Oral Oncology 49 (2013) 269–276

Contents lists available at SciVerse ScienceDirect

Oral Oncology

journal homepage: www.elsevier .com/locate /ora loncology

Volumetric staging in oropharyngeal cancer patients treated with definitive IMRT

Gabriela Studer ⇑, Christoph GlanzmannDepartment of Radiation Oncology, University Hospital Zurich, Zurich, Switzerland

a r t i c l e i n f o

Article history:Received 15 August 2012Received in revised form 22 September2012Accepted 26 September 2012Available online 22 October 2012

Keywords:Volumetric stagingTumor volume and outcomeVolumetric and cut off valuesOropharyngeal cancerMesopharyngeal cancer and SIB-IMRT

1368-8375/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.oraloncology.2012.09.014

⇑ Corresponding author. Address: Department of RaHospital Zurich, Raemistrasse 100, 8091 Zurich, CH, S39 31; fax: +41 44 255 45 47.

E-mail address: gabriela.studer@usz.ch (G. Studer)

s u m m a r y

Background: The superiority of volumetric staging (VS) over TNM/TNM-grouping system was previouslyprospectively tested in our head neck cancer population treated with intensity-modulated radiotherapy(IMRT); gross tumor volume (GTV) was the strongest predictor for disease control. Aim of this work wasto specifically assess the prognostic value of VS in oropharyngeal cancer (OC).Patients: Between 04/2002 and 12/2011, 277 consecutive OC patients underwent definitive IMRT. Mean/median follow-up was 33/27 months (3–113). Three volumetric cut-offs were used (resulting in 4 GTVsubgroups: 1–15 cc (14%), 16–70 cc (62%), 71–130 cc (20%), >130 cc (4%)).Methods: Outcome in the OC subgroup was prospectively assessed with VS and compared with thatresulting from TNM and AJCC staging.Results: Primary GTV was most reliably predicting local control (p < 0.0001), all other outcome parame-ters were predicted best by the total GTV (p < 0.0001).Conclusion: This is -to our knowledge- the first volumetric staging system for OC, and was found to bemost reliable in predicting outcome in OC patients treated with IMRT.

� 2012 Elsevier Ltd. All rights reserved.

Introduction

Former volumetric staging (VS) evaluations revealed the tumorvolume load being the most reliable predictor of disease controland survival following definitive IMRT of our head neck cancer tu-mor patients: our first analysis was based on 88 retrospectivelyand 84 prospectively analyzed patients (treated between 01/2002–12/2004 and 01–11/2005; n total = 172; 69 OC casesincluded),1 by using two cut off values (15 cc and 70 cc); a follow-ing prospective evaluation was focused on gross tumor volume(GTV) based prediction of distant spread (409 patients, 125 OCcases included),2 a third prospective evaluation aimed furtherGTV-based assessment of our patient subgroup with large tumorvolume load of >70 cc (112 patients; 62 OC), in order to try toidentify palliative patients out of this prognostically ‘unfavourable’subgroup.3 That latter analysis lead to a third cut-off (130 cc),translating in four GTV subgroups: patients with a primary or totalGTV of 1–15 cc (favorable), vs. 16–70 cc (intermediate), vs.71–130 cc (unfavorable), vs. >130 cc (very unfavorable). Whilethe primary GTV was used to predict local control rates, totalGTV (primary plus nodal disease) was shown to best predict nodaland local–regional control rates, disease free and overall survivalrates, respectively.

ll rights reserved.

diation Oncology, Universitywitzerland. Tel.: +41 44 255

.

The volume criterion has 2006 been applied prospectively inour definitive head neck IMRT cohort and is since in routine useas an integrated predictive parameter in decision making, dose def-inition and pre-therapeutic patient information.

The potential weakness of the above described analyses is thatall pharyngeal tumor sites were included in the evaluations (dueto, up to now, too small sample sizes for separate analyses fornaso-, oro-, and hypopharynx tumors).

