SEARCH

SEARCH BY CITATION

Keywords:

  • radiation;
  • cost;
  • quality-adjusted life years;
  • medulloblastoma;
  • children

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND

Radiation therapy is an important component in the treatment of medulloblastoma; however, in many patients, it is associated with risk of late adverse events. Proton radiation therapy has potential to reduce the risk of adverse events compared with conventional radiation, but it is associated with a higher treatment cost. The objective of the current study was to assess the cost-effectiveness of proton therapy compared with conventional radiation therapy in the treatment of childhood medulloblastoma.

METHODS

The consequences of radiation therapy were evaluated using a Markov simulation model. Children age 5 years with medulloblastoma were followed. The patients were at risk of several types of adverse events, including hearing loss, intelligence quotient (IQ) loss, hypothyroidism, growth hormone deficiency (GHD), osteoporosis, cardiac disease, and secondary malignancies. The patients also were at risk of death and were divided into risk groups for normal death, death due to tumor recurrence, treatment-related cardiac death, treatment-related subsequent tumor death, or treatment-related other death. A review of the literature was conducted to estimate the parameters in the model.

RESULTS

The base-case results showed that proton therapy was associated with €23,600 in cost savings and 0.68 additional quality-adjusted life-years per patient. The analyses showed that reductions in IQ loss and GHD contributed to the greatest part of the cost savings and were the most important parameters for cost-effectiveness.

CONCLUSIONS

The results of the current study indicated that proton radiation therapy can be cost-effective and cost-saving compared with conventional radiation therapy in the treatment of children with medulloblastoma if the appropriate patients are selected for the therapy. However, there have been few long-term follow-up studies, and more much information on the long-term consequences of radiation therapy is needed. Cancer 2005. © 2005 American Cancer Society.

Proton beam radiation for malignant tumors may offer clinical advantages for patients with some types of carcinomas.1, 2 The basis for the advantages is a better dose distribution of the proton radiation. This means that smaller volumes of normal tissues are irradiated, reducing the risk of adverse events, because the adverse events from radiation therapy are both dose-dependent and volume-dependant. An indirect effect of the reduced risk of side effects is the possibility of treating patients with higher radiation doses, which may lead to better tumor control in some patients.

Medulloblastoma in children is a severe form of malignant disease that is associated with substantial morbidity, mortality, and financial burden. Medulloblastoma accounts for approximately 20–25% of all central nervous system tumors in children.3 The 5-year and 10-year survival rates have been estimated at approximately 60% and 40%, respectively.3–6 Both radiation therapy and chemotherapy are used as treatments for medulloblastoma, but they are associated with a risk of late adverse events.3, 4, 6 The severity of treatment-related adverse events is linked to the age of the children, in that young children are more vulnerable and, thus, have greater risks.4 Radiation therapy is an essential component in their treatment, and the antitumor effect is dependant on the radiation dose.7 A higher radiation dose may increase the probability of tumor control in some patients; however, the maximum dose can be limited by dose-related adverse events (e.g., secondary tumors). Children who survive brain malignancies have an elevated risk of several types of new tumors compared with the general population (e.g., in the stomach, colon, breast, and lungs).8 Surviving children with brain tumors also have high risks for a variety of other late adverse events, which are caused at least in part by the radiation therapy. For example, Gurney et al. studied children with brain tumors who survived for ≥ 5 years9 and found that they had a significantly greater risk of hypothyroidism, growth hormone deficiency (GHD), osteoporosis, stroke, and angina-like symptoms. Radiation therapy also may cause significant ototoxicity in some children with medulloblastoma.10

Another important consequence of brain tumors is neuropsychological deficits, especially in young children. There is an increased risk of lower intelligence quotient (IQ) scores as a result of childhood medulloblastoma. Ris et al., for example, showed that children with medulloblastoma/primitive neuroectodermal tumors treated with radiotherapy and chemotherapy had a reduction in intellectual development,11 with a predicted full-scale IQ drop from 96.2 at baseline to 78.8 after 4 years. It is believed that radiation therapy is one source of these neuropsychological problems, but it remains unclear how much of the IQ loss is caused by radiation and how much is due to the primary tumor. The causality between radiation and late occurring events is a problem for many of the events, because large studies with long follow-up are needed to separate the consequences of radiation therapy from the consequences of either the tumor itself or other treatments.

