Mick Spillane Medical Formatted: Left Intensity-Modulated Radiotherapy for Sinonasal Tumors F Division of Radiotherapy, Department of Oncology (I. M., L. V., W. D. N.), and Division of Head and Neck Surgery, Department of Internal Medicine (K. B., T. B.), Ghent University Hospital, De Pintelaan 185, 9000 Ghent, Belgium From the the * Corresponding Author Division of Radiotherapy Department of Oncology Ghent University Hospital De Pintelaan 185, 9000 Ghent, Belgium Phone: (+32) 9 332 30 74 Fax: (+32) 9 240 38 63 e-mail: indira@krtkg.ugent.be Purpose: To report the long-term outcome of intensity-modulated radiotherapy (IMRT) for sinonasal tumors. Methods and Materials: Between July 1998 and November 2006, 84 patients with sinonasal tumors were treated with IMRT at a median dose of 70 Gy in 35 fractions. Seventy-three patients had a primary tumor and 11 patients had a local recurrence. Postoperative IMRT was performed in 75 patients and 9 patients received primary IMRT. Results: There was no difference in local control and survival between patients with primary and recurrent tumors. On multivariate analysis, invasion of the cribriform plate HENT Abstract, Underline, - ; 1
was associated with lower local control and overall survival. Local and distant recurrence was detected in 19 and 10 patients, respectively. One patient developed Grade 3 radiationinduced retinopathy and neovascular glaucoma. Nonocular late radiation-induced toxicity was comprised of complete lacrimal duct stenosis (n = 1) and brain necrosis (n = 3). Osteoradionecrosis of the maxilla and brain necrosis were detected in one of the five reirradiated patients. Conclusion: Intensity-modulated radiotherapy for sinonasal tumors provides low rates of radiation-induced toxicity without blindness at high local control and survival; thus, IMRT could be considered the treatment of choice. Sinonasal Tumors, Paranasal sinus tumors, Intensity-modulated radiotherapy, IMRT, Radiation-induced toxicity. INTRODUCTION Sinonasal tumors are a rare, histologically heterogeneous disease accounting for 5% of head-andneck tumors and less than 1% of all malignancies (1). The majority of patients remain asymptomatic for a long period but then present with locally advanced tumors extending into neighboring organs and tissues. Local tumor extension in the vicinity of critical organs such as the eyes, optic nerves and chiasm, pituitary gland, lacrimal glands, brainstem, and frontal and temporal lobes of the brain make treatment of sinonasal tumors extremely challenging. Patients commonly undergo surgery followed by radiotherapy. Conventional radiotherapy for sinonasal tumors results in local control and overall survival rates of 59% and 40% at 5 years (2). Up to 33% of patients develop radiation-induced blindness (3). The ability of IMRT to generate concave dose distributions around the critical structures with high-dose gradient outside the target supports our hypothesis that IMRT for sinonasal tumors could minimize dry-eye syndrome and preserve vision without compromising local control. We herein report our long-term experience assessing radiation-induced toxicity, local control, and survival in patients treated with IMRT between July 1998 and November 2006. g IMRT and could be Key words: ; S T ; ; I T Formatted: English (United States) Formatted: Normal [ ] are u i [ ] [ ] intensity modulated radiotherapy ( ) from to 2
METHODS AND MATERIALS Materials and methods Patients One hundred five patients with sinonasal tumors were treated with IMRT at the Department of Radiotherapy, Ghent University Hospital, and 84 were included in the analysis because of their histologic type (Table 1). Of 11 patients with local recurrence, the primary tumor had been treated with surgery alone (n = 6) or in combination with conventional postoperative radiotherapy (n = 5). Fourteen patients had invasion of one structure or fossa; in 20 patients, invasion was multiple (at least 2 invaded structures or fossae). Patients treated before 2002 were retrospectively restaged in accordance with the International Union against Cancer (UICC) TNM classification (4) or for esthesioneuroblastoma according