Aim of the present work was to assess the value of the describedvolumetric staging specifically for OC patients.

We hypothesized VS being superior in predicting disease con-trol in definitively irradiated SIB-IMRT OC patients compared withTNM/American Joint Committee on Cancer (AJCC) grouping stagingsystems.

Materials and methods (Table 1)

Patients: between 4/2002 and 12/2011, 277 consecutive OC pa-tients have been treated with definitive simultaneously integratedboost (SIB-)IMRT technique for squamous cell carcinoma at thedepartment of Radiation Oncology, University Hospital Zurich(USZ). Anatomical tumor sites were lateral oropharynx (lateralpharyngeal wall, tonsil, vallecula; 53%), and central oropharynx(base of tongue, soft palate; 47%). The mean/median follow up timewas 33/27 months (3–113). Concomitant systemic treatment wasgiven in 89%: cisplatin only (56%); cetuximab only (24%); switchfrom concomitant cisplatin to concomitant cetuximab, due to

Table 1Patient and disease characteristics.

Parameters Oropharynx

N patients 277Gender (f:m) 23%:77%Mean age (range) 62 (39–96)Mean FU (range), months 33 (3–113)Squamous cell carcinoma 277

Total tumor volumeMean (range) 50.5 cc (3–

216)V1 1–15 cc 38 (14%)V2 16–70 cc 175 (63%)V3 71–130 cc 54 (20%)V4 >130 cc 10 (4%)

T stageRecurrence 5 (2%)T1 26 (9%)T2 93 (33%)T3 64 (23%)T4 89 (32%)

N stageRecurrence 2 (1%)N0 47 (7%)N1–2b 139(50%)N2c 79 (29%)N3 10 (4%)

Stage groupingRecurrence 5 (2%)I 0II 23 (8%)III 36 (13%)IVA 189 (68%)IVB 24 (9%)

Cone, systemic therapyNone 31 (11%)Cisplatin 195 (70%)Cetuximab only 50 (18%)Cisplatin switched to

Cetuximab28 (10%)

Table 3Volumetric staging (total gross tumor volume (GTV-) based) related to TN groupstages.

Staging II III IVA IVB Recurrence Total

Total GTVV1 (1–15 cc) 12 10 15 0 1 38V2 (16–7 occ) 11 26 134 1 3 175V3 (71–13 occ) 0 0 39 14 1 54V4 (>13 occ) 0 0 1 9 0 10

Total 23 36 189 24 5 277

Table 4Volumetric staging (primary gross tumor volume (GTV-) based) related to T stages.

Staging T1 T2 T3 T4 Recurrence Total

Primary GTVV1 (1–15 cc) 26 50 5 5 1 87V2 (16–70 cc) 0 43 54 61 3 161V3 (71–13 occ) 0 0 5 23 1 29V4 (>13 occ) 0 0 0 0 0 0

Total 26 93 64 89 5 277

270 G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276

impaired renal or hearing function, tinnitus, skin reactions (9%). In11%, no systemic therapy was given, mainly due to co-morbidity orage. Cisplatin was given in weekly doses of 40 mg/m2 at 1 day aweek; cetuximab was given with 400 mg/m2 as loading dose, fol-lowed by 250 mg/m2 at 1 day a week. None of the includedpatients underwent induction chemotherapy. The pre-treatmentstandard imaging modality was positron emission tomography–

Table 2T/N stages of the analyzed cohort.

computed tomography (PET–CT, most with intravenous contrastagent) in >90% of the cohort, ± CT or magnetic resonance tomogra-phy (MRI). Planning CT images (2 mm slice thickness) wereacquired from the top of the oribta to the level of the carina withcontrast agent infusion in all eligible patients.

