherapy and facilitates further reduc-tion in high-dose treatment volumes, which may reduceacute and late toxicities of treatment.

The safety of implementing cone-down boost inpatients with PM RMS has been demonstrated inpreviously published institutional series,8,11 yet dataaddressing the impact of cone-down boost on disease con-trol as it relates to other patient and treatment related vari-ables, such as RT timing or primary disease site, islimited. We sought to evaluate the influence of theseparameters in an expanded and updated analysis ofchildren with both PM and non-PM primary RMS whowere treated at our institution.

MATERIALS AND METHODS

Patient Population

With institutional review board approval, children withRMS who received treatment with chemotherapy andRT with definitive intent at the Emory Clinic andChildren’s Healthcare of Atlanta from April 2000through January 2010 were identified for retrospectiveanalysis. Patients who had evidence of distant metastaticdisease at diagnosis were excluded. All patients had path-ologic confirmation of their RMS diagnosis from ourinstitution based on biopsy or surgical specimens. Thestaging evaluation included computed tomography (CT)and/or magnetic resonance imaging (MRI) studies of theprimary site. CT scans of the chest or fluorodeoxyglu-cose-positron emission tomography (FDG-PET) studies,technetium-99 bone scans, and bone marrow aspirateswere performed on all patients to evaluate for distantmetastasis. Cerebrospinal fluid cytology also was per-formed on patients who had PM primary sites. Patientswere assigned both a pretreatment stage according to theInternational Union Against Cancer staging system usedby the IRSG12 and a postoperative group according tothe IRSG system.13 All patients received systemic chem-otherapy after pathologic confirmation of their diagnosisand completion of systemic workup and staging. All but1 patient (98%) received intensive, multiagent, vincris-

tine-based chemotherapy, either enrolled on or followinga specified risk-group based protocol, and 1 patientreceived single-agent etoposide. RT was initiated at atime set according to chemotherapy protocol or variedbased on patient referral, delays in chemotherapy, orother social or patient-specific factors.

Radiation Therapy

CT simulation with immobilization specific to disease sitewas used for treatment planning. Patients received generalanesthesia for simulation and daily treatment when neces-sary for immobilization. The initial and postinductionchemotherapy diagnostic imaging studies, including MRI,FDG-PET, and/or CT scans, were registered to the simu-lation CT for delineation of the target volume. The initialgross tumor volume (GTV) was defined as the extent ofprechemotherapy gross soft tissue abnormality on contrast-enhanced CT or MRI studies or the area of focal increasedFDG uptake on FDG-PET studies. For patients whoreceived a cone-down boost, a second boost GTV was cre-ated based on the extent of gross soft tissue abnormality asdefined by the postinduction chemotherapy diagnosticimaging studies. The preinduction and postinductionchemotherapy clinical tumor volumes (CTVs) were cre-ated by expansion of the corresponding GTV with a 0.5-cm to 1.5-cm margin to account for potential microscopictumor extension. The margins were modified according tothe anatomic location of the tumor to respect natural bar-riers to tumor spread, such as bone. Initial and boost plan-ning target volumes (PTVs) were created to account forpatient setup uncertainty and target motion by the addi-tion of a 0.5-cm to 1-cm margin to the prechemotherapyand postinduction chemotherapy CTVs, respectively.

The RT dose was specified according to pretreatmentstage and postoperative group and ranged from 36 gray(Gy) to 54 Gy. For patients who received a cone-downboost, the field reduction was made after delivery of an ini-tial 36 Gy to 41.4 Gy to the prechemotherapy PTV, witha boost volume dose ranging from 9 Gy to 19.8 Gy. RTdelivery included external-beam RT with photons of �6MV using 3-dimenisonal conformal-field arrangements ormultifield IMRT. Electron-beam therapy and high-dose-rate brachytherapy also were used in selected patients. Con-touring and treatment planning were done using Eclipsesoftware (Varian Medical Systems, Palo Alto, Calif).

Data Collection

Primary tumors located in the nasal cavity, paranasalsinuses, nasopharynx, pytergopalantine fossa, infratem-poral fossa, mastoid region, and middle ear were classi-fied as PM RMS. All patients who had radiographic

RT Cone-Down Boost in Pediatric RMS/Eaton et al

Cancer April 15, 2013 1579

evidence of intracranial extension (ICE) or clinical signsof neurologic deficit (eg cranial nerve palsy) indicatingICE were classified as having ICE. Primary tumors in allother locations were classified as non-PM RMS accord-ing to primary tumor site.

