An insight into the pharmacoeconomics of carbon ion radiotherapy: a systematic review

Abstract

Purpose

Malignant diseases are among the most pressing public health challenges, exerting significant pressure on healthcare resources. Innovative cancer treatments like carbon ion radiation therapy (CIRT) by virtue of its advantages in physical properties, biological effectiveness, and dose distribution compared to photon and proton therapies stands out. This study aims to assess the cost-effectiveness of CIRT in cancer treatment by systematically reviewing existing economic evaluations.

Material and Methods

The protocol is registered with PROSPERO and employed PRISMA Guidelines. Systematic searches across PubMed, Embase, and Web of Science, between January 2000 and March 2025 were conducted and CIRT pharmacoeconomic articles were included. Screening of the search results, critical quality assessment using Drummond and CHEERS checklist and data extraction were performed.

Results

Out of the 10 studies included in this systematic review, seven analysed cost effectiveness and three analysed cost. Total cost for CIRT ranged from €16,937 (approx. USD 19,595) to €43,600 (approx. USD 50,443) and JPY 3,140,000 (approx. USD 20,450) to JPY 4,974,278 (approx. USD 32,396) in Germany and Japan, respectively. Seven studies assessed and reported increased effectiveness of CIRT. Reduction in CIRT technical fees, reirradiation with intensity-modulated radiation therapy, increased survival rate with CIRT, local control rate by 60% with CIRT were found to reduce incremental cost effectiveness ratio.

Conclusion

Nine studies show CIRT is cost effective in non–small cell lung cancer, adenoid cystic carcinoma, head and neck cancer, skull-based chordoma, recurrent rectal cancer, and hepatocellular carcinoma. The cost-effectiveness of CIRT is likely to improve more in real-world clinical practice due to enhanced efficacy, reduced toxicity, reduced fractionation, and cost reductions.

Introduction

Cancer imposes the greatest burden in terms of cause-specific Disability-Adjusted Life Years, surpassing all other diseases in clinical, social, and economic impact. The lifetime risk of developing cancer before the age of 75 is estimated at 20.2%. In 2018, approximately 18 million new cancer cases were diagnosed worldwide. Regarding mortality, cancer ranks as the second leading cause of death globally, with 8.97 million deaths, following ischemic heart disease; however, it is projected to become the leading cause by 2060, with an estimated 18.63 million deaths. Current epidemiological data, along with the projected increase in cancer incidence, prevalence, and mortality over the next four decades, indicate that the impact of malignant diseases is already significant and is expected to remain at epidemic levels for the foreseeable future. Cancer can be considered the leading public health challenge, consuming a substantial amount of economic resources [1]. In the United States, the medical expenses related to cancer are significant and are expected to rise sharply by 2030, driven primarily by demographic changes. National cancer care costs reached $183 billion in 2015 and are projected to grow by 34%, reaching $246 billion by 2030, highlighting the increasing financial burden of cancer care for survivors [2].

The cost of radiation therapy varies significantly both between countries and among different hospitals within the same country [3,4]. Greater price transparency can encourage decisions focused on value and effectiveness, while also helping to lessen the financial impact of radiation therapy. Emerging cancer treatments have become a key focus in health policy discussions across many high-income nations. Strong evidence supports that the latest carbon ion radiation therapy (CIRT) offers significant physical, biological, and dosimetric benefits compared to photon and proton therapies, with current clinical data demonstrating encouraging outcomes in various cancer types. CIRT possesses distinctive physical and biological characteristics that enable a gradual dose to increase with a sharp gradient. This allows for the delivery of high dose localized treatment while minimizing damage to surrounding healthy tissues. Additionally, tumours located close to critical organs can be treated more effectively using higher radiation doses [5]. Carbon ions offer several potential benefits over protons, including improved physical dose distribution due to reduced lateral scattering. They also exhibit greater relative biological effectiveness and a lower oxygen enhancement ratio—characteristics that are particularly advantageous for targeting hypoxic, radio-resistant tumours [6]. By 2025, 17 centers worldwide are offering CIRT [7]. Many are reporting encouraging safety and effectiveness outcomes from their initial patient groups. CIRT has been investigated across a wide range of cancers, including skin cancers, intracranial tumours, gynaecological malignancies, head and neck cancers, gastrointestinal tumours, prostate and other genitourinary cancers, sarcomas, primary and metastatic lung cancers, breast cancer and paediatric cancers [8]. CIRT is more expensive than other treatment options because it requires higher costs for the construction and operation of the accelerator system [9]. Although CIRT is expensive, it has the excellent therapeutic effects described above.

From a patient’s perspective, addressing the economic strain and financial hardships associated with cancer treatment is an urgent priority. For healthcare providers, it is essential to establish tools for assessing cost-effectiveness and to compare various diagnostic and treatment options to optimize care focused on the patient. At the public health level, prioritizing the proper allocation of resources and evaluating the effectiveness of various cancer care strategies are essential to maintain sustainable healthcare systems. Cutting-edge studies on comparative and cost-effectiveness analysis can help direct cancer treatment choices and shape health policy [10]. This systematic review offers insight into all the pharmacoeconomic aspects of CIRT and provides solutions for the above needs of patients, healthcare providers and health policymakers at this nick of the moment to determine whether it is appropriate to have more such centers around the world so that cost of cancer care can be reduced which might help society.

This study evaluates the efficiency of CIRT in cancer by systematically synthesizing evidence from economic evaluations. It compares CIRT with other radiotherapy options in terms of cost-effectiveness, total costs and incremental costs.