Methods

OC patients have been analyzed using a formerly defined andassessed tumor volume based staging system.1,3 The resulting dis-ease control rates were compared with those resulting from UICCTNM and group staging. Gross tumor volume (GTV) assessmenthas formerly been described in detail1: GTV delineation of all pa-tients was based on physical examination and endoscopy as wellas on diagnostic preoperative Magnetic Resonance Imaging(MRI), computed tomography (CT) and positron emission tomogra-phy (PET). All GTVs were contoured or reviewed by at least one ofthe authors (GS or CG) on all relevant axial computerized imageswithout interpolation, in most cases also by a third staff physician.In addition, the wide volumetric ranges (cut offs: 15 cc, 70 cc,

Table 5Disease control: comparison between the different staging systems.

Remark: recurrences (n = 5) not included in the TN-/grouping analysis.LC/NC/DMFS/DFS/OAS: local control/nodal control/distant metastasis free survival/overall survival.⁄The local control rate calculation bases on the primary tumor volume, while all other outcome parameters base on total tumor volumes.⁄⁄Statistically significant difference between the T-stages, however not in the expected way (OAS of T3 superior to T1/T2, T1 similar to T4).

G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276 271

130 cc) render the system quite robust with respect to inter-indi-vidual contouring differences. Volumetric three-dimensional mea-surements (cm3) of contoured structures were calculated by theVarian Treatment Planning System volume algorithm (Eclipse�

External Beam Planning System, Version 7.3.10 and PRO 8.9, AAA8.9, Varian Medical Systems).

The following principle has been followed (total GTV = primaryand nodal GTV):

Volumes

Outcome parameters

Primary GTV (cut-offs15/70/130 cc):

Local control

Total GTV (cut-offs 15/70/130cc):

Nodal and local-regional control,distant control, disease free survival,overall survival

Table 6Prognostic sub-division of T stages with help of volumetric staging.

272 G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276

All patients have been definitively irradiated using IMRT withsimultaneously integrated boost (SIB) technique,4 using IMRTschedules with 33 � 2.11 Gy to the boost planning target volume(boost PTV), or, -as a mild dose escalation-, with 33 � 2.2 Gy to largeGTVs, or with 35 � 2.0 Gy to the boost PTV in patients with centralnervous structures in the PTV, or with substantial parts of the larynxinvolved. The elective dose was 54 Gy, intermediate doses of 60–66 Gy were delivered depending on individual risk estimation.

All but eight patients are in a regular follow up program in ourHNC joint centre clinic or maxillofacial surgery clinic at the Univer-sity Hospital of Zurich. Routine tests included, besides the history,physical examination and endoscopy of the pharyngeal–laryngealregion. If these tests were of no evidence of disease, usually no fur-ther tests were done but a computed tomography scan or positronemission tomography (PET)–CT or magnetic resonance imaging at1 year post-treatment in the majority of patients.

SIB-IMRT

We used an extended-field SIB-IMRT technique, where the pri-mary tumor is covered in one phase along with the regional lymph

nodes. Irradiation was delivered with five or seven coplanar beamangles by a 6MV dynamic MLC system (Varian Medical Systems,Palo Alto, CA) using a sliding window technique, or using volumet-ric modulated rapid arc technique (VMAT, since 04/2010). Patientswere immobilized from head to shoulders with commerciallyavailable thermoplastic masks in the supine position.

Target volumes were delineated as follows: the gross tumorvolume (GTV) included the gross extent of the primary disease andinvolved lymph node metastases, taking clinical and radiologicalfindings into account; planning target volume 1 (PTV1) was definedby adding a (5–)10–20 mm margin to the GTV, dependent on theGTV proximity to critical structures (tight margins mainly if proxim-ity of the GTV to the mandible bone, spinal cord or brachial plexus,generous margins towards tongue and pharyngeal wall, or in caseswith difficult identification of the GTV based on the imaging, or ifclinical findings not entirely represented in the available imaging);PTV2 covered areas considered at high risk for potential microscopicdisease; PTV3 included the clinically negative cervical lymphaticpathways to the supraclavicular fossae (elective PTV coverage).

