- Email: [email protected]
Therapeutic Use of Filgrastim for Established Febrile Neutropenia Is Cost Effective Among Patients With Solid Tumors and Lymphomas
Clinical Therapeutics/Volume 39, Number 6, 2017
Therapeutic Use of Filgrastim for Established Febrile Neutropenia Is Cost Effective Among Patients With Solid Tumors and Lymphomas Xiao Jun Wang, BSc1,2,*; Wei Xiang Tong, BSc1,*; and Alexandre Chan, PharmD1,2 1
Department of Pharmacy, National University of Singapore, Singapore, Singapore; and 2Department of Pharmacy, National Cancer Centre Singapore, Singapore, Singapore
ABSTRACT Purpose: With the emergence of biosimilar filgrastim to the market, there is a gradual decrease in the listed price of the originator product of filgrastim over the years, and this could have an impact on the costeffectiveness of filgrastim in the treatment of febrile neutropenia (FN). A cost-effectiveness analysis would allow clinicians to make informed decision when considering the therapeutic filgrastim among low-risk FN patients. This study aims to evaluate the costeffectiveness of adding therapeutic filgrastim to antibiotics in the treatment of established FN among patients with solid tumors and lymphomas. Methods: A decision tree model was created to compare two treatment options for established FN as follows: (1) antibiotics alone (standard care) and (2) antibiotics with therapeutic filgrastim (comparator). The target population was a hypothetical cohort of adult cancer patients with solid tumors or lymphomas hospitalized with FN in Singapore. The analysis was performed from a hospital’s perspective over a 21-day time horizon. The main outcome measures included costs, quality-adjusted life year (QALY) and incremental cost-effectiveness ratio (ICER). One-way sensitivity analysis and probabilistic sensitivity analysis were conducted to evaluate the robustness of the results. Findings: Compared with antibiotics alone, the treatment strategy of antibiotics with therapeutic filgrastim was a dominant choice, incurring a cost saving of US$125 per patient (comparator versus standard care: US$9110 versus US$9235) and additional health benefit of 0.0007 QALY gained per patient (comparator versus standard care: 0.0450 versus 0.0443). Model results were robust against
*
These authors contributed equally to this work.
June 2017
the parameter variations in the one-way sensitivity analyses, but increasing the cost of filgrastim beyond US$87 per injection would increase the ICER to 4US $50,000/QALY. Furthermore, the strategy of antibiotics with therapeutic filgrastim was the preferred choice (dominant or cost-effective) in 83.7% of the model iterations at a willingness-to-pay threshold of US$50,000/QALY. Implications: From a hospital’s perspective, the therapeutic filgrastim, in conjunction with antibiotics, in the treatment of FN is cost effective. This provides evidence to support the routine use of filgrastim for the treatment of FN among adult cancer patients with solid tumors and lymphomas. (Clin Ther. 2017;39:1161–1170) & 2017 Elsevier HS Journals, Inc. All rights reserved. Key words: cost-effectiveness analysis, febrile neutropenia, filgrastim, treatment.
INTRODUCTION Febrile neutropenia (FN) is a serious oncologic emergency experienced by cancer patients receiving myelosuppressive chemotherapy. One recent study in Singapore identified that FN had an incidence of around 19.5% among patients with solid tumors or lymphomas and that FN was the most common reason for unwanted hospitalizations.1 FN is associated with a significant economic burden on patients. In Singapore, we have reported that the cost of inpatient FN management was around US $4193 per episode, and FN patients who had Accepted for publication May 11, 2017. http://dx.doi.org/10.1016/j.clinthera.2017.05.345 0149-2918/$ - see front matter & 2017 Elsevier HS Journals, Inc. All rights reserved.
1161
Clinical Therapeutics developed severe sepsis were associated with a significantly higher economic burden.2 Filgrastim is a granulocyte colony-stimulating factor (G-CSF) that stimulates neutrophil production. The prophylactic use of G-CSF, including filgrastim, has been supported by the current guideline to prevent FN.3 However, the therapeutic use of G-CSF for established FN is not recommended among low-risk FN patients, such as patients with solid tumors or lymphomas.3–5 One recent meta-analysis of randomized controlled trials suggested that patients receiving G-CSF plus antibiotics had shorter duration of FN and antibiotics use than patients receiving antibiotics alone.6 Therefore, the therapeutic use of filgrastim is expected to reduce patient’s hospital length of stay (LOS) and to provide benefits on cost savings associated with reduced hospitalization and antibiotics use. With the emergence of biosimilar filgrastim to the market, a gradual decrease in the listed price of the originator product of filgrastim has occurred over the years, and this could have an impact on the costeffectiveness of filgrastim in the treatment of FN.7 To our knowledge, cost-effectiveness studies that focus on the therapeutic use of filgrastim for established FN among patients with solid tumors and lymphomas are lacking. A cost-effectiveness analysis would allow clinicians to make informed decision when considering the therapeutic filgrastim among low-risk FN patients. Therefore, we designed this study with the primary objective to evaluate the cost-effectiveness of therapeutic filgrastim in the treatment of FN among adult patients with solid tumors and lymphomas.
PATIENTS AND METHODS A decision tree model was constructed by using TreeAge Pro 2013 (TreeAge Software, Inc, Williamstown, Massachusetts) to compare two treatment options for FN as follows: (1) antibiotics alone (standard care) and (2) antibiotics with therapeutic filgrastim (comparator). The antibiotic treatment for FN in both standard care and comparator arms followed the approaches recommended by the Infectious Disease Society of America guideline.8 Empiric antibiotic therapy included an antipseudomonal β-lactam agent, such as cefepime, carbapenem (meropenem or imipenem-cilastatin), or piperacillin-tazobactam.8 Other antibiotics, such as aminoglycosides, fluoroquinolones,
1162
and/or vancomycin, may be considered in patients with complicated presentations (eg, hypotension and pneumonia) or suspected/proven antimicrobial resistance.8 In the comparator group, the therapeutic filgrastim was administered daily until patients’ absolute neutrophil count (ANC) recovered (ANC Z2.0 109/L). In this model, the time horizon was set as 21 days to represent the interval between each chemotherapy cycle. No discounting was applied because of the short time horizon in this model. The perspective was taken from a Singaporean hospital, and only direct medical cost was considered in this study.
Definitions FN was defined as a single episode of fever Z38.31C or fever Z38.01C for at least 1 hour and an ANC of o0.5 109/L.8 Grade IV neutropenia was set as having an ANC of o0.5 109/L.9 The duration of grade IV neutropenia was the number of days with ANC o0.5 109/L. Time to ANC recovery was the duration from FN onset date until patients’ ANC increased to 2.0 109/L. In this study, serious complication was defined as the occurrence of severe sepsis during the hospitalization. LOS was the overall duration from FN onset date until patient was discharged from the hospital.
Model Structure The model structure is shown in Figure 1. The design of this model is based on the Infectious Disease Society of America guideline8 and clinicians’ discussion to reflect the local treatment pattern for established FN, and the face validity of this model was endorsed by the clinicians who manage FN. Several studies have reported the effectiveness of filgrastim on reducing the duration of grade IV neutropenia.6,10 Therefore, the duration of grade IV neutropenia was included in this model. The duration of grade IV neutropenia was categorized into the following two groups: (1) grade IV neutropenia r3 days and (2) grade IV neutropenia 43 days, which was based on a previous study in Singapore, showing that the mean duration of grade IV neutropenia was around 2 to 3 days.11 Serious complication was considered because prolonged duration of grade IV neutropenia would increase the risk of a patient developing serious complications.12
Volume 39 Number 6
X.J. Wang et al.
Alive Serious Complication Death
Grade IV neutropenia > 3days Alive Without Serious Complication Death
Antibiotics with Filgrastim Alive Serious Complication Death
Grade IV neutropenia <= 3days
Established FN
Alive Without Serious Complication Death Antibiotics alone
Figure 1. Decision tree model structure. FN ¼ febrile neutropenia.