Proton therapy is associated with higher treatment costs compared with conventional radiation, mainly due to the large investments required for building accelerators, beam transport systems, and gantries. Therefore, it is important to assess the magnitude of the clinical benefits from this technology and weight the benefits against the added treatment costs. The benefits from using proton therapy in patients with childhood medulloblastoma remain unclear, although comparative dose-planning studies have indicated a potential for reduced late effects.12–16 Proton radiation most likely will not have a large clinical advantage in all patients with medulloblastoma because conventional radiation using photons and electrons can be given to many patients, so that the risk of adverse event is very low. However, there still will be some patients with medulloblastoma in whom adverse events from conventional radiation can not be avoided. The objective of the current study was to assess the potential benefits of proton therapy in terms of costs, gained life years, and quality of life and to estimate its cost-effectiveness compared with conventional radiation therapy.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The consequences of radiation therapy in children with medulloblastoma were evaluated using a Markov cohort-simulation model.17 Two hypothetical cohorts of children receiving proton therapy or conventional radiation therapy in Sweden were compared. The model simulated the course of events in individual patients from diagnosis until death or until age 100 years. Patients in the model were divided into a number of different health states, each associated with a certain cost and utility. In the model, as time progresses, the patients move between the different health states according to a set of transition probabilities.

A population of children age 5 years who were diagnosed with medulloblastoma was analyzed in the base case. However, the typical age of a patient with medulloblastoma may be somewhat older, and analyses of patients ages 6 years and 8 years also were performed. The life time was divided into 1-year cycles. In each cycle, the patients were at risk of different adverse events or death. Seven types of adverse events were included in the model: hearing loss, IQ loss, hypothyroidism, GHD, osteoporosis, cardiac disease, and secondary malignancies. Deaths were divided into normal deaths, deaths due to tumor recurrence, treatment-related cardiac deaths, treatment-related subsequent tumor deaths, and treatment-related other deaths. Three outcomes measures were used in the model: cost, life years, and quality-adjusted life-years (QALYs). A review of the literature was conducted to estimate parameters in the model. One or more relevant studies were selected as basis for each parameter estimate, and the selection was based on factors like date of study, patient population, treatments used, and perceived quality of the study. Extensive sensitivity analyses were performed to assess the uncertainties and the impact of the individual assumptions on the results.

All costs are presented in the Euro 2002 value (€1 = 9.2 Swedish crowns), and costs and effects were discounted with 3% annually. The analyses were conducted from a societal perspective.

Estimation of Mortality Risks

The mortality risk during the first 10 years after the primary tumor was based on several studies that have estimated the 5-year and 10-year survival rates in children with medulloblastoma at between 40% and 80%.3–5, 13 We assumed a yearly mortality risk of 8% during the first 10 years in the base case. After the first 10 years, general population mortality rates were applied with the addition of long-term increases in mortality due to tumor recurrence and adverse effects of medulloblastoma treatment. Between 10 years and 20 years after diagnosis, a yearly risk of 0.6% for death due to recurrence of the medulloblastoma was assumed based on findings in a study by Mertens et al.,18 who analyzed the long-term risk of death in various survivors of childhood and adolescent malignancies.

The additional radiation-induced mortality was estimated from the same study by Mertens et al.18 The yearly absolute risks of death due to secondary malignancies, cardiac disease, and other deaths were estimated in the base case at 0.11%, 0.040%, and 0.016% per year, respectively. The patients in the study by Mertens et al. included children ages 5–20 years at diagnosis, and the study followed them for ≈ 10 years on average. In the current study, we simulated a cohort of children age 5 years in the base case, and it was assumed that the radiation-induced deaths appeared 10–20 years after primary diagnosis and treatment. It should be noted, however, that the risks in the study by Mertens et al. were the average risks of all patients, whether they received radiation therapy or not, which means that their assumptions probably underestimated the risk. However, the children in the study by Mertens et al. had different types of malignancies, not only medulloblastoma, and children who have medulloblastoma have a lower risk of developing secondary malignancies compared with children who have other types of primary malignancies. This means that the estimated risk also could be overestimated. Different mortality assumptions were tested in sensitivity analyses to assess the impact of this uncertainty on the current results.

Beyond 20 years after the tumor diagnosis, we applied normal age-specific mortality rates for the Swedish population. However, this may be a conservative estimate, because experience from studies in other tumors indicates that the risk of secondary malignancies continues beyond 20 years.19–21

Estimation of Adverse Event Risks

The risk of hearing loss was estimated at 13% in the base-case analysis, which was based on the incidence in patients receiving intensity-modulated radiation therapy (IMRT) in a study by Huang et al.12 However, assumptions with lower risks of hearing loss also were tested, because the risk of hearing loss is related highly to the radiation dose to the inner ear and, thus, to the beam arrangement; thus, the risk of hearing loss may vary substantially between patients. The average IQ loss was estimated at 17 points based on reported IQ losses in previous studies.11, 22–24 However, as discussed above, the extent to which IQ loss can be attributed to radiation or whether the risk of IQ loss is lower when using proton radiation remains unknown. In the base-case analysis, we assumed that 25% of the IQ loss was related to radiation therapy, but the assumption was varied in sensitivity analyses.