SIB-IMRT technique was performed using the followingschedules:

Figure 1 Local control rates according to T, stage grouping, volumetric staging (asno primary tumor measured >130 cc, there was no ‘V4’ in the volumetric graph atthe bottom).

Figure 2 Distant metastasis free survival rates according to T, stage grouping,volumetric staging.

G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276 273

– SIB2.00 (n = 8%): Daily dose 2.00 Gy (PTV1)/1.70 Gy (PTV2)/1.54 Gy (PTV3) to a total dose of 70.00 Gy (5 fractions/week).

– SIB2.11 (77%): Daily dose 2.11 Gy (PTV1)/1.80 Gy (PTV2)/1.64 Gy(PTV3) to a total dose of 69.60 Gy (5 fractions/week).

– SIB2.2 (15%): Daily dose 2.2 Gy (PTV1)/2.0 Gy (PTV2)/1.64 Gy(PTV3) to a total dose of 66.0 Gy (5 fractions/week).

The dose was normalized to the mean dose in PTV1. For inten-sity optimization, the prescribed dose encompassed at least 95% ofthe PTV. Additionally, no more than 20% of any PTV received >110%of its prescribed dose, whilst no more than 1% of any PTV received<93% of the prescribed dose. Hundred % of the prescription dose in-cludes the primary GTV in patients with a primary tumor volume>15 cc (since �2007).

Our interdisciplinary in-house guidelines include an electiveneck dissection in patients with initially diagnosed nodal metasta-sis >3 cm (n = 50, 18%).

Statistics

Statistical calculations were performed using the statistics pro-gram implemented in StatView� (Version 4.5). Proportions werecompared using the Chi-square test. P values < 0.05 were consid-ered statistically significant. Univariate analyses were performedwith a Cox proportional hazards regression model in StatView�.

Results

Five year local and nodal control rates were 75% and 82%, 5 yeardistant metastasis free, disease free, and overall survival of the

274 G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276

entire OC cohort were 85%, 62%, and 78%, respectively. 72% of allpatients were alive with no evidence of disease when last timeseen. Tables 2–4 show the cohort analyzed according to VS relatedto the TN, volumetric and AJCC group staging; in Table 5, diseasecontrol parameters are listed analyzed according to the three dif-ferent staging systems, with most significant outcome differencesfound for the VS. The comparison of VS with T1/2, T3/4 (as the en-tire cohort contains only 26 T1 cases, T1/T2 and T3/T4 were takentogether – T1 vs. T2 and T3 vs. T4 showed no relevant outcome dif-ferences among each other, see Table 5) and AJCC lV4 (the otherAJCC stages included too low number of patients) reveals the po-tential of the VS to significantly sub-divide T and AJCC stages intodifferent prognostic subgroups, Table 6. In the grouped T1/2 co-hort, VS reached only a significantly difference in OAS and DMFS,while in the grouped T3/4 cohort and mainly in AJCC stage lV4,all outcome parameters showed a highly significant difference byVS. V4 (>130 cc total GTV) got filtered out as poorest subgroupwith no 5 year survival.

Figs. 1–3 show Kaplan Meier survival curve comparisons be-tween the three staging systems.

In sum, the VS showed to be most reliable in predicting out-come after primary radiotherapy in this definitive IMRT cohort ascompared to the other staging systems.