Model Inputs Probabilities The model inputs used in this study are shown in Table I. Probability estimates needed for the analysis were mainly obtained from a database consisting of 305 adult cancer patients diagnosed with solid tumors or lymphomas at National Cancer Centre Singapore (NCCS) with documented FN-related hospitalizations between 2009 and 2012.10 Because of a lack of relevant data in Singapore, the probabilities of grade IV neutropenia for 43 days (with or without therapeutic filgrastim) were generated from a published randomized control trial.13 In that study, a total of 210 patients with solid tumors and developed FN were included to evaluate the efficacy of adding G-CSF to broad-spectrum antibiotic treatment for FN.13 For patients assigned to the intervention group, the additional G-CSF was initiated within 12 hours of the diagnosis of FN (G-CSF at a dose of 5 μg/kg/d) until patients’ ANC increased to 41000/mm3.13 Findings of that study suggested that adding G-CSF to antibiotic therapy could significantly reduce the duration of grade IV neutropenia compared with antibiotics alone.13
were adjusted to 2014 Singapore dollars (SG$) using the healthcare component of the Singapore Consumer Price Index.14 The cost data were then converted to 2014 US dollars (US$) using purchasing power parity conversion rate obtained from the World Bank (2014; SG$1 ¼ US$0.86).15
Utilities Two previous studies (both conducted in Singapore) were used to generate utility estimates.16,17 The first consisted of 521 adult patients with solid tumors and lymphomas without FN.17 The second consisted of 102 patients with solid tumors and lymphomas hospitalized for FN between 2014 and 2015, of which, 26 developed severe sepsis.16 Utilities were assessed using the three-level EuroQoL Group's fivedimension questionnaire and calculated using a Japanese value set.18 Baseline health utility value was obtained from the first database by matching with the second database in terms of cancer type and stage because these were the factors that had been shown to affect quality of life in cancer patients undergoing chemotherapy.19 Utility values of each health state were computed from the pooled data from these two databases.16,17
Costs Cost data were retrieved from a FN database at NCCS.2 The cost for each filgrastim injection (300 mg) was obtained from the Department of Pharmacy at Singapore General Hospital because patients with FN from NCCS were admitted to Singapore General Hospital for management. Components of direct medical cost included medication, laboratory, radiography, ward costs, and surgical costs. All costs
June 2017
Cost-effectiveness Analysis and Sensitivity Analyses Costs, quality-adjusted life year (QALY) gained, and the incremental cost-effectiveness ratio (ICER) were calculated in the base-case analysis. One-way sensitivity analyses were conducted on all parameters by varying each variable within its plausible range, at a willingness-to-pay (WTP) threshold of US$50,000/
1163
Clinical Therapeutics
1164
Table I. Model inputs. Parameter Costs (US$, 2014) Filgrastim 300 μg (per injection)
Mean 37
Volume 39 Number 6
Inpatient treatment 1100 Without complications (per day)* * With complications (per day) 1115 Time parameter (day) Time horizon 21 Time to ANC recovery Grade IV neutropenia 43 days (antibiotics 5.30 with filgrastim) Grade IV neutropenia r3 days (antibiotics 2.56 with filgrastim) Grade IV neutropenia 43 days (antibiotics 6.50 alone) Grade IV neutropenia r3 days (antibiotics 2.76 alone) Length of stay Without complications 7.31 With complications 13.37 Utility Adult cancer patients undergoing chemotherapy 0.85 without FN Hospitalized patient with FN without serious 0.73 complications Hospitalized patient with FN with serious 0.43 complications Probability Grade IV neutropenia 43 days (antibiotics alone) 0.47 Grade IV neutropenia 43 days (antibiotics with 0.26 filgrastim)
Range (95% CI) 29–44 (⫾20%) 858–1342 903–1326 NA
PSA Distribution Normal
Reference SGH
Gamma (α ¼ 79.98; λ ¼ 0.073) NCCS, Wang et al2 Gamma (α ¼ 117.19; λ ¼ 0.11) NCCS, Wang et al2 NA
Assumed
4.91–5.68
Gamma (α ¼ 761.99; λ ¼ 143.77) NCCS, Chan et al10
2.41–2.72
Gamma (α ¼ 10.24; λ ¼ 4.00)
NCCS, Chan et al10
3.48–9.52
Gamma (α ¼ 30.55; λ ¼ 4.70)
NCCS, Chan et al10
2.35–3.16
Gamma (α ¼ 192.36; λ ¼ 69.70) NCCS, Chan et al10
6.58–8.04 9.19–17.55
Gamma (α ¼ 388.23; λ ¼ 53.11) NCCS, Chan et al10 Gamma (α ¼ 42.29; λ ¼ 3.16) NCCS, Chan et al10
0.82–0.88
Beta (α ¼ 422.49; β ¼ 74.56)
NCCS, Wang et al16, Chan et al17
0.66–0.80
Beta (α ¼ 131.40; β ¼ 48.60)
NCCS, Wang et al16, Chan et al17
0.20–0.67
Beta (α ¼ 8.78; β ¼ 11.63)
NCCS, Wang et al16, Chan et al17
0.38–0.57 0.18–0.34
Beta (α ¼ 46.53; β ¼ 51.47) Beta (α ¼ 26.74; β ¼ 76.26)
Garcia-Carbonero et al13 Garcia-Carbonero et al13 (continued)
June 2017
ANC ¼ absolute neutrophil count; FN ¼ febrile neutropenia; NA ¼ not applicable; NCCS ¼ National Cancer Centre Singapore; PSA ¼ probabilistic sensitivity analysis; SGH ¼ Singapore General Hospital. * Inpatient treatment cost per day was calculated by subtracting total cost of filgrastim from total health care cost and subsequently divided by the corresponding length of stay.
NCCS, Chan et al10 NCCS, Chan et al10 Beta (α ¼ 4.82; β ¼ 236.16) Beta (α ¼ 8.88; β ¼ 25.26) 0.004–0.04 0.11–0.40 0.02 0.26
NCCS, Chan et al10 Beta (α ¼ 16.28; β ¼ 46.33) 0.16–0.37 0.26
NCCS, Chan et al10 Beta (α ¼ 20.31; β ¼ 182.77) 0.05–0.14 0.10
Serious complications (grade IV neutropenia r3 days) Serious complications (grade IV neutropenia 43 days) Case fatality rate FN without complications FN with complications
Parameter
Table I. (continued).
Mean
Range (95% CI)
PSA Distribution
Reference
X.J. Wang et al. QALY. A 10,000-iteration probabilistic sensitivity analysis (PSA) was also performed. The costeffectiveness acceptability curve was generated from the PSA.
Threshold Analysis Threshold analysis for the cost of G-CSF was conducted from a WTP threshold of US$50,000/ QALY. The cost of filgrastim (per injection) was varied upward from US$0 until the ICER of comparator (antibiotics with filgrastim) versus standard care (antibiotics alone) was equal to US$50,000/QALY.
Key Assumptions Several assumptions were made in this study. (1) Duration of FN and duration of therapeutic use of filgrastim were assumed to be equivalent to the duration of ANC recovery. (2) Duration of serious complication was assumed to be equivalent to the LOS after recovery from FN. (3) The therapeutic use of filgrastim did not affect the mortality rate. (4) The utility for those patients who died remained the lowest value of their health state throughout their hospitalization. (5) Patients returned to their baseline utility value once they recovered from low ANC and/or serious complications.
RESULTS Base-case Analysis Costs associated with comparator group and standard care group were US$9110 and US$9235, respectively (Table II). The QALYs associated with these two groups were 0.0450 and 0.0443, respectively. A cost saving of US$125 and an additional 0.0007 QALYs gained were generated with the comparator, compared with the standard care. The treatment strategy of antibiotics with therapeutic filgrastim was a dominant choice for managing FN compared with antibiotics alone.
Sensitivity Analyses In one-way sensitivity analyses, variations in all the variables except for probability of grade IV neutropenia 43 days and LOS for patients with serious complications yielded an ICER oUS$50,000/QALY (Figure 2). The use of filgrastim remained cost effective as long as the LOS in patients with serious complications is not shorter than 9.5 days.