The risk of hypothyroidism was estimated at 33% based on the incidence in a previous study that followed patients for 6 years after their primary diagnosis.25 It was assumed that all episodes of hypothyroidism occurred 4 years after the primary diagnosis.26 The risk of GHD and osteoporosis were estimated at 18.7% and 2.4%, respectively, based on the incidences reported in a study by Gurney et al.9 The risk of osteoporosis was applied at age 20 years.

It is plausible that radiation, along with fatal secondary malignancies, also is associated with nonfatal secondary malignancies. Miralbell et al. estimated that the yearly risk of different radiation-induced malignancies was 0.43% for conventional X-rays and 0.05% for protons.15 We estimated that the risk of nonfatal malignancies was 0.32%, which is the difference between the overall risk of malignancy from the study by Miralbell et al. and the risk of fatal malignancy from the study by Mertens et al. The risk of nonfatal malignancies was applied between 10 years and 20 years after diagnosis (i.e., the same period as the risk of fatal secondary malignancies). The risk of nonfatal secondary malignancies, like the risk of fatal secondary malignancies, is somewhat uncertain; therefore, different assumptions were tested in the sensitivity analyses.

Estimation of the Cost of Adverse Events

The calculation of the costs of adverse events in children with medulloblastoma is complex. It is plausible that many patients will have several adverse events, but is not certain that the costs of the individual adverse events will be additive in patients who have multiple events. The most obvious case for this is the cost for lost production, in which more than one of the adverse events may lead to lost productivity. The cost for lost productivity, therefore, only was assumed for IQ loss in the model to avoid double counting of this cost. Other costs for the different adverse events were assumed to be additive. Mortality was not assumed to incur any costs, only reductions in life years and QALYs.

The yearly cost of severe hearing loss was estimated at €16,957 based on findings in a previous study by Mohr et al.27 However, the patients in the study by Mohr et al. had rather severe hearing loss. Therefore, it was assumed that only 25% of the patients with radiation-induced hearing loss would incur the cost estimated in the study by Mohr et al., and it was assumed that the remaining 75% of patients would incur a cost of €1087 based on the cost of hearing aid estimated in a previous Swedish study.28

Costs for IQ loss were calculated using a methodology applied in previous studies, for example. by Schwartz29 and Salkever.30 The method is based on long-term follow-up studies of toxic exposures and their effects on schooling results, work performance, etc. Schwartz estimated the costs for IQ reduction to an earnings loss of 1.763% per IQ point decrease. That calculation was based on two percentage changes: decreased earnings for individuals who work and a probability of not working at all. Based on this estimate, we calculated the yearly cost of IQ loss to €576 per IQ point (1.763% of average labor cost in Sweden, estimated at €32,826).29 The cost was applied only from age 20 years until age 65 years, because it was based on reduced earnings. However, it is plausible that children with IQ loss will incur substantial costs during childhood and adolescence due to increased needs for personal assistants, special school training, etc.; however, no data on these increased needs were available. In the sensitivity analyses, we applied the same costs for IQ loss from age 7 years (starting school) and, thus, assumed that the costs before age 20 years were due to costs for personal assistants and special school training instead of lost production.

The yearly cost of hypothyroidism was estimated at €114.0 based on a cost of €28.3 for thyroxin substitution,31 a cost of €6.6 for a yearly test for thyroid-stimulating hormone,32 and a cost of €79.1 for a yearly physician visit.33 There also may be a risk of long-term need for thyroid replacement, but this was not included in the current analyses, because no data regarding this were available, and the costs for this also are expected to be low compared with other costs in the analyses.

The average yearly cost of GHD was estimated at €13,478 up to the age of 19 years and €1348 for patients age > 19 years.31, 34 The cost was based on the assumption that all children need growth hormone substitute up to age 19 years and that 10% of patients will need life-long substitution.34

The yearly cost for osteoporosis was estimated at €363.0 based on results from a previously developed simulation model of osteoporosis.35–38 It was assumed that osteoporosis would become symptomatic at age 20 years and would continue until death.

The cost of nonfatal secondary malignancies was estimated at €19,565 based on the average total health care costs for cancer treatments in Sweden.39 No specific costs were available for malignancies occurring in young individuals.

Estimation of Utility Loss of Adverse Events

The baseline utility for healthy individuals was based on a Swedish study.40 It was assumed that patients who had different types of side effects would have a percentage reduction in utility.