Discussion

This work was focused on volumetric staging in our large singlecentre cohort of OC patients treated with definitive IMRT(-chemo-therapy) – to our knowledge the first prospectively tested volu-metric staging system for OC. The used VS had previouslyprospectively been tested for head neck cancer patients and hadbeen shown to highly reliably predict outcome.1–3 Choosing theused cut offs was based on a quite pragmatic retrospective ap-proach tested in our very first cohort back in 2006: all tumor vol-umes of that subgroup (n = 88) have been ranked from smallest tolargest depicted as bars, and locally failed cases were marked. Twocut-off values (15/70 cc) were then visually chosen based on thefrequency of failures in the so ranked tumor volumes, translatinginto three groups (favorable, intermediate and unfavorable out-come).1 A next step using the same visual method lead to the iden-tification of a ‘very unfavourable’ fourth group out of the‘unfavorable’ subgroup with GTV > 70 cc (3rd cut off: 130 cc).3

However, those former VS analyses included all head neck squa-mous cell tumor sites (but glottis cancer). The VS could herewithspecifically be confirmed as a useful staging system for non-surgi-cal OC patients treated with definitive IMRT, resulting in highlystatistically significant differences of OC tumor volume related dis-ease control rates (Figs. 1–3 and Table 5).

Its main advantage – beside the fact of a reliable outcome pre-diction- may be its potential to improve the comparability of irra-diated cohorts, as the key parameter in radiation therapy is, mainlyin the era of IMRT, number of tumor cells rather than the anatomicdisease extension. In addition, given T and AJCC stages seem to ren-der sub-dividable by the VS into subgroups with different progno-sis, Table 6.

The weakness of the work is the missing HPV status as animportant predictive parameter in OC patients.

Figure 3 Overall survival rates according to T, stage grouping, volumetric staging.

Disease control following IMRT

An overview over the IMRT literature on OC patients is shown inTable 7. Outcome following IMRT in OC as the most frequent headneck cancer site has, among all head neck sites, most extensivelybeen analyzed over the past decade.5–22 However, from the radia-tion oncology perspective, also in the IMRT era, comparison of

IMRT cohorts unfortunately remains limited due to retrospectiveevaluations based on TN staging. The percentage of advanced stagepatients referred for definitive radiation therapy differs among dif-ferent centers. In addition, the threshold to treat patients withloco-regionally very advanced TN stage with curative intention isnot necessarily objective but rather institution or individually de-fined, making inter-institutional cohort comparisons limitedlyinterpretable. To summarize the available OC IMRT data as listedin Table 7, disease control seems generally good, and own resultsin the published range, taking into account that all other cohortsbut one6 contained less advanced T stages (0–48% T3/4, vs. 55%in the own cohort) or no T stage information, all others but onereport21 presented survival rates <5 year observation time, and inhalf the reports local control rates and distant metastasis free sur-vival were not provided.

Table 7Oropharyngeal cancer: selected publications on definitive IMRT.

cons.: consecutively treated patients.aPostoperative and definitive IMRT patients (identical outcome).b4 Postoperative and 62 definitive IMRT treated patients.c10/66 PEG-dependent’ at ’last follow up’: no minimal time since RT indicated.dMost with definitive IMRT.e25% PEG dependent at 2 years, 6% G3 xerostomia.fGrade 3 reactions only in part related to the assessed tumor entity (oropharynx/larynx).gUpdate of Ref. 7.

G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276 275

IMRT for OC was reported to translate in �10% higher diseasecontrol than that reached following former historic three-dimen-sional radiation techniques at two single institutions,10,22 however,by retrospective comparison and as such of limited value.

Literature on the impact of tumor volume in OC

With respect to the volumetric approach to predict outcome inIMRT(-chemotherapy) OC patients, we found two other reportspublished with focus on the impact of tumor volume in IMRT OCpatients: Chao et al.6 found GTV and nodal GTV independent riskfactors for loco-regional and disease free survival in a multivariateanalysis of 31 patients; Lok et al.23 reported recently on the so farlargest published cohort of 340 OC patients treated with definitiveIMRT(-chemotherapy), finding the primary GTV associated withoverall survival, local and distant metastatic failure when analyzedby dichotomizing the primary GTV size (5 year overall survival�60% vs. 94% for >/<32.78 cm3). The emphasis of both analyseswas on the impact of tumor volume; tumor volume was not pro-vided to serve as a generally usable volumetric staging system withevaluated cut off values – similarly as in several reports publishedon OC patients treated in the historic non-IMRT era: Nathu et al.observed a large variation in tumor volume within a given T stage,with tumor volume having marginal influence on local control

