Cost and effectiveness of image-guided radiotherapy for non-operated localized lung cancer: a population-based propensity score-matched analysis

Introduction

Radiotherapy (RT) is one of the treatment modalities in lung cancer treatment, especially in localized non-operated lung cancer (1-4). Image-guided RT (IGRT) is a new technology to enhance RT delivery accuracy via imaging in the treatment room (1,5,6). Although IGRT hold great potential in improving patient outcome, currently there were few available clinical data to prove this concept (5,6). In the field of lung cancer, IGRT had changed the face of lung cancer RT (7). However, in the era of increasing emphasize on the affordable care (8,9), the cost-effectiveness of a new technology is also important as in the case of IGRT (10). To our knowledge, although IGRT had been reported to be associated with improving pathological response rate for patients receiving neoadjuvant RT in the field of lung cancer (11), the effectiveness of IGRT regarding harder endpoint like survival is less clear in the literature. Therefore, the aim of our study is to compare the cost and effectiveness (survival) of curative conventional fractionated RT for non-operated localized lung cancer delivered with vs. without IGRT via this population-based propensity score (PS) matched analysis.

Material and methods

Data source

The Application Center for Health and Welfare statistics database is a set of databases with complete information regarding cancer registry, death registration, and reimbursement data from National Health Insurance (NHI) for the whole Taiwanese population. The cancer registry provides details regarding individual demographics, tumor histology, cancer primary sites, stage of disease, and primary surgical, radiation, and systemic therapy. NHI is a single compulsory payer with universal coverage in Taiwan and provides a comprehensive services package “All medically necessary services are covered. The package covers inpatient, outpatient, dental services, traditional Chinese medicine, etc.” (12). NHI’s reimbursement data files also provide information including the income of the insured and the characteristics of health care providers.

Study population and study design

Our study flow chart is depicted in Figure 1. Our target populations were non-operated localized lung cancer patients received curative conventional fractionated RT, via either with IGRT or without IGRT within 2007-2010. In brief, the date of diagnose was used as the index date. We set the duration of interest (DOI) as 2-year within the index date. We then decided the explanatory variable of interest (IGRT vs. non-IGRT) based on the cancer registry record. We collected other covariables for the adjustment of potential non-randomized treatment selection and cost and effectiveness data from the Application Center for Health and Welfare statistics (see next sub-section “other explanatory covariables”). Finally, we constructed a PS matched sample based on PS estimated through the above covariables to compare the cost and effectiveness of IGRT vs. non-IGRT within the DOI. In PS analysis, we modeled the use of IGRT (vs. non-IGRT) as the dependent variable and the covariables as independent variables, and used logistic regression to model the probability of receiving IGRT as commonly used in the literatures (13,14). We then used the logit of the probability as the PS, as commonly used in the literature (14). This study had been approved by Research Ethics Committee in our institute [CMUH103-REC-005].

Figure 1 Study flow chart. ¹, We only included those treated (class 1-2) by any single institution to ensure data consistency; ², 6^th American Joint Committee on Cancer staging clinical stage I-III (but not cT4NxM0) [2007-2009] or 7^th stage I-III [2010]; ³, 50-70 Gy in 1.8-2 Gy/fraction, within +/− 10% in dose and treatment duration; ⁴, lower lobe vs. upper/middle lobe; ⁵, hospitals were classified as medical center or regional hospital. RT, radiotherapy; IMRT, intensity modulated radiotherapy; IGRT, image-guided radiotherapy; SES, social-economic status; PET, positron emission tomography; PS, propensity score; DOI, duration of interest; ICER, incremental cost-effectiveness ratio; OOP, out-of-pocket.

