1. Aetna considers proton beam radiotherapy (PBRT) medically necessary in any of the following radiosensitive tumors:

   1. Chordomas or chondrosarcomas arising at the base of the skull or along the axial skeleton without distant metastases; or
   2. Central nervous system (CNS) lesions including but not limited to, primary CNS malignancies, malignancies metastatic to the CNS, or intracranial arterio-venous malformations, adjacent to critical structures such as the optic nerve, globe, brain stem or spinal cord and not amenable to conventional surgical or endovascular techniques; or
   3. Pituitary neoplasms; or
   4. Postoperative use in retroperitoneal soft tissue sarcoma not amenable to other radiotherapy (e.g., IMRT, stereotactic body radiotherapy) in persons who have not received preoperative radiotherapy; or
   5. Malignancies in children (21 years of age and younger); or
   6. Uveal melanomas confined to the globe (i.e., not distant metastases) (the uvea is comprised of the iris, ciliary body, and choroid [the vascular middle coat of the eye]).

2. Aetna considers proton beam radiotherapy for treatment of prostate cancer not medically necessary for individuals with localized prostate cancer because it has not been proven to be more effective than other radiotherapy modalities for this indication. Proton beam therapy for metastatic prostate cancer is considered experimental and investigational.

3. Aetna considers proton beam radiotherapy experimental and investigational for all other indications, including the following indications in adults (over age 21) (not an all inclusive list) because its effectiveness for these indications has not been established:

   1. Adenoid cystic carcinoma
   2. Age-related macular degeneration
   3. Bladder cancer
   4. Breast cancer
   5. Carotid body tumor
   6. Cavernous hemangioma
   7. Cervical cancer
   8. Choroidal hemangioma
   9. Desmoid fibrosarcoma
   10. Esophageal cancer
   11. Ewing's sarcoma
   12. Fibrosarcoma of the extremities
   13. Hepatocellular carcinoma
   14. Hodgkin's lymphoma
   15. Leiomyosarcoma of the extremities
   16. Liver metastases
   17. Lung cancer (including non-small-cell lung carcinoma)
   18. Nasopharyngeal tumor not adjacent to critical structures
   19. Non-uveal melanoma
   20. Pancreatic cancer
   21. Parotid gland tumor
   22. Rectal cancer
   23. Seminoma
   24. Sino-nasal carcinoma not adjacent to critical structures
   25. Small bowel adenocarcinoma
   26. Soft tissue sarcoma
   27. Squamous cell carcinoma of the tongue/glottis, the head and neck not adjacent to critical structures
   28. Submandibular gland tumor
   29. Thymoma
   30. Tonsillar cancer
   31. Vestibular schwannoma.

4. Aetna considers neutron beam therapy medically necessary for the treatment of any of the following salivary gland tumors:

   1. Inoperable tumor; or
   2. Locally advanced tumor especially in persons with gross residual disease; or
   3. Unresectable tumor.
       

5. Aetna considers neutron beam therapy experimental and investigational for all other indications including malignancies listed below (not an all inclusive list) because its effectiveness for these indications has not been established:

   1. Colon cancer
   2. Glioma
   3. Kidney cancer
   4. Laryngeal cancer
   5. Lung cancer
   6. Pancreatic cancer
   7. Prostate cancer
   8. Rectal cancer
   9. Soft tissue sarcoma.

Proton Beam Therapy:

Proton beams have less scatter than other sources of energy such as gamma rays, x-rays, or electrons.  Because of this feature, proton beam radiotherapy (PBRT) has been used to escalate radiation dose to diseased tissues while minimizing damage to adjacent normal tissues.  Proton beams have been used in stereotactic radiosurgery of intracranial lesions; the gamma knife and linear accelerator have also been used in stereotactic radiosurgery.  Proton beam radiotherapy has been shown to be particularly useful in treating radiosensitive tumors that are located next to vital structures, where complete surgical excision or administration of adequate doses of conventional radiation is difficult or impossible.  Examples include uveal melanomas, chordomas and chondrosarcomas at the base of the skull, and inoperable arterio-venous malformations.  There is inadequate data on the application of PBRT for the treatment of non-uveal melanoma.

The American Society of Radiation Oncology (ASTRO, 2013) has stated: "At the present time, ASTRO believes the comparative efficacy evidence of proton beam therapy with other prostate cancer treatments is still being developed, and thus the role of proton beam therapy for localized prostate cancer within the current availability of treatment options remains unclear."

The emerging technology committee of the American Society of Radiation Oncology (ASTRO) concluded that current evidence provides a limited indication for proton beam therapy (Allen, et al., 2012). The ASTRO report concluded that current data do not provide sufficient evidence to recommend proton beam therapy in lung cancer, head and neck cancer, gastrointestinal malignancies, and pediatric non-CNS malignancies. The ASTRO report stated that, in hepatocellular carcinoma and prostate cancer, there is evidence for the efficacy of proton beam therapy but no suggestion that it is superior to photon based approaches. In pediatric central nervous system (CNS) malignancies, proton beam therapy appears superior to photon approaches but more data is needed. The report found that, in large ocular melanomas and chordomas, there is evidence for a benefit of proton beam therapy over photon approaches. The ASTRO report stated that more robust prospective clinical trials are needed to determine the appropriate clinical setting for proton beam therapy.

A systematic evidence review (Lodge et al, 2007) compared the efficacy and cost-effectiveness of PBRT and other types of hadron therapy (neutron and heavy and light ion therapy) with photon therapy.  The authors concluded that, overall, the introduction or extension of PBRT and other types of hadron therapy as a major treatment modality into standard clinical care is not supported by the current evidence base.  The authors stated, however, that the efficacy of PBRT appears superior to that of photon therapy for some ocular and skull base tumours.  The authors found that, for prostate cancer, the efficacy of PBRT seems comparable to photon therapy.  The authors stated that no definitive conclusions can be drawn for the other cancer types.  The authors also noted that they found little evidence on the relative cost-effectiveness of PBRT and other types of hadron therapy compared to photon therapy or with other cancer treatments.  Other systematic evidence reviews of PBRT have reached similar conclusions (Lance, 2010; Brada et al, 2009; Efstathiou et al, 2009; ICER, 2008; Wilt et al, 2008; Brada et al, 2007; Olsen et al, 2007).