The utility reduction for patients who had severe hearing loss was estimated at 18% based on a study by Bichey et al.41 However, it was assumed that only 25% of patients would have severe radiation-induced hearing loss leading to the assumed utility reduction (i.e., the same patients who also incurred the higher costs due to their hearing loss). No utility reduction for patients with IQ loss was assumed in the base case, although it is plausible that both children with IQ loss and their relatives may have reduced utility. In the sensitivity analyses, we tested assumptions in which the IQ loss was associated with a life-long 5% reduction in utility. The utility reduction for patients with hypothyroidism was estimated at 10% based on a study by Bona et al.42 The utility reduction, however, only was applied during the first year after hypothyroidism was diagnosed, because it was assumed that patients would return to normal health after they received treatment.

Several studies have shown that GHD is associated with reduced quality of life, but no studies have quantified the utility reduction. The utility reduction for patients with GHD in the current study was estimated at 20% in the base case based on the quality-of-life reductions reported in previous studies.43–49 The utility reduction for patients with osteoporosis was estimated at 2% based on results from a previously developed simulation model of osteoporosis.35–38 Patients with osteoporosis had a reduction in utility from age 20 years until death (i.e., the same period during which the costs of osteoporosis were incurred).

Risk Reduction with Proton Therapy

Mertens et al. calculated that the relative risks for patients who were receiving radiation therapy compared with the risks for patients who were not receiving radiation were approximately 2.5 for subsequent malignancies and 2.2 for treatment-related cardiac and other deaths.18 However, how much of that radiation-induced mortality could be avoided if proton therapy was used remains unknown. A recent dose-planning study compared the percentage of prescribed dose received by different organs in patients who were receiving standard IMRT or proton radiation.16 The relative risk that 5% of the radiation would reach the heart was estimated at approximately 0.05, whereas the relative risk that radiation would reach other organs varied from 0.02 to 1.5. In the base case, we assumed relative mortality risks of 0.05 for treatment-related cardiac death and 0.6 for other deaths and secondary malignancies in patients who were receiving proton radiation compared with conventional radiation. The relative risks of hearing loss, hypothyroidism, GHD, IQ loss, and osteoporosis all were estimated at 0.12 based on a study by Miralbell et al.14, 15

Cost of Radiation

The cost of radiation therapy depends heavily on the investment costs for the radiation equipment. A study by Goitein and Jermann in Switzerland estimated the costs of proton and X-ray radiation therapy.50 Those authors estimated the total operation cost per patient, including fixed business costs, at €25,600 and €10,600 for proton radiation and X-ray radiation, respectively. If business costs were excluded, then the costs per patient were €14,700 and €7600 for proton radiation and X-ray radiation, respectively (i.e., a difference of €7100 or a 93% increase in costs). There may be less expensive conventional radiation facilities; however, in the current study, we only compared proton radiation with a modern IMRT facility used in developed countries.

A Swedish study estimated the operating costs of radiation, excluding capital costs, at €123.0 per fraction.51 That estimate was used in the base case as the operating cost for conventional radiation. The increased operation cost for proton therapy was estimated at 93% based on the study by Goitein and Jermann.50

The investment costs for proton and conventional radiation therapy facilities have been calculated at approximately €62.5 million and €16.8 million, respectively.50 To distribute the investment cost to each patient, the one-time investment has to be translated to a yearly cost during the life time of the proton facility. This was done by estimating an annuity cost. The capital cost for a proton therapy facility was calculated at €4239 per patient based on a facility lifetime of 30 years, a 5% interest rate, and the assumption that a facility would treat 960 patients per year. The corresponding cost for a conventional radiation facility was €1141 per patient. The calculations were based on the assumption that the capital costs would be distributed evenly among all patients treated at the facility. We assumed that the patients would be treated with 25 fractions, which means that the total radiation costs were estimated at €4239 for conventional radiation therapy and at €10,218 for proton radiation therapy.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The results are presented for a model cohort of children age 5 years with medulloblastoma who were treated with either conventional radiation therapy or proton radiation therapy in addition to surgery and chemotherapy, which were identical in the 2 groups. The model predicted the that average undiscounted life expectancy of children who were treated with proton radiation therapy and with conventional radiation therapy was 31.1 years and 30.2 years, respectively. Table 1 shows the average results per patient. The base-case results show that proton therapy dominated conventional radiation (i.e., had both lower costs and better effects).

Table 1. Cost and Clinical Outcome per Patient for the Base-Case Assumptions
VariableProton radiationConventional radiationDifference
  1. LYG: life-years gained; QALY: quality-adjusted life-years.

Radiation cost (€)10217.94239.15978.8
Cost from adverse events (€)4231.833857.1−29625.3
Total cost (€)14449.738096.2−23646.5
LYG13.86613.6000.266
QALY12.77812.0950.683

GHD and hypothyroidism were the most common radiation-induced events, and osteoporosis was the least common (Table 2). However, it was assumed that osteoporosis was given after age 20 years, when a large proportion of the children already had died.

