Proton Beam Radiation Therapy

results of a long-term follow-up study of 64 patients with skull base chordomas or chondrosarcomas treated with spot-scanning proton beam radiotherapy


Background
Charged particle radiotherapy uses beams of protons or other particles such as helium or carbon instead of photons. In contrast to conventional photon radiotherapy, in which the greatest energy release is at the surface of the tissue and decreases exponentially the deeper the radiation travels, the energy of a proton beam is released near the end of its path, resulting in a sharp and localized dose peak, referred to as the Bragg peak. This allows for better dose distribution when compared to photon beam radiotherapy, thereby decreasing the dose to normal surrounding tissues, and reducing the risk of both acute and long-term side effects 1,2 . To date, there are no published controlled comparative studies describing outcomes from patients treated with proton beam radiotherapy versus other therapies; thus, the advantage of protons over conventional photon therapy is based on the dosimetric advantage of protons over photons for tumours that are in immediate proximity to critical structures. Most of the published literature is in the form of prospective or retrospective case series and cohort studies; there is also significant variation in the types and stages of cancer for which treatment with proton beam radiotherapy has been reported, as well as the reported doses and fractionation schedules. Published reports focus mainly on toxicity, and include pediatric, adolescent, and adult patients treated with proton beam radiotherapy primarily for ocular cancers, spine or skull base chordomas and chondrosarcomas, spinal and paraspinal bone and soft tissue sarcomas, head and neck cancers, and brain tumours. In addition, there is some literature reporting the use of proton beam radiotherapy for patients with lymphoma, non-small cell lung cancer, prostate cancer, and some gastrointestinal cancers.
By December 2017, 170,556 patients worldwide had been treated with proton beam radiotherapy 3 . Historically, the high capital cost of proton facilities equipped with rotational gantries has limited the number of facilities in operation; however, that number is now increasing rapidly. Between 2004 and 2014, the number of facilities in the United States alone grew to 13 from 2 and from 13 to 32 between 2014 and 2017 3 . Compact single-room cyclotron centres have drastically reduced the capital cost and account for 10 of the 22 facilities under construction worldwide 3 . Gantry-equipped facilities capable of treating a broad range of tumour sites are not currently available in Canada.
In early 2012, the Cancer Care Alberta Proton Therapy Guideline Working Group and Guideline Advisory Group met to evaluate the most current evidence for the use of proton beam radiotherapy in pediatric and adult patients with cancer, and to develop a de novo guideline with recommendations based on an expert review of the available literature. The resulting evidence review, guideline document, and accompanying documents were presented to the out of country Health Services Committee, which operates at arm's length from Alberta Health, to establish a process to identify which patients are appropriate candidates to receive out of country treatment with proton beam radiation therapy. In 2019, the evidence was reviewed, and this guideline was updated. In 2021, the Alberta Health working group set up a special program unit to review proton beam radiation therapy requests, which means the requests no longer go through the out of country Health Services Committee.

Target Population
The recommendations in this guideline are for pediatric, adolescent, young adult and adult patients who are residents of Alberta and may qualify to receive proton beam radiotherapy at a facility outside of Canada for treatment of their cancer. Summary of Recommendations 1. Factors other than diagnosis should be taken into account in assessing whether proton beam radiotherapy will confer a significant advantage for the patient over photon therapies available in Alberta such as photon intensity-modulated radiotherapy (IMRT), stereotactic radiotherapy, and brachytherapy. 2