to Kadish et al (5). None of the patients had distant metastases at the time of diagnosis. Forty-six patients (55%) exposed to wood dust had adenocarcinoma. Wood dust exposure was found in 13 out of 18 patients with invasion of the cribriform plate and in 11 out of 13 patients with T4b tumors. Of 6 patients with clinically enlarged lymph nodes, 4 patients underwent lymph node dissection, with pathological confirmation in 1 patient who had adenocarcinoma of the ethmoid sinus (pn2c). In the other 2 patients, nodal disease was clinically staged as cn1 in esthesioneuroblastoma and cn2b in squamous cell carcinoma. Seventy-five patients (89%) underwent surgery : lateral rhinotomy (n = 25), craniofacial resection (n = 13), maxillectomy (n = 10), ethmoidectomy (n = 7), functional endoscopic sinus surgery (n = 7), ethmoidosphenoidomaxillectomy (n = 6), and tumor resection (n = 4). Of 13 patients with T4b tumors, 10 underwent surgery, including seven cases of craniofacial resection. None of the patients received chemotherapy. Target definition Contrast enhanced CT scanning of the head and neck region was performed with 2-mm-wide slices. Additionally, MRI images were manually coregistered with planning CT images on a Pinnacle treatment planning system, version 6.2b (Philips Medical Systems, Andover, MA). The targets and organs at risk (retina, optic nerves, optic chiasm, brainstem, brain, spinal cord, the major lacrimal glands, mandible, and parotid glands) were outlined on each CT slice. The target Formatted: Indent: First line: 0.5" on the basis of y - A [ ] [ ] were operated - - 7 D computer tomography ( ) agnetic resonance imaging ( ) 3
definition was based on the compartment-related clinical target volume (CTV) approach (6). A 3- mm isotropic expansion of the CTV resulted in the planning target volume (PTV). Clinically enlarged lymph nodes were included in the PTV in 2 patients. From December 2004, we included ipsilateral level Ib-II elective neck dissection in the second PTV in patients with Stage T3-T4 squamous cell carcinoma of the maxillary sinus, given the high probability of occult metastases in neck lymph nodes (7). [ 6)] A 3-mm isotropic expansion of... Dose specification, treatment planning, and delivery Dose specification, treatment planning, and delivery have been previously described (8). In our study, the median prescription dose was 70 Gy, range 42 70 Gy, delivered in 35 fractions of 2.0 Gy. At least 95% of the PTV was to receive a dose of 66.5 Gy, while a dose of 74.9 Gy was restricted to 5% of the PTV. Dose inhomogeneity, defined as (D 2 D 98 )/D 50, did not exceed 18% in the PTV. D 2, D 50, and D 98 are dose levels on the dose volume histograms above which lay 2%, 50%, and 98% of the contoured volume, respectively. Treatment planning was based on Ghent University Hospital s class solution for sinonasal tumors, which incorporated seven noncoplanar beams with a single isocenter (Bouckaert 2001). Two to four additional beams were used for irradiating clinically enlarged lymph nodes and elective neck dissection. Two planning tools an anatomy-based segmentation tool (9) and a segment outline and wieght-adapting tool (10) were used to (1) generate initial multileafcollimated beam apertures and multileaf-collimated segments for those beam apertures; and (2) perform a direct multileaf collimator aperture optimization of segment weight and leaf position using a biophysical objective function (10,11). Optimization was performed using an in-house developed extension of the GRATIS software package (12). S pecification, T reatment P lanning,... previously [ 8)] In our study, the... Formatted: Indent: First line: 0.5" Comment [MS1]: AU: Should it be >18%? - Comment [MS2]: AU: Can you please check this Reference. It does not appear in the "References." 7 Between 2 Two and to four 4... References 1. Parsons T, Mendenhall W, Stringer S, et al. Nasal cavity and paranasal sinuses. In: Perez CA and LW Brady, eds. Principles and practice of radiation oncology 3rded.. Philadelphia: Lippincott-Raven: 1997; p. 941 959. 2. Dulguerov P, Jacobsen MS, Allal AS, et al. Nasal and paranasal sinus carcinoma: Are we making progress? A series of 220 patients and a systematic review. Cancer 2001; 92: 3012 29. CA itor. Principles and practice of..., Philadelphia, 997; p. 941 -... a - 4