Table 2. Radiation-Induced Events per 100 Patients
VariableHearing lossHypothyroidismOsteoporosisGHDNonfatal secondary malignanciesFatal events
  1. GHD: growth hormone deficiency.

Conventional radiation11.916.30.417.11.21.91
Proton radiation1.42.70.12.00.70.38
Difference10.513.60.315.10.51.53

Table 3 shows the differences in costs and utilities between the two radiation therapies divided into the different sources. The results show that IQ loss, hearing loss, and GHD contributed to the greatest part of the cost difference and that GHD, fatal adverse events, and hearing loss contributed to the greatest part of the utility difference.

Table 3. Distribution of Cost and Utility Differences: Proton Radiation versus Conventional Radiation
Cost sourceCost difference (€)Utility difference
  1. IQ: intelligence quotient; GHD: growth hormone deficiency.

Total difference−23646.50.683
Radiation5978.8
IQ loss−12206.9
Hearing loss−2735.50.057
GHD−14263.20.367
Hypothyroidism−202.00.009
Osteoporosis−18.30.001
Fatal and nonfatal secondary malignancies95.60.021
Other fatal adverse events0.230

Sensitivity analyses were performed in which the parameters in the model were varied. Those analyses showed that that the results were stable (Table 4). The adverse events that contributed to the greatest part of the costs (i.e., IQ reduction and GHD) also were the most important parameters for cost-effectiveness. A life-long reduction in utility for IQ loss led to a large increase in the number of gained QALYs.

Table 4. Results from Sensitivity Analyses: Per Patient
Sensitivity scenarioΔ Cost (€)aΔ LYGbΔ QALYb
  • LYG: life-years gained; QALY: quality-adjusted life-years; IQ: intelligence quotient; GHD: growth hormone deficiency.

  • a

    The cost of proton radiation minus the cost of conventional radiation.

  • b

    Outcome after proton radiation minus outcome after conventional radiation.

Base case results−23646.50.2660.683
Start age (base case 5 yrs)   
 6 yrs−23416.50.2650.679
 8 yrs−22932.20.2610.672
Radiation-induced deaths and carcinomas applied for 20 yrs (base case, 10 yrs)−23203.51.6452.377
Low proton radiation cost (80% of base case)−25690.10.2660.683
High proton radiation cost (120% of base case)−21602.90.2660.683
Relative risk of subsequent carcinomas (base case, 0.6)   
 0.8−23594.60.2550.672
 1.0−23594.00.2410.660
Relative risk of nonfatal adverse events (base case, 0.12)   
 0.3−17446.20.2660.590
 0.5−10568.00.2660.487
 0.7−3701.80.2660.384
Relative risk of cardiac death (base case, 0.05)   
 0.0−23643.30.2790.694
 0.2−23658.60.2280.649
 0.4−23673.90.1780.603
Risk of hearing loss (base case, 13%)   
 0%−20911.60.2660.626
 7%−22384.60.2660.657
IQ loss from radiation (base case, 4.25 points)   
 0 points−11439.70.2660.683
 3 points−20056.70.2660.683
 10 points−40163.00.2660.683
5% Utility reduction from IQ loss  1.288
Cost of IQ loss from age 7 yrs (starting school)−34493.70.2660.683
GHD: 50% of base case risk−16515.50.2660.500
Hearing loss: 50% of base case risk−22279.4 0.655
Cost of hearing loss (base case, 100%)   
 0% of base case−20858.60.2660.683
 50% of base case−22252.80.2660.683
Cost of GHD (base case, 100%)   
 0% of base case−9307.50.2660.683
 50% of base case−16477.30.2660.683
Cost of IQ loss (base case, 100%)   
 50% of base case−17543.40.2660.683

Model Validation

The model was based on several uncertain assumptions; therefore, it was important to validate the outcome. The quality and validity of the analyses were assessed by comparing the outcome with previous studies. Extensive sensitivity analyses also were performed to assess the uncertainty of the results. The model predicted 5-year and 10-year survival rates of 66% and 43%, respectively. These figures are in line with findings in epidemiological studies.3–5 Furthermore, the model predicted that 1.9% of the patients treated with conventional radiation therapy would die from radiation-induced events. A study by Mertens et al.18 found that 1.9% of children had died from treatment-related consequences, but that was after a mean follow-up of ≈ 10 years. However, it is not known how many of those deaths were attributable to radiation therapy.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

In this report, we present an assessment of the potential economic benefits and cost-effectiveness of proton therapy compared with conventional radiation in the treatment of childhood medulloblastoma. It is important to explore the cost-effectiveness of proton therapy as a basis for decisions regarding investments in this new technology. However, there are very few studies available assessing the consequences of radiation-induced adverse events, limiting the possibly of making accurate estimates. The available studies also differ substantially in patient characteristics, types of radiation, doses, etc. This makes it difficult to base all assumptions on a certain type of patient who receives a certain type of radiation treatment. It is recognized that the risks of most adverse events included in the analyses are highly dependent on the beam arrangement, the radiation dose, etc. and that proton radiation in some patients may not offer a clinically relevant advantage compared with conventional radiation. However, in some patients, proton radiation will offer advantages, and the assumed risks and therefore the results are only relevant in these high-risk patients.