Other explanatory covariables

Firstly, we searched the literature regarding potential factors that might influence the cost of lung cancer patients treated with RT. We used the following balanced search filters regarding costs or economics in the PubMed “(“costs and cost analysis”[MeSH] OR costs[Title/Abstract] OR cost effective*[Title/Abstract]) OR (cost*[Title/Abstract] OR “costs and cost analysis”[MeSH:noexp] OR cost benefit analysis*[Title/Abstract] OR cost-benefit analysis[MeSH] OR health care costs[MeSH:noexp])” as in the literatures (15,16). We combined the above keywords with “(lung cancer) AND ((radiotherapy) OR (radiation therapy))” and found that after the use of positron emission tomography (PET) during peri-diagnostic period was a potential factor (17). Secondly, we collected other factors that were not reported in the literature but that might affect the cost based on our clinical and research experiences. In this regard, we also included patient demographic factors [age, gender, residency region, social-economic status (SES)], patient characteristics (comorbidity), disease characteristics (tumor location, histology, clinical stage & period), treatment (RT method & dose, systemic therapy), and health service provider characteristics (hospital level) based on our clinical experiences and prior NHI and the Application Center for Health and Welfare statistics related studies (18-24). Age was classified as ≥65 years old or not. Patient residency was classified as northern Taiwan or elsewhere. SES was classified as high (income greater than minimal wage) or not. Tumor location was classified as lower vs. upper/middle. Histology was classified as small cell or non-small cell. Stage was classified as stage (I-II vs. III) while period was classified as 2007-2009 (6^th staging edition) vs. 2010 (7^th edition). Hospital was classified as medical center or not.

Cost and effectiveness assessment

We obtained survival status according to the death registry and used survival duration as effectiveness. The cost and cost-effectiveness were conducted from a payers’ perspective (i.e., charges to NHI). The cost was converted to 2014 USD by purchasing the power parity and consumer price indexes (25).

Statistical & supplementary analysis (SA)

Tabulation and standardized difference were used to assess the balance of covariates between PS-matched groups. We used the incremental cost-effectiveness ratio (ICER) to evaluate the cost-effectiveness and used the cost effectiveness acceptability curve (CEAcC) to represent the uncertainty of cost-effectiveness at various willing-to-pay (WTP) thresholds (26). We also compared the survival during the entire follow-up period (censored on 1 January 2013) using a robust variance estimator (14). We performed two supplementary analyses to evaluate the robustness of our finding. In the 1^st SA (SA-1) we assumed an out-of-pocket (OOP) charge of 3,861 USD (60,000 National Taiwan Dollar by purchasing power parity index (PPP) in 2014, the charge in our hospital for IGRT) and re-calculate the CEAcC. In the 2^nd SA (SA-2), we estimated the cost-effectiveness if DOI was set at 4 years via weighted estimator (27). SAS 9.3 (SAS Institute, Cary, NC) was used for all the analysis.

Results

Identification of the study cases (Figure 1 & Table 1)

Table 1 Patient characteristics of the propensity-score matched final study population
Full table

As revealed in Figure 1, 596 localized lung cancer patients treated with curative RT via either IGRT or non-IGRT were identified as the initial study population. After exclusion of those with missing data and matching by PS, the final study population included 124 patients. The characteristics of these patients are described in Table 1. A good balance of covariables and small standardized differences (<0.1) were seen for all covariables except hospital and systemic therapy (0.108; 0.102).

Cost and effectiveness (Figures 2,3 & Table 2)

Figure 2 Kaplan-Meier survival curve (in days). IGRT in dotted line vs. non-IGRT in solid line. IGRT, image-guided radiotherapy.

Figure 3 Cost-effectiveness and used the cost effectiveness acceptability curve. (A) Primary analysis; (B) supplementary analysis-1 (take out-of-pocket payment into consideration).

Table 2 Cost-effectiveness results*
Full table

Within DOI (2-year), both the mean cost (2014 USD) and survival (year) were higher for IGRT ($60,774 vs. $60,554; 1.43 vs. 1.37). The ICER when IGRT was compared to non-IGRT was 3,667 [USD/life-year (LY)]. For the entire follow-up period, the survival rate of IGRT group was better but was not of statistical significance (hazard ratio of death =0.903, P value =0.63). The Kaplan-Meier survival curve is depicted in Figure 2. The CEAcC in Figure 3A revealed that the chance for IGRT to be cost-effective was around 68% & 70% at WTP threshold 50,000 USD/LY and 150,000 USD/LY respectively. In SA-1 as seen in Figure 3B, the chance for IGRT to be cost-effective was lower if potential OOP was considered, but was still higher than a half (61%) when the WTP threshold was 150,000 USD/LY. In the SA-2, the estimated incremental cost and effectiveness were 0.11 (LY) and −3,372 (USD). Therefore, if DOI was set as 4-year, IGRT was less costly and more effective, still cost-effective as well.

Discussion

In this population-based PS matched analysis, we found that when used in curative conventional fractionated RT for non-operated localized lung cancer, IGRT was in average cost-effective when compared with non-IGRT.