Pediatric Tumours
Radiation therapy has played an important role in the treatment and cure of children diagnosed with malignant tumours over the past 30 years; given that approximately 70% of children with malignancies can now expect to be cured, the late effects of treatment have now become a major focus 4 . Proton beam therapy is associated with a reduction in acute and long-term toxicities, as well as lower rates of radiation-induced second malignancies, and potentially less acute and long-term damage to developing organs in pediatric and adolescent patients with cancer 4-6 ; therefore, the benefits of proton beam radiotherapy are potentially the greatest in this population. Although limited, published reports of pediatric and adolescent patients treated with proton beam radiotherapy have mostly addressed tumours arising on or near critical structures, as well as tumours where organs in the exit path of photon radiotherapy present significant secondary malignancy risk or risk of impaired development and function; such tumours include medulloblastomas, gliomas, ependymomas, retinoblastomas, rhabdomyosarcomas, and pelvic sarcomas 4,7 .
In one of the larger pediatric series published to date, MacDonald and colleagues reported the results of 17 pediatric patients with WHO grade II or III intracranial ependymomas treated with 3-dimensional conformal proton beam radiotherapy 8 . At a median follow-up of 26 months, the rates of local control, progression free survival, and overall survival rates were 86%, 80%, and 89%, respectively. Similarly, Amsbaugh and colleagues published the results of a prospective case series of eight pediatric patients with WHO grade I or II spinal ependymomas treated with proton beam radiotherapy 9 . Five patients were treated with proton beam radiotherapy as part of their primary management, while three were treated for recurrence of their disease. The entire vertebral body was treated in all but two patients, with a mean dose of 51.1 cobalt grey equivalents (CGE). The rates of local control, eventfree survival, and overall survival were all 100% at a mean follow-up of 26 months, and the investigators reported minimal acute side effects.
In a case series of three pediatric patients with medulloblastoma, Yuh and colleagues reported minimal acute side effects, excellent sparing of surrounding critical structures, and no treatment disruptions when the patients were treated with 36 CGE to the craniospinal axis and 18 CGE to the posterior fossa 10 . Similarly, St. Clair and colleagues compared treatment plans from standard photon therapy to intensity modulated radiotherapy (IMRT) and proton beam radiotherapy for craniospinal axis irradiation and posterior fossa boost in a pediatric patient with medulloblastoma. They reported substantial normal-tissue dose sparing with IMRT and proton treatment of the posterior fossa and spinal column, and also that protons were superior to IMRT in this case 11 . In another case series, six pediatric medulloblastoma patients were treated with vertebral-body-sparing proton craniospinal irradiation after maximal safe resection 12 . These patients received a dose of 23.4 Gy or 36 Gy to the craniospinal axis followed by a boost to the posterior fossa and any metastatic lesions. After a median follow-up of 13.6 years, the overall survival and disease-free survival were 83%, treatment did not increase spinal abnormalities compared with historical data, and it allowed for vertebral body growth outside of radiation field and bone marrow sparing. In a modeling study evaluating the consequences of radiotherapy in five year old patients with medulloblastoma, Lundkvist and colleagues reported that proton radiation therapy was cost-effective and was associated with an additional 0.68 qualityadjusted life-years per patient compared with conventional photon radiotherapy 13 . Reductions in IQ loss and growth hormone deficiency contributed most to the cost savings and were the most important parameters for cost-effectiveness. This analysis suggests that proton beam radiotherapy may be cost saving in comparison to photon radiotherapy for pediatric patients with medulloblastoma due to the reduction of late side-effects. A reduction in side-effects for pediatric patients with medulloblastoma treated with proton beam radiotherapy was also demonstrated in a prospective study of 23 pediatric patients who were followed for high-grade ototoxicity 14 , although a cost-benefit analysis was not reported for this trial.

Young Adult Tumours
Adolescent and young adult patients (age range  can be particularly vulnerable to complications from radiotherapy. With increased survival, the long term complications of treatment can have a major impact on growth, fertility, and emotional well-being 15 . Particularly the risk of radiation-induced secondary cancer is of concern in these patients. Zhang and colleagues reported on 17 medulloblastoma patients treated with passively scattered proton or field-in-field photon craniospinal irradiation (CSI) 16 . They found that proton CSI reduced the predicted risks of radiogenic second cancer incidence and cardiac mortality by six times compared with photon CSI. Yoon et al. retrospectively review 10 randomly selected cancer patients treated with either photon or proton radiotherapy 17 . The secondary cancer risks to lungs, stomach, liver and pancreas were at least five fold higher in patients treated with photon therapy compared to proton beam therapy. Another retrospective review studied 86 patients with retinoblastoma treated with proton or photon radiotherapy 18 . The 10-year cumulative incidence of radiation-induced or in-field second malignancies was 0% for patients treated with proton beam therapy versus 14% for patients treated with photon therapy (p=0.015). The 10-years cumulative incidence of all second malignancies was 5 % after proton therapy and 14% after photon therapy (p=0.120). Chung and colleagues performed a retrospective cohort study where they matched 558 proton patients with 558 photon patients 19 . Secondary malignancies occurred in 42 (7.5%) photon patients versus only 29 (5.2%) proton patients. They found that proton therapy was not associated with an increased risk of secondary cancer with an adjusted hazard ratio of 0.52 [95% confidence interval, 0.32-0.85]; p=0.009.

