Proton Beam Radiation Therapy for Skull Base Adenoid Cystic Carcinoma

Objective  To determine the treatment outcome and prognostic factors in patients with adenoid cystic carcinoma of the skull base treated with proton beam radiation therapy.

Design  Retrospective analysis.

Setting  Massachusetts General Hospital, Massachusetts Eye and Ear Infirmary, and Harvard Medical School, Boston.

Patients  From 1991 to 2002, 23 patients with newly diagnosed adenoid cystic carcinoma with skull base extension were treated with combined proton and photon radiotherapy. There was tumor involvement of the sphenoid sinus in 61% of patients (14), nasopharynx in 61% (14), clivus in 48% (11), and cavernous sinus in 74% (17). The extent of surgery was biopsy alone in 48% (11), partial resection in 39% (9), and gross total resection with positive margins in 13% (3). The median total dose to the primary site was 75.9 cobalt-gray equivalent. The median follow-up of all surviving patients was 64 months.

Main Outcome Measures  Locoregional control and disease-free survival and overall survival rates.

Results  Tumors recurred locally in 2 patients at 33 and 68 months, respectively. No patients developed neck recurrence. Eight patients had distant metastasis as the first site of recurrence. The local control rate at 5 years was 93%. The rate of freedom from distant metastasis at 5 years was 62%. The disease-free and overall survival rates at 5 years were 56% and 77%, respectively. In multivariate analysis, significant adverse factors predictive for overall survival were change in vision at presentation (P = .02) and involvement of sphenoid sinus and clivus (P = .01).

Conclusions  High-dose conformal proton beam radiation therapy results in a very encouraging local control rate in patients with adenoid cystic carcinoma of the skull base. Changes in vision at presentation and tumor involvement of the sphenoid sinus and clivus are important prognostic factors.

Adenoid cystic carcinoma has been recognized as an aggressive malignant tumor with a unique natural history since its first description by 2 Frenchmen, Charles Robin and Alexandre Laboulbene, in 1853.¹ It is characterized by having an indolent but aggressive clinical course, the presence of early perineural invasion, frequent local recurrence, and a high rate of delayed distant metastasis. It is an uncommon tumor arising mainly from the major and minor salivary glands, most often in the head and neck region. En bloc removal of the tumor is the mainstay treatment. However, when the tumor originates in or extends to the skull base, adequate oncologic surgery is challenging. Despite the technical advances of skull base surgery and the use of postoperative radiotherapy, the treatment outcome has remained very suboptimal, with many patients experiencing local recurrences.^2-13 Patients with inoperable tumors or gross residual disease have the worst outcomes, with local control rates of 0% to 30%.^2,4,10-14 Alternative treatment strategies are clearly needed for adenoid cystic carcinoma with skull base involvement. Although it has been shown that an increased radiation dose is associated with improved local control,^8,12,15 the surrounding critical healthy tissues in the skull base precludes the delivery of adequate tumoricidal dose.

Protons are charged particles with biological effectiveness similar to conventional photon radiation. Owing to the defined range of protons exhibited by the Bragg peak, a larger radiation dose can be delivered to the tumor while significantly lowering the dose to the surrounding healthy tissues. This superior physical property of protons could potentially lead to improved local control and decreased acute and late toxic effects.¹⁶ The purpose of this study is to determine the treatment outcome and prognostic factors of the use of dose escalation with proton beam radiation therapy in the treatment of adenoid cystic carcinoma with skull base involvement. To our knowledge, this is the first report on the use of proton beam radiation therapy in the treatment of this disease.

From November 1991 to December 2002, 24 patients with newly diagnosed adenoid cystic carcinoma of the skull base were treated with proton beam radiation therapy at the Massachusetts General Hospital, Boston. One patient was lost to follow-up; the remaining 23 patients are the basis of this retrospective analysis. The study was approved by the institutional review board. All available histopathologic slides were reviewed at our institution before treatment. Each patient was seen jointly by radiation oncologists and otolaryngologists, and the decision was made jointly for patients to undergo proton beam radiation therapy with or without surgery. The median duration of follow-up of all patients in the study was 62 months. The median duration of follow-up of all surviving patients was 64 months.