3. Shukovsky LJ, Fletcher GH. Retinal and optic nerve complications in a high dose irradiation technique of ethmoid sinus and nasal cavity. Radiology 1972;104:629 34. 4. Sobin LH, Wittekind C, eds., for the International Union Against Cancer. TNM classification of malignant tumors. 6th ed. New York: Wiley-Liss; 2002. Fig, 1. (A) Overall survival for patients with primary and recurrent tumors, and (B) overall survival by Tstage for all patients, except patients with esthesioneuroblastoma. Fig. 2. (A) Local control for patients with primary and recurrent tumors and (B) with and without invasion of the cribriform plate in all patients. - Formatted: Line spacing: single, Don't adjust space between Latin and Asian text, Don't adjust space between Asian text and numbers Formatted: Font: (Default) Times New Roman, 12 pt International Union Against Cancer: TNM classification of malignant tumors (6 th edition). LH Sobin, C Wittekind, editors. Wiley-Liss, New York, 2002. Formatted: Font: (Default) Times New Roman, 12 pt Formatted: Font: (Default) Times New Roman, 12 pt Formatted: Font: (Default) AdvP641C, 9 pt, English (United States) IGURE (A) & (B) o - IGURE (A) & 5
TABLE 1 Patient characteristics Characteristic n (%) Sex: Male 70 83 Female 14 17 Age (years): Range 30-80 Median 63 Tumor: Primary 73 87 Local recurrence 11 13 Karnofsky 100 2 2 performance 90 47 56 status: 80 8 10 70 2 2 Unknown 25 30 Site: Ethmoid sinus 47 56 Maxillary sinus 19 23 Nasal cavity 16 19 Multiple 2 2 Histologic type: Adenocarcinoma 54 64 Squamous cell carcinoma 17 20 Esthesioneuroblastoma 9 11 Adenoid cystic carcinoma 4 5 T Stage: T1 4 5 T2 24 29 T3 14 17 T4a 20 24 T4b 13 15 Kadish: B 3 4 C 6 7 N Stage: N0 81 97 N1 1 1 N2 2 2 Invasion: Cribriform plate 18 21 Orbit 14 17 Pterygoid fossa 12 14 Dura 9 11 Brain 6 7 Infratemporal fossa 3 4 Surgery: Yes 75 89 No 9 11 TABLE 1 Formatted C. s f r p l u e n y a s e a c - - c o p d b i 6
TABLE 2. Actuarial 3- and 5-year local control, survival, and freedom from distant metastases in all patients and patients with primary and recurrent tumors TABLE 2. 3-year rates (%) 5-year rates All Primary Recurrent All Primary Recurrent patients tumor (n = tumor (n = patients (n tumor (n = tumor (n = (n = 84) 73) 11) = 84) 73) 11) Local control 74.9 77.1 60 70.7 71.9 60 (n = (n = (n (n = (n = Overall survival 70.2 69.5 72.7 58.5 57.9 60.6 Disease-specific survival Disease-free survival 71.9 71.5 72.7 67 65.5 61.4 61.4 59.3 51.5 Freedom from distant 84.8 82.5 82.2 79.2 metastases 7
TABLE 3 Table 3. Summary of the studies reporting treatment outcome and late severe (Grade 3) visual impairment after IMRT for Formatted g sinonasal tumors. The most frequent histological type in the whole patient cohort and it s proportion in parenthesis are indicated. Reference Pati ent s(n) Histologic type Treatment Median Dose Follow-up, month Local Control (year) Overall Survival (year) Grade 3 Visual Impairment Claus (6) 32 ADC (53%) S+IMRT or 70 Gy 15 NR 80% (1) None IMRT Duthoy et al. (8) 39 ADC (79%) S+IMRT 70 Gy 31 73% (2) 68% (2) 2 68% (4) 59% (4) Hoppe et al. (17) * 30 S+IMRT 60 Gy 23 NR NR None Combs et al. (16) 36 SCC (33%) S+IMRT or 70 Gy 39 62% (2) 69% (2) None IMRT 58% (5) 45% (5) Chen et al. (15) 23 S+IMRT or 70 Gy 44 65% (5) 47% (5) None IMRT Present study 84 ADC (64%) S+IMRT or IMRT 70 Gy 40 74.9% (3) 70.7% (5) 70.2% (3) 58.5% (5) 1 Abbreviations: n = number of patients; ADC = adenocarcinoma; SCC = squamous cell carcinoma; ACC = adenoid cystic carcinoma; S = surgery; IMRT = intensity modulated radiotherapy. *Reported treatment outcome in 85 patients treated with three-dimensional conformal radiotherapy and IMRT. Reported treatment outcome in 127 patients treated with conventional three-dimensional radiotherapy and IMRT. Abbreviations: n = number of patients; ADC = adenocarcinoma; ACC = adenoid cystic carcinoma; SCC = squamous cell carcinoma; S = surgery; IMRT = intensity modulated radiotherapy. y Formatted [ ] [ ] [ ] Formatted [ ] [ ]* Formatted: Superscript *