Central Nervous System Tumours
Similar to pediatric patients with favourable prognoses, adult patients with benign lesions and indolent malignant tumours also benefit from proton beam radiotherapy due to a decreased risk of late neurologic toxicities.

Meningiomas.
Meningiomas are the second most common intracranial tumour reported in adults, accounting for 13 to 26% of all primary brain tumours in this population 20,21 . The management of a patient with a meningioma depends on the signs and symptoms produced by the tumour, the age of the patient, and the location and size of the tumour 22 . In the Meningiomas clinical practice guideline, the Alberta Provincial CNS Tumour Team members recommend surgery as the primary treatment for patients who are not candidates for management by watch-and-wait. This guideline also states that radiotherapy offers reasonable control for patients who are not candidates for surgery, patients whose tumour location or shape is not amenable to surgery (such as a cavernous sinus meningioma), patients who have symptomatic residual disease, or for the treatment of recurrence. Boskos and colleagues reported on 24 patients with either atypical or metastatic meningioma treated at the Centre Protonthérapie d'Orsay with a postoperative combination of proton and photon radiotherapy 23 .
The mean local relapse-free survival was 28.3 months (range 10-50) for atypical meningioma and 23 months (range 13-33) for malignant meningioma; ten tumours recurred locally. Overall survival rates were 100% at one year, 80.4% ± 8.8% at three years, 53.2% ± 11.6% at five years, and 42.6% ± 13% at eight years. Noël et al. also reported on the outcomes of 51 patients treated for base of skull meningiomas with a combination of proton and photon radiotherapy 24 . The four-year local control rate was 98%, and the four-year overall survival rate was 100%. The investigators reported tumour stabilization in 38 patients (72%), volume reduction in ten patients (20%), and intra-tumour necrosis in three patients; grade three side effects were observed in two patients. Similarly, Wenkel et al.
reported on 46 patients with partially resected, biopsied, or recurrent meningiomas who were treated with combined proton and photon radiotherapy at the Massachusetts General Hospital and the Harvard Cyclotron Laboratory 25 . Five-and ten-year overall survival rates were 93% and 77%, respectively, and the five-and ten-year recurrence-free rates were 100% and 88%, respectively. Survival without severe toxicity was 80% at both five and ten years. The use of stereotactic and hypofractionated stereotactic proton beam radiotherapy were reported by Vernimmen and colleagues in a study of 27 patients with large or complex intracranial meningiomas 26 . Of the 18 patients treated with hypofractionated stereotactic proton beam radiotherapy, 16 remained clinically stable or improved. Of the five patients treated with stereotactic proton beam radiotherapy, two showed clinical improvement and three remained clinically stable. The radiologic control rate was 88% for the patients treated with hypofractionation, and 100% for patients treated without hypofractionation. In a study investigating the use of spot-scanning proton beam radiotherapy, Weber and colleagues reported on 13 patients treated postoperatively for incomplete resection or recurrence and three patients treated radically after presumptive diagnosis based on imaging 27 . The cumulative three-year local control, progression-free survival and overall survival rates were 91.7%, 91.7%, and 92.7%, respectively, and no patient died from recurrent disease. The cumulative three-year toxicity-free survival rate was 76.2%; radiation-induced side effects included grade III optic neuropathy, grade II retinopathy, and grade IV brain necrosis. El Shafie et al. reported on 110 patients with meningiomas of the skull base treated with particle radiotherapy (proton: 95%, carbon ion: 5%) 28 . Progression-free survival was 100% after 36 months and 96.6% after 60 months. The overall survival was 96.2% after 60 months and 92% after 72 months.