Materials and Methods

The protocol was registered with PROSPERO with the number CRD42023395821 (accessible from: https://www.crd.york.ac.uk/PROSPERO/view/CRD42023395821). This review was carried out following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta Analyses) framework to maintain transparency and methodological accuracy. Each phase of the review process identification, screening, eligibility evaluation, and final inclusion was systematically documented using the PRISMA flowchart. We conducted systematic searches across three online bibliographic databases, PubMed, Embase, and Web of Science, between January 2000 and March 2025, adopting a specific search strategy. Details of the search strategy used are provided in the supplementary material. The following research articles were included: Original research articles of cost-effectiveness, cost-minimization, cost-utility, and cost-benefit evaluations that incorporate CIRT and any radiation therapy treatments, irrespective of the comparator/s: for treating cancer; cost comparison evaluations that incorporate CIRT and any radiation therapy treatments, irrespective of the comparator/s: for treating cancer; Studies from any nation published over the aforementioned time frame were considered along with cost evaluations that solely take CIRT while treating cancer. The following research articles were excluded: studies that are distinct from cost comparison or economic evaluations (budget impact evaluations, disease burden, and illness costs); no primary research articles (comments, systematic reviews, or editorials); research evaluating screening or diagnostics as well as other research studies (genetics, animals, etc.) that are not categorized under the previously listed categories.

The first step involved two authors independently reviewing the titles and abstracts of the articles. Following that, the full texts of the qualified studies were screened for relevance. The coding for the inclusion criteria and the exclusion criteria for each step was defined and documented using the Microsoft Excel spreadsheet. Lastly, all the shortlisted studies’ references were reviewed for any other research studies that satisfied the inclusion criteria and exclusion criteria. Using the Drummond Checklist, two authors impartially and critically assessed the included studies [11]. The CHEERS (Consolidated Health Economic Evaluation Reporting Standards) statement was also used to assess the quality [12]. Two authors individually extracted the primary outcomes, such as cost/treatment, cost/QALY (quality-adjusted life year), and “incremental cost-effectiveness ratios” (ICERs), into the standard pre-piloted format. A similar approach was taken to obtain additional outcomes such as cost/treatment, methods and source used for estimating the cost, efficacy or effect of the therapy, methods and source used to estimating effectiveness, discount rate, benefits and timeline; standard outcomes for the cost-utility analysis and cost effectiveness analysis, sensitivity analysis, uncertainty measures or “willingness to pay” (WTP) threshold, “expected reported value of perfect value,” the pharmacoeconomic evaluation method, analysis method and utilities. The third author resolved all discrepancies during the screening and data extraction process.

Results

The PRISMA flow chart for systematic review is illustrated in Fig. 1. Ten studies were finally included in this systematic review.

Fig. 1.

PRISMA flow chart for systematic review.

1. Characteristic features of CIRT pharmacoeconomic literatures

One study was done in each of the following years 2007, 2009, 2010, 2014, 2018, 2019, 2022, and 2024 (Table 1) [13-22]. Three studies were done in 2010 alone. Four studies were done in Japan, three in Germany, two in the Netherlands and one recently in Belgium. Out of 10 studies, seven studies had patients with say non-small lung cancer, adenoid cystic cancer, head and neck cancer, skull-based chordoma, recurrent rectal cancer, prostate cancer, lung cancer, and hepatocellular cancer. Other three studies were done on all types of cancer. Seven studies were conducted in universities, and three studies were conducted in CIRT centers. CIRT only or CIRT+ intensity-modulated radiation therapy (IMRT)/external beam radiotherapy or Hadron therapy is the intervention. Out of ten studies, five studies had comparators namely stereotactic body radiation therapy (SBRT), IMRT, protons, photons, boron neutron capture therapy (BNCT), and transarterial chemoembolization (TACE). Nine of the studies were retrospective studies. Among them, three were case-control observational studies. Actual measurement, one decision-analytic Markov model and activity-based costing and business model were used by one study each. Seven of the studies were done in payer perspective, two of them were done in hospital perspective and one was done in both societal and payer perspective. Most of the studies showing cost effectiveness to CIRT were done in payer’s perspective.

Table 1.

Characteristic features of CIRT pharmacoeconomic literature

Author	Country	Population	Setting	Intervention	Comparator	Study design	Economic perspective
Okazaki et al. (2022) [13]	Japan	Clinical stage I NSCLC	Gunma University: CIRT was carried out as a clinical study, whereas SBRT was undertaken in a clinical practice setting	CIRT (matched: 15)	SBRT (matched: 15)	Case-control retrospective observational study	Payer
Jensen and Debus (2019) [14]	Germany	Adenoid cystic carcinoma of the head and neck (>90% of population-T4, 60% of skull base invasion)	Heidelberg University Hospital	IMRT plus C12 boost (30 fractions photon IMRT plus 6 fractions C12): 58 patients	IMRT (32 fractions photon IMRT): 37 patients	Case-control retrospective observational study	Payer
Sprave et al. (2018) [15]	Germany	Skull base chordoma	CIRT center	CIRT (45 GyE) in 15 fractions from the skull base to the bottom of the second cervical vertebra, followed by a 15–21 GyE boost (delivered in 5–7 fractions) to the initial extent of tumor)	Photon RT modality of Gamma Knife stereotactic RT	Case-control retrospective observational study	Payer
Mobaraki et al. (2010) [16]	Japan	Locally recurrent rectal cancer (25 patients)	The National Institute of Radiological Science and Gunma University Hospital	CIRT (a total radiation dose of 73.6 Gy [n = 13] or 70.4 Gy [n = 1] in 16 fractions over 4 weeks) (14 patients)	Conventional multimodality therapy (three-dimensional conformal radiotherapy (25–29 fractions of 2.0 Gy, delivered 5 days a week), chemotherapy (5-fluorouracil and leucovorin), and hyperthermia) (11 patients)	Retrospective observational study	Payer
Grutters et al. (2010) [17]	The Netherlands	Hypothetical cohort of inoperable and operable stage I NSCLC patients	Maastricht University Medical Centre	Inoperable stage I NSCLC: carbon-ion and proton therapy; operable stage I NSCLC: carbon ion and proton therapy	Inoperable stage I NSCLC: CRT and SBRT; operable stage I NSCLC: SBRT	Retrospective study	Payer
Peeters et al. (2010) [18]	The Netherlands	Prostate, lung, head and neck, and skull-base chordoma	A combined carbon ion and proton, a proton-only, and a photon facility	Combined carbon ion and proton	Proton-only and photon	Retrospective study	Hospital
Jakel et al. (2007) [19]	Germany	Skull-base chordoma 96 patients	The University Hospital of Heidelberg, Department of Radiation Oncology	CIRT	Conventional photon RT	Retrospective study	Payer
Nakagawa et al. (2009) [20]	Japan	Cancer patients	Heavy ion medical accelerator centers, proton centers and reactors	Carbon ion	Conventional radiotherapy, proton therapy, and BNCT	Retrospective study	Hospital
Okazaki et al. (2024) [21]	Japan	Localized hepatocellular carcinoma (34 patients)	Gunma University	Carbon ion 14 patients	TACE 15 patients	Retrospective study	Payer
Vanderstraeten et al. (2014) [22]	Belgium	Cancer patients	Ghent University	Carbon only	Proton only, combined carbon and proton	Prospective study	Societal, payer