The current analyses indicated that proton therapy had both lower total cost and better effect than conventional radiation. In the base-case analysis, proton therapy was associated with €23,600 cost savings, 0.27 additional life years, and 0.68 additional QALYs per patient compared with conventional radiation. Thus, the additional costs for radiation therapy were offset by reduced costs for adverse events. If only patients with medulloblastoma were to be treated at the facility, then approximately 110 patients per year would have to be treated to make the proton facility cost neutral compared with conventional radiation. However, proton therapy still may be considered cost-effective even if it is not cost-saving, because it also leads to improved health for the patients. Although the analyses were performed in a Swedish setting, it is reasonable to generalize the results to other countries, because most of the model parameters were based on international studies.

The results showed that the cost of GHD and IQ loss were two of the most important variables for cost-effectiveness. The costs of these adverse events mainly include costs for pharmacologic treatment of GHD and for lost productivity due to reduced IQ. We assumed that only 25% of the observed reduction in IQ was attributable to radiation therapy. This assumption was uncertain, however, because no data on this appear to be available. No quality-of-life reduction was assumed for IQ loss in the base case, an assumption that most likely is too conservative. Indirect cost for lost production was assumed only for IQ loss, although studies have indicated that patients with GHD also have productivity losses.52 Although the risk reductions of GHD and IQ loss from using proton therapy still are uncertain in many patients, the results indicate that proton radiation may provide valuable clinical benefits in a cost-effective manner among groups of children in whom proton therapy is expected to reduce the risk of these events. Identifying risk factors for these adverse events, therefore, can provide valuable information for the selection of patients who would benefit the most from proton therapy.

The current results indicate that proton radiation can be cost-effective and cost-saving compared with conventional radiation in the treatment of patients with medulloblastoma. However, the advantage of proton radiation is dependent on the possibility of selecting patients who may have a reduced risk of adverse events with proton radiation compared with conventional radiation. If it would be possible to select patients with an even higher risk of adverse events than assumed in this study, then the cost-effectiveness of proton therapy would improve even more. The data included in the current analysis are uncertain, and we generally applied conservative assumptions. This means that the economic benefit may be even greater than estimated in this study.