Arteriovenous malformations.
Arteriovenous malformations (AVMs) are benign brain lesions that occur in approximately 0.1% of the population, and can cause intracerebral hemorrhage, seizures, and focal neurological deficits 29 . Standard interventions for brain AVMs include resection for surgically accessible lesions and embolization; stereotactic radiosurgery with either photons or protons can be used for patients with unresectable lesions, or those who are poor candidates or refuse surgery. In a review of 68 patients with cerebral AVMs treated with proton-beam stereotactic radiosurgery, Seifert et al. reported symptoms control in 85.7% in patients with Spetzler-Martin grades I and II AVMs, 54.2% of patients with grade III AVMs, and 24% of patients with grade IV AVMs 30 . Vernimmen and colleagues followed 64 patients with predominantly large AVMs who were treated with hypofractionated stereotactic proton radiosurgery, and reported obliteration rates of 67% for patients with AVM volumes less than 14cc and 43% for AVM volumes greater than 14cc 31 . Grade IV acute complications were observed in 3% of patients, and transient delayed effects were seen in 23% of patients. A recent review of 248 patients with 254 cerebral AVMs (23% in deep locations) treated with single-fraction proton stereotactic radiosurgery reported a 31 month median time to total obliteration, with 70% and 90% respective five-and ten-year cumulative incidences of total obliteration 32 . Similarly, in 59 patients with high-risk cerebral AVMs treated with two-fraction proton stereotactic radiosurgery because of large tumour size or eloquent brain location, the median time to total obliteration was 62 months and the five-year actuarial total obliteration rate was 33% 33 . Complications associated with radiosurgery include radiation necrosis, seizures, and new neurologic deficits, therefore the efficacy of radiosurgery must be weighed against the potential toxicities of treatment 34 .
Acoustic neuromas. Acoustic neuromas, also known as vestibular schwannomas, are benign slowgrowing tumours that commonly arise from the vestibular portion of the eighth cranial nerve and account for approximately 8% of intracranial tumours in adults 35 . Options for treatment depend on tumour size, tumour growth rate, symptoms, health status, and patient preference, and may include observation, single-session stereotactic radiosurgery, fractionated conventional radiotherapy, fractionated stereotactic radiotherapy, proton beam radiotherapy, or surgery 36,37 . Although a distinct advantage of proton therapy over photon therapy for acoustic neuromas has not been demonstrated to date, larger tumours may benefit from proton beam radiotherapy because of the reduction of exit dose to the non-targeted brain and surrounding normal tissues 34 . In a review of 88 patients treated with proton beam radiotherapy at the Harvard Cyclotron Laboratory, Weber and colleagues reported two-and five-year actuarial control rates of 95.3% and 93.6% respectively in patients treated with stereotactic proton beam radiotherapy 38 . Facial and trigeminal nerve function was preserved in 91% and 89% of patients, respectively, and the five-year cumulative radiologic response rate was 94.7%. Salvage therapy was required in five patients post-radiotherapy. In a trial of 30 patients with acoustic neuromas treated with fractionated proton beam radiotherapy at the Loma Linda Medical Center, Bush et al. reported no disease progression at a mean follow-up of 34 months, and radiographic regression in 11 patients 39 . The rate of hearing preservation was 31%, however only 13 patients had useful hearing prior to radiotherapy. No transient or permanent treatment-related trigeminal or facial nerve dysfunction was observed.