1165
Clinical Therapeutics
Table II. Results of base-case analysis. Incremental Incremental Cost Effectiveness Cost Effectiveness ICER (US$, 2014) (QALYs) (US$, 2014) (QALYs) (US$/QALY)
Strategy Standard care (antibiotics alone) Comparator (antibiotics with therapeutic filgrastim)
9235 9110
0.0443 0.0450
— 125
— 0.0007
— Dominant
ICER ¼ incremental cost-effective ratio; QALY ¼ quality-adjusted life year.
Similarly, when the risk of developing serious complications in patients with grade IV neutropenia 43 days is 416.3%, the strategy of antibiotics with therapeutic filgrastim remains a cost-effective choice. The PSA has shown that the strategy of antibiotics with therapeutic filgrastim was the dominant choice in 76.9% of the model iterations (Figure 3). In addition, the therapeutic use of filgrastim remains as a preferred strategy (either cost-effective or dominant) in 83.7%
of the model iterations, compared with the standard care (antibiotics alone), at a WTP of US$50,000/ QALY (Figure 4).
Threshold Analysis Threshold analysis showed that the therapeutic use of filgrastim remained cost effective at a threshold of US$50,000/QALY if the cost of filgrastim was oUS $87 per injection.
Tornado Analysis (ICER)
WTP threshold of US$50,000
Length of stay; with complications Probability of serious complications (grade IV neutropenia > 3days) Inpatient treatment cost; with complications (per day) Probability of grade IV neutropenia > 3days (antibiotics alone) Probability of grade IV neutropenia > 3days (antibiotics with filgrastim) Inpatient treatment cost; without complications (per day) Utiliy for hospitalized patient with FN without serious complications Probability of serious complications (grade IV neutropenia ≤ 3days) Length of stay; without complications Case fatality rate for FN with complications Filgrastim 300μg (per injection) Utility for adult cancer patients undergoing chemotherapy without FN Utility for hospitalized patient with FN with serious complications Time to ANC recovery; Grade IV neutropenia > 3days (antibiotics alone) Time to ANC recovery; Grade IV neutropenia ≤ 3days (antibiotics alone) Case fatality rate for FN without complications Time to ANC recovery; Grade IV neutropenia > 3days (antibiotics with filgrastim) Time to ANC recovery; Grade IV neutropenia ≤ 3days (antibiotics with filgrastim)
EV: -175258.26244 -400000
-350000
-300000
-250000
-200000
-150000
-100000
-50000
0
50000
100000
Figure 2. Tornado diagram for one-way sensitivity analyses. ANC ¼ absolute neutrophil count; EV ¼ expected value; FN ¼ febrile neutropenia; ICER ¼ incremental cost-effectiveness ratio; WTP ¼ willingness to pay.
1166
Volume 39 Number 6
X.J. Wang et al.
Incremental Cost-Effectiveness, Comparator (antibiotics with filgrastim) vs. Standard Care (antibiotics alone) 250.00
(3)
200.00 150.00 100.00
(2)
50.00
0
0. 000
0.00 -50.00
P=
WT
50
-100.00 -150.00 -200.00
Incremental Cost
-250.00 -300.00 -350.00 -400.00 -450.00 -500.00
(1)
-550.00 -600.00 -650.00 -700.00 -750.00 -800.00 -850.00 -900.00 -950.00 -1000.00 -0.01
0.00
0.01
Incremental Effectiveness
Figure 3. Results of probabilistic sensitivity analyses showing scatter plots on the cost-effectiveness plane: incremental costs and quality-adjusted life year (QALY) of comparator (antibiotics with filgrastim) versus standard care (antibiotics alone) for the treatment of febrile neutropenia (FN). This graph is divided into several regions: (1) comparator dominates standard care (76.9%); (2) comparator is more costly and effective, and its ICER is less than or equal to the willingness to pay (WTP), so comparator is cost effective (6.8%); (3) comparator is more costly and effective, but its ICER is greater than the WTP, so comparator is not cost effective (15.8%).
DISCUSSION In view of the recent reduction on the listed price of filgrastim, this study was conducted to evaluate the cost-effectiveness of therapeutic filgrastim in the treatment of FN among patients diagnosed with solid tumors and lymphomas. Results suggested that the therapeutic use of filgrastim is a dominant strategy (cost saving and more QALYs gained) and should be used in most patients for the treatment of FN. In addition, as shown in the one-way sensitivity analysis, treatment with therapeutic filgrastim would remain as the preferred strategy when the patient’s risk of experiencing prolonged grade IV neutropenia (43 days) was 416.3% or the length of hospitalization for FN with serious complication was expected to be
June 2017
49.5 days. Therefore, the routine use of therapeutic filgrastim in FN patients with solid tumors and lymphomas appears reasonable on both clinical and economic grounds. Our suggestions on the routine use of filgrastim for treatment of FN differ from the recommendations provided by published guidelines and reports.3–5 This is partly due to the reduction of the listed price of filgrastim over the years. The mean cost of filgrastim in the early studies was about US$200 per dose compared with the current US$37 per injection of filgrastim 300 mg.13,20 The recommendations from guidelines were last updated in 2011,3–5 and they seemed acceptable at that time because the use of filgrastim for the treatment of FN may not be cost
1167
Clinical Therapeutics
CE Acceptability Curve 100% 95% 90%
83.7%
85%
WTP threshold of US$50,000
80% 75%
% Iterations Cost-Effective
70% 65% 60% 55% 50% 45% 40% 35% 30% 25% 20%
Therapeutic G-CSF Without Therapeutic G-CSF
15% 10% 5% 0% 0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
220000
240000
260000
280000
300000
Willingness-to-Pay
Figure 4. Cost-effectiveness acceptability curves. CE ¼ cost-effectiveness; G-CSF ¼ granulocyte colonystimulating factor; WTP ¼ willingness to pay. effective in patients at a low risk of FN complications, because the cost exceeded the threshold value of US $87 per injection of filgrastim 300 mg. However, in view of the decrease in the cost of filgrastim, it may be prudent to review these recommendations. In fact, the recently updated National Comprehensive Cancer Network guideline recommended the use of therapeutic G-CSF in combination with antibiotics in certain circumstances during the treatment of FN21: (1) if a patient received a short-acting G-CSF (such as filgrastim) as prophylaxis but developed FN, the shortacting G-CSF should be continued during the treatment of FN; (2) if a patient did not receive any G-CSF as prophylaxis before the development of FN, therapeutic G-CSF should be considered in patients with risk factors for serious complications. Furthermore, nivestim, a biosimilar filgrastim, was recently approved for clinical use in Singapore. This has major impact because the price of nivestim is lower than the current listed price of filgrastim, resulting in further cost savings. However, the long-term safety profile
1168
and efficacy data of nivestim are still unknown. Additional cost-effectiveness analyses are necessary to compare the cost-effectiveness of nivestim with its reference product (filgrastim). To the best of our knowledge, there is a lack of studies evaluating the cost-effectiveness of using G-CSF as treatment for established FN. However, a few studies have examined and found the cost-saving potential of adding therapeutic G-CSF to antibiotic treatment for FN.13,20,22,23 One recently published study assessed the potential cost implications of therapeutic G-CSF for hospitalized patients with FN.20 It was shown that the addition of daily G-CSF until the patient’s neutrophil recovery led to an estimated savings of US$1046 per FN episode compared with using antibiotics alone. This result remained robust in the PSAs, with 73.5% of 10,000 iterations indicating that therapeutic G-CSF was the cost-saving strategy. These findings advocate our main findings that therapeutic use of filgrastim is a dominant strategy compared with the standard care
Volume 39 Number 6
X.J. Wang et al. (antibiotics alone) and should be used among patients with solid tumor and lymphoma who require treatment of FN. It has been well recognized that not all FN patients are at the same level of risk of developing serious complications, and outpatient management can be considered among patients who are at low risk of FN complications, such as patients diagnosed with solid tumors and lymphomas.24 However, because of a lack of high-level evidence, outpatient management among low-risk patients was not strongly recommended by the current guidelines,4,24 and the adoption of outpatient management among low-risk FN patients is still relatively poor in the current clinical grounds. In a recently published study involving patients with solid tumors and lymphomas in the United States, 480% of the FN episodes among these low-risk patients were initially managed in an inpatient setting.25 Hence, early discharge is a main goal for the hospitalized low-risk FN patients because prolonged hospitalization could lead to unnecessary medical cost, reduced quality of life, and increased risk of developing hospital-acquired infections, especially considering that cancer incidence is expected to increase in the future, which can further strain limited health care resources.26 It is extremely important to identify appropriate ways for allocation of resources on managing cancer and chemotherapy-associated toxicities. Because therapeutic use of G-CSF (filgrastim) is cost effective, as found in this study, and allows patients to become clinically stable in a shorter time frame,13 routine use of therapeutic filgrastim for managing FN should be considered in this group of patients because this could promote early discharge among the low-risk FN patients and provide tremendous benefits to them. There are a number of limitations in this study. First, our model presumed daily filgrastim use until recovery of ANC (defined as 42.0 109/L), which may exceed current clinical practice. However, the duration of ANC recovery has little impact on the results as shown by one-way sensitivity analysis. Second, the adverse reactions associated with filgrastim, such as mild bone pain, were not included in the analysis. Note that these adverse reactions could lead to a reduction of a patient’s QALYs gained. However, considering that the cost of administrating nonopioid analgesics to manage these adverse reactions are relatively low and in our one-way sensitivity analyses
June 2017
the variations on utility parameters have not influenced our results, we expect these adverse effects would not affect the major findings in this study. In the best-case scenario, future studies should evaluate the impact of filgrastim-associated adverse reactions in treatment of FN and their impact of costeffectiveness.