CIRT, carbon ion radiation therapy; NSCLC, non–small cell lung cancer; SBRT, stereotactic body radiation therapy; IMRT, intensity-modulated radiation therapy; C12, carbon ion; GyE, gray equivalents; RT, radiotherapy; CRT, conventional radiotherapy; BNCT, boron neutron capture therapy; TACE, transarterial chemoembolization.

2. Critical quality assessment of CIRT pharmacoeconomic literatures

Using Drummond checklist, it was found out that out of score ten, two studies scored nine, five studies scored eight and three studies scored seven (Table 2). Common points that caused reduction in Drummond checklist’s score were effectiveness were not established, all relevant cost and consequences were not identified, cost and consequences were not adjusted to differential timings, incremental analysis was not performed.

Table 2.

Drummond checklist assessing the quality of CIRT pharmacoeconomic literature

Study	Drummond score (maximum score = 10)	Was a well-defined question posed in answerable form? (yes/no/can't tell)	Was a comprehensive description of the competing alternatives given (i.e., can you tell who did what to whom, where, and how often)? (yes/no/can't tell)	Was the effectiveness of the programme or services established? (yes/no/can't tell)	Were all the important and relevant costs and consequences for each alternative identified? (yes/no/can't tell)	Were costs and consequences measured accurately in appropriate physical units (yes/no/can't tell)	Were costs and consequences valued credibly? (yes/no/can't tell)	Were costs and consequences adjusted for differential timing? (yes/no/can't tell)	Was an incremental analysis of costs and consequences of alternatives performed? (yes/no/can't tell)	Was allowance made for uncertainty in the estimates of costs and consequences? (yes/no/can't tell)	Did the presentation and discussion of study results include all issues of concern to users? (yes/no/can't tell)
Okazaki et al. [13]	9	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes
Jensen and Debus [14]	8	Yes	Yes	Yes	No	Yes	Yes	Yes	No	Yes	Yes
Sprave et al. [15]	8	Yes	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes
Mobaraki et al. [16]	8	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Yes
Grutters et al. [17]	8	Yes	Yes	Yes	Can't tell	Yes	Yes	Yes	No	Yes	Yes
Peeters et al. [18]	7	Yes	Yes	No	Yes	Yes	Yes	No	No	Yes	Yes
Jakel et al. [19]	7	Yes	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes
Nakagawa et al. [20]	7	Yes	Yes	No	Yes	Yes	Yes	Yes	No	No	Yes
Okazaki et al. [21]	9	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes
Vanderstraeten et al. [22]	8	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	No

CIRT, carbon ion radiation therapy.

Using CHEERS checklist, it was found out that of score 28, six studies scored 20 and above. Four studies scored below 20. It was noted that three studies which scored only seven in Drummond checklist also scored less than 20 while using CHEERS checklist. Common points that caused reduction in CHEERS checklist were study population characteristic was not described, why the perspective was chosen was not described, time horizon was not stated, discount rate was not reported, outcomes were not measured, rational of selecting the model not reported, characterizing heterogeneity and distributional effects were not described, approaches to engage patients were not reported. There was no report on difference in stakeholder involvement in the finding of study, no funding and conflict of interest were reported (Table 3).

Table 3.