To our knowledge there have been few long-term follow-up studies of surviving children with medulloblastoma; thus, the assumptions about long-term consequences of radiation are uncertain. The lack of data makes it important to determine whether the potential benefits and the assumptions used in the current analyses are valid. Although the model and the assumed consequences of radiation were based on limited data, the current results provide valuable information regarding the potential benefits of proton radiation in patients with medulloblastoma. The GHD and IQ reduction were two particularly important consequences that could be distinguished. It is expected that these adverse events will incur large costs to the health care system and to society; therefore, they are interesting subjects for further research.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Suit H, Goldberg S, Niemierko A, et al. Proton beams to replace photon beams in radical dose treatments. Acta Oncol. 2003; 42: 800808.
  • 2
    Glimelius B, Isacsson U, Blomquist E, Grusell E, Jung B, Montelius A. Potential gains using high-energy protons for therapy of malignant tumours. Acta Oncol. 1999; 38: 137145.
  • 3
    Freeman CR, Taylor RE, Kortmann RD, Carrie C. Radiotherapy for medulloblastoma in children: a perspective on current international clinical research efforts. Med Pediatr Oncol. 2002; 39: 99108.
  • 4
    Kiltie AE, Lashford LS, Gattamaneni HR. Survival and late effects in medulloblastoma patients treated with craniospinal irradiation under three years old. Med Pediatr Oncol. 1997; 28: 348354.
  • 5
    Sun LM, Yeh SA, Wang CJ, et al. Postoperative radiation therapy for medulloblastoma—high recurrence rate in the subfrontal region. J Neurooncol. 2002; 58: 7785.
  • 6
    Gilbertson RJ. Medulloblastoma: signalling a change in treatment. Lancet Oncol. 2004; 5: 209218.
  • 7
    Thomas PR, Deutsch M, Kepner JL, et al. Low-stage medulloblastoma: final analysis of trial comparing standard-dose with reduced-dose neuraxis irradiation. J Clin Oncol. 2000; 18: 30043011.
  • 8
    Garwicz S, Anderson H, Olsen JH, et al. Second malignant neoplasms after cancer in childhood and adolescence: a population-based case-control study in the five Nordic countries. The Nordic Society for Pediatric Hematology and Oncology. The Association of the Nordic Cancer Registries. Int J Cancer. 2000; 88: 672678.
  • 9
    Gurney JG, Kadan-Lottick NS, Packer RJ, et al. Endocrine and cardiovascular late effects among adult survivors of childhood brain tumors: Childhood Cancer Survivor Study. Cancer. 2003; 97: 663673.
  • 10
    Grau C, Overgaard J. Postirradiation sensorineural hearing loss: a common but ignored late radiation complication. Int J Radiat Oncol Biol Phys. 1996; 36: 515517.
  • 11
    Ris MD, Packer R, Goldwein J, Jones-Wallace D, Boyett JM. Intellectual outcome after reduced-dose radiation therapy plus adjuvant chemotherapy for medulloblastoma: a Children's Cancer Group study. J Clin Oncol. 2001; 19: 34703476.
  • 12
    Huang E, Teh BS, Strother DR, et al. Intensity-modulated radiation therapy for pediatric medulloblastoma: early report on the reduction of ototoxicity. Int J Radiat Oncol Biol Phys. 2002; 52: 599605.
  • 13
    Archambeau JO, Slater JD, Slater JM, Tangeman R. Role for proton beam irradiation in treatment of pediatric CNS malignancies. Int J Radiat Oncol Biol Phys. 1992; 22: 287294.
  • 14
    Miralbell R, Lomax A, Russo M. Potential role of proton therapy in the treatment of pediatric medulloblastoma/primitive neuro-ectodermal tumors: spinal theca irradiation. Int J Radiat Oncol Biol Phys. 1997; 38: 805811.
  • 15
    Miralbell R, Lomax A, Cella L, Schneider U. Potential reduction of the incidence of radiation-induced second cancers by using proton beams in the treatment of pediatric tumors. Int J Radiat Oncol Biol Phys. 2002; 54: 824829.
  • 16
    Clair W, Adams J, Bues M, et al. Advantage of protons compared to conventional X-ray or IMRT in the treatment of a pediatric patient with medulloblastoma. Int J Radiat Oncol Biol Phys. 2004; 48: 727734.
  • 17
    Sonnenberg F, Beck J. Markov models in medical decision making: a practical guide. Med Decis Making. 1993; 13: 322338.
  • 18
    Mertens AC, Yasui Y, Neglia JP, et al. Late mortality experience in five-year survivors of childhood and adolescent cancer: the Childhood Cancer Survivor Study. J Clin Oncol. 2001; 19: 31633172.
  • 19
    Travis LB, Gospodarowicz M, Curtis RE, et al. Lung cancer following chemotherapy and radiotherapy for Hodgkin's disease. J Natl Cancer Inst. 2002; 94: 182192.
  • 20
    Travis LB. Therapy-associated solid tumors. Acta Oncol. 2002; 41: 323333.
  • 21
    Fossa SD, Aass N, Harvei S, Tretli S. Increased mortality rates in young and middle-aged patients with malignant germ cell tumours. Br J Cancer. 2004; 90: 607612.
  • 22
    Palmer SL, Goloubeva O, Reddick WE, et al. Patterns of intellectual development among survivors of pediatric medulloblastoma: a longitudinal analysis. J Clin Oncol. 2001; 19: 23022308.
  • 23
    Walter AW, Mulhern RK, Gajjar A, et al. Survival and neurodevelopmental outcome of young children with medulloblastoma at St. Jude Children's Research Hospital. J Clin Oncol. 1999; 17: 37203728.