Time from the start of chemotherapy to the initia-tion of RT was recorded in weeks and days for eachpatient. PTV initial and boost volumes were calculatedand recorded. Relative PTV reduction was calculated asthe percentage volume reduction from the initial PTVvolume to the boost PTV volume for all available patients.

Local failure was defined as tumor recurrence orprogression at the primary site, and distant failure wasdefined as disease progression at any site distant from theprimary tumor, as evident by radiographic imaging orphysical examination. Diagnostic imaging at the time oflocal failure was fused with an initial treatment planningCT for evaluation of the spatial relation between the siteof disease failure and the initial and boost PTVs, if appli-cable. Central nervous system failure with leptomeningealmetastasis was specifically labeled for independent analy-sis. Survival analysis was calculated from the time ofchemotherapy initiation until the first evidence of localfailure, distant failure, or death or was censored at date oflast follow-up for patients who remained alive with no evi-dence of local or distant relapse of RMS.

Statistical Analysis

Patient, disease, and treatment characteristics were sum-marized for patients with PM RMS and non-PM RMS.The mean and standard deviation of the continuous varia-bles, RT timing and relative PTV reduction, were com-pared using the Wilcoxon rank-sum test. Other categoricvariables, such as group, stage, histology, IMRT, ICE, andPTV reduction, were compared using the chi-square test orthe Fisher exact test. Survival analysis included Kaplan-Meier estimates of LC, disease-free survival (DFS), freedomfrom leptomeningeal failure (LMF), and overall survival(OS). The effect of RT timing, PTV reduction, IMRT, tu-mor histology, stage, grade, and PM versus non-PM diseasesite were assessed using a Cox proportional hazards model.The significance levels were set at .05 for all tests. All statisti-cal analyses were performed using the SAS statistical pack-age (version 9.2; SAS Institute, Inc., Cary, NC).

RESULTS

Patient Population

Fifty-five consecutive patients with RMS who had no evi-dence of metastatic disease were evaluated, including 18patients (33%) with PM RMS and 37 patients (67%) with

non-PM RMS of both head and neck and nonhead andneck sites. Of the 18 patients who had PM RMS, 12 (67%)had evidence of ICE. Patient demographics and tumor vari-ables are listed in Table 1. The majority of patients hadembryonal histology (64%) and group 3 disease (84%).

Radiation Therapy Variables

RT variables for the PM and non-PM groups are detailedin Table 2. For the subgroup of patients with PM whohad ICE, the median RT timing was 3 weeks (range, 0-23weeks), and a cone-down boost was received by 9 patients(75%). Among all patients, a cone-down boost wasreceived by 28 (51%) with a median relative reduction inPTV volume of 56% (range, 5%-91%). Figure 1 illus-trates a plan and dose-volume histogram comparison for 2patients who received a cone-down boost and the reduceddose to critical structures by way of reducing the high-dose treatment volume.

Overall Survival

After a median follow-up of 41 months, 49 patients (89%)remained alive. The 3-year Kaplan-Meier OS rate forpatients with PM RMS was 76% (95% CI, 0.49%-0.90%) compared with 96% (95% CI, 0.77%-0.99%)for patients with non-PM RMS (P ¼ .01) (Fig. 2). Noassociation was observed between OS and RT timing orthe receipt of a cone-down boost among patients withPM RMS or non-PM RMS. However, for the subset of12 patients who had PM RMS with ICE, delayed RTtiming was associated with decreased OS (P ¼ .055).There was no association between OS and the receipt of acone-down boost among patients who had PM RMSwith ICE (P ¼ .883).

Disease-Free Survival

The 3-year Kaplan-Meier estimates of DFS were 69%(95% CI, 0.41%-0.86%) for the PM subgroup and 88%(95% CI, 0.72%-0.95%) for the non-PM subgroup (Fig.2). No significant association was observed between DFSand RT timing or the receipt of a cone-down boost.Among the 12 patients who had PM RMS and evidence ofICE, 5 patients (42%) experienced disease progression,including 1 patient (12%) with local failure and 4 patients(30%) who had LMF as their first site of disease progres-sion. Three of the 4 patients with LMF died at a medianof 2 months (range, 2-3 months) from LMF, and 1 patientwas lost to follow-up. All 4 patients with LMF had receiveda cone down boost. The 3-year Kaplan-Meier estimate offreedom from LMF was 50% (95% CI, 0.152%-0.775%)for patients who had received a cone down boost, althoughthe association between LMF and cone-down boost wasnot statistically significant (P ¼ .21) (Fig. 3). Among the 4

Original Article


patients with LMF, 2 patients had embryonal histologywith RT timing of 3 weeks and 17 weeks, 1 patient had al-veolar histology with RT timing of 3 weeks, and 1 patienthad RMS not otherwise specified with RT timing of 8weeks. At the time of diagnosis, these patients had both adiagnostic MRI that was negative for leptomeningealenhancement and a cerebrospinal fluid examination thatwas negative for evidence of malignant cells.