CHEERS checklist assessing quality of CIRT pharmacoeconomic literature

	Okazaki et al. [13]	Jensen and Debus [14]	Sprave et al. [15]	Mobaraki et al. [16]	Grutters et al. [17]	Peeters et al. [18]	Jakel et al. [19]	Nakagawa et al. [20]	Okazaki et al. [21]	Vanderstraeten et al. [22]
CHEERS score (maximum score = 28)	22	23	20	16	23	18	17	12	25	23
Title: Identify the study as an economic evaluation and specify the interventions being compared. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Abstract: Provide a structured summary that highlights context, key methods, results and alternative analyses. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes
INTRODUCTION: Background and objectives—Give the context for the study, the study question and its practical relevance for decision making in policy or practice. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
METHODS: Health economic analysis plan—Indicate whether a health economic analysis plan was developed and where available. (yes/no/can't tell)	No	No	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Study population: Describe characteristics of the study population (such as age range, demographics, socioeconomic, or clinical characteristics). (yes/no/can't tell)	Yes (only clinical characteristics)	Yes (only clinical characteristics)	No	Yes	No	No	No	No	Yes	No
Setting and location: Provide relevant contextual information that may influence findings. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Comparators: Describe the interventions or strategies being compared and why chosen. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Perspective: State the perspective(s) adopted by the study and why chosen. (yes/no/can't tell)	No	Yes	Yes	No	No	Yes	No	No	Yes	Yes
Time horizon: State the time horizon for the study and why appropriate. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	No	No	No	Yes	Yes
Discount rate: Report the discount rate(s) and reason chosen. (yes/no/can't tell)	No	Yes	Yes	No	Yes	No	No	No	No	No
Selection of outcomes: Describe what outcomes were used as the measure(s) of benefit(s) and harm(s). (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Yes	No
Measurement of outcomes: Describe how outcomes used to capture benefit(s) and harm(s) were measured. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Yes	No
Valuation of outcomes: Describe the population and methods used to measure and value outcomes. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Yes	No
Measurement and valuation of resources and costs: Describe how costs were valued. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Currency, price date, and conversion: Report the dates of the estimated resource quantities and unit costs, plus the currency and year of conversion. (yes/no/can't tell)	No	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Rationale and description of model: If modelling is used, describe in detail and why used. Report if the model is publicly available and where it can be accessed. (yes/no/can't tell)	Yes	Yes	Can't tell	No	Yes	No	NO	No	Yes	Yes
Analytics and assumptions: Describe any methods for analysing or statistically transforming data, any extrapolation methods, and approaches for validating any model used. (yes/no/can't tell)	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes
Characterizing heterogeneity: Describe any methods used for estimating how the results of the study vary for sub-groups. (yes/no/can't tell)	Yes	No	No	No	Yes	Yes	No	No	Yes	Yes
Characterizing distributional effects: Describe how impacts are distributed across different individuals or adjustments made to reflect priority populations. (yes/no/can't tell)	Yes	No	No	Yes	Yes	Yes	No	No	Yes	Yes
Characterizing uncertainty: Describe methods to characterize any sources of uncertainty in the analysis. (yes/no/can't tell)	Yes	Yes	Yes	No	Yes	Yes	Yes	No	Yes	Yes
Approach to engagement with patients and others affected by the study: Describe any approaches to engage patients or service recipients, the general public, communities, or stakeholders (e.g., clinicians or payers) in the design of the study. (yes/no/can't tell)	No	No	No	No	No	No	No	No	Yes	Yes
RESULTS: Study parameters—Report all analytic inputs (e.g., values, ranges, references) including uncertainty or distributional assumptions. (yes/no/can't tell)	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes
Summary of main results: Report the mean values for the main categories of costs and outcomes of interest and summarise them in the most appropriate overall measure. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Effect of uncertainty: Describe how uncertainty about analytic judgments, inputs, or projections affect findings. Report the effect of choice of discount rate and time horizon, if applicable. (yes/no/can't tell)	Yes	Yes	Yes	No	Yes	Yes	Yes	No	Yes	Yes
Effect of engagement with patients and others affected by the study: Report on any difference patient/service recipient, general public, community, or stakeholder involvement made to the approach or findings of the study. (yes/no/can't tell)	No	No	No	No	No	Yes	No	No	No	Yes
Discussion: Study findings, limitations, generalizability, and current knowledge—Report key findings, limitations, ethical or equity considerations not captured, and how these could impact patients, policy, or practice. (yes/no/can't tell)	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Source of funding: Describe how the study was funded and any role of the funder in the identification, design, conduct, and reporting of the analysis. (yes/no/can't tell)	Yes	Yes	Can't tell	No	No	No	No	Yes	Yes	Yes
Conflicts of interest: Report authors conflicts of interest according to journal or International Committee of Medical Journal Editors requirements. (yes/no/can't tell)	Yes	Yes	Yes	No	Yes	No	No	No	No	Yes

CIRT, carbon ion radiation therapy.

3. Cost and outcome aspects of CIRT pharmacoeconomic literature

Out of 10 studies, seven were cost effectiveness analysis and three studies were analyzing only cost (Table 4). Total cost for CIRT ranged from €16,937 (approx. USD 19,595) to €43,600 (approx. USD 50,443) in Germany. The cost was least for adenoid cystic carcinoma of head and neck cancer and high for skull-based chordoma. The cost of CIRT ranged from JPY 3,140,000 (approx. USD 20,450) to JPY 4,974,278 (approx. USD 32,396) in Japan. The cost was least for non–small cell lung cancer (NSCLC) when compared to recurrent rectal cancer and hepatocellular carcinoma (HCC). It is noted that the adjusted cost of CIRT can bring the price from €31,538.21 (approx. USD 36,488) to €19,957.78 (approx. USD 23,090). It is also noted that by changing the perspective from payer to societal the cost drops from €29,450 (approx. USD 34,072) to €16,059 (approx. USD 18,579). Yet another finding is that overall cost (€43,600 [approx. USD 50,443]) is nearly half the cost to treat the recurrences (€81,470 [approx. USD 94,256]). Out of 10 studies, only seven studies assessed effectiveness. And all of them reported increased effectiveness of CIRT compared to the comparators. Medical records and publications were used as sources of cost and effectiveness by most of the studies. The time horizon ranged from 2 years to lifetime and none of the studies were comparable in terms of time horizon. Two studies used discount rate of 3% and only one study considered the discount rate of 4%.

Table 4.