Our finding was compatible with the literature in that IGRT was associated with higher pathological response rate when used in neoadjuvant RT in lung cancer (11), but that our study provided a more clinically meaningful endpoint (survival) rather than the surrogate endpoint (response rate) in the literature.

The interpretation of our finding is likely to be consistent as in the literature in that IGRT improved the accuracy of RT delivery (5,6). However, our result should also be interpreted with caution given the non-randomized nature of our study and the limit in generalizability to health care systems other than Taiwan.

There were also limitations in our study. Firstly, the intervention in our study was not randomized. Therefore, potential unobserved confounding variable was possible although we had done our best as suggested in the literature (28). In addition, the use of registry in our study was a reasonable alternative to the randomized controlled study as suggested in the literature (29). Secondly, results in our primary endpoints (2-year cost-effectiveness) might be changed in the long term although we had estimated the 4-year results in our supplemental analysis.

Conclusions

In this population-based PS matched cost-effectiveness analysis, we provide the first empirical evidence that when compared to non-IGRT, IGRT was potentially cost-effective in the mid-term (2-year) and probably still cost-effectiveness at longer follow-up (4 years). However, the result should be interpreted with caution given the non-randomized design and the uncertainty regarding applicability in other health care systems.

Acknowledgements

The data analyzed in this study was provided by the Application Center for Health and Welfare statistics, Ministry of Health and Welfare, Executive Yuan, Taiwan. The author would like to thank the funding agencies [Health and welfare surcharge of tobacco products, China Medical University Hospital Cancer Research Center of Excellence (MOHW104-TDU-B-212-124-002, Taiwan)] for their financial support. The corresponding author would like to thank Dr. Ya-Chen Tina Shih for her mentoring.

Funding: This work was supported by the Health and welfare surcharge of tobacco products, China Medical University Hospital Cancer Research Center of Excellence, Taiwan [MOHW104-TDU-B-212-124-002 and Ministry of Science and Technology [MOST 104-2314-B-039-041-].

Footnote

Conflicts of Interest: The authors have no conflicts of interest to declare.