The only randomized controlled clinical trial comparing PBRT to conventional radiotherapy published to date found no advantage of PBRT in overall survival (OS), disease-specific survival, or total recurrence-free survival (Shipley, 1995).  A total of 202 patients with stage T3-T4 prostate cancer were randomly assigned to a standard dose of conventional radiotherapy plus a 25.2 Gy equivalent PBRT boost or to a standard dose of conventional radiotherapy with a 16.8 Gy boost of conventional radiotherapy.  After a median follow-up of 61 months, there were no significant differences between the 2 groups in OS, disease-specific survival, total recurrence-free survival, or local control.  Local control was better with the proton beam boost only among the subgroup of patients with poorly differentiated carcinoma.  Patients receiving the proton beam boost had increased rates of late radiation sequelae.

Loma Linda University’s experience with PBRT of prostate cancer was reported in an article published in 1999 by Rossi et al.  These investigators reported the results of an uncontrolled study of PBRT treatment of 319 patients with biopsy-proven early-stage prostate cancer, with no patient having an initial PSA of greater than 15.  Because the study was uncontrolled, one is unable to determine whether the results of PBRT are superior to conventional forms of radiation therapy.  In addition, the definitions of success and failure used in this study are not comparable to those used in other recent studies of prostate cancer treatments.  In the study by Rossi et al, patients were considered to have an adequate response if their PSA level fell below 1.0; most other recent studies define an adequate response as PSA level below 0.5.  In the study by Rossi et al, patients were considered treatment failures if they had 3 consecutive rises of PSA of 10 % or more, measured at 6-month intervals.  In other words, for a patient to be considered a treatment failure, it would take at least 18 months, and patients would have to have 3 consecutive rises in PSA, each greater than 10 %.  By contrast, other reported studies of prostate cancer radiotherapy have defined failure as any PSA elevation over a target PSA nadir.  Finally, Rossi et al defined clinical disease free survival as having "no symptoms and no evidence of disease upon physical examination or radionuclide scans".  These are very gross tests to determine success, and one would expect these tests to be negative in a high number of patients who harbored occult disease.

Of the 319 patients included in the study by Rossi et al, only 288 patients (91 %) who had achieved a nadir (any nadir) or who had been followed for at least 24 months were included in the analysis.  This would indicate that 31 (9 %) of the patients originally included in the study either had persistently rising PSA levels without a nadir despite treatment, had dropped out of the study, or had not been followed for a sufficient length of time for some unspecified reason.  Only 187 patients (59 % of the original 319 patients) achieved a PSA nadir of 0.5 or less, 66 (21 %) achieved a PSA nadir of 0.51 to 1.0, and 35 (11 %) achieved a PSA nadir of 1.0 and above.  Thus, only 59 % of patients would be considered to have had an adequate response by the measure most commonly used in other recent prostate cancer treatment studies.  In addition, because of the peculiar way the results are reported, there is no way of knowing how many patients' PSA nadirs were maintained.

In a randomized, prospective, sham-controlled, double-masked study (n = 37), Ciulla et al (2002) examined the effect of PBRT on subfoveal choroidal neovascular membranes associated with age-related macular degeneration.  These investigators concluded that with the acceptance of photodynamic therapy, future studies will require more complex design and larger sample size to determine whether radiation can play either a primary or adjunctive role in treating these lesions.

In a phase II clinical study (n = 30), Kawashima and colleagues (2005) assessed the safety and effectiveness PBRT for patients with hepatocellular carcinoma (HCC).  Eligibility criteria for this study were: solitary HCC; no indication for surgery or local ablation therapy; no ascites; age of 20 years or older; Zubrod performance status of 0 to 2; no serious co-morbidities other than liver cirrhosis; written informed consent.  Proton beam radiotherapy was administered in doses of 76 cobalt gray equivalent in 20 fractions for 5 weeks.  No patients received transarterial chemoembolization or local ablation in combination with PBRT.  All patients had liver cirrhosis, the degree of which was Child-Pugh class A in 20, and class B in 10 patients.  Acute reactions of PBRT were well-tolerated, and PBRT was completed as planned in all patients.  Four patients died of hepatic insufficiency without tumor recurrence at 6 to 9 months; 3 of these 4 patients had pre-treatment indocyanine green retention rate at 15 minutes of more than 50 %.  After a median follow-up period of 31 months (range of 16 to 54 months), only 1 patient experienced recurrence of the primary tumor, and 2-year actuarial local progression-free rate was 96 %.  Actuarial overall survival rate at 2 years was 66 %.  These investigators concluded that PBRT showed excellent control of the primary tumor, with minimal acute toxicity.  They stated that further study is warranted to scrutinize adequate patient selection in order to maximize survival benefit of this promising modality.

In a phase II prospective trial, Bush et al (2011) evaluated the safety and effectiveness of PBRT for HCC.  Patients with cirrhosis who had radiological features or biopsy-proven HCC were included in the study.  Patients without cirrhosis and patients with extra-hepatic metastasis were excluded.  The mean age was 62.7 years.  The mean tumor size was 5.5 cm.  Eleven patients had multiple tumors, and 46 % were within the Milan criteria.  Patients received 63 Gy delivered over a 3-week period with PBRT.  A total of 76 patients were treated and followed prospectively.  Acute toxicity was minimal; all patients completed the full course of treatment.  Radiation-induced liver disease was evaluated using liver enzyme, bilirubin, and albumin levels; no significant change supervened 6 months post-treatment.  Median progression-free survival for the entire group was 36 months, with a 60 % 3-year progression-free survival rate for patients within the Milan criteria.  Eighteen patients subsequently underwent liver transplantation; 6 (33 %) explants showed pathological complete response and 7 (39 %) showed only microscopic residual.  The authors concluded that PBRT was found to be a safe and effective local-regional therapy for inoperable HCC.  They noted that a randomized controlled trial to compare its efficacy to a standard therapy has been initiated.