Cost and outcome aspects of CIRT pharmacoeconomic literature

Author	Type of economic evaluation	Cost per treatment	Effectiveness/Efficacy of the treatment as results reported	Source/Methods to estimate costs	Source/Methods to estimate effectiveness, benefits	Time horizon, discount rate, and standard reporting outcomes for cost-effectiveness or cost-utility analysis
Okazaki et al. [13]	CEA	CIRT technical fee was JPY 3,140,000 (approx. USD 20,366)	The LY was 4.515 in the CIRT group and 3.358 in the SBRT group, with an LY difference of 1.157.	Medical records	The Kaplan-Meier curve	Time horizon: preparatory period to 5 years after treatment initiation
		CIRT mean total cost was JPY 5,719,532 (approx. USD 37,098). SBRT mean total cost was JPY 1,428,994 (approx. USD 9,268) in the SBRT group
Jensen and Debus [14]	CEA	Costs of IMRT + C12 boost was €16,877.67 (approx. USD 19,518). Costs of IMRT was €5,675 (approx. USD 6,563). Mean difference in total cost was €16, 937 (approx. USD 19,586).	Median overall survival in the IMRT + C12 group was 102.1 months and 73.7 months in the IMRT group. Mean overall survival in the IMRT + C12 group is 78.3 and 68.4 months in the IMRT group. At the time of evaluation, 30 patients in the IMRT and 26 patients in the IMRT + C12 group were deceased. 62% in the IMRT group and 50% in the IMRT + C12 group l developed locally recurrent disease.	2015 Fee schedules, uniform reimbursement specialist catalogues 2015, German DRG reimbursement scheme, AiDKlinik search engine, and statutory sickness funds.	Medical records	Lifetime horizon, mean difference in total costs is €16,035 (USD 18,518) and €15,900 (USD 18,362) discounted at 3% and 3.5%, respectively.
Sprave et al. [15]	CEA	CIRT treatment total cost was €31,538.21 (approx. USD 36,472). Adjusted CIRT cost after removing financing costs was €19,690.88 (approx. USD 22,771). Adjusted CIRT cost after removing follow-up cost was €18,957.78 (approx. USD 21,923).	The QALYs were 6.65 for photon RT and 8.26 for CIRT.	Institution and published sources	Institution and published sources	34-Year time horizon, QALYs were discounted at 3% per year
Mobaraki et al. [16]	CEA	Mean direct CIRT cost was JPY 4,803,946 (approx. USD 31,159). Mean direct cost in multimodality treatment was JPY 4,611,100 (approx. USD 29,908).	The 2-year overall survival rate was 85% for CIRT and 55% for multimodality treatment. The risk of dying in the multimodality treatment group was 1.4 of that in the carbon ion group. The median hospitalization duration was 66 days for the multimodality treatment group and 37 days for the CIRT group.	Medical records	Medical records	2003 to 2005
Grutters et al. [17]	CEA	CIRT cost was €19,215 (approx. USD 22,244)/patient over five years in inoperable stage I NSCLC and was €14,620 (approx. USD 16,924) for operable stage I NSCLC. SBRT cost was €13,781 (approx. USD 15,953)/patient over 5 years in inoperable stage I NSCLC and was €8,485 (approx. 9,822 USD) for operable stage I NSCLC	For inoperable stage I NSCLC, CIRT yielded 2.67 QALYs per patient and SBRT yielded 2.59. For operable stage I NSCLC, carbon ion therapy yielded 3.16 QALYs and SBRT yielded 3.20.	Published studies	Published studies, personal communication with the author	Five-year time horizon, future effects were discounted to their present value by a rate of 1.5%. Converted costs to the 2007 price level. Future costs were discounted to their present value by a rate of 4%.
Peeters et al. [18]	Partial economic evaluation comparing only cost	A combined facility (carbon ion and proton) total cost per year was €36.7 million (approx. USD 42,485,938), proton total cost per year was €24.9 million (approx. USD 24,900,000), photon total cost per year was €9.6 million (approx. USD 11,113,488). Combined facility cost per fraction was €1,128 (approx. USD 1,305). Proton cost per fraction was €743 (approx. USD 860) and photon cost per fraction was €233 (approx. USD 269).	Not mentioned	Published studies, business plan Maastro Clinic, a Belgian report on hadrontherapy, Turner and Townsend, construction and management consultants	-	Not mentioned
Jakel et al. [19]	CEA	Primary treatment overall CIRT cost was €43,600 (approx. USD 50,473) and photon RT was €27,100 (approx. USD 31,372). Recurrences CIRT treatment cost was €81,470 (approx. USD 94,314) and photon RT was €94,670 (approx. USD 109,595).	The control rate after conventional radiotherapy was between 23% and 38%, photons was 50% and CIRT was at 70%.	Medical records, standard reimbursement	Published studies	Not mentioned
Nakagawa et al. [20]	Partial economic evaluation comparing only cost	The CIRT cost was 3.14 million yen (approx. USD 20,307), proton therapy cost was JPY 2.883 million (approx. USD 18,644), and BNCT cost was >JPY 2.5 million (approx. USD 16,168).	Not mentioned	Published studies	Not mentioned	Not mentioned
Okazaki et al. [21]	CEA	CIRT total cost mean was <JPY 4,974,278 (approx. USD 32,169), TACE total cost mean was JPY 5,284,524 (approx. USD 34,176).	LY of CIRT group was 2.75 and TACE was 2.41	Accounting records maintained by Gunma University	Medical records	3 years
Vanderstraeten et al. [22]	Partial economic evaluation comparing only cost	€29,450 (approx. USD 34,092) and €16,059 (approx. USD 18,590) for carbon only center as per private financing and public sponsoring, respectively €42,749 (approx. USD 49,488) and €21,507 (approx. USD 24,897) for carbon ion in combined center (carbon and proton) as per private financing and public sponsoring respectively. €43,842 (approx. USD 50,753) and €27,217 (approx. USD 31,507) for proton only center as per private financing and public sponsoring, respectively.	Not mentioned	Not mentioned	Not mentioned	Time horizon of 16-year post setup

CIRT, carbon ion radiation therapy; CEA, cost effectiveness analysis; LY, life years; SBRT, stereotactic body radiation therapy; IMRT, intensity-modulated radiation therapy; C12, carbon ion; QALY, quality-adjusted life year; RT, radiotherapy; NSCLC, non–small cell lung cancer; BNCT, boron neutron capture therapy; TACE, transarterial chemoembolization.