References

1. Goldstraw P, Ball D, Jett JR, et al. Non-small-cell lung cancer. Lancet 2011;378:1727-40. [PubMed]
2. van Meerbeeck JP, Fennell DA, De Ruysscher DK. Small-cell lung cancer. Lancet 2011;378:1741-55. [PubMed]
3. National Comprehensive Cancer Network (NCCN) Guidelines for Non-Small Cell Lung Cancer, version 7.2015. (Accessed 25 Jun 2015). Available online: . [Free registration required].http://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf
4. National Comprehensive Cancer Network (NCCN) Guidelines for Small Cell Lung Cancer, version 1.2016. (Accessed 25 Jun 2015). Available online: . [Free registration required].http://www.nccn.org/professionals/physician_gls/pdf/sclc.pdf
5. Bujold A, Craig T, Jaffray D, et al. Image-guided radiotherapy: has it influenced patient outcomes? Semin Radiat Oncol 2012;22:50-61. [PubMed]
6. Chen AB. Comparative effectiveness research in radiation oncology: assessing technology. Semin Radiat Oncol 2014;24:25-34. [PubMed]
7. Christodoulou M, Bayman N, McCloskey P, et al. New radiotherapy approaches in locally advanced non-small cell lung cancer. Eur J Cancer 2014;50:525-34. [PubMed]
8. Sullivan R, Peppercorn J, Sikora K, et al. Delivering affordable cancer care in high-income countries. Lancet Oncol 2011;12:933-80. [PubMed]
9. Institute of Medicine. Delivering High-Quality Cancer Care: Charting a New Course for a System in Crisis. Washington, DC: The National Academies Press, 2013.
10. Baumann M, Hölscher T, Zips D. The future of IGRT - cost benefit analysis. Acta Oncol 2008;47:1188-92. [PubMed]
11. Shumway D, Corbin K, Salgia R, et al. Pathologic response rates following definitive dose image-guided chemoradiotherapy and resection for locally advanced non-small cell lung cancer. Lung Cancer 2011;74:446-50. [PubMed]
12. Bureau of National Health Insurance (NHI). Universal Health Coverage in Taiwan. Available online: http://www.nhi.gov.tw/Resource/webdata/21717_1_20120808UniversalHealthCoverage.pdf
13. Rosenbaum PR. Basic Tools of Multivariate Matching. In: Rosenbaum PR, editor. Design of Observational Studies (Springer Series in Statistics). New York: Springer, 2010:163-86.
14. Austin PC, Chiu M, Ko DT, et al. Propensity Score Matching for Estimating Treatment Effects. In: Faries DE, Leon AC, Haro JM, et al. editors. Analysis of Observational Health Care Data Using SAS®. Cary, North Carolina: SAS Institute Inc., 2010:51-84.
15. Glanville J, Paisley S. Identifying economic evaluations for health technology assessment. Int J Technol Assess Health Care 2010;26:436-40. [PubMed]
16. Smieliauskas F, Chien CR, Shen C, et al. Cost-effectiveness analyses of targeted oral anti-cancer drugs: a systematic review. Pharmacoeconomics 2014;32:651-80. [PubMed]
17. Dinan MA, Curtis LH, Carpenter WR, et al. Redistribution of health care costs after the adoption of positron emission tomography among medicare beneficiaries with non-small-cell lung cancer, 1998-2005. J Thorac Oncol 2014;9:512-8. [PubMed]
18. Lin CC, Hsia TC, Chien CR. 3rd line Erlotinib for lung cancer in Asia may be as cost-effective as in the Western world. Lung Cancer 2012;76:499-500. [PubMed]
19. Chien CR, Su SY, Cohen L, et al. Use of Chinese medicine among survivors of nasopharyngeal carcinoma in Taiwan: a population-based study. Integr Cancer Ther 2012;11:221-31. [PubMed]
20. Hsia TC, Tu CY, Chen HJ, et al. Effectiveness of intensity-modulated radiotherapy for lung cancer. Clin Oncol (R Coll Radiol) 2013;25:447-8. [PubMed]
21. Chien CR, Shih YC. Use of personalized decision analysis in decision making for Palliative vs. surgical management of the oldest-old patients with localized skin cancer in a culturally sensitive environment: a case study of a 96-year-old male Taiwanese patient. J Pain Symptom Manage 2013;45:792-7. [PubMed]
22. Chien CR, Lin HW, Yang CH, et al. High case volume of radiation oncologists is associated with better survival of nasopharyngeal carcinoma patients treated with radiotherapy: a multifactorial cohort analysis. Clin Otolaryngol 2011;36:558-65. [PubMed]
23. Chien CR, Hsia TC, Chen CY. Cost-effectiveness of chemotherapy combined with thoracic radiotherapy versus chemotherapy alone for limited stage small cell lung cancer: A population-based propensity-score matched analysis. Thoracic Cancer 2014;5:530-6.
24. Ke TW, Liao YM, Chiang HC, et al. Effectiveness of neoadjuvant concurrent chemoradiotherapy versus up-front proctectomy in clinical stage II-III rectal cancer: A population-based study. Asia Pac J Clin Oncol 2014. [Epub ahead of print]. [PubMed]
25. International Monetary Fund (IMF). Available online: http://www.imf.org/external/pubs/ft/weo/2009/02/weodata/weorept.aspx?sy=2007&ey=2014&scsm=1&ssd=1&sort=country&ds=.&br=1&pr1.x=34&pr1.y=10&c=528%2C111&s=PPPEX%2CPCPI&grp=0&a=
26. Hoch JS, Briggs AH, Willan AR. Something old, something new, something borrowed, something blue: a framework for the marriage of health econometrics and cost-effectiveness analysis. Health Econ 2002;11:415-30. [PubMed]
27. Zhao H, Wang H. Cost and Cost-Effectiveness Analysis with Censored Data. In: Faries DE, Leon AC, Haro JM, et al. editors. Analysis of Observational Health Care Data Using SAS®. Cary, North Carolina: SAS Institute Inc., 2010:363-84.
28. Kreif N, Grieve R, Sadique MZ. Statistical methods for cost-effectiveness analyses that use observational data: a critical appraisal tool and review of current practice. Health Econ 2013;22:486-500. [PubMed]
29. van Loon J, Grutters J, Macbeth F. Evaluation of novel radiotherapy technologies: what evidence is needed to assess their clinical and cost effectiveness, and how should we get it? Lancet Oncol 2012;13:e169-77. [PubMed]