CONCLUSIONS In conclusion, our study revealed that the therapeutic filgrastim, in conjunction with antibiotics, in the treatment of FN is cost effective. This provides evidence to support the routine use of filgrastim for the treatment of FN among adult cancer patients with solid tumors and lymphomas.
ACKNOWLEDGMENTS This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors. X.J. Wang and W.X. Tong contributed to study design, data collection, data analysis, writing the original draft, and revision to the manuscript. A. Chan contributed to study design, software, writing review and editing, supervision, and project administration. All authors have read and approved the final version of the manuscript.
CONFLICT OF INTEREST The authors declared no conflict of interest. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
REFERENCES 1. Chan A, Soh D, Ko Y, et al. Characteristics of unplanned hospital admissions due to drug-related problems in cancer patients. Supportive Care Cancer. 2014;22:1875–1881. 2. Wang XJ, Wong M, Hsu LY, Chan A. Costs associated with febrile neutropenia in solid tumor and lymphoma patients - an observational study in Singapore. BMC Health Serv Res. 2014;14:434. 3. Aapro MS, Bohlius J, Cameron DA, et al. 2010 update of EORTC guidelines for the use of granulocyte-colony stimulating factor to reduce the incidence of chemotherapy-induced febrile neutropenia in adult patients with lymphoproliferative disorders and solid tumours. Eur J Cancer. 2011;47:8–32.
1169
Clinical Therapeutics 4. de Naurois J, Novitzky-Basso I, Gill MJ, et al. Management of febrile neutropenia: ESMO Clinical Practice Guidelines. Ann Oncol. 2010;21 Suppl 5:v252–256. 5. Smith TJ, Khatcheressian J, Lyman GH, et al. 2006 update of recommendations for the use of white blood cell growth factors: an evidence-based clinical practice guideline. J Clin Oncol. 2006;24: 3187–3205. 6. Mhaskar R, Clark OA, Lyman G, et al. Colony-stimulating factors for chemotherapy-induced febrile neutropenia. Cochrane Database Syst Rev. 2014;(10):CD003039. 7. Amgen. 2015 Trends in Biosimilars Report. http://www.amgenbiotech. com/resources/2015_Trends_in_Bio similars_Repor t-83531R1V1.pdf [Accessed June 12, 2016]. 8. Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis. 2011;52:e56–e93. 9. National Cancer Institute and Department of Health and Human Services (2010) Common terminology criteria for adverse events (CTCAE) v4.03. http://evs.nci.nih. gov/f tp1/CTC AE/CTC AE_4. 0 3 _ 2 010 - 0 6 - 14 _ Q u i c k R e f e r e n ce_5x7.pdf [Accessed February 28, 2016]. 10. Chan A, Wong QX, Ali MK, et al. Clinical efficacy of adjunctive G-CSF on solid tumor and lymphoma patients with established febrile neutropenia. Support Care Cancer. 2014; 22:1105–1112. 11. Holmes FA, O'Shaughnessy JA, Vukelja S, et al. Blinded, randomized, multicenter study to evaluate single administration pegfilgrastim once per cycle versus daily filgrastim as an adjunct to chemotherapy in patients with high-risk stage II or stage III/IV breast cancer. J Clin Oncol. 2002;20:727–731.
1170
12. Bodey GP, Buckley M, Sathe YS, Freireich EJ. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann Intern Med. 1966;64:328–340. 13. Garcia-Carbonero R, Mayordomo JI, Tornamira MV, et al. Granulocyte colony-stimulating factor in the treatment of high-risk febrile neutropenia: a multicenter randomized trial. J Natl Cancer Inst. 2001;93:31–38. 14. Department of Statistics Singapore. Consumer Price Index. http://www.sing stat.gov.sg/statistics/browse-by-theme/ prices [Accessed February 28, 2016]. 15. The World Bank. PPP conversion factor, GDP. http://data.worldbank.org/ indicator/PA.NUS.PPP [Accessed February 28, 2016]. 16. Wang XJ, Wong CM, Chan A. Psychometric Properties of the Functional Assessment of Cancer Therapy-Neutropenia in Asian cancer patients with chemotherapyinduced neutropenia. J Pain Symptom Manage. 2016;52:428–436. 17. Chan A, Chiang YY, Low XH, et al. Affordability of cancer treatment for aging cancer patients in Singapore: an analysis of health, lifestyle, and financial burden. Support Care Cancer. 2013;21:3509–3517. 18. Oppe M, Devlin NJ, Szende A. EQ5D value sets: inventory, comparative review and user guide. New York: Springer; 2007. 19. Heydarnejad MS, Hassanpour DA, Solati DK. Factors affecting quality of life in cancer patients undergoing chemotherapy. Afr Health Sci. 2011; 11:266–270. 20. Cosler LE, Eldar-Lissai A, Culakova E, et al. Therapeutic use of granulocyte colony-stimulating factors for
21.
22.
23.
24.
25.
26.
established febrile neutropenia: effect on costs from a hospital perspective. Pharmacoeconomics. 2007;25:343–351. National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology (NCCN Guidelines). Myeloid Growth Factors. v2. 2016. https://www. nccn.org/professionals/physician_gls/ pdf/myeloid_growth.pdf [Accessed June 12, 2016]. Mayordomo JI, Rivera F, DiazPuente MT, et al. Improving treatment of chemotherapy-induced neutropenic fever by administration of colony-stimulating factors. J Natl Cancer Inst. 1995;87:803–808. Mitchell PL, Morland B, Stevens MC, et al. Granulocyte colonystimulating factor in established febrile neutropenia: a randomized study of pediatric patients. J Clin Oncol. 1997;15:1163–1170. Flowers CR, Seidenfeld J, Bow EJ, et al. Antimicrobial prophylaxis and outpatient management of fever and neutropenia in adults treated for malignancy: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2013;31:794–810. Weycker D, Barron R, Kartashov A, et al. Incidence, treatment, and consequences of chemotherapy-induced febrile neutropenia in the inpatient and outpatient settings. J Oncol Pharm Pract. 2014;20:190–198. Fitzmaurice C, Allen C, Barber RM, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 32 cancer groups, 1990 to 2015: a systematic analysis for the global burden of disease study. JAMA Oncol. 2017;3:524–548.