Olfactory neuroblastoma (ONB) is a rare disease, and a standard treatment strategy has not been established.  Radiation therapy for ONB is challenging because of the proximity of ONB to critical organs.  Nishimura et al (2007) analyzed the feasibility and effectiveness of PBRT for ONB.  A retrospective review was performed on 14 patients who underwent PBRT for ONB as definitive treatment.  The total dose of PBRT was 65 cobalt Gray equivalents (Gy(E)), with 2.5-Gy(E) once-daily fractionations.  The median follow-up period for surviving patients was 40 months.  One patient died from disseminated disease.  There were 2 persistent diseases, 1 of which was successfully salvaged with surgery.  The 5-year overall survival rate was 93 %, the 5-year local progression-free survival rate was 84 %, and the 5-year relapse-free survival rate was 71 %.  Liquorrhea was observed in 1 patient with Kadish's stage C disease (widely destroying the skull base).  Most patients experienced grade 1 to 2 dermatitis in the acute phase.  No other adverse events of grade 3 or greater were observed according to the RTOG/EORTC acute and late morbidity scoring system.  The authors concluded that these preliminary findings of PBRT for ONB achieved excellent local control and survival outcomes without serious adverse effects.  They stated that PBRT is considered a safe and effective modality that warrants further study.

Proton beam radiotherapy represents a special case for children for several reasons (Wilson et al, 2005; Hall, 2006; Merchant, 2009).  It has been shown in dosimetric planning studies to have a potential advantage over conventional photon therapy because of the ability to confine the high-dose treatment area to the tumor volume and minimize the radiation dose to the surrounding tissue.  This especially important in children, as children are more sensitive to radiation-induced cancer than adults.  An increased risk of second cancers in long-term survivors is more important in children than older adults. In addition to second malignant neoplasms, late effects of radiation to normal tissue can include developmental delay.  Also, radiation scattered from the treatment volume is more important in the small body of the child.  Finally, the question of genetic susceptibility arises because many childhood cancers involve a germline mutation.

An assessment of proton beam radiotherapy by the Veterans Health Administration Technology Assessment Program (VATAP) (Flynn, 2010) found that available English-language reviews for proton therapy generally concur on the state of the literature as consisting primarily of observational studies from which conclusions about the relative effectiveness of proton therapy versus alternatives cannot validly be made.  The assessment reported that available reviews reflect the state of the literature in that they attempt to cover so much territory (multiple poor-prognosis inoperable tumors in both children and adults) that the reviews themselves are cumbersome to read, not well organized, and provide only diffuse or equivocal conclusions by individual diagnoses.  "In other words, the literature reflects the early clinical investigation status of proton therapy, where observational studies are framed in terms of potential benefits, reasoning from pathophysiology, dose-finding, refinement of treatment protocols, and baseline safety of the entire approach" (Flynn, 2010).  The assessment noted that only prostate cancer is represented by randomized controlled clinical trials, and in that case two small ones primarily concerned with refinement of protocol/dose escalation.

Regarding cost-effectiveness analyses of PBRT, the VATAP assessment found that the availability of studies titled by their authors as “economic evaluations” is misleading (Flynn, 2010).  The assessment stated that such studies require cost and efficacy data about both the intervention and its alternatives (costs and consequences of alternative interventions), hence should be conducted only after efficacy data from randomized controlled trials are available.  The assessment noted that, in the case of proton therapy, the economic studies are premature, really should be labeled simple cost rather than cost-effectiveness analyses, and their conclusions based on unwarranted efficacy assumptions.  Cost data have been carefully collected and reported, but these are only one element of decision making about investment in proton therapy.  The VATAP assessment concluded that there are no indications for which proton therapy has been shown unequivocally to be effective, or more effective than its alternatives.  The VATAP assessment also concluded that no research published subsequent to the searches conducted for available systematic reviews has changed the conclusions of those reviews.

Regarding research implications, the VATAP assessment concluded that, in order to obtain the next generation of data, explicit decisions need to be made about which malignancies are amenable to/should require randomized trials (e.g., prostate cancer is sufficiently common) and which malignancies are sufficiently rare or difficult to treat with surgery or conventional radiotherapy (e.g., ocular tumors, tumors of the optical nerve, spinal cord, or central nervous system) that observational studies with larger cohorts than studies to date are the best approach (Flynn, 2010).  The VATAP assessment also concluded that future studies should strongly consider valid and reliable embedded collection of cost data in order to inform better quality economic evaluation than currently available.

An assessment prepared for the Agency for Healthcare Research and Quality (Trikalinos, et al., 2009) found that a large number of scientific papers on charged particle radiotherapy for the treatment of cancer currently exist. However, these studies do not document the circumstances in contemporary treatment strategies in which radiotherapy with charged particles is superior to other modalities. Comparative studies in general, and randomized trials in particular (when feasible), are needed to document the theoretical advantages of charged particle radiotherapy to specific clinical situations. The assessment noted that most eligible studies were noncomparative in nature and had small sample sizes. The report stated that it is likely that focused systematic reviews will not be able to provide a definitive answer on the effectiveness and safety of charged particle beam radiotherapies compared with alternative interventions. This is simply because of the relative lack of comparative studies in general, and randomized trials in particular. The report stated that comparative studies (preferably randomized) are likely necessary to provide meaningful answers on the relative safety and effectiveness of particle beam therapy versus other treatment options in the context of current clinical practice. This is especially true for the treatment of common cancers. The report stated that, especially for many common cancers, such as breast, prostate, lung, and pancreatic cancers, it is essential that the theorized advantages of particle beam therapy versus contemporary alternative interventions are proven in controlled clinical trials, along with concomitant economic evaluations.

An assessment of the comparative effectiveness and value of management options in low-risk prostate cancer by the Institute for Clinical and Economic Review (ICER) (Ollendorf et al, 2008) found that the evidence on the comparative effectiveness and harms of proton beam therapy is limited to relatively small, highly selective case series of short duration, making any judgments about its relative benefit or inferiority to other options premature.  The uncertainty regarding PBRT is accentuated because this technology involves delivery of a novel form of radiation, and there remain important questions about the full spectrum of possible effects.  ICER rated PBRT's comparative clinical effectiveness as "insufficient", indicating that there is not enough evidence to allow a reasonable judgment of the likely balance of harms and benefits of PBRT in comparison to radical prostatectomy or other management options.  ICER judged the comparative value of PBRT to be low compared to other options.  The ICER reported explained, that, while ICER does not always provide a comparative value rating for technologies with insufficient evidence on comparative clinical effectiveness, the decision was made to rate the comparative value of PBRT as “low” relative to radical prostatectomy, based on current levels of reimbursement that are more than 3-fold higher for PBRT.