The median age was 46 years (range, 25-66 years). There were 5 men and 18 women. The median Karnofsky Performance Scale score at the time of radiation was 90. The median duration from onset of symptoms to diagnosis was 12 months (range, 0-248 months). The main symptoms at presentation were facial numbness or pain in 35% (8) of the patients, vision change in 22% (5), nasal congestion or obstruction in 43% (10), headache in 29% (6), cranial nerve deficits in 48% (11), and dental problems in 29% (6). None of the patients had received previous irradiation to the primary site. The pathologic findings of 1 patient confirmed the presence of a solitary pulmonary nodule at presentation. None of the patients had nodal metastasis clinically or radiographically at presentation.

The primary site and the extent of the tumor were evaluated by computed tomography studies, magnetic resonance imaging studies and operative findings (Table 1). The primary site of the tumor was defined by the epicenter of the tumor mass. The sphenoid sinus was the most common site of origin, followed by the maxillary sinus and nasopharynx. The base of the skull was involved in all patients. Tumor involved the orbital soft tissues in 39% (9) of the patients, sphenoid sinus in 61% (14), frontal sinus in 17% (4), nasopharynx in 61% (14), clivus in 48% (11), pterygopalatine fossa in 78% (18), infratemporal fossa in 39% (9), cavernous sinus in 74% (17), Meckel's cave in 57% (13), and brain parenchyma in 52% (n = 12). All but 3 patients had gross tumor prior to the initiation of radiation.

The extent of surgery was gross total resection with positive margins in 3 patients, partial resection in 9, and biopsy alone in 11. The surgical approach was transfacial alone in 9 patients and craniofacial in 3. One patient underwent orbital exenteration prior to radiation. No patient underwent neck dissection. Of those who had undergone surgery, 67% (8) were referred for proton beam radiation therapy following surgery at other institutions. The median time interval between surgery and the initiation of radiation was 2.48 months (range, 1-16 months).

Most of the patients were immobilized by means of a custom-made dental fixation technique and an integrated thermoplastic head mask.¹⁷ This immobilization device limited the mean net 3-dimensional patient motion during the treatment to less than 1 mm. A thin-cut high-resolution computed tomographic scan with contrast was obtained in the treatment position. Magnetic resonance imaging was obtained to assist in target delineation. The gross tumor volume, clinical target volume, and surrounding critical structures were outlined. Dose-volume histograms were generated for the gross tumor volume, clinical target volume, and surrounding critical structures.

For proton treatment planning, a patch combination (split-target volume) technique was used to optimize the proton dose distribution within an irregular volume in close proximity to critical healthy structures.¹⁸ The target volume was divided into multiple segments with each treated by a separate radiation field. Using the sharp dose fall-off of the Bragg peak, each field was designed to stop in the penumbra of the other fields. Each treatment field was shaped by an individually designed brass aperture and a Lucite range compensator. An appropriate modulator wheel was selected to spread out the Bragg peak. Daily pretreatment alignment radiographs were obtained and compared with computed tomography–generated, digitally reconstructed radiographs to ensure precise positioning of the treatment fields. Proton beam radiation therapy was delivered at the Harvard Cyclotron Laboratory, Cambridge, Mass, or the Francis H. Burr Proton Therapy Center at the Massachusetts General Hospital using 160-MeV and 230-MeV beams, respectively.

For photon radiotherapy, a 5-field graduated block technique,¹⁹ consisting of an anterior and 2 sets of right and left lateral beams targeting the clinical target volume, was used to design the treatment plan. These fields were matched to a lower anteroposterior neck field if required. Photon beam radiation therapy was delivered mostly with either 4 or 6 MV.

All patients were treated with definitive radiotherapy with curative intent. The median prescribed dose to gross tumor volume or high-risk clinical target volume was 75.9 cobalt-gray equivalent (CGE) (range, 70.0-76.8 CGE). Before March 1997, there was a calibration adjustment of more than 6.5% to the proton dose. The total median dose delivered to the gross tumor volume after calibration adjustment was 76.4 CGE (range, 70.0-79.1 CGE). The median percentage of proton was 59.6% (range, 30.8%-71.7%). The median duration of combined proton and photon radiotherapy of all patients was 38 days (range, 27-58 days).