Spinal and Skull Base Chordomas and Chondrosarcomas
Chordomas are slow growing, locally aggressive bone tumours arising from the remnants of the notochord and most frequently occurring in the sacrococcygeal region or at the base of the skull near the sphenooccipital region 40,41 . Chordomas are rare in both adults and children, accounting for only three to four percent of all primary bone tumours 41 . Chondrosarcomas are malignant cartilaginous tumours that can occur anywhere in the skeletal system, and most commonly in the long bones and pelvis; in the skull base, chondrosarcomas account for six percent of all tumours, and most commonly occur in the middle, posterior, or anterior fossae 42 . The clinical features of chordomas and chondrosarcomas are quite similar, as is the optimal treatment strategy, which involves maximal surgical resection [41][42][43] . Adjuvant postoperative radiotherapy is commonly used for patients with chordomas, and high doses have been associated with lower rates of local recurrence or treatment failure 44,45 . Postoperative radiotherapy has also been reported for the management of patients with chondrosarcomas. As a result of their proximity to critical neural structures, however, chordomas and chondrosarcomas of the skull base and spine are difficult to manage with conventional radiotherapy techniques, therefore making these tumours one of the main applications for proton beam radiotherapy. Despite the relative rarity of these tumours, the results of over  41 . Late toxicity was reported in the range of 5-17%, with some grade III and IV toxicities reported. Similarly, in another systematic review of four studies involving 254 patients with chondrosarcomas of the skull base treated post-operatively with proton beam radiotherapy, Amichetti and colleagues reported a five-year overall survival rate of 99-100% and a five-year local control rate of 75-99% 42 .
The use of proton beam radiotherapy has also been described for soft tissue sarcomas of the spine. DeLaney and colleagues reported the results of a prospective phase II trial involving 50 patients mostly with spinal and paraspinal chordomas and chondrosarcomas, but patients with liposarcoma, osteosarcoma, Ewing's sarcoma, giant cell tumour, angiosarcoma, malignant schwannoma, and spindle cell sarcoma were also included 4 . Patients were treated post-operatively with high-dose proton-photon radiotherapy at a median dose of 76.6 CGE and select patients with high-grade tumours also received chemotherapy. Thirty-six patients were treated for primary tumours, and 14 for recurrent disease. With 7.3-year median follow-up, the respective five-and eight-year actuarial local control rates were 94% and 85% for primary tumours and 81% and 74% for the entire group. Local recurrence was less common for primary tumours, compared to recurrent tumours, (11% vs. 50%, p=0.002). The eight-year actuarial risk of grade 3-4 late RT morbidity was 13% 48 . Guttmann and colleagues studied 23 recurrent soft tissue sarcoma patients treated with proton reirradiation, prospectively 49 . All sarcomas were in a previously irradiated field. The three-year progression-free survival was found to be 43% (95% CI 21-62%) and the three-year overall survival was 64% (95% CI 39-81%). Weber et al. retrospectively reviewed 38 Ewing sarcoma patients treated with pencil beam scanning proton therapy 50 . The five-year actuarial rate of local control, distant metastasis-free survival and overall survival were 81.5%, 76.4% and 83.0% respectively. This treatment was associated with a low rate of high -grade late toxicity. Demizu and colleagues retrospectively reviewed 91 patients with unresectable or incompletely resected bone and soft tissue sarcomas, treated with 70.4 GyRBE proton or carbon ion particle therapy in 16 or 32 fractions 51 . They found that particle therapy was effective with a 3-year overall survival of 83%, a 3-year progression free survival of 72% and a 3-year local control of 92%. The 16 fractions protocol was associated with significantly (p<0.001) more late grade ≥ 3 toxicities.

Ocular Melanoma
Uveal melanoma, which includes tumours in the iris, ciliary body, and choroid, is the most common type of primary ocular tumour in adults, accounting for 95% of all cases 52,53 . Depending on the size and location of the tumour, treatment can range from local ablative treatments to complete removal of the eye. The use of proton beam radiation therapy for larger tumours has been reported extensively in the literature, with five recent comparative studies and numerous non-comparative studies showing high rates of local control and minimal adverse effects [54][55][56][57][58][59][60][61][62][63][64][65][66][67][68] . However, a publication based on data from the SEER database in the United States reported that survival rates for patients with uveal melanoma have not improved from 1973 to 2008, despite a shift toward more conservative treatments such as proton beam radiotherapy during this time period 69 .