The Blue Cross and Blue Shield Association Medical Advisory Panel (BCBSA, 2010) concluded that proton beam radiation therapy for treatment of non-small-cell lung cancer at any stage or for recurrent non-small-cell lung cancer does not meet the Technology Evaluation Center criteria. The TEC assessment stated that, overall, evidence is insufficient to permit conclusions about the results of proton beam therapy for any stage of non-small-cell lung cancer. The report found that all proton beam therapy studies are case series; there are no studies directly comparing proton beam therapy and stereotactic body radiotherapy. Among study quality concerns, no study mentioned using an independent assessor of patient reported adverse events, adverse events were generally poorly reported, and details were lacking on several aspects of proton beam therapy treatment regimens. The proton beam therapy studies had similar patient ages, but there was great variability in percent within stage Ia, sex ratio, and percent medically inoperable. There is a high degree of treatment heterogeneity among the proton beam therapy studies, particularly with respect to planning volume, total dose, number of fractions, and number of beams. Survival results are highly variable. It is unclear if the heterogeneity of results can be explained by differences in patient and treatment characteristics. Indirect comparisons between proton beam therapy and stereotactic body radiotherapy, comparing separate sets of single-arm studies on proton beam therapy and stereotactic body radiotherapy, may be distorted by confounding. In the absence of randomized, controlled trials, the comparative effectiveness of proton beam therapy and stereotactic body radiotherapy is uncertain.

Mizumoto et al (2010) evaluated the efficacy and safety of PBRT for locoregionally advanced esophageal cancer.  The subjects were 51 patients with esophageal cancer who were treated between 1985 and 2005 using proton beams with or without X-rays.  All but 1 had squamous cell carcinoma.  Of the 51 patients, 33 received combinations of X-rays (median of 46 Gy) and protons (median of 36 GyE) as a boost.  The median total dose of combined X-rays and proton radiation for these 33 patients was 80 GyE (range of 70 to 90 GyE).  The other 18 patients received PBRT alone (median of 79 GyE, range of 62 to 98 GyE).  Treatment interruption due to radiation-induced esophagitis or hematologic toxicity was not required for any patient.  The overall 5-year actuarial survival rate for the 51 patients was 21.1 % and the median survival time was 20.5 months (95 % confidence interval [CI]: 10.9 to 30.2).  Of the 51 patients, 40 (78 %) showed a complete response within 4 months after completing treatment and 7 (14 %) showed a partial response, giving a response rate of 92 % (47/51).  The 5-year local control rate for all 51 patients was 38.0 % and the median local control time was 25.5 months (95 % CI: 14.6 to 36.3).  The authors concluded that these findings suggested that PBRT is an effective treatment for patients with locally advanced esophageal cancer.  Moreover, they stated that further studies are needed to determine the optimal total dose, fractionation schedules, and best combination of PBRT with chemotherapy.  Furthermore, the NCCN guideline on esophageal cancer (2011) does not mention the use of PBRT as a therapeutic option for this condition.

Bassim et al (2010) reviewed the literature on radiation therapy for the treatment of vestibular schwannoma (VS).  PubMed searches for English language articles on radiation treatment of VS published from January 2002 to July 2007 were conducted.  Studies presenting outcomes were selected, yielding 56 articles (58 studies) in journals of neurosurgery (30), oncology (18), otolaryngology (6), and other (2).  Data included type of study, number of subjects, demographics, follow-up times, type of radiation, tumor size, tumor control definition, control rates, facial nerve function measure and outcome, type of hearing and vestibular testing and outcomes, and complications.  Descriptive statistics were performed.  Studies (72.9 %) were retrospective reviews with stated sample sizes ranging from 5 to 829.  Gamma-knife (49.2 %), linear accelerator (35.6 %), and proton beam (6.8 %) were used with various doses.  Average follow-up was less than 5 years in 79.6 % of studies, and 67.4 % included patients at less than or equal to 1 year.  Tumor size was reported as diameter (23.7 %), volume (49.2 %), both (11.9 %), other (3.4 %), or not reported (11.9 %).  Definition of tumor control varied: less than or equal to 2 mm growth (22.0 %), no visible/measurable change (16.9 %), required surgery (10.2 %), other (17.0 %), and not clearly specified (33.9 %).  Facial nerve outcome was reported as House-Brackmann (64.4 %), normal/abnormal (11.9 %), other (1.7 %), or was not reported (22 %).  The authors concluded that the lack of uniform reporting criteria for tumor control, facial function and hearing preservation, and variability in follow-up times make it difficult to compare studies of radiation treatment for VS.  They recommended consideration of reporting guidelines such as those used in otology for reporting VS resection results.

Mizumoto et al (2011) evaluated the safety and effectiveness of hyper-fractionated concomitant boost proton beam therapy (PBT) for patients with esophageal cancer.  The study participants were 19 patients with esophageal cancer who were treated with hyperfractionated photon therapy and PBT between 1990 and 2007.  The median total dose was 78 GyE (range of 70 to 83 GyE) over a median treatment period of 48 days (range of 38 to 53 days).  Ten of the 19 patients were at clinical T Stage 3 or 4.  There were no cases in which treatment interruption was required because of radiation-induced esophagitis or hematologic toxicity.  The overall 1- and 5-year actuarial survival rates for all 19 patients were 79.0 % and 42.8 %, respectively, and the median survival time was 31.5 months (95 % limits: 16.7 to  46.3 months).  Of the 19 patients, 17 (89 %) showed a complete response within 4 months after completing treatment and 2 (11 %) showed a partial response, giving a response rate of 100 % (19/19).  The 1- and 5-year local control rates for all 19 patients were 93.8 % and 84.4 %, respectively.  Only 1 patient had late esophageal toxicity of Grade 3 at 6 months after hyperfractionated PBT.  There were no other non-hematologic toxicities, including no cases of radiation pneumonia or cardiac failure of Grade 3 or higher.  The authors concluded that these findings suggested that hyperfractionated PBT is safe and effective for patients with esophageal cancer.  They stated that further studies are needed to establish the appropriate role and treatment schedule for use of PBT for esophageal cancer.