The radiation protocol to the primary site combining proton and photon radiotherapy evolved over a period of time, such that varying fractionation schedules were used (Table 2). Nineteen patients received twice-daily accelerated fractionated radiation without any planned break. Of these 19 patients, 13 received 1.6 CGE twice daily and 6 received 1.8 CGE once daily and a concomitant boost of 1.4 to 1.5 CGE for 2 to 3 weeks. For the patients who received radiation twice daily, the median delivered total dose was 76.40 CGE (range, 75.60-79.06 CGE) with a median duration of 36 days (range, 27-48 days). For the 4 patients who received 1.8 or 2.0 CGE once-daily radiation, the median total dose delivered was 73.34 CGE (range, 70.00-77.62 CGE) with a median duration of 53 days (range, 42-58 days). Seventy percent of patients received elective irradiation to the uninvolved high jugular node and low jugular lymphatic and supraclavicular node to a median dose of 44 Gy (range, 40-46 Gy) using photon radiotherapy.

Only 1 patient with unresectable tumor received induction chemotherapy that consisted of 2 cycles of doxorubicin hydrochloride (Adriamycin; Pharmacia Inc, Milan, Italy) and fluorouracil with progression. No patient received concurrent or adjuvant chemotherapy.

Locoregional control was measured from the end of radiation treatment to the date of local or regional relapse. Survival time was measured from the end of radiotherapy until death or last follow-up. The Kaplan-Meier product-limit method was used to estimate the probabilities of tumor control and survival rates. Local control and survival probabilities were compared in univariate analysis, using log-rank test for discrete variables or likelihood ratio test for proportional hazards model for continuous variables. The variables with P values less than .10 were then entered into the multivariate analysis using Cox proportional hazards model.

Patient-related factors that were entered into analysis were age at diagnosis, duration of symptoms, vision change at presentation, sex, and cranial nerve deficits at presentation. Tumor-related factors were site of primary tumor, tumor size, and the extent of tumor involvement of the skull base (orbital wall, orbital content, frontal sinus, sphenoid sinus, nasopharynx, clivus, pterygopalatine fossa, infratemporal, cavernous sinus, Meckel's cave, and brain parenchyma). Treatment-related factors used were extent of surgery, duration of radiotherapy, time interval from surgery to initiation of radiation, time interval from surgery to end of radiation, radiation dose, and percentage of proton beam radiation therapy.

Two patients had a local recurrence at 33 and 68 months, respectively, and both subsequently died from local progression. The primary site of tumor was ethmoid sinus in one and maxillary sinus in the other. Both had undergone partial resection before radiation, including orbital exenteration in 1 patient. Both failed within the high-dose regions. The local control rates at 5 and 8 years were 93% and 82%, respectively (Figure 1). There were no neck recurrences. None of the patients underwent planned neck dissection after radiation.

Distant metastasis was the predominant pattern of relapse, occurring in 35% of the patients at the first site of relapse. Two patients (8%) had isolated local recurrence as the first site of relapse whereas 8 patients (35%) had isolated distant metastasis as the first site of relapse. There was no regional recurrence and no synchronous local and distant recurrence. The sites of distant metastasis were lung in 30% of patients (7), bone in 9% (2), liver in 4% (1), and leptomeninges in 4% (1). The median time to development of distant metastasis after the completion of radiotherapy was 7.5 months (range, 0-46 months). The median time of survival after development of distant metastasis was 26 months (range, 4-81 months). The rate of freedom from distant metastasis at both 5 and 8 years was 62% (Figure 1).

At the last follow-up of the patients, 2 had died from local progression and 3 from distant metastasis. The disease-free survival rates at 5 and 8 years were 56% and 31%, respectively. The overall survival rates at 5 and 8 years were 77% and 59%, respectively.

The 2 patients who had local recurrence did not undergo any salvage treatment. One patient with distant metastasis received gemcitabine, cisplatin, fluorouracil, and doxorubicin chemotherapy without response. Three patients received palliative radiation to their metastases. The median survival after palliative radiation was 9 months (range, 1-81 months).

In univariate analysis, change in vision at presentation (P = .01) and optic nerve invasion (P<.001) were predictive for local failure. Two of the 3 patients with gross tumor invasion of the optic nerve developed local failure in this study.