Address correspondence to: Alexandre Chan, PharmD, Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543. E-mail: [email protected]
Volume 39 Number 6
Therapeutic Use of Filgrastim for Established Febrile Neutropenia Is Cost Effective Among Patients With Solid Tumors and Lymphomas Xiao Jun Wang, BSc1,2,*; Wei Xiang Tong, BSc1,*; and Alexandre Chan, PharmD1,2 1
Department of Pharmacy, National University of Singapore, Singapore, Singapore; and 2Department of Pharmacy, National Cancer Centre Singapore, Singapore, Singapore
ABSTRACT Purpose: With the emergence of biosimilar filgrastim to the market, there is a gradual decrease in the listed price of the originator product of filgrastim over the years, and this could have an impact on the costeffectiveness of filgrastim in the treatment of febrile neutropenia (FN). A cost-effectiveness analysis would allow clinicians to make informed decision when considering the therapeutic filgrastim among low-risk FN patients. This study aims to evaluate the costeffectiveness of adding therapeutic filgrastim to antibiotics in the treatment of established FN among patients with solid tumors and lymphomas. Methods: A decision tree model was created to compare two treatment options for established FN as follows: (1) antibiotics alone (standard care) and (2) antibiotics with therapeutic filgrastim (comparator). The target population was a hypothetical cohort of adult cancer patients with solid tumors or lymphomas hospitalized with FN in Singapore. The analysis was performed from a hospital’s perspective over a 21-day time horizon. The main outcome measures included costs, quality-adjusted life year (QALY) and incremental cost-effectiveness ratio (ICER). One-way sensitivity analysis and probabilistic sensitivity analysis were conducted to evaluate the robustness of the results. Findings: Compared with antibiotics alone, the treatment strategy of antibiotics with therapeutic filgrastim was a dominant choice, incurring a cost saving of US$125 per patient (comparator versus standard care: US$9110 versus US$9235) and additional health benefit of 0.0007 QALY gained per patient (comparator versus standard care: 0.0450 versus 0.0443). Model results were robust against
*
These authors contributed equally to this work.
June 2017
the parameter variations in the one-way sensitivity analyses, but increasing the cost of filgrastim beyond US$87 per injection would increase the ICER to 4US $50,000/QALY. Furthermore, the strategy of antibiotics with therapeutic filgrastim was the preferred choice (dominant or cost-effective) in 83.7% of the model iterations at a willingness-to-pay threshold of US$50,000/QALY. Implications: From a hospital’s perspective, the therapeutic filgrastim, in conjunction with antibiotics, in the treatment of FN is cost effective. This provides evidence to support the routine use of filgrastim for the treatment of FN among adult cancer patients with solid tumors and lymphomas. (Clin Ther. 2017;39:1161–1170) & 2017 Elsevier HS Journals, Inc. All rights reserved. Key words: cost-effectiveness analysis, febrile neutropenia, filgrastim, treatment.
INTRODUCTION Febrile neutropenia (FN) is a serious oncologic emergency experienced by cancer patients receiving myelosuppressive chemotherapy. One recent study in Singapore identified that FN had an incidence of around 19.5% among patients with solid tumors or lymphomas and that FN was the most common reason for unwanted hospitalizations.1 FN is associated with a significant economic burden on patients. In Singapore, we have reported that the cost of inpatient FN management was around US $4193 per episode, and FN patients who had Accepted for publication May 11, 2017. http://dx.doi.org/10.1016/j.clinthera.2017.05.345 0149-2918/$ - see front matter & 2017 Elsevier HS Journals, Inc. All rights reserved.
1161
Clinical Therapeutics developed severe sepsis were associated with a significantly higher economic burden.2 Filgrastim is a granulocyte colony-stimulating factor (G-CSF) that stimulates neutrophil production. The prophylactic use of G-CSF, including filgrastim, has been supported by the current guideline to prevent FN.3 However, the therapeutic use of G-CSF for established FN is not recommended among low-risk FN patients, such as patients with solid tumors or lymphomas.3–5 One recent meta-analysis of randomized controlled trials suggested that patients receiving G-CSF plus antibiotics had shorter duration of FN and antibiotics use than patients receiving antibiotics alone.6 Therefore, the therapeutic use of filgrastim is expected to reduce patient’s hospital length of stay (LOS) and to provide benefits on cost savings associated with reduced hospitalization and antibiotics use. With the emergence of biosimilar filgrastim to the market, a gradual decrease in the listed price of the originator product of filgrastim has occurred over the years, and this could have an impact on the costeffectiveness of filgrastim in the treatment of FN.7 To our knowledge, cost-effectiveness studies that focus on the therapeutic use of filgrastim for established FN among patients with solid tumors and lymphomas are lacking. A cost-effectiveness analysis would allow clinicians to make informed decision when considering the therapeutic filgrastim among low-risk FN patients. Therefore, we designed this study with the primary objective to evaluate the cost-effectiveness of therapeutic filgrastim in the treatment of FN among adult patients with solid tumors and lymphomas.
PATIENTS AND METHODS A decision tree model was constructed by using TreeAge Pro 2013 (TreeAge Software, Inc, Williamstown, Massachusetts) to compare two treatment options for FN as follows: (1) antibiotics alone (standard care) and (2) antibiotics with therapeutic filgrastim (comparator). The antibiotic treatment for FN in both standard care and comparator arms followed the approaches recommended by the Infectious Disease Society of America guideline.8 Empiric antibiotic therapy included an antipseudomonal β-lactam agent, such as cefepime, carbapenem (meropenem or imipenem-cilastatin), or piperacillin-tazobactam.8 Other antibiotics, such as aminoglycosides, fluoroquinolones,
1162
and/or vancomycin, may be considered in patients with complicated presentations (eg, hypotension and pneumonia) or suspected/proven antimicrobial resistance.8 In the comparator group, the therapeutic filgrastim was administered daily until patients’ absolute neutrophil count (ANC) recovered (ANC Z2.0 109/L). In this model, the time horizon was set as 21 days to represent the interval between each chemotherapy cycle. No discounting was applied because of the short time horizon in this model. The perspective was taken from a Singaporean hospital, and only direct medical cost was considered in this study.
Definitions FN was defined as a single episode of fever Z38.31C or fever Z38.01C for at least 1 hour and an ANC of o0.5 109/L.8 Grade IV neutropenia was set as having an ANC of o0.5 109/L.9 The duration of grade IV neutropenia was the number of days with ANC o0.5 109/L. Time to ANC recovery was the duration from FN onset date until patients’ ANC increased to 2.0 109/L. In this study, serious complication was defined as the occurrence of severe sepsis during the hospitalization. LOS was the overall duration from FN onset date until patient was discharged from the hospital.
Model Structure The model structure is shown in Figure 1. The design of this model is based on the Infectious Disease Society of America guideline8 and clinicians’ discussion to reflect the local treatment pattern for established FN, and the face validity of this model was endorsed by the clinicians who manage FN. Several studies have reported the effectiveness of filgrastim on reducing the duration of grade IV neutropenia.6,10 Therefore, the duration of grade IV neutropenia was included in this model. The duration of grade IV neutropenia was categorized into the following two groups: (1) grade IV neutropenia r3 days and (2) grade IV neutropenia 43 days, which was based on a previous study in Singapore, showing that the mean duration of grade IV neutropenia was around 2 to 3 days.11 Serious complication was considered because prolonged duration of grade IV neutropenia would increase the risk of a patient developing serious complications.12
Volume 39 Number 6
X.J. Wang et al.
Alive Serious Complication Death
Grade IV neutropenia > 3days Alive Without Serious Complication Death
Antibiotics with Filgrastim Alive Serious Complication Death
Grade IV neutropenia <= 3days
Established FN
Alive Without Serious Complication Death Antibiotics alone
Figure 1. Decision tree model structure. FN ¼ febrile neutropenia.