Local Control

The 3-year Kaplan-Meier estimates of LC were 92%(95% CI, 0.5%4-0.99%) for the PM subgroup and 91%(95% CI, 0.75%-0.97%) for the non-PM subgroup (Fig.2). In total, 5 local failures were observed, including 2

patients (11%) with PM RMS and RT timing of 1 weekand 3 patients (8%) with non-PM RMS and RT timingof 11 weeks, 13 weeks, and 14 weeks. All 5 of the localfailures occurred in patients with stage III, group 3 diseaseat unfavorable sites, including tumors of the head andneck with embryonal histology (2 patients), prostate withembryonal histology (2 patients), and extremity with alve-olar histology (1 patient). Local failure occurred in 2patients (7%) who had received a cone-down boost and in4 patients (11%) who had received IMRT. In bothpatients who had received a cone-down boost and experi-enced local failure, the local failure occurred within theboost PTV volume (Fig. 4). In the other 3 patients whohad local failure for which there was no cone-down boost,

TABLE 2. Radiation Therapy Variables

Variable PM RMS, N ¼ 18 Non-PM RMS, N ¼ 36 P

RT timing: Median [range], wk 3.4 [0-23] 15.8 [1-43] .0001

PTV cone-down boost: No. of patients (%) 12 (67) 16 (43) .103

Relative PTV reduction: Median [range], %a 56 [5-81] 54.5 [21-91] .966

IMRT: No. of patients (%) 17 (94) 20 (54) .003

Abbreviations: IMRT, intensity-modulated radiotherapy; PM-RMS, parameningeal rhabdomyosarcoma; PTV, planning target volume; RT, radiotherapy.a The relative PTV volume reduction from the initial PTV to the boost PTV.

TABLE 1. Patient and Tumor Characteristics

No. of Patients (%)

Characteristic Total, N ¼ 55 PM, N ¼ 18 Non-PM, N ¼ 37 P

Sex

Boys 27 10 17 .391

Girls 28 8 20

Age: Median [range], y 6 [1-18] 4 [1-18] 6 [1-16] .766

Primary site

Nasopharynx 8 (44)

Nasal cavity/paranasal sinuses 4 (22)

Pytergopalatine/infratemporal fossa 4 (22)

Middle ear 2 (11)

Total PM 18 (33)

Head and neck, non-PM 13 (35) NA

Trunk/retroperitoneum/etc 9 (24)

Extremity 6 (16)

Bladder/prostate 5 (14)

GU: Nonbladder/nonprostate 5 (14)

Total non-PM 37 (67)

Histologic type

Embryonal 35 (64) 12 (67) 23 (62) .785

Alveolar 18 (33) 5 (28) 13 (35)

NOS 2 (3) 1 (5) 1 (3)

ISRG stage

I 15 (27) 1 (6) 14 (38) .002

II 13 (24) 9 (50) 4 (11)

III 26 (49) 8 (44) 19 (51)

ISRG group

2 9 (16) 2 (11) 7 (19) .702

3 46 (84) 16 (89) 30 (81) .702

Abbreviations: GU, genitourinary; IRSG, International Rhabdomyosarcoma Study Group; NA, not applicable; NOS, not otherwise specified; PM,

parameningeal.



the local recurrence developed within the initial PTV. Norecurrences developed marginal to the treatment volumes.No significant association was observed between LC andRT timing or the receipt of a cone-down boost.

DISCUSSIONTo our knowledge, this is the largest series to date assess-ing the effect of a PTV reduction by cone-down boost onpatterns of disease recurrence in patients with pediatricRMS and is the first to include all primary disease sites.We demonstrated excellent LC with no marginal failures,although 50% of patients had received a cone-down boostin response to tumor regression with induction chemother-apy. A local failure rate of 9% was observed, and all localfailures occurred in patients who had group 3 and stage IIIdisease of unfavorable sites, which is comparable to the 7%to 19% local failure rate observed in the IRSG IV studyamong similar patients.14 Thus, the receipt of a response-based boost in conjunction with delayed RT timing in thisseries had no discernible impact on LC and allowed for aconsiderable reduction in the boost PTV volume.