In a phase I clinical study, Hong et al (2011) evaluated the safety of 1 week of chemo-radiation with proton beam therapy and capecitabine followed by early surgery on 15 patients with localized resectable, pancreatic ductal adenocarcinoma of the head.  Patients received radiation with proton beam.  In dose level 1, patients received 3 GyE × 10 (week 1, Monday to Friday; week 2, Monday to Friday).  Patients in dose levels 2 to 4 received 5 GyE × 5 in progressively shortened schedules: level 2 (week 1, Monday, Wednesday, and Friday; week 2, Tuesday and Thursday), level 3 (week 1, Monday, Tuesday, Thursday, and Friday; week 2, Monday), level 4 (week 1, Monday through Friday).  Capecitabine was given as 825 mg/m(2) b.i.d.  Weeks 1 and 2 Monday through Friday for a total of 10 days in all dose levels.  Surgery was performed 4 to 6 weeks after completion of chemotherapy for dose levels 1 to 3 and then after 1 to 3 weeks for dose Level 4.  Three patients were treated at dose levels 1 to 3 and 6 patients at dose level 4, which was selected as the MTD.  No dose limiting toxicities were observed.  Grade 3 toxicity was noted in 4 patients (pain in 1; stent obstruction or infection in 3).  Eleven patients underwent resection.  Reasons for no resection were metastatic disease (3 patients) and unresectable tumor (1 patient).  Mean post-surgical length of stay was 6 days (range of 5 to 10 days).  No unexpected 30-day post-operative complications, including leak or obstruction, were found.  The authors concluded that pre-operative chemo-radiation with 1 week of PBRT and capecitabine followed by early surgery is feasible.  A phase II study is underway.

UpToDate reviews on "Management of locally advanced and borderline resectable exocrine pancreatic cancer" (Ryan and Mamon, 2012) and "Surgery in the treatment of exocrine pancreatic cancer and prognosis" (Fernandez-del Castillo et al, 2012) do not mention the use of proton beam therapy.  Furthermore, the NCCN's clinical practice guideline on "Pancreatic adenocarcinoma" (2011) does not mention the use of proton beam.

Available peer-reviewed published evidence does not support the use of PBRT for squamous cell carcinomas of the head and neck.  There is a lack of clinical outcome studies comparing PBRT  to stereotactic radiosurgery or other photon-based methods.  What few comparative studies exist are limited to dosimetric planning studies and not studies of clinical outcomes.  Current guidelines from the NCCN and the National Cancer Institute (PDQ) include no recommendation for use of PBRT for squamous cell carcinoma of the head and neck. A report from the American Society for Therapeutic and Radiation Oncology (ASTRO) (2012) concludes that there is insufficient evidence to support the use of proton beam therapy for head and neck cancers, and conclude that “current data do not provide sufficient evidence to recommend PBT in  ... head and neck cancer… “.  An AHRQ comparative effectiveness review (2010) on radiotherapy for head and neck cancer reached the following conclusions regarding proton beam therapy versus other radiotherapy treatments for head and neck cancer: “The strength of evidence is insufficient as there were no studies comparing proton beam therapy to any other radiotherapy modality. Therefore, no conclusions can be reached regarding the comparative effectiveness of proton beam therapy for any of the four key questions.”

An UpToDate review on "Clinical presentation and management of thymoma and thymic carcinoma" (Salgia, 2012) does not mention the use of proton beam therapy.  Also, the NCCN's clinical practice guideline on "Thymoma and thymic carcinomas" (2011) does not mention the use of proton beam therapy.

Guidelines on soft tissue sarcoma from the National Comprehensive Cancer Network (2012) indicate a potential role for proton therapy in retroperitoneal soft tissue sarcomas in persons who did not receive preoperative radiotherapy. The guidelines state: "Postoperative RT using newer techniques such as IMRT, 3D conformal proton therapy, and intensity modulated proton therapy (IMPT) may allow tumor target coverage and acceptable clinical outcomes within normal tissue dose constraints to adjacent organs at risk in some patients with retroperitoneal STS who did not receive pre-operative radiotherapy. Multicenter randomized controlled trials are needed to address the toxicities and therapeutic benefits of adjuvant RT techniques in patients with retroperitoneal STS."

A recent BCBS TEC assessment found insufficient evidence for PBRT in the treatment of non-small-cell lung cancer.  In addition, the American Society for Radiation Oncology (ASTRO) guidelines (Allen et al, 2012) found insufficient evidence for PBRT in lung cancer.

An UpToDate review on "Malignant salivary gland tumors: Treatment of recurrent and metastatic disease" (Laurie, 2012) stated that "The most common malignant salivary gland tumors include mucoepidermoid carcinoma, adenoid cystic carcinoma, polymorphous low grade adenocarcinoma, carcinoma ex pleomorphic adenoma, acinic cell carcinoma, and adenocarcinoma not otherwise specified".  However, it does not mention the use of PBRT as a therapeutic option.

UpToDate reviews on “Treatment of early (stage I and II) head and neck cancer: The larynx” (Koch and Machtay, 2012) and “Treatment of locoregionally advanced (stage III and IV) head and neck cancer: The larynx and hypopharynx” (Brockstein et al, 2012) do not mention the use of NBT.

Given concerns of excess malignancies following adjuvant radiation for seminoma, Efstathiou et al (2012) evaluated photon beam therapy and PBRT treatment plans to assess dose distributions to organs at risk and model rates of second cancers.  A total of 10 stage I seminoma patients who were treated with conventional para-aortic AP-PA photon radiation to 25.5 Gy at Massachusetts General Hospital had PBRT plans generated (AP-PA, PA alone).  Dose differences to critical organs were examined.  Risks of second primary malignancies were calculated.  Proton beam radiotherapy plans were superior to photons in limiting dose to organs at risk; PBRT decreased dose by 46 % (8.2 Gy) and 64 % (10.2 Gy) to the stomach and large bowel, respectively (p < 0.01).  Notably, PBRT was found to avert 300 excess second cancers among 10,000 men treated at a median age of 39 and surviving to 75 (p < 0.01).  The authors concluded that in this study, the use of protons provided a favorable dose distribution with an ability to limit unnecessary exposure to critical normal structures in the treatment of early-stage seminoma.  It is expected that this will translate into decreased acute toxicity and reduced risk of second cancers, for which prospective studies are warranted.  Furthermore, UpToDate reviews on “Treatment of stage I seminoma” (Beard, 2012) and “Treatment of stage II seminoma” (Beard and Oh, 2012) do not mention the use of PBRT.

