Cost associated with hematopoietic stem cell transplantation: a retrospective claims data analysis in Germany

Acute lymphoblastic leukemia (ALL) is a cancer of the blood characterized by an uncontrolled proliferation of lymphoblasts. It is the most common type of leukemia in children (with an incidence peak between 2 and 5 years of age). ALL is usually treated with chemotherapy, comprising an induction phase, a consolidation phase and a maintenance phase that can stretch over several years. For high-risk patients or patients with an early relapse, allogeneic HSCT is a treatment option. However, because of its serious side effects, allogeneic HSCT is generally not recommended for patients older than 65 years [4].

Diffuse large B-cell lymphoma (DLBCL) is the most common type of aggressive non-Hodgkin lymphoma. The average age at diagnosis is 65–70 years; however, there is considerable variation in patient age, and even children can be affected. DLBCL is usually treated with chemotherapy (often in combination with rituximab). For patients who are heavily pretreated, have refractory disease or are experiencing failure after autologous HSCT, allogeneic HSCT is the main treatment option [5].

Follicular lymphoma (FL) constitutes the most common type of slow-growing non-Hodgkin lymphoma. Its progression is often asymptomatic and may not require immediate treatment. However, overall survival for patients with FL is highly variable and can range from a few years to more than 30 years. FL can develop into more aggressive types of lymphoma such as DLBCL. Radiation therapy is a potentially curative treatment during the early stage of the disease. Systemic treatments generally involve chemotherapy in combination with rituximab. Autologous HSCT is a treatment option for relapsed chemosensitive patients after treatment with two to three lines of previous chemoimmunotherapies. For younger patients who relapse following autologous HSCT, allogeneic HSCT is a potential treatment option [5,6].

Given the lack of previously published results on HSCT-associated costs in patients with leukemia and lymphoma in Germany, this study aims to quantify costs associated with HSCT in patients with ALL, DLBCL and FL from a German study population. This information can inform policy makers about the burden of HSCT. As a reference, costs were also calculated for age- and sex-matched controls who were diagnosed with ALL, DLBCL or FL; received chemotherapy; and had no HSCT (this was an exploratory objective of the study).

To the best of our knowledge, this is the first study to investigate costs associated with HSCT in patients with ALL, DLBCL and FL. Results showed that HSCT in the treatment of ALL is associated with substantial healthcare costs. These findings are generally consistent with other recent investigations done in EU countries. Among patients with relapsed/refractory Philadelphia-negative B-cell relapsed/refractory ALL, costs associated with HSCT were €133,965 (including €26,337 for HSCT itself) in Belgium; estimated costs were €118,672 in the HSCT period and €68,344 during chemotherapy before HSCT in France [15,16]. Of note, there were differences in design between the studies. For example, in the French study, the index date was the diagnosis date for ALL, whereas in our study, the index was the HSCT quarter. The French study was based on chart review and separated the chemotherapy and HSCT periods, whereas our study was based on claims data and did not differentiate between the chemotherapy and HSCT periods. Additionally, there may be differences in measurement periods due to data limitations (differences in time to patient death or loss to follow-up). However, we found the results from these studies were generally consistent with our study.

Costs are even higher in the USA. In patients with DLBCL, the average total healthcare cost was estimated to be US$248,390 during the index hospitalization and $455,741 1 year after the index date [17]. In patients with ALL, during the first year following HSCT, the total cost was estimated to be US$683,099 based on two US nationally representative, large claims databases [18]. Consistent with our analysis, hospitalization represented the largest component of the overall cost. Another more recent US claims-based study estimated that the total cost of allogeneic HSCT for patients with ALL was US$669,711 (US$275,642 per patient for the index hospitalization period and US$394,069 during the year following the index period) [19]. In a US study that used a national private claims database (including patients across multiple diagnoses [ICD-9 diagnosis codes]), the median total cost reported for allogeneic HSCT was US$203,026, with the majority of costs incurred during the initial transplant hospitalization [20]. Consistent with our analysis, costs were higher for pediatric patients versus adults and for allogeneic HSCT versus autologous HSCT.