The clinical impetus for using this strategy of adapt-ive treatment planning with highly conformal RT is toreduce the volume of normal tissues exposed to the fullRT treatment dose, thereby reducing treatment-relatedmorbidity. Indeed, the prevalence of long-term sequelae

in survivors of pediatric RMS who received treatmentwith definitive RT is well recognized,15-17 and the dose-dependent nature of such sequelae also has been demon-strated.18,19 Although toxicity data were not presentedhere, and further investigation is needed to estimate theclinical benefit of cone-down boost, reducing the high-dose RT volume may be expected to reduce both acuteand late treatment toxicity in this vulnerable pediatricpopulation. Experience with the use of tight margin pro-ton RT with a steep dose fall has demonstrated expectedclinical outcomes with favorable rates of late effects in pe-diatric RMS and further supports this assumption.20

The 3-year Kaplan-Meier estimates of LC and OSamong patients with PM RMS in our series were 92% and76%, respectively, which are comparable to the results forpatients with PMRMS treated on prior IRSG studies.3 BothDFS and OS were shorter among patients with PM RMSthan in those with non-PM RMS, as expected given theknown negative prognostic impact of PMprimary sites.21,22

Although the use of a cone-down boost in conjunc-tion with delayed RT timing was not associated with adetriment to disease outcomes in all PM patients, amongthe subgroup of patients who had PM RMS with ICE,delayed RT timing was associated with decreased survival.It is important to note that this association represents atrend that was observed in a small subset of patients (n ¼

Figure 1. Two plans for radiotherapy (RT) illustrate the reduced dose to nearby critical structures with the receipt of a cone-down boost. (a) An RT plan without a cone-down boost (a1,b1; triangle) is compared with an RT plan that includes a cone-downboost (a2,b2; square), indicating a reduced dose to the brain (pink), temporal lobe (light blue), hypothalamic-pituitary-adrenalaxis (orange), optic chiasm (hot pink), optic nerve (dark blue), right eye (green dash), and lacrimal gland (yellow) in a patientwith optic rhabdomyosarcoma (RMS). (b) In a patient with prostate RMS, the plan with a cone-down boost reduced the dose tothe right and left femur (purple), bladder (yellow), and rectum (brown).


Original Article

12) and should be interpreted with caution. Four of the12 patients with ICE (30%) developed LMF as their firstsite of disease failure, and their median survival was 2months from the time of disease progression. All 4

patients who had LMF received a cone-down boost andinitiated RT at 3 weeks or more from the time of chemo-therapy initiation. This represents an unexpected highrate of LMF in a cohort of patients with PM RMS andICE who received treatment with a cone-down boost anddelayed RT timing.

In a comprehensive review of 595 patients with PMRMS who received treatment on IRSG studies II, III, andIV, Michalski and colleagues reported improved LC withthe early initiation of RT among patients with ICE (5-year LC rate: 84% for �2 weeks vs 63% for >2 weeks; P¼ .07) and a central nervous system failure rate of 9%.3

Compared with that review, our patient population con-sisted of a proportionately larger number of patients withICE (68% vs 38%) with a higher rate of central nervoussystem failure (30% vs 9%). Although we did not observedecreased LC with delayed RT, all patients with LMF hadRT timing >2 weeks; thus, our results similarly supportthe use of early RT for this high-risk population. For thepatients with ICE who received treatment on IRSG stud-ies II, III, and IV, margins from 2 cm to 5 cm around theGTV were used to define the PTV, and some patientsreceived whole-brain RT. Although no decrease in centralnervous system failure was observed with the addition ofwhole-brain RT, the GTV to PTV expansion marginswere larger than the margin expansions used in ourpatients. The higher rate of LMF observed in our patientswho also had a response-based PTV reduction may sug-gest that larger volume coverage is necessary to control

Figure 2. These Kaplan-Meier curves illustrate overall survival,disease-free survival, and local control for patients with para-meningeal (red line) and nonparameningeal (blue line)rhabdomyosarcoma.