4. Salient aspects of CIRT pharmacoeconomic literature

In Japan, there are varied ICER as their denominators and time (Table 5). In 2010, the ICER was JPY 6,428 (approx. USD 41)/1% increase in rate of survival, in 2021 the ICER was JPY 3,708,330 (approx. USD 24,151)/life years (LY), and in 2024 it was JPY 1,77,259 (approx. USD 1,154)/LY. This significant drop in ICER with CIRT across years 2021 and 2024 can be explained by time and distinct patient population say NSCLC and HCC, respectively. The ICER in Germany and the Netherlands were also distinct say €20,854 (approx. USD 24,116)/LY, €8,855.76 (approx. USD 10,241)/QALY, €67,257 (approx. USD 77,778)/QALY and €16,500 (approx. 19,081 USD)/year of disease-free survival. They are quite varied because of distinct time, denominators, and patient population. The ICER was much less when used in when used in skull chordoma patients (€8,855.76 [approx. USD 10,241]/QALY) in Germany than NSCLC patients (€67,257 [approx. USD 77,778]/QALY) in the Netherlands. Out of 10 studies, five studies mentioned WTP threshold. And CIRT was found acceptable by all of them. Sensitivity analysis was done in eight out of ten studies. Reduction in CIRT technical fees, reirradiation with IMRT, increased survival rate with CIRT, local control rate by 60% with CIRT were found to reduce ICER. Out of 10 studies, nine show CIRT is cost effective say in NSCLC, Adenoid cystic carcinoma, head and neck cancer, skull-based chordoma, recurrent rectal cancer, and hepatocellular carcinoma. Among these, two studies that used SBRT, proton-only, and photon as comparators reported that the cost difference for CIRT further decreases with lower fractionation, especially when specific treatment costs and total annual costs were analyzed. Only one study among them shows uncertainty and suggests particle facility investment should be focused by future researchers. Out of 10 studies, only one study suggests that BNCT is more cost-effective than CIRT in the treatment of cancer.

Table 5.

Salient aspects of the CIRT pharmacoeconomic literature

Author	Incremental cost-effectiveness ratios for CIRT	Willingness to pay threshold	Performance of probabilistic sensitivity analysis or other measure of uncertainty	EVPI if reported	Conclusion
Okazaki et al. [13]	JPY 3,708,330 (approx. USD 23,982)/LY for matched patients	At JPY 4,000,000 (approx. USD 25,868) and JPY 8,000,000 (approx. USD 51,737) WTP, the acceptability for CIRT was 49.7% and 81.6%.	CIRT technical fee ranged between JPY 1,600,000 (approx. USD 10,347) and JPY 3,140,000 (approx. USD 20,307), the ICER fluctuated between JPY 3,028,660 (approx. USD 19,587)/LY and JPY 4,754,755 (approx. USD 30,750)/LY.	67.9% and 87.6%	CIRT was a cost-effective treatment. However, it can be more cost-effective by considering the validity and necessity of examinations and hospitalizations and reducing high costs as much as possible.
Jensen and Debus [14]	Annual discount rate of 3% and 3.5% respectively gives rise to ICER of €20,854 (approx. USD 24,141)/LY and €22,078 (approx. USD 25,558)/LY; adjusted values are €26,929 (approx. USD 31,174)/LY (3.0% discount) and €26,863 (approx. USD 31,098)/LY (3.5% discount).	No explicit ICER threshold in Germany. When willingness to pay exceeds €26,863 (approx. USD 31,098), the net monetary benefits turn positive, and the experimental treatment becomes acceptable.	Calculations were performed for two scenarios: the first scenario assumes that all patients undergoing re-irradiation received IMRT. The second scenario assumes re-irradiation always as C12. Both scenarios show a lower ICER with scenario one showing the lowest value at €20,638 (approx. USD 23,891)/LY.	Not mentioned	There is no explicit ICER threshold in Germany. Hence, though IMRT + C12 boost increased initial and overall treatment cost, it is acceptable because of less options.
Sprave et al. [15]	€8,855.76 (approx. USD 10,251)/QALY	Thresholds for “cost effectiveness” and “high-cost effectiveness” were €105,135 (approx. USD 121,710)/QALY and €35,045 (approx. USD 40,570)/QALY, respectively.	Deviance of ±10% for our pooled survival rate for the sensitivity analyses was used. Change in ICER was €8,855.76 (approx. USD 10,251)/QALY	Not mentioned	CIRT was found to be a highly cost-effective option for the treatment of skull base chordoma.
Mobaraki et al. [16]	JPY 6,428 (approx. USD 41) per 1% increase in survival	Not mentioned	Not mentioned	Not mentioned	CIRT paid per treatment could be a cost-effective due to much lower hospitalization costs and treatment-related morbidity.
Grutters et al. [17]	For inoperable stage I NSCLC, carbon ion therapy had an ICER of €67,257 (approx. USD 77,908) per QALY gained compared to SBRT. For a ceiling ratio of €80,000 (approx. USD 92,669) CIRT had the highest probability of being cost-effective (52%), followed by SBRT (47%), proton therapy (2%) and CRT (0%). For operable stage I NSCLC at a ceiling ratio of €80,000 (approx. USD 92,669), SBRT had a 78% probability of being cost-effective, versus 22% for carbon ion therapy.	€80,000 (approx. USD 92,612) per QALY	CIRT in inoperable stage I NSCLC resulted in an acceptable ICER of €36,017 (approx. USD 41,695) per QALY gained. proton therapy, CRT and SBRT were dominated by carbon ion therapy.	Population EVPI €22 million (approx. USD 254,684,100)	Due to the considerable uncertainty in stage I NSCLC, it is recommended not to adopt particle therapy as standard treatment in NSCLC yet.
Peeters et al. [18]	Not mentioned	Not mentioned	A change in capital costs, operational costs, the greater number of treated patients, shortening the time, treatment room utilization reduction and exclusion of the interest payment for financing reduces the cost per fraction.	Not mentioned	Investment costs are highest for the combined carbon ion/proton facility and lowest for the photon facility. Cost differences become smaller when total costs per year and specific treatment costs are compared. Lower fractionation schedule of particle therapy might further reduce its costs.
Jakel et al. [19]	€16,500 (approx. USD 19,101)/year of disease-free survival, €2,539 (approx. USD 2,939) per 1% increase in survival rate, €7,692 (approx. USD 8,904)/year	Not mentioned	A local control rate reduction and moderate reduction of the total number of fractions reduce overall treatment costs	Not mentioned	The analysis therefore shows clearly that carbon ion RT is more cost-effective than a conventional RT treatment for chordoma patients.
Nakagawa et al. [20]	Not mentioned	Not mentioned	Not mentioned	Not mentioned	If more than 200 patients use BNCT in a year, this would be much more cost-effective than carbon ion and proton therapy because of lower accelerator and building cost.
Okazaki et al. [21]	JPY 177,259 (approx. USD 1,146)/LY	<JPY 4,000,000 (approx. USD 25,868)/LY	CIRT total cost maximization or TACE total cost minimization or when WTP close to JPY 0/life years demonstrated that CIRT’s cost effectiveness acceptability	Not mentioned	CIRT is a cost-effective treatment option for localized HCC cases unsuitable for surgical resection.
Vanderstraeten et al. [22]	Not mentioned	Not mentioned	Commissioning phase delay, increases in long-term interest rates and various investment or personnel costs, and a shortening of the amortization period	Not mentioned	A large proportion of adult patients experience more clinical benefits from the higher biological effect obtained with carbon ion treatments. The lower number of fractions for carbon ion treatments compensates for the higher investment costs.