Within the first 2 years after transplantation, healthcare costs were two- to four-times higher compared with age- and sex-matched patients who did not receive HSCT for the same indication. In this study the cost impact of HSCT (i.e., the difference in cost utilized for patients with HSCT vs those without) observed in pediatric ALL was on average almost €200,000 covering an observation period of 1 year before and 2 years after transplant. The cost impact of allogeneic HSCT in adult ALL and DLBCL/FL was €160,000 and €175,000, respectively. Patients receiving autologous HSCT in DLBCL/FL were associated with additional costs of €50,000.

The highest healthcare costs and most intensive healthcare resource utilization in Germany were observed within the first year following the HSCT, which is consistent with results from other studies [18]. The main driver of high economic burden is hospitalizations, which accounted for more than 80% of direct healthcare costs. In addition to direct healthcare cost, the indirect cost in adult patients induced by sick pay and productivity loss add another 15–20% to the potential overall cost.

This analysis was subject to certain limitations. Retrospective claims data analyses are subject to potential coding errors or data omissions. Although the sample size was relatively small, our findings are likely to be representative of the economic burden of allogeneic HSCT in ALL, DLBCL and FL in Germany. To avoid underestimation of the overall economic burden, indirect cost was calculated in terms of sick pay and productivity loss. This calculation did not include further nonmedical and nonpharmacy costs incurred by patients or caregivers, such as temporary housing or transportation. Further research would be warranted to assess the nonmedical and nonpharmacy costs. Selection of patients for the control group was a further limitation of the study. Our control groups included patients diagnosed with ALL, DLBCL or FL who did not receive HSCT. Matching of controls was based on age and sex because the claims database did not include additional information that could be used to match cases with controls (e.g., disease characteristics, comorbidities, prior lines of therapy, time to remission/duration of remission). Because there may have been important differences between the control and study groups that have not been accounted for, the control groups should be regarded as descriptive.

Hematopoietic stem cell transplantation is associated with substantial cost in the treatment of malignant hematopoietic diseases. Indirect cost due to sick leave and consecutive productivity loss constitute relevant additional cost factors in Germany. Novel and effective therapies that reduce HSCT use could potentially reduce total medical costs.

A small portion of the research (in 30 pediatric patients with ALL) was presented at the 58th Annual Meeting of the American Society of Hematology in 2016 (Lehne M, Hickstein L, Salimullah T et al. Costs of allogeneic hematopoietic stem cell transplantation [HSCT] in pediatric patients with acute lymphoblastic leukemia [ALL] – an analysis of German claims data. Abstract #5932).

All authors were involved in the conception and design, or analysis and interpretation of the data; L Mayerhoff and J Zhang were involved in the drafting of the paper or revising it critically for intellectual content; and all authors were involved in the final approval of the version to be published. All authors agree to be accountable for all aspects of the work.

This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/

This study was financed by Novartis Pharmaceuticals Corporation. Elsevier Health Analytics has received consultancy fees from Novartis for this project. Novartis was involved in the study planning and interpretation of the results. Novartis was not involved in the execution of the analysis. T Salimullah is an employee of Novartis. SK Thomas was an employee of Novartis at the time of the study and is a current employee of Alexion Pharmaceuticals. J Zhang is an employee of and stock holder of Novartis. L Mayerhoff, M Lehne and S Prieur are employees of Elsevier Health Analytics (EHA). EHA received payment from Novartis for analyzing the study data. L Hickstein is an employee of the Health Risk Institute. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Editorial assistance was provided by Rozena Varghese, PharmD, of C4 MedSolutions, LLC (Yardley, PA, USA), a CHC Group company and was funded by Novartis Pharmaceuticals Corporation.

Anonymized SHI data were used for this research. The data are provided in accordance with paragraphs 287 SGB V and 75 Sozialgesetzbuch (SGB) X of German law. In accordance with German data protection requirements, data on patients and physicians are anonymized, as are the providers and the sickness funds. No regions smaller than federal states or cohorts with less than 100 patients are identified. The use of such data is not subject to ethics committee approval in Germany. All InGef studies need approval from the internal InGef Scientific Research Group, which considers both ethical and scientific aspects of the intended study.