Head and Neck Tumours
Tumours of the paranasal sinuses and nasal cavity are rare, accounting for 2-3% of all head and neck tumours 70 . Optimal treatment involves surgical resection and postoperative radiotherapy; chemotherapy, either alone or in combination with radiotherapy and/or surgery, is also an option for patients who are at high risk of recurrence or for patients with inoperable tumours 71,72 . For patients with paranasal sinus and nasal cavity tumours who are good candidates, proton beam radiotherapy is the ideal form of radiotherapy, owing to the irregular shape of many of these tumours, the relative radioresistance of some of these tumours, requiring high physical and biologically effective doses, the high risk of recurrence associated with these tumours, and the proximity to critical normal tissues in the ocular globes, optic nerves, and brain 34 73 . Combined photon and proton beam radiotherapy was given as definitive treatment in 32 patients, as adjuvant treatment after partial resection in 50 patients, and as adjuvant treatment after complete resection in 20 patients. Five-year local control rates were 95% for patients with a complete resection, 82% for patients with a partial resection, and 87% for patients with biopsy only. Five-year overall survival rates were 90% for patients with a complete resection, 53% for patients with a partial resection, and 49% for patients with a biopsy only. Also from the Massachusetts General Hospital, Truong and colleagues described 20 patients with primary sphenoid sinus tumours treated with proton beam radiotherapy at a median dose of 76 CGE 74 . The two-year rates of local control, regional control, and freedom from distant metastasis were 86%, 86%, and 50%, respectively. The two-year disease-free and overall-survival rates were 31% and 53%, respectively. The investigators identified brain invasion and involvement of the oropharynx and anterior cranial fossa as important prognostic factors. In a systematic review and meta-analysis which included the Resto et al. study, Ramaekers and colleagues reported a significantly higher pooled estimated five-year local control rate for patients with paranasal and sinonasal tumours treated with proton beam radiotherapy compared to intensity modulated photon therapy (88% vs. 66%;p=0.035) 75 . Tokuuye and colleagues reported the results of a retrospective review of 33 patients with head and neck tumours and no history of surgery who were treated with proton beam radiotherapy with or without photon beam radiotherapy at the University of Tsukuba Hospital in Japan 76 . The five-year overall survival rate was 44%, and the five-year local control rate was 74%; acute toxicity greater than grade III was reported in one patient, and late toxicity greater than grade III was reported in six patients. Similarly, in a retrospective analysis of 39 adult patients with unresectable malignant tumours of the nasal cavity, paranasal sinus, or skull base who were treated with proton beam radiotherapy at a dose of 60 CGE or higher at the National Cancer Center Hospital East in Japan, Zenda and colleagues reported a one-year local control rate of 77%, and three-year progression-free and overall survival rates of 49.1% and 59.3%, respectively 77 . Acute toxicity included grade II dermatitis; long-term grade III-IV toxicity was reported in five patients, and one treatment-related death was reported. Romesser and colleagues retrospectively reviewed 92 patients with recurrent head and neck cancer treated with proton beam therapy at a median dose of 60.6 GyRBE 78 . The actuarial 12 month freedom-from-distance metastasis and overall survival rates were 84% and 65.2% respectively. Grade III or greater toxicities were noted in 10.8% of patients.

Lymphoma
Cure rates of early Hodgkin lymphoma are high, and the avoidance of late complications and second malignancies has become increasingly important for these patients; proton beam radiotherapy may therefore offer an advantage over conventional methods for patients with lymphoma requiring radiotherapy. In a study of ten consecutive patients with mediastinal masses from lymphoma who were treated with three-dimensional proton beam radiotherapy, Li et al. reported a complete metabolic response rate of 86% 79 . The patients ranged in age from 26 to 45 years of age; eight of these patients had nodular sclerosing stage II Hodgkin lymphoma, one had stage II diffuse large B cell lymphoma, and one had disseminated T-cell lymphoblastic lymphoma. At the time of referral for proton beam radiotherapy, seven patients had either primary refractory disease or recurrent disease. When compared to conventional radiotherapy, proton beam radiotherapy resulted in mean lower doses to the lungs, esophagus, and heart, but not to the breasts 79 . Hoppe and colleagues also recently reported the results of a phase II trial involving dosimetric comparisons of three-dimensional conformal radiotherapy, intensity modulated radiotherapy, and proton beam radiotherapy for ten patients with stage IA-IIIB Hodgkin lymphoma with mediastinal (bulky or nonbulky) involvement after chemotherapy 80 . Proton beam radiotherapy resulted in the lowest mean dose to the heart, lungs, and breasts for all ten patients compared with both the three-dimensional and intensity modulated radiotherapy.