CIRT, carbon ion radiation therapy; EVPI, expected value of perfect information; LY, life years; WTP, willingness to pay; ICER, incremental cost-effectiveness ratio; IMRT, intensity-modulated radiation therapy; C12, carbon ion; QALY, quality-adjusted life year; NSCLC, non–small cell lung cancer; SBRT, stereotactic body radiation therapy; CRT, conventional radiotherapy; BNCT, boron neutron capture therapy; TACE, transarterial chemoembolization; RT, radiotherapy.

Discussion and Conclusion

Ten studies published between 2007 and 2024 were finally included in the systematic review. SBRT, IMRT, protons, photons, BNCT, and TACE were the comparators used with CIRT. Effectiveness was compared in seven out of ten studies. Across studies, CIRT consistently outperformed all the comparators. Okazaki et al. [13] reported higher life years, while Jensen and Debus [14] observed longer survival and lower recurrence. Sprave et al. [15] showed better QALYs and Mobaraki et al. [16] reported improved survival and lower mortality risk. Grutters et al. [17] reported higher QALYs in inoperable NSCLC. Jakel et al. [19] reported superior local control. Okazaki et al. [21] also demonstrated better life years. Mobaraki et al. [16] also reported that CIRT caused only grade 1 skin toxicity in contrast to grade 3 gastrointestinal toxicity caused by multimodality treatment. Carbon ions have a greater Bragg peak relative to the entry dosage. In contrast to photon or proton beams, CIRT beams exhibit greater linear transfer of energy, providing greater relative biologic effectiveness and reduced toxicity [23]. This systematic review points out the CIRT's comparative cost-effectiveness data. With this robust economic evaluation, policymakers and clinicians will get the evidence needed to guide resource allocation and reimbursement decisions.

The time horizon has a considerable impact on cost-effectiveness assessments. The ideal time horizon varied depending on the type of cancer and its therapies, such as the disease's stage, patient age and time to recurrence [24]. Since they reflect the longer-term costs and benefits of the interventions, longer time horizons frequently result in favourable ICERs [25]. Jensen and Debus [14], Sprave et al. [15], and Vanderstraeten et al. [22] used longer timelines say lifetime, 34 years, and 16 years, respectively. All three studies showed CIRT acceptable in terms of cost effectiveness [14,15,22].

In this systematic review, the included CIRT pharmacoeconomic literature were assessed for quality using Drummond and CHEERS checklist. The Drummond checklist prioritizes methodological rigor, whereas the CHEERS checklist places more emphasis on comprehensive transparency in reporting. Coupled together, they offer a robust basis for assessing the quality and lucidity of pharmacoeconomic literature [26-28]. According to Drummond checklist seven studies scored >7 and three equal to 7. But according to CHEERS list six of the studies scored >20 and four <20. The study by Nakagawa et al. [20] scored the least score of 12 out of 28 as per CHEERS checklist. Interestingly that is the only study which concludes CIRT is not cost effective and found it inferior to BNCT in terms of cost.