Figure 3. This Kaplan-Meier curve illustrates freedom fromleptomeningeal failure and overall survival for patients withparameningeal rhabdomyosarcoma and intracranial extensionseparated according to receipt (yes/no) of radiation therapywith a cone-down boost to reduce the planning target vol-ume (PTV).



adjacent areas at high risk for meningeal tumor extensionin patients who have evidence of ICE.

The observed trend toward increased central nerv-ous system failure with the receipt of a cone-down boostand decreased survival with delayed RT timing amongpatients with PM RMS and ICE, although they representa small subset, is nevertheless hypothesis-generating. Chil-dren with PM RMS who have ICE pose a unique chal-lenge for disease control. Our data suggest that, in thishigh-risk subset, early initiation of RT and wider marginRT may be more important for adequate disease controlthan in other primary sites. The importance of compre-hensive RT may be related to the reduced efficacy of sys-temic therapy for targeting ICE of disease and maywarrant additional strategies for central nervous system-directed therapy. Pending further investigation, we rec-ommend maintaining coverage of adjacent areas at highrisk for meningeal extension within the boost PTV vol-ume when this practice would not otherwise compromisethe recognized dose limitations of nearby criticalstructures.

In addition to the relatively small patient numbersand heterogeneous patient population included in thecurrent study, as with any retrospective series, interpreta-tion of our data is limited by the possibility that addi-tional, uncontrolled variables and biases may haveinfluenced outcomes. Patients with group 2 disease, forwhom the utility and benefit of cone-down boost may notbe applicable, were included in this study. However, it

would be expected that the use of cone-down boost in thehigher risk subset of group 3 patients would then favor amore conservative estimate of any ill effects from thereduced volume technique and, thus, the inclusion ofgroup 2 patients does not alter our conclusion regardingthe safety of reducing high-dose treatment volume inpatients without ICE. Poor documentation of limitedpatient follow-up has prevented analysis of toxicity data incorrelation with reduced volume RT, and any potentialbenefit to reduced treatment toxicity with the use of cone-down boost is not directly supported by the data pre-sented here.

In conclusion, among patients with non-PM RMS,delayed initiation of RT and use of a cone-down boostallowed for significant reductions in treatment volumeswith no apparent compromise in LC or survival. How-ever, among patients with PM RMS and radiographic orclinical evidence of ICE, a high rate of intracranial pro-gression with decreased survival was observed among thepatients who received a cone-down boost and who hadRT delayed. Further analysis is required to define thepotential clinical benefit from a cone-down boost inpatients with non-PM RMS and to assess the relationbetween reduced PTV volumes and RT timing and therisk of leptomeningeal dissemination in patients with PMRMS and ICE.

FUNDING SOURCESNo specific funding was disclosed.

Figure 4. Two local failures occurred in patients who received radiation therapy with a cone-down boost, including (Left) 1patient with prostate rhabdomyosarcoma (RMS) and (Right) 1 patient with nasopharyngeal RMS. In both patients, the local failure(yellow) occurred within the boost planning target volume (PTV) (blue) and the corresponding 95% isodose line (red).

Original Article


CONFLICT OF INTEREST DISCLOSURESThe authors made no disclosures.

REFERENCES1. Wolden SL, Anderson JR, Crist WM, et al. Indications for radio-

therapy and chemotherapy after complete resection in rhabdomyo-sarcoma: a report from the Intergroup Rhabdomyosarcoma StudiesI to III. J Clin Oncol. 1999;17:3468-3475.

2. Regine WF, Fontanesi J, Kumar P, et al. Local tumor control inrhabdomyosarcoma following low-dose irradiation: comparison ofgroup II and select group III patients. Int J Radiat Oncol Biol Phys.1995;31:485-491.

3. Michalski JM, Meza J, Breneman JC, et al. Influence of radiationtherapy parameters on outcome in children treated with radiationtherapy for localized parameningeal rhabdomyosarcoma in Inter-group Rhabdomyosarcoma Study Group trials II through IV. Int JRadiat Oncol Biol Phys. 2004;59:1027-1038.

4. Douglas JG, Arndt CA, Hawkins DS. Delayed radiotherapy follow-ing dose intensive chemotherapy for parameningeal rhabdomyosar-coma (PM-RMS) of childhood. Eur J Cancer. 2007;43:1045-1050.

5. Smith SC, Lindsley SK, Felgenhauer J, Hawkins DS, Douglas JG.Intensive induction chemotherapy and delayed irradiation in themanagement of parameningeal rhabdomyosarcoma. J Pediatr Hema-tol Oncol. 2003;25:774-779.