Prostate Cancer
Several studies have suggested that proton beam radiotherapy may be beneficial for patients with locally advanced prostate cancer, due to the low rate of radiation scattering to adjacent structures. The largest report to date comes from an analysis of 1255 patients treated with either proton beam radiotherapy alone or in combination with photons between 1991 and 1997 at Loma Linda University 81 . The authors reported minimal morbidity, and a 5-year overall biochemical disease-free survival rate of 73%, which is comparable to rates reported with other modalities in this population of patients. The long-term results of the only randomized trial examining proton beam radiotherapy comparing conventional and high-dose radiotherapy were recently published by Zietman and colleagues 82,83 . Three hundred and ninety three patients with stage T1b through T2b prostate cancer and PSA levels less than or equal 15 ng/mL received photon beam radiotherapy at a fixed dose 50.4 Gy, and were then randomized to receive boost proton beam radiotherapy to a total dose of either 70.2 Gy (conventional dose) or 79.2 Gy (high dose). For patients with low-risk disease (N=228), the 10-year biochemical failure rate was significantly different between the two treatment groups (28.2% conventional dose versus 7.1% high dose, p<0.0001). There were no differences in the 10-year overall survival rates among the treatment arms (78.4% versus 83.4%; p=0.41). Treatment with higher-dose radiotherapy was not associated with an increase in acute or late patient-reported prostate cancer symptoms 84 . Although these results are promising, this trial was not designed to test whether proton beam radiotherapy is more or less efficacious than other conformal techniques or brachytherapy for the treatment of patients with prostate cancer. In an analysis of data from the Surveillance, Epidemiology and End Results (SEER) database, patients treated for primary prostate cancer with IMRT (N=684) were compared to patients treated with proton beam radiotherapy (N=684) 85 . Patients treated with IMRT had a lower rate of gastrointestinal morbidity (relative risk=0.66, 95% CI=0.55-0.79), but there were no significant differences in rates of other morbidities between IMRT and proton therapy. AIM Specialty Health published proton bean therapy clinical appropriateness guidelines in 2019 86 . After reviewing the evidence, it was concluded that proton beam therapy is not medically necessary for the treatment of prostate cancer. The evidence quality is low and insufficient to determine how proton therapy and photon-based therapies differ.

Non-Small Cell Lung Cancer
A limited number of studies have reported an 80-90% rate of local control for patients with stage I non-small cell lung cancer (NSCLC) treated with hypofractionated proton beam radiotherapy [87][88][89][90] . However, in a meta-analysis comparing photon, proton, and carbon-ion radiotherapy for patients with inoperable stage I NSCLC, Grutters and colleagues concluded that proton beam radiotherapy did not offer a statistically significant improvement in overall survival when compared to photon-based stereotactic radiotherapy, although both modalities were significantly better than conventional photon radiotherapy 91 . Proton beam radiotherapy may offer an advantage for patients with advanced inoperable NSCLC, and may spare lung volumes from receiving low-dose irradiation from exiting photon beams. In a retrospective review by Nakayama et al., the progression-free survival rate of 35 patients with stage II and III NSCLC was 59.6% at one year and 29.2% at two years, and the overall survival rate was 81.8% at one year and 58.9% at two years 92 . Similarly, in a retrospective review of 57 patients with inoperable stage IIIA-IIIB NSCLC, Oshiro and colleagues reported one-and two-year progression-free survival rates of 36.2% and 24.9% respectively, and one-and two-year overall survival rates of 65.5% and 39.4%, respectively 93 . While these results are promising, it is important to note that proton beam radiotherapy planning is particularly challenging for lung tumours, owing to organ motion and changes to lung density during respiration 94 .
Members of the working group do not currently recommend that patients with prostate cancer, nonsmall cell lung cancer, or most lymphomas be referred for proton beam radiotherapy, due to an insufficient evidence base. However, individual patient cases should be discussed by the multidisciplinary team during a Tumour Board meeting.