The results of the univariate and multivariate analysis of disease-free survival rates are shown in Table 3. In univariate analysis, an age of 46 years or younger, change in vision at presentation, pterygopalatine fossa involvement, and sphenoidal and clival involvement were predictive for decreased disease-free survival rates. The disease-free survival rate at 5 years was 20% for patients with vision change and 68% for patients without vision change at presentation (P = .02). The 5-year disease-free survival rates were 44% and 100% for patients with and without pterygopalatine fossa involvement (P = .02), respectively (Figure 2). The 5-year disease-free survival rates were 38% and 71% for patients with and without sphenoidal and clival involvement, respectively (P = .02) (Figure 3). In multivariate analysis, only sphenoidal and clival involvement was predictive for disease-free survival (P = .01; adjusted hazard ratio, 5.17).

The results of the univariate and multivariate analysis of overall survival rates are shown in Table 4. In univariate analysis, an age of 46 years or younger, change in vision at presentation, and sphenoidal and clival involvement were predictive for decreased overall survival rates. The overall survival rate at 5 years was 20% for patients with vision change and 94% for patients without vision change at presentation (P<.001) (Figure 4). The 5-year overall survival rates were 60% and 90% for patients with and without sphenoidal and clival involvement, respectively (P = .02) (Figure 5). In multivariate analysis, change in vision at presentation (P = .02; adjusted hazard ratio, 22.4) and involvement of the sphenoid sinus and clivus were predictive for overall survival (P = .01; adjusted hazard ratio, 22.1).

Treatment toxicity was scored using the Common Terminology Criteria for Adverse Events (version 3.0) of the National Cancer Institute.²⁰ All patients tolerated radiation treatment without any treatment break. One patient required hospitalization during radiation for gastrostomy tube placement for mucositis.

During radiation therapy, none of the patients experienced an acute grade 3 ocular or visual toxic effect (symptomatic conjunctivitis, keratitis, tearing, epiphora, photophobia, or visual change that interferes with the activities of daily living). There were also no acute grade 4 or 5 toxic effects (blindness, disability, or death). Thirteen percent of patients (3) experienced grade 2 (symptomatic, but not interfering with activities of daily living) as the highest toxic effect during radiation treatment. Tearing was the most frequent grade 2 toxic effect.

None of the patients underwent orbital exenteration after radiation. None of the patients developed any grade 5 ocular or visual complication. One patient developed chronic grade 4 retinopathy. Three patients developed chronic grade 3 toxic effects requiring surgical intervention: dacryocystorhinostomy in 1, reconstruction for ectropion in 1, and lens replacement for cataract in 1. Fifty-two percent of patients (12) developed a chronic grade 2 toxic effect requiring conservative treatment only, at a median time of 8 months (range, 0-88 months): epiphora in 4, dry eye in 6, ectropion in 1, cataract in 1, retinopathy in 1, and nasolacrimal duct obstruction in 3 patients.

Radiographic brain change was defined as any enhancement seen in delayed hyperintensity on T1-weighted magnetic resonance imaging at any follow-up. Radiographic brain change was observed in 12 patients at a median time of 15 months (range, 5-85 months) after radiation. Of these 12 patients, 2 had grade 2 (asymptomatic, radiographic findings only) and 10 had grade 3 (symptomatic, not interfering with activities of daily living; medical intervention needed) toxic effects. The types of grade 3 toxic effects were seizures in 7 and decreased short-term memory in 3 patients. All seizures were controlled with medication. Two patients required permanent antiseizure medication.

Five patients with radiographic brain changes were treated with short-term dosages of corticosteroids: 3 had a complete resolution,1 had partial resolution, and 1 had stable radiographic findings. One patient developed grade 5 toxic effects from radiation-induced injury of the temporal lobe at 61 months after radiation treatment. One patient with tumor involvement of the sphenoid sinus, cavernous sinus, carotid canals, Meckel's cave, optic chiasm, nasopharynx, and clivus developed grade 5 toxic effects at 9 months after radiation treatment from infectious meningitis associated with repeated cerebrospinal fluid leaks. The cerebrospinal fluid leak resulted from a fistula that was formed between the pontine cistern and nasal cavity through the clivus and sphenoid sinus. Multiple graft procedures were attempted without success.

Six patients developed grade 2 nonsymptomatic hypothyroid (not interfering with activities of daily living, thyroid replacement needed) at a median time of 14 months (range, 0-57 months). There was no other pituitary or hypothalamic dysfunction.