Model Inputs Probabilities The model inputs used in this study are shown in Table I. Probability estimates needed for the analysis were mainly obtained from a database consisting of 305 adult cancer patients diagnosed with solid tumors or lymphomas at National Cancer Centre Singapore (NCCS) with documented FN-related hospitalizations between 2009 and 2012.10 Because of a lack of relevant data in Singapore, the probabilities of grade IV neutropenia for 43 days (with or without therapeutic filgrastim) were generated from a published randomized control trial.13 In that study, a total of 210 patients with solid tumors and developed FN were included to evaluate the efficacy of adding G-CSF to broad-spectrum antibiotic treatment for FN.13 For patients assigned to the intervention group, the additional G-CSF was initiated within 12 hours of the diagnosis of FN (G-CSF at a dose of 5 μg/kg/d) until patients’ ANC increased to 41000/mm3.13 Findings of that study suggested that adding G-CSF to antibiotic therapy could significantly reduce the duration of grade IV neutropenia compared with antibiotics alone.13
were adjusted to 2014 Singapore dollars (SG$) using the healthcare component of the Singapore Consumer Price Index.14 The cost data were then converted to 2014 US dollars (US$) using purchasing power parity conversion rate obtained from the World Bank (2014; SG$1 ¼ US$0.86).15
Utilities Two previous studies (both conducted in Singapore) were used to generate utility estimates.16,17 The first consisted of 521 adult patients with solid tumors and lymphomas without FN.17 The second consisted of 102 patients with solid tumors and lymphomas hospitalized for FN between 2014 and 2015, of which, 26 developed severe sepsis.16 Utilities were assessed using the three-level EuroQoL Group's fivedimension questionnaire and calculated using a Japanese value set.18 Baseline health utility value was obtained from the first database by matching with the second database in terms of cancer type and stage because these were the factors that had been shown to affect quality of life in cancer patients undergoing chemotherapy.19 Utility values of each health state were computed from the pooled data from these two databases.16,17
Costs Cost data were retrieved from a FN database at NCCS.2 The cost for each filgrastim injection (300 mg) was obtained from the Department of Pharmacy at Singapore General Hospital because patients with FN from NCCS were admitted to Singapore General Hospital for management. Components of direct medical cost included medication, laboratory, radiography, ward costs, and surgical costs. All costs
June 2017
Cost-effectiveness Analysis and Sensitivity Analyses Costs, quality-adjusted life year (QALY) gained, and the incremental cost-effectiveness ratio (ICER) were calculated in the base-case analysis. One-way sensitivity analyses were conducted on all parameters by varying each variable within its plausible range, at a willingness-to-pay (WTP) threshold of US$50,000/
1163
Clinical Therapeutics
1164
Table I. Model inputs. Parameter Costs (US$, 2014) Filgrastim 300 μg (per injection)
Mean 37
Volume 39 Number 6
Inpatient treatment 1100 Without complications (per day)* * With complications (per day) 1115 Time parameter (day) Time horizon 21 Time to ANC recovery Grade IV neutropenia 43 days (antibiotics 5.30 with filgrastim) Grade IV neutropenia r3 days (antibiotics 2.56 with filgrastim) Grade IV neutropenia 43 days (antibiotics 6.50 alone) Grade IV neutropenia r3 days (antibiotics 2.76 alone) Length of stay Without complications 7.31 With complications 13.37 Utility Adult cancer patients undergoing chemotherapy 0.85 without FN Hospitalized patient with FN without serious 0.73 complications Hospitalized patient with FN with serious 0.43 complications Probability Grade IV neutropenia 43 days (antibiotics alone) 0.47 Grade IV neutropenia 43 days (antibiotics with 0.26 filgrastim)
Range (95% CI) 29–44 (⫾20%) 858–1342 903–1326 NA
PSA Distribution Normal
Reference SGH
Gamma (α ¼ 79.98; λ ¼ 0.073) NCCS, Wang et al2 Gamma (α ¼ 117.19; λ ¼ 0.11) NCCS, Wang et al2 NA
Assumed
4.91–5.68
Gamma (α ¼ 761.99; λ ¼ 143.77) NCCS, Chan et al10
2.41–2.72
Gamma (α ¼ 10.24; λ ¼ 4.00)
NCCS, Chan et al10
3.48–9.52
Gamma (α ¼ 30.55; λ ¼ 4.70)
NCCS, Chan et al10
2.35–3.16
Gamma (α ¼ 192.36; λ ¼ 69.70) NCCS, Chan et al10
6.58–8.04 9.19–17.55
Gamma (α ¼ 388.23; λ ¼ 53.11) NCCS, Chan et al10 Gamma (α ¼ 42.29; λ ¼ 3.16) NCCS, Chan et al10
0.82–0.88
Beta (α ¼ 422.49; β ¼ 74.56)
NCCS, Wang et al16, Chan et al17
0.66–0.80
Beta (α ¼ 131.40; β ¼ 48.60)
NCCS, Wang et al16, Chan et al17
0.20–0.67
Beta (α ¼ 8.78; β ¼ 11.63)
NCCS, Wang et al16, Chan et al17
0.38–0.57 0.18–0.34
Beta (α ¼ 46.53; β ¼ 51.47) Beta (α ¼ 26.74; β ¼ 76.26)
Garcia-Carbonero et al13 Garcia-Carbonero et al13 (continued)
June 2017
ANC ¼ absolute neutrophil count; FN ¼ febrile neutropenia; NA ¼ not applicable; NCCS ¼ National Cancer Centre Singapore; PSA ¼ probabilistic sensitivity analysis; SGH ¼ Singapore General Hospital. * Inpatient treatment cost per day was calculated by subtracting total cost of filgrastim from total health care cost and subsequently divided by the corresponding length of stay.
NCCS, Chan et al10 NCCS, Chan et al10 Beta (α ¼ 4.82; β ¼ 236.16) Beta (α ¼ 8.88; β ¼ 25.26) 0.004–0.04 0.11–0.40 0.02 0.26
NCCS, Chan et al10 Beta (α ¼ 16.28; β ¼ 46.33) 0.16–0.37 0.26
NCCS, Chan et al10 Beta (α ¼ 20.31; β ¼ 182.77) 0.05–0.14 0.10
Serious complications (grade IV neutropenia r3 days) Serious complications (grade IV neutropenia 43 days) Case fatality rate FN without complications FN with complications
Parameter
Table I. (continued).
Mean
Range (95% CI)
PSA Distribution
Reference
X.J. Wang et al. QALY. A 10,000-iteration probabilistic sensitivity analysis (PSA) was also performed. The costeffectiveness acceptability curve was generated from the PSA.
Threshold Analysis Threshold analysis for the cost of G-CSF was conducted from a WTP threshold of US$50,000/ QALY. The cost of filgrastim (per injection) was varied upward from US$0 until the ICER of comparator (antibiotics with filgrastim) versus standard care (antibiotics alone) was equal to US$50,000/QALY.
Key Assumptions Several assumptions were made in this study. (1) Duration of FN and duration of therapeutic use of filgrastim were assumed to be equivalent to the duration of ANC recovery. (2) Duration of serious complication was assumed to be equivalent to the LOS after recovery from FN. (3) The therapeutic use of filgrastim did not affect the mortality rate. (4) The utility for those patients who died remained the lowest value of their health state throughout their hospitalization. (5) Patients returned to their baseline utility value once they recovered from low ANC and/or serious complications.
RESULTS Base-case Analysis Costs associated with comparator group and standard care group were US$9110 and US$9235, respectively (Table II). The QALYs associated with these two groups were 0.0450 and 0.0443, respectively. A cost saving of US$125 and an additional 0.0007 QALYs gained were generated with the comparator, compared with the standard care. The treatment strategy of antibiotics with therapeutic filgrastim was a dominant choice for managing FN compared with antibiotics alone.
Sensitivity Analyses In one-way sensitivity analyses, variations in all the variables except for probability of grade IV neutropenia 43 days and LOS for patients with serious complications yielded an ICER oUS$50,000/QALY (Figure 2). The use of filgrastim remained cost effective as long as the LOS in patients with serious complications is not shorter than 9.5 days.