III. Referral and Funding Process for Out of Country Treatment
A standalone document outlining this process, in Alberta, is available on Insite on the Cancer Care Alberta resource page.

Patient Selection Criteria.
Health care providers must take into account a wide range of factors in assessing if proton beam radiotherapy will confer a significant advantage for the patient over standard radiotherapy -the diagnosis alone is often not sufficient 95 . Based on the data published to date, combined with the expert clinical experiences of the working group members, we recommend proton beam radiotherapy be considered for pediatric, adolescent, young adult and adult patients who are residents of Alberta, are covered by the Alberta Health Care Insurance Plan (AHCIP), and meet the criteria outlined in the table below.

Referral Process (Appendix A).
The Radiation Oncologist presents the patient case, which meets the criteria in Table 1, at the Proton Therapy Referral Rounds. A patient deemed borderline for meeting the criteria, requires the referring Radiation Oncologist to contact the Florida Proton Beam Program and request a comparative protonphoton plan. The borderline patient's case and comparative plan is then discussed by members of the Proton Therapy Referral Rounds.
The Proton Therapy Referral Rounds is a multidisciplinary tumour board meeting. At this meeting, all members can provide input on patient cases presented. The team also discusses possible additional treatments, travel arrangements and follow-up care. If a referral is recommended, a summary note of the meeting is generated (checklist in Appendix B), and the Radiation Oncologist contacts the Florida Proton Beam Program to confirm if they have capacity. If the Florida Proton Beam Program does not have capacity, a referral to a non-contracted vendor is required. The Radiation Oncologist must make the request through the out of country Health Services Committee.
Once capacity is confirmed, the referring Radiation Oncologist completes the Request for Approval Form (available on Insite on the Cancer Care Alberta Resource page), which is then sent, along with all items on the Alberta Health Checklist (Appendix C), via FAX or mail to the Senior Medical Director of Cancer Care Alberta. If approved, the Senior Medical Director's office sends the referral via FAX or mail to the Alberta Health Special Program Unit. The referral letter must include the following: • Referring physician's contact information: name, specialty, facility address, cell phone, email, fax • Patient information: name, date of birth, address, email, phone, Alberta personal health card number, and if the patient is pediatric-parent's/guardians contact information • Approved diagnosis • Proton Therapy Referral Rounds date, summary note and treatment recommendation • Notification of access to Proton Beam Therapy-standard process or expediated process and the rationale behind the choice • Verification of capacity at the University of Florida Proton Beam Institute • Admission date of treatment • All medically necessary prerequisite investigations/assessments required to receive Proton Therapy have been completed in Alberta The Alberta Health Special Program Unit does not have access to any electronic medical records. All details on the case must be included in the referral letter. The application is considered complete when all the required information has been submitted, and the Alberta Health Special Program Unit Chair notifies the referring Radiation Oncologist, in writing, that the application has been scheduled for review at an upcoming meeting. The Alberta Health Special Program Unit will assess the application and decide within 30 days. If the Alberta Health Special Program Unit has any additional questions regarding the referral, they will direct them to the lead of the Proton Beam Therapy Referral Rounds.
Once decided, the Alberta Health Special Program Unit will contact the referring Radiation Oncologist. The approval letter will: 1. Stipulate approval for payment of the services requested 2. Outline the required next steps for the patient 3. Provide details on what services are covered (all medical treatment and transportation to and from) and what is not covered (all accommodation and special costs).
The patient must complete two release forms and return them to the manager of the Alberta Health Special Program Unit. These releases allow the medical management company to assist both the patient and the referring physician with the details of the approved medical treatment. The referral letter to the proton beam radiotherapy centre should clearly state that the patient has been approved by the Alberta Health Special Program Unit.