6. Eaton B, Katzenstein H, Sutter A, et al. Delayed radiation therapytiming and use of intensity-modulated radiation therapy in non-head and neck pediatric rhabdomyosarcoma [published online aheadof print July 20, 2012]. J Radiat Oncol. 2012.

7. Lin C, Donaldson SS, Meza JL, et al. Effect of radiotherapy techni-ques (IMRT vs 3D-CRT) on outcome in patients with intermedi-ate-risk rhabdomyosarcoma enrolled in COG D9803—a reportfrom the Children’s Oncology Group. Int J Radiat Oncol Biol Phys.2012;82:1764-1770.

8. McDonald MW, Esiashvili N, George BA, et al. Intensity-modu-lated radiotherapy with use of cone-down boost for pediatric head-and-neck rhabdomyosarcoma. Int J Radiat Oncol Biol Phys.2008;72:884-891.

9. Wolden SL, Wexler LH, Kraus DH, Laquaglia MP, Lis E, MeyersPA. Intensity-modulated radiotherapy for head-and-neck rhabdomy-osarcoma. Int J Radiat Oncol Biol Phys. 2005;61:1432-1438.

10. Curtis AE, Okcu MF, Chintagumpala M, Teh BS, Paulino AC.Local control after intensity-modulated radiotherapy for head-and-

neck rhabdomyosarcoma. Int J Radiat Oncol Biol Phys.2009;73:173-177.

11. Chen C, Shu HK, Goldwein JW, Womer RB, Maity A. Volumetricconsiderations in radiotherapy for pediatric parameningeal rhabdo-myosarcomas. Int J Radiat Oncol Biol Phys. 2003;55:1294-1299.

12. Lawrence W Jr, Gehan EA, Hays DM, Beltangady M, MaurerHM. Prognostic significance of staging factors of the UICC stagingsystem in childhood rhabdomyosarcoma: a report from the Inter-group Rhabdomyosarcoma Study (IRS-II). J Clin Oncol. 1987;5:46-54.

13. Maurer HM, Beltangady M, Gehan EA, et al. The IntergroupRhabdomyosarcoma Study-I. A final report. Cancer. 1988;61:209-220.

14. Raney RB, Maurer HM, Anderson JR, et al. The Intergroup Rhab-domyosarcoma Study Group (IRSG): major lessons from the IRS-Ithrough IRS-IV studies as background for the current IRS-V treat-ment protocols. Sarcoma. 2001;5:9-15.

15. Punyko JA, Mertens AC, Gurney JG, et al. Long-term medicaleffects of childhood and adolescent rhabdomyosarcoma: a reportfrom the Childhood Cancer Survivor Study. Pediatr Blood Cancer.2005;44:643-653.

16. Raney RB, Asmar L, Vassilopoulou-Sellin R, et al. Late complica-tions of therapy in 213 children with localized, nonorbital soft-tis-sue sarcoma of the head and neck: a descriptive report from theIntergroup Rhabdomyosarcoma Studies (IRS)-II and IRS-III. IRSGroup of the Children’s Cancer Group and the Pediatric OncologyGroup. Med Pediatr Oncol. 1999;33:362-371.

17. Paulino AC, Simon JH, Zhen W, Wen BC. Long-term effects inchildren treated with radiotherapy for head and neck rhabdomyosar-coma. Int J Radiat Oncol Biol Phys. 2000;48:1489-1495.

18. Merchant TE, Goloubeva O, Pritchard DL, et al. Radiation dose-volume effects on growth hormone secretion. Int J Radiat OncolBiol Phys. 2002;52:1264-1270.

19. Merchant TE, Rose SR, Bosley C, Wu S, Xiong X, Lustig RH.Growth hormone secretion after conformal radiation therapy in pe-diatric patients with localized brain tumors. J Clin Oncol.2011;29:4776-4780.

20. Childs SK, Kozak KR, Friedmann AM, et al. Proton radiotherapyfor parameningeal rhabdomyosarcoma: clinical outcomes and lateeffects. Int J Radiat Oncol Biol Phys. 2012;82:635-642.

21. Maurer HM, Gehan EA, Beltangady M, et al. The IntergroupRhabdomyosarcoma Study-II. Cancer. 1993;71:1904-1922.

22. Crist W, Gehan EA, Ragab AH, et al. The Third Intergroup Rhab-domyosarcoma Study. J Clin Oncol. 1995;13:610-630.



Documents

Radiation therapy target volume reduction in pediatric rhabdomyosarcoma