Proton beam radiotherapy has been used as therapeutic option for choroidal hemangiomas.  However, available evidence on its effectiveness for this indication is mainly in the form of retrospective reviews with small sample size and a lack of comparison to standard therapies.  Furthermore, a review on “Choroidal hemangioma” (Finger, 2013) from the Eye Cancer Network’s website does not mention PBRT as a therapeutic option.  Thus, PBRT is not an established treatment for patients with choroidal hemangiomas.

In a retrospective study, Hocht et al (2006) compared the results of therapy in patients with uveal hemangioma treated with photon or proton irradiation at a single center.  From 1993 to 2002, a total of 44 patients were treated.  Until 1998 radiotherapy was given with 6 MV photons in standard fractionation of 2.0 Gy 5 times per week.  In 1998 PBRT became available and was used since then.  A dose of 20 to 22.5 Cobalt Gray Equivalent (CGE; CGE = proton Gy x relative biological effectiveness 1.1) 68 MeV protons was given on 4 consecutive days.  Progressive symptoms or deterioration of vision were the indications for therapy.  Of the 44 patients treated, 36 had circumscribed choroidal hemangiomas (CCH) and 8 had diffuse choroidal hemangiomas (DCH) and Sturge-Weber syndrome.  Of the patients, 19 were treated with photons with a total dose in the range of 16 to 30 Gy.  A total of 25 patients were treated with PBRT.  All patients with DCH but 1 were treated with photons.  Stabilization of visual acuity was achieved in 93.2 % of all patients.  Tumor thickness decreased in 95.4 % and retinal detachment resolved in 92.9 %.  Late effects, although generally mild or moderate, were frequently detected.  In all, 40.9 % showed radiation-induced optic neuropathy, maximum Grade I.  Retinopathy was found in 29.5 % of cases, but only 1 patient experienced more than Grade II severity.  Retinopathy and radiation-induced optic neuropathy were reversible in some of the patients and in some resolved completely.  No differences could be detected between patients with CCH treated with protons and photons; treatment was less effective in DCH patients (75 %).  The authors concluded that radiotherapy is effective in treating choroidal hemangiomas with respect to visual acuity and tumor thickness; but a benefit of PBRT could not be detected.

In a retrospective review, Levy-Gabriel et al (2009) evaluated the long-term effectiveness and outcome of low-dose PBRT in the treatment of symptomatic CCH.  A total of 71 patients with symptomatic CCH were treated by PBRT between September 1994 and October 2002 using a total dose of 20 CGE.  The median follow-up was 52 months (range of 8 to 133 months).  Retinal re-attachment was obtained in all cases.  Tumor thickness decreased in all cases and a completely flat scar was obtained in 65 patients (91.5 %).  Visual acuity was improved by 2 lines or more in 37 of the 71 patients (52 %), and in 30 of the 40 patients (75 %) treated within 6 months after onset of the first symptoms.  The main radiation complications detected during follow-up were cataract (28 %) and radiation-induced maculopathy (8 %).  None of the 71 patients developed eyelid sequelae or neovascular glaucoma.  The authors concluded that PBRT with a total dose of 20 CGE appeared to be a valid treatment for CCH, inducing definitive retinal re-attachment and decreasing tumor thickness.  However, delayed radiation-induced maculopathy may occur.  A successful functional outcome is dependent on a short interval between onset of the first symptoms and initiation of therapy.

In a retrospective chart review, Chan et al (2010) described the clinical outcomes of patients (n = 19) with CCH and DCH treated by PBRT using a non-surgical light-field technique.  Choroidal hemangiomas were treated with PBRT using a light-field technique and doses ranging from 15 to 30 CGE in 4 fractions.  Patients with at least 6 months' follow-up were included in the study.  Tumor regression, visual acuity, absorption of sub-retinal fluid, and treatment-associated complications were evaluated by clinical examination and ultrasonography.  Visual acuity improved or remained stable in 14 of 18 eyes (78 %).  Sub-retinal fluid was initially present in 16 of 19 eyes (84 %), and completely resolved in all 16 eyes.  Tumor height, as measured by B-scan ultrasonography, decreased in 18 of 19 eyes and remained stable in 1 of 19, as of the last examination.  Complications of radiation developed in 9 of 19 eyes (47 %) with the total applied dose ranging from 15 to 30 CGE for these 9 eyes.  The authors concluded that PBRT using a light-field technique without surgical tumor localization is an effective treatment option in managing both CCH and DCH associated with Sturge-Weber syndrome.  A total proton dose as low as 15 CGE applied in 4 fractions appeared to be sufficient to reduce tumor size, promote absorption of sub-retinal fluid, and improve or stabilize vision in most patients.

Neutron Beam Therapy:

Most radiation therapies utilize photons -- lightweight particles that damage cancerous cells.  Neutron beam therapy (NBT) uses neutrons, which are much heavier than photons and appear to be more effective in destroying very dense tumors.  Compared to roentgen ray (X-ray), neutrons are characterized by several properties: (i) reduced oxygen enhancement factor, (ii) less or no repair of sub-lethal or potentially lethal cell damage, and (iii) less variation of sensitivity through cell cycle.

Neutron beam therapy entails the use of a particle accelerator; protons from the accelerator are deflected by a magnet to a target which creates the neutron beam.  Neutron bean therapy has been employed mainly for the treatment of the salivary gland cancers.  It has also been used to treat other malignancies such as soft tissue sarcoma (STS) as well as lung, pancreatic, colon, kidney and prostate cancers.  Nevertheless, NBT has not gained wide acceptance because of the clinical difficulty in generating neutron particles.  It should be noted that NBT is different from boron neutron capture therapy (BNCT), which is a radiotherapy based on the preferential targeting of tumor cells with non-radioactive isotope (10)B and subsequent activation with thermal neutrons to produce a highly localized radiation, and is often used to treat brain tumors.  In BNCT, the patient is given a drink containing boron, which is taken up by tumor cells.  The tumor is then irradiated with a neutron beam, causing the boron to split into two highly energetic particles (helium and lithium) that destroy the cancerous cells while largely sparing adjacent healthy cells.