In this article, we report the results of the use of highly conformal, high-dose proton beam radiation therapy in the treatment of adenoid cystic carcinoma with skull base involvement. Our study cohort consisted of a high-risk group; all patients had tumors with skull base invasion with 48% to 74% of patients (11-17) having involvement of the nasopharynx, clivus, sphenoid sinus, or cavernous sinus. In addition, only 3 patients had undergone gross total resection. Our local control rate of 93% and 82% at 5 and 8 years, respectively, in this group of patients with advanced skull base involvement is very encouraging. Longer follow-up will be required to determine if our high local control rates are durable.

For patients with adenoid cystic carcinoma of the head and neck, tumor involvement of the skull base through perineural spread is frequent. The treatment outcome of adenoid cystic carcinoma with skull base involvement, especially when there are unresectable or residual gross tumors after surgery, is very poor, with the reported local control rate of 0% to 30%.^2,4,10-14 Owing to absence of any successful salvage treatment, patients eventually die of local progression. Mendenhall et al⁴ reported one of the best local control rates in the literature of the use of photon radiotherapy in the treatment of the adenoid cystic carcinoma of the head and neck. In this retrospective analysis of 101 patients with a median follow-up of 6.6 years, of whom 27% of the tumor originated in the paranasal sinus, nasal cavity, or nasopharynx, the local control rate with radiotherapy alone (median dose, 72.4 Gy) was reported to be 44% at 5 years and 30% at 10 years. Vikram et al¹² reported a local control rate of 6.5% in 47 patients with adenoid cystic carcinoma of various head and neck sites treated with radiotherapy alone (12-100 Gy) with all recurrences occurring within 5 years after radiotherapy. Wiseman et al¹³ and Kim et al² reported local control rates of 20% and 0%, respectively, for patients treated with radiotherapy alone for adenoid cystic carcinoma of the paranasal sinuses. Improved treatment strategies are clearly needed in the treatment of this aggressive malignancy. In the study reported herein, the local control rate for patients treated with radiotherapy alone with a median dose of 76.4 CGE was 100% at 5 and 10 years. These data suggest that dose escalation to a median dose of 76.4 CGE with proton beam radiation therapy results in an improved local control rate compared with published rates using a lower prescribed dose.

The high local control rate in our study has not resulted in an improved survival rate compared with the historical controls. Distant metastasis occurred in 35% (8) of our patients as the first site of relapse and in fact is the main pattern of failure. Laramore et al²¹ reported the final results of the randomized trial conducted by the Radiation Therapy Oncology Group and the Medical Research Council that compared photon radiotherapy with fast neutron radiotherapy for patients with unresectable malignant salivary tumors. The local control rate was 17% with photon vs 56% with fast neutron. Distant metastasis accounted for most failures on the neutron arm, and locoregional failures accounted for most failures in the photon arm.

Radiation planning of the skull base is extremely challenging. The radiation tolerance of the critical structures in or near to the skull base is in the range of 54 to 63 Gy.^22,23 With the use of conventional photon radiation, the dose delivered to the tumor is therefore limited to approximately 60 Gy without causing significant morbidity. The use of this dose in the treatment of sinus malignancy with gross disease is ineffective. Dose escalation with other radiation modalities has been an important area of investigation.

Currently, neutron radiotherapy is an accepted treatment option for patients with unresectable or residual adenoid cystic carcinoma of the head and neck. In a randomized study of 32 patients, the local control rate improved from 17% to 56% with the use of fast neutron radiotherapy compared with photon radiotherapy.²¹ Douglas et al,²⁴ in a retrospective analysis of 159 adenoid cystic carcinoma patients treated with neutron therapy, reported a locoregional control rate of 23% in patients with base of skull involvement compared with 68% in patients without base of skull involvement. The 23% locoregional control rate for skull base adenoid cystic carcinoma reported by these investigators is similar to some of the locoregional control rates observed after conventional radiation for unfavorable tumors.^4,13

Carbon ion therapy, a heavy charged particle with higher intrinsic biological and physical properties than photon, has been used in the treatment of skull base tumors in recent years. Schulz-Ertner et al²⁵ reported a local control rate of 62% in 21 patients with locally advanced adenoid cystic carcinoma treated with carbon ion therapy. The median follow-up was only 14 months.