(n = 114).24 Been et al. did not find a significant correlation be-tween tumor volume and loco-regional control in 79 irradiatedOC patients.25 Hermans et al. investigated 112 patients, finding tu-mor volumes significantly correlated with local recurrence rates,but not significantly related to locoregional outcome.26 Chung etal. found statistically significant outcome differences in 42 OC pa-tients by using a retrospectively cut off value of 35 cc.27

In contrast, the here presented VS may be taken as a distinctvolumetric staging system, and, in addition, was found availableto sub-divide outcome in T(N) and AJCC lVA stages (Table 6).

Conclusion

The presented volumetric staging system showed to be mostreliable to predict outcome in our large non-surgical OC patient co-hort treated with IMRT. Volumetric staging combined with TNM/AJCC staging may offer better comparability of irradiated cohorts,and may offer the potential to divide T stages into subgroups withdifferent outcome.

Conflict of interest statement

None declared.

276 G. Studer, C. Glanzmann / Oral Oncology 49 (2013) 269–276

References

1. Studer G, Lutolf UM, El-Bassiouni M, et al. Volumetric staging (VS) is superior toTNM and AJCC staging in predicting outcome of head and neck cancer treatedwith IMRT. Acta Oncol 2007;46(3):386–94.

2. Studer G, Seifert B, Glanzmann C. Prediction of distant metastasis in head neckcancer patients: implications for induction chemotherapy and pre-treatmentstaging? Strahlenther Onkol 2008;184(11):580–5.

3. Studer G, Rordorf T, Glanzmann C. Impact of tumor volume and systemictherapy on outcome in patients undergoing IMRT for large volume head neckcancer. Radiat Oncol 2011;6:120.

4. Studer G, Huguenin PU, Davis JB, et al. IMRT using simultaneously integratedboost (SIB) in head and neck cancer patients. Radiat Oncol 2006;1:7.

5. Eisbruch A, Marsh LH, Dawson LA, et al. Recurrences near base of skullafter IMRT for head-and-neck cancer: implications for target delineation inhigh neck and for parotid gland sparing. Int J Radiat Oncol Biol Phys2004;59(1):28–42.

6. Chao KS, Ozyigit G, Blanco AI, et al. Intensity-modulated radiation therapy fororopharyngeal carcinoma: impact of tumor volume. Int J Radiat Oncol Biol Phys2004;59(1):43–50.

7. Yao M, Nguyen T, Buatti JM, et al. Changing failure patterns in oropharyngealsquamous cell carcinoma treated with intensity modulated radiotherapy andimplications for future research. Am J Clin Oncol 2006;29(6):606–12.

8. Lawson JD, Otto K, Chen A, et al. Concurrent platinum-based chemotherapy andsimultaneous modulated accelerated radiation therapy for locally advancedsquamous cell carcinoma of the tongue base. Head Neck 2008;30(3):327–35.

9. Garden AS, Morrison WH, Wong PF, et al. Disease-control rates followingintensity-modulated radiation therapy for small primary oropharyngealcarcinoma. Int J Radiat Oncol Biol Phys 2007;67(2):438–44.

10. Hodge CW, Bentzen SM, Wong G, et al. Are we influencing outcome inoropharynx cancer with intensity-modulated radiotherapy? An inter-eracomparison. Int J Radiat Oncol Biol Phys 2007;69(4):1032–41.

11. Schoenfeld GO, Amdur RJ, Morris CG, et al. Patterns of failure and toxicity afterintensity-modulated radiotherapy for head and neck cancer. Int J Radiat OncolBiol Phys 2008;71(2):377–85.