Salivary Gland Cancer

In the treatment of patients with salivary gland cancer, primary radiation including NBT may play a role in certain histological types or non-operative patients (Day, 2004).  Neutron beam therapy has been most extensively used either for an incompletely excised primary tumor or for recurrent disease.

In a randomized clinical study, Laramore and associates (1993) compared the effectiveness of fast neutron radiotherapy versus conventional photon and/or electron radiotherapy for unresectable, malignant salivary gland tumors.  Eligibility criteria included either inoperable primary or recurrent major or minor salivary gland tumors.  Patients were stratified by surgical status (primary versus recurrent), tumor size (less than or greater than 5 cm), and histology (squamous or malignant mixed versus other).  After a total of 32 patients were entered into this study, it appeared that the group receiving fast neutron radiotherapy had a significantly improved local/regional control rate and also a borderline improvement in survival and the study was stopped earlier than planned for ethical reasons.  Twenty-five patients were study-eligible and analyzable.  Ten-year follow-up data for this study was presented.  On an actuarial basis, there was a statistically-significant improvement in local/regional control for the neutron radiotherapy group (56 % versus 25 %, p = 0.009), but there was no statistically significant improvement in OS (15 % versus 25 %).  Patterns of failure were analyzed and it was demonstrated that distant metastases account for the majority of failures on the neutron radiotherapy arm and local/regional failures account for the majority of failures on the photon/electron radiotherapy arm.  Long-term, treatment-related morbidity was analyzed and while the incidence of morbidity graded "severe" was greater on the neutron arm, there was no significant difference in "life-threatening" complications.  These investigators concluded that fast neutron radiotherapy appeared to be the treatment-of-choice for patients with inoperable primary or recurrent malignant salivary gland tumors.

Prott et al (2000) reported their findings of fast neutron therapy in 72 patients with adenoid cystic carcinoma (ACC) of the salivary glands.  The median age was 54 years; and the median follow-up was 50 months.  This study showed that 39.1 % of the patients achieved a complete remission and 48.6 % achieved partial remission.  The survival probability was 86 % after 1 year, 73 % after 2 years and 53 % after 5 years.  The recurrence-free survival was 83 % after 1 year, 71 % after 2 years and 45 % after 5 years.  These investigators concluded that NBT appeared to have been an effective treatment in these selected patients.

Huber and colleagues (2001) compared retrospectively radiotherapy with neutrons, photons, and a photon/neutron mixed beam in patients (n = 75) with advanced ACC of the head and neck.  Local control, survival, distant failure, and complications were analyzed.  Follow-up ranged from 1 to 160 months (median 51 months), and the surviving patients had a minimum follow-up of 3 years at the time of analysis.  The actuarial 5-year local control was 75 % for neutrons, and 32 % for both mixed beam and photons (p = 0.015, log-rank).  This advantage for neutrons in local control was not transferred to significant differences in survival (p > 0.1).  In multi-variate analysis post-operative radiotherapy (p = 0.003) and small tumor size (p = 0.01) were associated with high local control, while primary therapy (p = 0.006) and negative lymph nodes (p = 0.01) were associated with longer survival.  While acute toxicity was similar in all 3 radiotherapy groups, severe late grade 3 and 4 toxicity tended to be more prevalent (p > 0.1) with neutrons (19 %) than with mixed beam (10 %) and photons (4 %).  These researchers concluded that fast neutron radiotherapy provides higher local control rates than a mixed beam and photons in advanced, recurrent or not completely resected ACC of the major and minor salivary glands.  Neutron radiotherapy can be recommended in patients with bad prognosis with gross/macroscopic residual disease (R2 resection), with unresectable tumors, or inoperable tumors.

Douglas et al (2003) evaluated the effectiveness of fast neutron radiotherapy for the treatment of salivary gland neoplasms.  Of the 279 patients, 263 had evidence of gross residual disease at the time of treatment, while16 had no evidence of gross residual disease; 141 had tumors of a major salivary gland, and 138 had tumors of minor salivary glands.  The median follow-up period was 36 months (range of 1 to 142 months).  The main outcome measures were local-regional control, cause-specific survival, and freedom from metastasis.  The 6-year actuarial cause-specific survival rate was 67 %.  Multi-variate analysis revealed that low group stage (I - II) disease, minor salivary sites, lack of skull base invasion, and primary disease were associated with a statistically significant improvement in cause-specific survival.  The 6-year actuarial local-regional control rate was 59 %.  Multi-variate analysis revealed size 4 cm or smaller, lack of base of skull invasion, prior surgical resection, and no previous radiotherapy to have a statistically significant improved local-regional control.  Patients without evidence of gross residual disease had a 100 % 6-year actuarial local-regional control.  The 6-year actuarial freedom from metastasis rate was 64 %.  Factors associated with decreased development of systemic metastases included negative lymph nodes at the time of treatment and lack of base of skull involvement.  The 6-year actuarial rate of development of grade 3 or 4 long-term toxicity (using the Radiation Therapy Oncology Group and European Organization for Research on the Treatment of Cancer criteria) was 10 %.  No patient experienced grade 5 toxic effects.  The authors concluded that NBT is an effective treatment for patients with salivary gland neoplasms who have gross residual disease and achieves excellent local-regional control in patients without evidence of gross disease.

Other Types of Cancer

Russell et al (1994) evaluated the effectiveness of fast neutron radiation therapy in treatment of locally advanced carcinomas of the prostate (n = 178).  Median follow-up was 68 months (range of 40 to 86 months).  The 5-year actuarial clinical local-regional failure rate was significantly better for neutron-treated patients than photon-treated patients (11 % versus 32 %).  When findings of routine post-treatment prostate biopsies were incorporated, the resulting "histological" local-regional tumor failure rates were 13 % for the neutron-treated group versus 32 % for the photon-treated group (p = 0.01).  Moreover, actuarial survival and cause-specific survival rates were statistically indistinguishable for the 2 patient cohorts, with 32 % of the neutron-treated patient deaths and 41 % of the photon-treated patient deaths caused by prostate cancer.  Prostate specific antigen values were elevated in 17 % of neutron-treated patients and 45 % of photon-treated patients at 5 years (p < 0.001).  Severe late complications of treatment were higher for the neutron-treated patients (11 % versus 3 %), and were inversely correlated with the degree of neutron beam shaping available at the participating institutions.  The authors concluded that high energy fast neutron radiotherapy is safe and effective when adequate beam delivery systems and collimation are available, and it is significantly superior to external beam photon radiotherapy in the local-regional treatment of large prostate tumors.