Okazaki et al. [13] in his study found that at WTP JPY 4,000,000 (approx. USD 25,944) and JPY 8,000,000 (approx. USD 51,889) CIRT was acceptable 49.7% and 81.6%, respectively in clinical stage I NSCLC patients of Japan. Okazaki et al. [21] in his study found that at WTP <JPY 4,000,000 (approx. USD 25,944)/LY CIRT was cost-effective in HCC patients of Japan. Jensen and Debus [14] found that at implicit WTP €26,863 (approx. USD 31,065) CIRT is acceptable because of less options in adenoid cystic carcinoma of the head and neck patients in Germany. Sprave et al. [15] found that at WTP of €105,135 (approx. USD 1,21,582)/QALY and €35,045 (approx. USD 40,473)/QALY CIRT was be a highly cost-effective for skull base chordoma patients in Germany. Grutters et al. [17] in his study found that at WTP of €80,000 (approx. USD 92,515) CIRT was 52% cost effective in inoperable stage I NSCLC patients in the Netherlands and SBRT was 78% cost effective in operable stage I NSCLC in the Netherlands.

In a systematic review on pharmacoeconomic evaluations on protons variation in ICER was explained by patient age, pharmacoeconomic perspective, human papillomavirus disease/p16 status, and prevalence of adverse reactions [29]. In our present review, the ICER reduces due to reduction CIRT technical fee, re-irradiation received IMRT and not C12, increase in survival rate, inoperable cancer type, reduction in capital costs, reduction in operational costs, higher number of treated patients, time shortening, reduction in therapy room utilization and payment exclusion of interest, high rate of local control and reduction in number of total fractions, reduction in CIRT total cost, delay in commissioning phase, long-term increase in interest rates and less investment/personnel costs and shortened amortization period.

CIRT has shown to be a cost-efficient option, especially for treating conditions such as skull base chordoma and localized HCC in patients who are not eligible for surgery [15,21]. Its high biological effectiveness and reduced number of treatment sessions help balance the higher initial setup costs [22]. Although CIRT involves substantial investment, its overall cost-effectiveness can be improved by avoiding unnecessary tests and hospital stays [13]. In Germany, where no official cost-effectiveness threshold exists, approaches like IMRT combined with a carbon ion boost are still considered acceptable due to limited alternatives [14]. However, for early-stage NSCLC, the current uncertainty around clinical outcomes suggests that particle therapy should not yet be considered standard care [17]. Overall, when applied appropriately, CIRT remains a viable and economically reasonable treatment option.

When comparing the cost-effectiveness of CIRT for stage I NSCLC across countries, we identified two key studies: one from Japan by Okazaki et al. (2022) [13] and another from the Netherlands by Grutters et al. (2010) [17]. The Japanese study concluded that CIRT was more cost-effective than SBRT. In contrast, the Netherlands study reported that for inoperable stage I NSCLC, the probability of CIRT being cost-effective was 52%, while for operable stage I NSCLC, this probability dropped to 22% compared with SBRT. This difference can be explained by variations in CIRT’s efficacy between the studies. In the Japanese study, the efficacy was higher than in the Netherlands study. In Japan, life-years were 4.515 for CIRT versus 3.358 for SBRT (difference = 1.157), whereas in the Netherlands, CIRT yielded 2.67 QALYs versus 2.59 for SBRT in inoperable cases, and 3.16 versus 3.20 in operable cases, showing minimal benefit [13,17].

The cost per treatment session is typically higher during the initial phase before a facility reaches full operational capacity. Another important factor is the learning curve as healthcare teams gain experience with new technologies, operational efficiency tends to improve, leading to cost reductions over time. This trend has already been observed in radiotherapy, including with techniques like IMRT [30]. The number of sessions a facility can deliver annually is influenced by the types of patients treated, as this determines the fractionation protocols. The treatment schedules referenced in this study are based on current clinical trials, but it is likely that future advancements will allow for significantly fewer treatment sessions, which would lower the overall cost of particle therapy. Additionally, the average time required per session varies depending on the complexity of the case; more intricate procedures take longer than standard ones. Once a facility is established, it is expected to serve only those patients who meet specific clinical criteria. While photon centers handle a mix of standard and complex cases, particle therapy centers are more likely to treat complex cases, at least for now. Further integrating carbon ion and proton therapy in a single facility is the technical benefit of utilizing a shared accelerator system [18].

Insurance coverage significantly influences the pharmacoeconomic evaluation of CIRT. In Japan, stage I NSCLC treatment with CIRT is not covered by national insurance and is offered only as advanced medical care, leading to high out-of-pocket costs [13]. Conversely, Germany provides partial reimbursement for specific indications with fixed rates (€20,000 [approx. USD23,134] per course), altering cost structures [14]. These disparities affect cost-effectiveness comparisons, and most studies do not consistently adjust for such variations, limiting the standardization and generalizability of their conclusions.

On the flip side of the coin, a phase III randomized trial is currently underway in Japan to assess the effects of dose escalation in SBRT [31]. If higher doses lead to better local tumour control, they could potentially replace the current standard SBRT regimen. This change would likely reduce the LY difference between CIRT and SBRT, resulting in a less favourable ICER for CIRT. Should SBRT outcomes surpass those of CIRT in terms of LY, SBRT would become the more cost-effective or "dominant" option, as indicated by sensitivity analysis [32]. Meanwhile, a new dose-fractionation approach for CIRT has also been proposed, which could improve outcomes in the CIRT group. As a result, ICER values are expected to evolve over time, highlighting the need for ongoing cost-effectiveness evaluations [13].

A precise quantitative analysis was not possible because the studies included did not use consistent criteria to measure treatment costs and outcomes. Most evaluations also overlooked initial investment expenses and equipment utilization rates. Utilities were not accounted for as per protocol, as they were not discussed in any of the included studies. Not all indications of CIRT were discussed because of the limited studies in this advanced field.

To conclude, almost all studies identified CIRT as a cost-effective treatment when compared to the standard treatments. Its cost-effectiveness may improve further in real-world settings due to increased efficacy, lower toxicity, reduction in number of total fractions and potential cost reductions in real clinical settings compared to controlled clinical trial environments across various cancer types.