1165
Clinical Therapeutics
Table II. Results of base-case analysis. Incremental Incremental Cost Effectiveness Cost Effectiveness ICER (US$, 2014) (QALYs) (US$, 2014) (QALYs) (US$/QALY)
Strategy Standard care (antibiotics alone) Comparator (antibiotics with therapeutic filgrastim)
9235 9110
0.0443 0.0450
— 125
— 0.0007
— Dominant
ICER ¼ incremental cost-effective ratio; QALY ¼ quality-adjusted life year.
Similarly, when the risk of developing serious complications in patients with grade IV neutropenia 43 days is 416.3%, the strategy of antibiotics with therapeutic filgrastim remains a cost-effective choice. The PSA has shown that the strategy of antibiotics with therapeutic filgrastim was the dominant choice in 76.9% of the model iterations (Figure 3). In addition, the therapeutic use of filgrastim remains as a preferred strategy (either cost-effective or dominant) in 83.7%
of the model iterations, compared with the standard care (antibiotics alone), at a WTP of US$50,000/ QALY (Figure 4).
Threshold Analysis Threshold analysis showed that the therapeutic use of filgrastim remained cost effective at a threshold of US$50,000/QALY if the cost of filgrastim was oUS $87 per injection.
Tornado Analysis (ICER)
WTP threshold of US$50,000
Length of stay; with complications Probability of serious complications (grade IV neutropenia > 3days) Inpatient treatment cost; with complications (per day) Probability of grade IV neutropenia > 3days (antibiotics alone) Probability of grade IV neutropenia > 3days (antibiotics with filgrastim) Inpatient treatment cost; without complications (per day) Utiliy for hospitalized patient with FN without serious complications Probability of serious complications (grade IV neutropenia ≤ 3days) Length of stay; without complications Case fatality rate for FN with complications Filgrastim 300μg (per injection) Utility for adult cancer patients undergoing chemotherapy without FN Utility for hospitalized patient with FN with serious complications Time to ANC recovery; Grade IV neutropenia > 3days (antibiotics alone) Time to ANC recovery; Grade IV neutropenia ≤ 3days (antibiotics alone) Case fatality rate for FN without complications Time to ANC recovery; Grade IV neutropenia > 3days (antibiotics with filgrastim) Time to ANC recovery; Grade IV neutropenia ≤ 3days (antibiotics with filgrastim)
EV: -175258.26244 -400000
-350000
-300000
-250000
-200000
-150000
-100000
-50000
0
50000
100000
Figure 2. Tornado diagram for one-way sensitivity analyses. ANC ¼ absolute neutrophil count; EV ¼ expected value; FN ¼ febrile neutropenia; ICER ¼ incremental cost-effectiveness ratio; WTP ¼ willingness to pay.
1166
Volume 39 Number 6
X.J. Wang et al.
Incremental Cost-Effectiveness, Comparator (antibiotics with filgrastim) vs. Standard Care (antibiotics alone) 250.00
(3)
200.00 150.00 100.00
(2)
50.00
0
0. 000
0.00 -50.00
P=
WT
50
-100.00 -150.00 -200.00
Incremental Cost
-250.00 -300.00 -350.00 -400.00 -450.00 -500.00
(1)
-550.00 -600.00 -650.00 -700.00 -750.00 -800.00 -850.00 -900.00 -950.00 -1000.00 -0.01
0.00
0.01
Incremental Effectiveness
Figure 3. Results of probabilistic sensitivity analyses showing scatter plots on the cost-effectiveness plane: incremental costs and quality-adjusted life year (QALY) of comparator (antibiotics with filgrastim) versus standard care (antibiotics alone) for the treatment of febrile neutropenia (FN). This graph is divided into several regions: (1) comparator dominates standard care (76.9%); (2) comparator is more costly and effective, and its ICER is less than or equal to the willingness to pay (WTP), so comparator is cost effective (6.8%); (3) comparator is more costly and effective, but its ICER is greater than the WTP, so comparator is not cost effective (15.8%).
DISCUSSION In view of the recent reduction on the listed price of filgrastim, this study was conducted to evaluate the cost-effectiveness of therapeutic filgrastim in the treatment of FN among patients diagnosed with solid tumors and lymphomas. Results suggested that the therapeutic use of filgrastim is a dominant strategy (cost saving and more QALYs gained) and should be used in most patients for the treatment of FN. In addition, as shown in the one-way sensitivity analysis, treatment with therapeutic filgrastim would remain as the preferred strategy when the patient’s risk of experiencing prolonged grade IV neutropenia (43 days) was 416.3% or the length of hospitalization for FN with serious complication was expected to be
June 2017
49.5 days. Therefore, the routine use of therapeutic filgrastim in FN patients with solid tumors and lymphomas appears reasonable on both clinical and economic grounds. Our suggestions on the routine use of filgrastim for treatment of FN differ from the recommendations provided by published guidelines and reports.3–5 This is partly due to the reduction of the listed price of filgrastim over the years. The mean cost of filgrastim in the early studies was about US$200 per dose compared with the current US$37 per injection of filgrastim 300 mg.13,20 The recommendations from guidelines were last updated in 2011,3–5 and they seemed acceptable at that time because the use of filgrastim for the treatment of FN may not be cost
1167
Clinical Therapeutics
CE Acceptability Curve 100% 95% 90%
83.7%
85%
WTP threshold of US$50,000
80% 75%
% Iterations Cost-Effective
70% 65% 60% 55% 50% 45% 40% 35% 30% 25% 20%
Therapeutic G-CSF Without Therapeutic G-CSF
15% 10% 5% 0% 0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
220000
240000
260000
280000
300000
Willingness-to-Pay
Figure 4. Cost-effectiveness acceptability curves. CE ¼ cost-effectiveness; G-CSF ¼ granulocyte colonystimulating factor; WTP ¼ willingness to pay. effective in patients at a low risk of FN complications, because the cost exceeded the threshold value of US $87 per injection of filgrastim 300 mg. However, in view of the decrease in the cost of filgrastim, it may be prudent to review these recommendations. In fact, the recently updated National Comprehensive Cancer Network guideline recommended the use of therapeutic G-CSF in combination with antibiotics in certain circumstances during the treatment of FN21: (1) if a patient received a short-acting G-CSF (such as filgrastim) as prophylaxis but developed FN, the shortacting G-CSF should be continued during the treatment of FN; (2) if a patient did not receive any G-CSF as prophylaxis before the development of FN, therapeutic G-CSF should be considered in patients with risk factors for serious complications. Furthermore, nivestim, a biosimilar filgrastim, was recently approved for clinical use in Singapore. This has major impact because the price of nivestim is lower than the current listed price of filgrastim, resulting in further cost savings. However, the long-term safety profile
1168
and efficacy data of nivestim are still unknown. Additional cost-effectiveness analyses are necessary to compare the cost-effectiveness of nivestim with its reference product (filgrastim). To the best of our knowledge, there is a lack of studies evaluating the cost-effectiveness of using G-CSF as treatment for established FN. However, a few studies have examined and found the cost-saving potential of adding therapeutic G-CSF to antibiotic treatment for FN.13,20,22,23 One recently published study assessed the potential cost implications of therapeutic G-CSF for hospitalized patients with FN.20 It was shown that the addition of daily G-CSF until the patient’s neutrophil recovery led to an estimated savings of US$1046 per FN episode compared with using antibiotics alone. This result remained robust in the PSAs, with 73.5% of 10,000 iterations indicating that therapeutic G-CSF was the cost-saving strategy. These findings advocate our main findings that therapeutic use of filgrastim is a dominant strategy compared with the standard care
Volume 39 Number 6
X.J. Wang et al. (antibiotics alone) and should be used among patients with solid tumor and lymphoma who require treatment of FN. It has been well recognized that not all FN patients are at the same level of risk of developing serious complications, and outpatient management can be considered among patients who are at low risk of FN complications, such as patients diagnosed with solid tumors and lymphomas.24 However, because of a lack of high-level evidence, outpatient management among low-risk patients was not strongly recommended by the current guidelines,4,24 and the adoption of outpatient management among low-risk FN patients is still relatively poor in the current clinical grounds. In a recently published study involving patients with solid tumors and lymphomas in the United States, 480% of the FN episodes among these low-risk patients were initially managed in an inpatient setting.25 Hence, early discharge is a main goal for the hospitalized low-risk FN patients because prolonged hospitalization could lead to unnecessary medical cost, reduced quality of life, and increased risk of developing hospital-acquired infections, especially considering that cancer incidence is expected to increase in the future, which can further strain limited health care resources.26 It is extremely important to identify appropriate ways for allocation of resources on managing cancer and chemotherapy-associated toxicities. Because therapeutic use of G-CSF (filgrastim) is cost effective, as found in this study, and allows patients to become clinically stable in a shorter time frame,13 routine use of therapeutic filgrastim for managing FN should be considered in this group of patients because this could promote early discharge among the low-risk FN patients and provide tremendous benefits to them. There are a number of limitations in this study. First, our model presumed daily filgrastim use until recovery of ANC (defined as 42.0 109/L), which may exceed current clinical practice. However, the duration of ANC recovery has little impact on the results as shown by one-way sensitivity analysis. Second, the adverse reactions associated with filgrastim, such as mild bone pain, were not included in the analysis. Note that these adverse reactions could lead to a reduction of a patient’s QALYs gained. However, considering that the cost of administrating nonopioid analgesics to manage these adverse reactions are relatively low and in our one-way sensitivity analyses
June 2017
the variations on utility parameters have not influenced our results, we expect these adverse effects would not affect the major findings in this study. In the best-case scenario, future studies should evaluate the impact of filgrastim-associated adverse reactions in treatment of FN and their impact of costeffectiveness.