In a review on the use of fast neutron radiation for the treatment of prostatic adenocarcinomas, Lindsley et al (1998) stated that the Radiation Therapy Oncology Group performed a multi-institutional randomized trial comparing mixed beam (neutron plus photon) irradiation to conventional photon irradiation for the treatment of locally advanced prostate cancer.  A subsequent randomized trial by the Neutron Therapy Collaborative Working Group compared pure neutron irradiation to standard photon irradiation.  Both studies reported significant improvement in loco-regional control with neutron irradiation compared to conventional photon irradiation in the treatment of locally advanced prostate carcinoma.  To date, only the mixed beam study has demonstrated a significant survival benefit.  Future analysis of the larger Neutron Therapy Collaborative Working Group trial at the 10-year follow-up should confirm whether or not improved loco-regional control translates into a survival advantage.

Lindsley et al (1996) noted that a phase III clinical study comparing NBT to photon radiotherapy for inoperable regional non-small cell lung cancer showed no overall improvement in survival.  However, a statistically significant improvement in survival was observed in the subset of patients with squamous cell histology.  Engenhart-Cabillic and colleagues (1998) discussed the use of NBT in the management of locally advanced non-resectable primary or recurrent rectal cancer.  They noted that the value of radiation therapy in managing such patients is being appreciated, although up to 40 % of the treated patients have no symptomatic response.  The authors also stated that over 350 patients were entered in studies comparing NBT alone and mixed-beam treatments.  At present, no therapeutic gain for long-lasting survival has been achieved.  However, local control and pain improvement seems to be better with NBT than with photon therapy.  There is insufficient evidence regarding the effectiveness of NBT for rectal and lung cancers.

Strander et al (2003) stated that there is some evidence that adjuvant radiation therapy in combination with conservative surgery improves the local control rate in the treatment of STS of extremities and trunk in patients with negative, marginal or minimal microscopic positive surgical margins.  A local control rate of 90 % has been achieved.  Improvement is obtained with radiation therapy added in the case of intralesional surgery, but the local control rate is somewhat lower.  More studies are needed on this issue.  For STS in other anatomical sites, retroperitoneum, head and neck, breast and uterus, there is only weak indication of a benefit for the local control rate, with the use of adjuvant radiation therapy.  There is still insufficient data to establish that pre-operative radiotherapy is favorable compared to post-operative radiotherapy for local control in patients presenting primarily with large tumors.  One small study has shown a possible survival benefit for pre-operative radiotherapy.  There is fairly good evidence to suggest that the pre-operative setting results in more wound complications.  There is no randomized study comparing external beam radiotherapy and brachytherapy.  The data suggested that external beam radiotherapy and low-dose rate brachytherapy result in comparable local control for high-grade tumors.  Some patients with low-grade STS benefit from external beam radiotherapy in terms of local control.  Brachytherapy with low-dose rate for low-grade tumors seems to be of no benefit, but data are sparse.  The available data are inconclusive concerning the effect of intra-operative high-dose rate radiotherapy for retroperitoneal STS.  Further studies are needed.  Neutron radiotherapy might be beneficial for patients with low-grade and intermediate-grade tumors considered inoperable and for those operated with intralesional margins.  More severe adverse effects for NBT have been reported.

Murray (2004) noted that the commonest STS of the upper extremity are the epithelioid sarcoma, synovial cell sarcoma, and malignant fibrous histiocytoma.  Limb salvage surgery is the treatment of choice for STS to preserve upper extremity function.  Following wide tumor resection, adjuvant therapies such as chemotherapy, external beam radiation therapy, and brachytherapy may lessen local recurrence rates, but their effect on overall survival remains unclear.

A review by Hassen-Khodja and Lance (2003) stated that the efficacy of NBT is well-established only for the treatment of inoperable or unresectable salivary gland tumors, regardless of their degree of malignancy or stage of progression, and for the treatment of large residual tumors after surgical resection.  The authors also examined the data on the effectiveness of for NBT in the treatment of malignant prostate tumors, STS and central nervous system tumors.  However, these data are insufficient to rule on its therapeutic efficacy.

An assessment of the evidence for neutron beam radiotherapy prepared by the Australia and New Zealand Horizon Scanning Network (Purins et al, 2007) found that NBT is a promising technology.  The assessment cautioned, however, that "[t]he studies identified in this prioritising summary were not of high quality and, as such, the conclusions must be taken as preliminary in nature."

In a phase I study, Kankaanranta and colleagues (2011) examined the safety of BNCT in the treatment of malignant gliomas that progress after surgery and conventional external beam radiation therapy.  Adult patients who had histologically confirmed malignant glioma that had progressed after surgery and external beam radiotherapy were included in this study, provided that greater than 6 months had elapsed from the last date of radiation therapy.  The first 10 patients received a fixed dose, 290 mg/kg, of l-boronophenylalanine-fructose (l-BPA-F) as a 2-hour infusion before neutron irradiation, and the remaining patients were treated with escalating doses of l-BPA-F, either 350 mg/kg, 400 mg/kg, or 450 mg/kg, using 3 patients on each dose level.  Adverse effects were assessed using National Cancer Institute Common Toxicity Criteria version 2.0.  A total of 22 patients entered the study.  Twenty subjects had glioblastoma, and 2 patients had anaplastic astrocytoma, and the median cumulative dose of prior external beam radiotherapy was 59.4 Gy.  The maximally tolerated l-BPA-F dose was reached at the 450 mg/kg level, where 4 of 6 patients treated had a grade 3 adverse event.  Patients who were given more than 290 mg/kg of l-BPA-F received a higher estimated average planning target volume dose than those who received 290 mg/kg (median of 36 versus 31 Gy [W, i.e., a weighted dose]; p = 0.018).  The median survival time following BNCT was 7 months.  The authors concluded that BNCT administered with an l-BPA-F dose of up to 400 mg/kg as a 2-hour infusion is feasible in the treatment of malignant gliomas that recur after conventional radiation therapy.