  • 24
    Copeland DR, deMoor C, Moore BD III, Ater JL. Neurocognitive development of children after a cerebellar tumor in infancy: a longitudinal study. J Clin Oncol. 1999; 17: 3476386.
  • 25
    Ricardi U, Corrias A, Einaudi S, et al. Thyroid dysfunction as a late effect in childhood medulloblastoma: a comparison of hyperfractionated versus conventionally fractionated craniospinal radiotherapy. Int J Radiat Oncol Biol Phys. 2001; 50: 12871294.
  • 26
    Paulino AC. Hypothyroidism in children with medulloblastoma: a comparison of 3600 and 2340 cGy craniospinal radiotherapy. Int J Radiat Oncol Biol Phys. 2002; 53: 543547.
  • 27
    Mohr PE, Feldman JJ, Dunbar JL. The societal costs of severe to profound hearing loss in the United States. Policy Anal Brief H Ser. 2000; 2: 14.
  • 28
    Brorssan B. Hörapparat för vuxna—nytta och kostnader, report 164. Stockholm: The Swedish Council on Technology Assessment in Health Care, 2003.
  • 29
    Schwartz J. Societal benefits of reducing lead exposure. Environ Res. 1994; 66: 105124.
  • 30
    Salkever DS. Updated estimates of earnings benefits from reduced exposure of children to environmental lead. Environ Res. 1995; 70: 16.
  • 31
    LINFO. FASS Läkemedel i Sverige [Swedish drug prices]. Oslo: Läkemedelsinformation AB, 2001.
  • 32
    [No authors]. Price list. Labmedicin, Huddinge Hospital. Stockholm: 2003.
  • 33
    [No authors]. Kostnad per intagen patient. Cooperation of county councils. Stockholm, 2003.
  • 34
    Läkemedelsboken. Apoteket AB, 2001.
  • 35
    Johnell O, Kanis JA, Oden A, et al. Fracture risk following an osteoporotic fracture. Osteoporos Int. 2004; 15: 175179.
  • 36
    Johnell O, Kanis JA, Oden A, et al. Mortality after osteoporotic fractures. Osteoporos Int. 2004; 15: 3842.
  • 37
    Kanis JA, Johnell O, Oden A, et al. Long-term risk of osteoporotic fracture in Malmö. Osteoporos Int. 2000; 11: 669674.
  • 38
    Zethraeus N. Costs and quality of life associated with osteoporosis related fractures—based on a Swedish Pilot Study. Stockholm: Centre for Health Economics, Stockholm School of Economics, 2002.
  • 39
    Glimelius B, Bergh J, Brandt L, et al. The Swedish Council on Technology Assessment in Health Care (SBU) systematic overview of chemotherapy effects in some major tumour types—summary and conclusions. Acta Oncol. 2001; 40(2–3): 135154.
  • 40
    Lundberg L. Health-related quality of life in Sweden [dissertation]. Uppsala: Faculty of Pharmacy, Uppsala University, 1999.
  • 41
    Bichey BG, Hoversland JM, Wynne MK, Miyamoto RT. Changes in quality of life and the cost-utility associated with cochlear implantation in patients with large vestibular aqueduct syndrome. Otol Neurotol. 2002; 23: 323327.
  • 42
    Bona M, Santini F, Rivolta G, Grossi E, Grilli R. Cost effectiveness of screening for subclinical hypothyroidism in the elderly. A decision-analytical model. Pharmacoeconomics. 1998; 14: 209216.
  • 43
    Burman P, Broman JE, Hetta J, et al. Quality of life in adults with growth hormone (GH) deficiency: response to treatment with recombinant human GH in a placebo-controlled 21-month trial. J Clin Endocrinol Metab. 1995; 80: 35853590.
  • 44
    Lagrou K, Xhrouet-Heinrichs D, Massa G, et al. Quality of life and retrospective perception of the effect of growth hormone treatment in adult patients with childhood growth hormone deficiency. J Pediatr Endocrinol Metab. 2001; 14( Suppl 5): 12491260; discussion, 1261–1262.
  • 45
    Li Voon Chong JS, Groves T, Foy P, Wallymahmed ME, MacFarlane IA. Elderly people with hypothalamic-pituitary disease and untreated GH deficiency: clinical outcome, body composition, lipid profiles and quality of life after 2 years compared to controls. Clin Endocrinol (Oxf). 2002; 56: 175181.
  • 46
    Rosen T, Wiren L, Wilhelmsen L, Wiklund I, Bengtsson BA. Decreased psychological well-being in adult patients with growth hormone deficiency. Clin Endocrinol (Oxf). 1994; 40: 111116.
  • 47
    Wallymahmed ME, Foy P, MacFarlane IA. The quality of life of adults with growth hormone deficiency: comparison with diabetic patients and control subjects. Clin Endocrinol (Oxf). 1999; 51: 333338.
  • 48
    Wallymahmed ME, Foy P, Shaw D, Hutcheon R, Edwards RH, MacFarlane IA. Quality of life, body composition and muscle strength in adult growth hormone deficiency: the influence of growth hormone replacement therapy for up to 3 years. Clin Endocrinol (Oxf). 1997; 47: 439446.
  • 49
    Wiren L, Whalley D, McKenna S, Wilhelmsen L. Application of a disease-specific, quality-of-life measure (QoL-AGHDA) in growth hormone-deficient adults and a random population sample in Sweden: validation of the measure by rasch analysis. Clin Endocrinol (Oxf). 2000; 52: 143152.
  • 50
    Goitein M, Jermann M. The relative costs of proton and X-ray radiation therapy. Clin Oncol (R Coll Radiol). 2003; 15: S37S50.
  • 51
    Norlund A. Cost of radiotherapy. Acta Oncol. 2003; 42: 411415.
  • 52
    Hakkaart-van Roijen L, Beckers A, Stevenaert A, Rutten FF. The burden of illness of hypopituitary adults with growth hormone deficiency. Pharmacoeconomics. 1998; 14: 395403.