In recent years, intensity-modulated radiation therapy has been accepted as a treatment modality in the treatment of head and neck cancers, especially nasopharyngeal and oropharyngeal carcinomas. However, to our knowledge, to date, no study has reported the intensity-modulated radiation therapy treatment outcome of adenoid cystic carcinoma of the paranasal sinus with or without skull base involvement.

Although the comparisons of results with different studies are limited by the inherent caveat of retrospective studies, our local control rates of 93% at 5 years and 82% at 8 years represent the most successful outcomes. Because randomized study is not very feasible with this rare malignancy, single or multi-institutional studies with risk stratifications are necessary to define the optimal treatment.

Studies have shown that base of skull involvement confers the worst prognosis in patients with adenoid cystic carcinoma, with decreased local control rates and overall survival rates.^3,4,7,9,24 In the present study, we further defined the prognostic significance of the extent of skull base involvement. We have found that tumor involvement of the sphenoid sinus and clivus was associated with decreased disease-free and overall survival rates (Table 3 and Table 4). Patients with sphenoidal and clival involvement were 22 times more likely to die than patients without involvement. Involvement of the pterygopalatine fossa was associated with decreased disease-free survival but not with overall survival. Involvement of the nasopharynx, cavernous sinus, infratemporal fossa, orbital soft tissues, brain, or frontal sinus was not of prognostic significance.

In the study described herein, with delivery of a total dose of 70.0 to 79.1 CGE, none of our patients developed radiation-induced optic neuropathy. We report a risk of visual or ocular toxic effects that is significantly lower than that reported by others.^4,24,26 We attributed this to the sharp dose fall-off exhibited by the proton's Bragg peak and to our tight constraints of radiation tolerance of the optic structures at the time of radiation planning. However, these constraints might have been responsible for our 2 local failures.

The risk of neurologic toxic effects in this study was slightly higher than expected. In recent years, we have made significant modifications in our proton beam radiation therapy techniques, such as the use of gantry-based technique instead of fixed beams in our new proton facility, once-daily radiation instead of twice-daily radiation in most patients, refined target delineation, daily pretreatment using high-resolution digital portals, and the use of more sophisticated radiation treatment planning software. With these modifications, the risk of temporary or permanent radiation-induced brain injury had decreased significantly. The in-depth analysis of the radiation-induced brain injury was reported separately.^27,28

In conclusion, our study suggests that dose escalation with conformal proton beam radiation therapy is an effective treatment for adenoid cystic carcinoma in patients with advanced tumors at the skull base. Vision change at presentation and involvement of the sphenoid sinus and clivus are adverse prognostic factors. Prospective and multi-institutional studies are necessary to further study the use of proton beam radiation therapy with or without chemotherapy or biological therapy in the treatment of this rare and aggressive malignancy.

Correspondence: Annie W. Chan, MD, Department of Radiation Oncology, Massachusetts General Hospital, 55 Fruit St, Cox 3, Boston, MA 02114 (awchan@partners.org).

Submitted for Publication: September 5, 2005; final revision received April 10, 2006; accepted May 31, 2006.

Author Contributions: All authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Pommier, Liebsch, and Chan. Acquisition of data: Pommier, Liebsch, Deschler, Lin, McIntyre, Barker, Adams, Lopes, Varvares, Loeffler, and Chan. Analysis and interpretation of data: Pommier, Liebsch, Deschler, Lin, McIntyre, Barker, Adams, Lopes, Varvares, Loeffler, and Chan. Critical revision of the manuscript for important intellectual content: Pommier, Liebsch, Deschler, Lin, McIntyre, Barker, Adams, Lopes, Varvares, Loeffler, and Chan. Statistical analysis: Barker, Lopes, and Chan. Obtained funding: Pommier, Liebsch, Adams, Loeffler, and Chan. Administrative, technical, and material support: Adams and Lopes. Study supervision: Chan.

Financial Disclosure: None reported.

Funding/Support: This study was supported in part by the National Institute of Health grant NCI-PO1-CA21239 and the Phillipe Foundation Award.

Previous Presentation: This study was presented in part at the 17th Annual Meeting of the North America Skull Base Society; February 16-19, 2006; Phoenix, Ariz.

Acknowledgment: We acknowledge Herman Suit, MD, a pioneer in proton beam radiation therapy, and the late C. C. Wang, MD, a pioneer in head and neck radiation oncology. Their training, inspiration, and contributions form a firm base on which our current study is built.