CONCLUSIONS In conclusion, our study revealed that the therapeutic filgrastim, in conjunction with antibiotics, in the treatment of FN is cost effective. This provides evidence to support the routine use of filgrastim for the treatment of FN among adult cancer patients with solid tumors and lymphomas.
ACKNOWLEDGMENTS This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors. X.J. Wang and W.X. Tong contributed to study design, data collection, data analysis, writing the original draft, and revision to the manuscript. A. Chan contributed to study design, software, writing review and editing, supervision, and project administration. All authors have read and approved the final version of the manuscript.
CONFLICT OF INTEREST The authors declared no conflict of interest. The authors have indicated that they have no other conflicts of interest regarding the content of this article.
REFERENCES 1. Chan A, Soh D, Ko Y, et al. Characteristics of unplanned hospital admissions due to drug-related problems in cancer patients. Supportive Care Cancer. 2014;22:1875–1881. 2. Wang XJ, Wong M, Hsu LY, Chan A. Costs associated with febrile neutropenia in solid tumor and lymphoma patients - an observational study in Singapore. BMC Health Serv Res. 2014;14:434. 3. Aapro MS, Bohlius J, Cameron DA, et al. 2010 update of EORTC guidelines for the use of granulocyte-colony stimulating factor to reduce the incidence of chemotherapy-induced febrile neutropenia in adult patients with lymphoproliferative disorders and solid tumours. Eur J Cancer. 2011;47:8–32.
1169
Clinical Therapeutics 4. de Naurois J, Novitzky-Basso I, Gill MJ, et al. Management of febrile neutropenia: ESMO Clinical Practice Guidelines. Ann Oncol. 2010;21 Suppl 5:v252–256. 5. Smith TJ, Khatcheressian J, Lyman GH, et al. 2006 update of recommendations for the use of white blood cell growth factors: an evidence-based clinical practice guideline. J Clin Oncol. 2006;24: 3187–3205. 6. Mhaskar R, Clark OA, Lyman G, et al. Colony-stimulating factors for chemotherapy-induced febrile neutropenia. Cochrane Database Syst Rev. 2014;(10):CD003039. 7. Amgen. 2015 Trends in Biosimilars Report. http://www.amgenbiotech. com/resources/2015_Trends_in_Bio similars_Repor t-83531R1V1.pdf [Accessed June 12, 2016]. 8. Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the Infectious Diseases Society of America. Clin Infect Dis. 2011;52:e56–e93. 9. National Cancer Institute and Department of Health and Human Services (2010) Common terminology criteria for adverse events (CTCAE) v4.03. http://evs.nci.nih. gov/f tp1/CTC AE/CTC AE_4. 0 3 _ 2 010 - 0 6 - 14 _ Q u i c k R e f e r e n ce_5x7.pdf [Accessed February 28, 2016]. 10. Chan A, Wong QX, Ali MK, et al. Clinical efficacy of adjunctive G-CSF on solid tumor and lymphoma patients with established febrile neutropenia. Support Care Cancer. 2014; 22:1105–1112. 11. Holmes FA, O'Shaughnessy JA, Vukelja S, et al. Blinded, randomized, multicenter study to evaluate single administration pegfilgrastim once per cycle versus daily filgrastim as an adjunct to chemotherapy in patients with high-risk stage II or stage III/IV breast cancer. J Clin Oncol. 2002;20:727–731.
1170
12. Bodey GP, Buckley M, Sathe YS, Freireich EJ. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann Intern Med. 1966;64:328–340. 13. Garcia-Carbonero R, Mayordomo JI, Tornamira MV, et al. Granulocyte colony-stimulating factor in the treatment of high-risk febrile neutropenia: a multicenter randomized trial. J Natl Cancer Inst. 2001;93:31–38. 14. Department of Statistics Singapore. Consumer Price Index. http://www.sing stat.gov.sg/statistics/browse-by-theme/ prices [Accessed February 28, 2016]. 15. The World Bank. PPP conversion factor, GDP. http://data.worldbank.org/ indicator/PA.NUS.PPP [Accessed February 28, 2016]. 16. Wang XJ, Wong CM, Chan A. Psychometric Properties of the Functional Assessment of Cancer Therapy-Neutropenia in Asian cancer patients with chemotherapyinduced neutropenia. J Pain Symptom Manage. 2016;52:428–436. 17. Chan A, Chiang YY, Low XH, et al. Affordability of cancer treatment for aging cancer patients in Singapore: an analysis of health, lifestyle, and financial burden. Support Care Cancer. 2013;21:3509–3517. 18. Oppe M, Devlin NJ, Szende A. EQ5D value sets: inventory, comparative review and user guide. New York: Springer; 2007. 19. Heydarnejad MS, Hassanpour DA, Solati DK. Factors affecting quality of life in cancer patients undergoing chemotherapy. Afr Health Sci. 2011; 11:266–270. 20. Cosler LE, Eldar-Lissai A, Culakova E, et al. Therapeutic use of granulocyte colony-stimulating factors for
21.
22.
23.
24.
25.
26.
established febrile neutropenia: effect on costs from a hospital perspective. Pharmacoeconomics. 2007;25:343–351. National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology (NCCN Guidelines). Myeloid Growth Factors. v2. 2016. https://www. nccn.org/professionals/physician_gls/ pdf/myeloid_growth.pdf [Accessed June 12, 2016]. Mayordomo JI, Rivera F, DiazPuente MT, et al. Improving treatment of chemotherapy-induced neutropenic fever by administration of colony-stimulating factors. J Natl Cancer Inst. 1995;87:803–808. Mitchell PL, Morland B, Stevens MC, et al. Granulocyte colonystimulating factor in established febrile neutropenia: a randomized study of pediatric patients. J Clin Oncol. 1997;15:1163–1170. Flowers CR, Seidenfeld J, Bow EJ, et al. Antimicrobial prophylaxis and outpatient management of fever and neutropenia in adults treated for malignancy: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2013;31:794–810. Weycker D, Barron R, Kartashov A, et al. Incidence, treatment, and consequences of chemotherapy-induced febrile neutropenia in the inpatient and outpatient settings. J Oncol Pharm Pract. 2014;20:190–198. Fitzmaurice C, Allen C, Barber RM, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 32 cancer groups, 1990 to 2015: a systematic analysis for the global burden of disease study. JAMA Oncol. 2017;3:524–548.
Address correspondence to: Alexandre Chan, PharmD, Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543. E-mail: [email protected]
Volume 39 Number 6