12. Huang K, Xia P, Chuang C, et al. Intensity-modulated chemoradiation fortreatment of stage III and IV oropharyngeal carcinoma: the University ofCalifornia-San Francisco experience. Cancer 2008;113(3):497–507.

13. Sanguineti G, Gunn GB, Endres EJ, et al. Patterns of locoregional failure afterexclusive IMRT for oropharyngeal carcinoma. Int J Radiat Oncol Biol Phys2008;72(3):737–46.

14. Rusthoven KE, Raben D, Ballonoff A, et al. Effect of radiation techniques intreatment of oropharynx cancer. Laryngoscope 2008;118(4):635–9.

15. Eisbruch A, Harris J, Garden AS, et al. Multi-institutional trial of acceleratedhypofractionated intensity-modulated radiation therapy for early-stageoropharyngeal cancer (RTOG 00-22). Int J Radiat Oncol Biol Phys 2010;76(5):1333–8.

16. Daly ME, Le QT, Maxim PG, et al. Intensity-modulated radiotherapy in thetreatment of oropharyngeal cancer: clinical outcomes and patterns of failure.Int J Radiat Oncol Biol Phys 2010;76(5):1339–46.

17. Saba NF, Edelman S, Tighiouart M, et al. Concurrent chemotherapy withintensity-modulated radiation therapy for locally advanced squamous cellcarcinoma of the larynx and oropharynx: a retrospective single-institutionanalysis. Head Neck 2009;31(11):1447–55.

18. Feng FY, Kim HM, Lyden TH, et al. Intensity-modulated chemoradiotherapyaiming to reduce dysphagia in patients with oropharyngeal cancer: clinical andfunctional results. J Clin Oncol 2010;28(16):2732–8.

19. Shoushtari A, Meeneghan M, Sheng K, et al. Intensity-modulated radiotherapyoutcomes for oropharyngeal squamous cell carcinoma patients stratified byp16 status. Cancer 2010;116(11):2645–54.

20. Setton J, Caria N, Romanyshyn J, et al. Intensity-modulated radiotherapy in thetreatment of oropharyngeal cancer: an update of the Memorial Sloan-KetteringCancer Center experience. Int J Radiat Oncol Biol Phys 2012;82(1):291–8.

21. Mendenhall WM, Amdur RJ, Morris CG, et al. Intensity-modulated radiotherapyfor oropharyngeal squamous cell carcinoma. Laryngoscope 2010;120(11):2218–22.

22. Clavel S, Nguyen DH, Fortin B, et al. Simultaneous integrated boost usingintensity-modulated radiotherapy compared with conventional radiotherapyin patients treated with concurrent carboplatin and 5-fluorouracil for locallyadvanced oropharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2012;82(2):582–9.

23. Lok BH, Setton J, Caria N, et al. Intensity-modulated radiation therapy inoropharyngeal carcinoma: effect of tumor volume on clinical outcomes. Int JRadiat Oncol Biol Phys 2012;82(5):1851–7.

24. Nathu RM, Mancuso AA, Zhu TC, Mendenhall WM. The impact of primary tumorvolume on local control for oropharyngeal squamous cell carcinoma treatedwith radiotherapy. Head Neck 2000;22(1):1–5.

25. Been MJ, Watkins J, Manz RM, et al. Tumor volume as a prognostic factor inoropharyngeal squamous cell carcinoma treated with primary radiotherapy.Laryngoscope 2008;118(8):1377–82.

26. Hermans R, Op de beeck K, Van den Bogaert W, et al. The relation of CT-determined tumor parameters and local and regional outcome of tonsillarcancer after definitive radiation treatment. Int J Radiat Oncol Biol Phys2001;50(1):37–45.

27. Chung EJ, Lee NJ, Baek SK, Kwon SY, Woo JS, Jung KY. Clinical efficacy of primarytumor volume measurements: comparison of different primary sites. Clin ExpOtorhinolaryngol 2009;2:78–84.