- Email: [email protected]
Cost-Effectiveness of Extended Adjuvant Rituximab for US Patients Aged 65–70 Years with Follicular Lymphoma in Second Remission
Original Contribution Cost-Effectiveness of Extended Adjuvant Rituximab for US Patients Aged 65-70 Years with Follicular Lymphoma in Second Remission John W. Hayslip,1 Kit N. Simpson2
Abstract Purpose: A recent Intergroup trial showed that receiving adjuvant rituximab after having a second remission from follicular lymphoma (FL) improved progression-free and overall survival (OS). The current study was conducted to determine the incremental cost-effectiveness ratio of this strategy in the United States. Patients and Methods: We constructed a transition state model to estimate the incremental cost-effectiveness ratio of extended adjuvant rituximab for 2 years for patients having a second FL remission. Event-free and OS rates were estimated from the recently published Intergroup trial. These were adjusted to reflect the contribution of non-cancer–specific mortality for patients aged 65-70 years, a more commonly affected age group than in the Intergroup trial, which had a median age of 54 years. Previously reported quality of life and cost estimates were obtained from peer-reviewed sources. Results: Five years after a second induction with R-CHOP (rituximab with cyclophosphamide/doxorubicin/vincristine/ prednisone), disease-free survival is expected to be 47% and 22%, and the OS rates are estimated to be 73% and 61% for extended adjuvant rituximab and observation, respectively, during the second remission. The discounted incremental cost-effectiveness ratio for the addition of adjuvant rituximab is estimated to be $19,522 per quality-adjusted life-years gained. Conclusion: Extended adjuvant rituximab offers a clinical benefit to patients aged 65-70 years who have a second remission from FL at a cost generally acceptable to the US healthcare system.
Clinical Lymphoma & Myeloma, Vol. 8, No. 3, 166-170, 2008; DOI: 10.3816/CLM.2008.n.020 Key words: CHOP, Health state utility, Indolent lymphoma, Salvage therapy
Introduction Follicular lymphoma (FL) is the most common indolent non-Hodgkin lymphoma. Follicular lymphoma has an average annual incidence of 2.6 per 100,000 people, afflicting an estimated 8000 new Americans in 2007.1 Many patients will have a remission with their initial treatment, but eventual relapse is expected as part of the natural history of FL. Although many treatment options exist, the median survival of patients diagnosed with FL is 7.8 years.2 Despite the numerous available treatment options for patients with FL, there is little consensus regarding the optimal treatment algorithm for patients as they progress through the various stages of their disease course. Indeed, the National Comprehensive Cancer Network practice guidelines endorse approaches ranging from observation to multiagent 1Markey Cancer Center, University of Kentucky, Lexington 2Center for Health Economic and Policy Studies, College of
chemotherapy for the initial treatment of FL and include autologous and allogeneic bone marrow transplantation (BMT) as reasonable second-line treatments.3 The heterogeneity of these recommendations reflects the variability of the disease and our limited understanding of how to optimally use our current treatments to maximize health while minimizing illness and expense. Recently, a new strategy of extended adjuvant rituximab given to patients with FL in their second remission was reported to improve disease-free survival (DFS) and overall survival (OS) compared with standard observation.4 Patients, physicians, and payers will have to struggle to incorporate these new findings as they consider the myriad options for treatment of patients with FL. To aid in these complex decisions, we constructed a transition state model to estimate the cost and patient benefit of extended adjuvant rituximab therapy during second remission for patients with FL aged 65-70 years in the United States.
Health Professions,
Medical University of South Carolina, Charleston
Patients and Methods
Submitted: Mar 27, 2008; Accepted: Apr 7, 2008
We restricted this study to US patients aged 65-70 years (the age range with the most cases of FL5) in their second remission from FL. A transition state model was created to calculate the incremental cost-effectiveness ratio of adjuvant rituximab
Address for correspondence: John W. Hayslip, MD, MSCR, Markey Cancer Center, University of Kentucky, 800 Rose St, CC450, Lexington, KY 40536 Fax: 859-257-7715; e-mail: [email protected]
Electronic forwarding or copying is a violation of US and International Copyright Laws. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by CIG Media Group, LP, ISSN #1557-9190, provided the appropriate fee is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA 978-750-8400.
166
•
Clinical Lymphoma & Myeloma June 2008
Figure 1 Transition State Model Describing Patients with Follicular Lymphoma Who Successfully Have a Second Remission Disease-Free Survivor in Second Remission
Relapse
Table 1 Estimated Utilities and Costs for Health States Health State
Utility
Cost
Disease Free
0.7320
$3233
Salvage
0.6621
$27,233
Subsequent Remissions
0.7320
$3233
Transplantation
0.6522
$89,916
Refractory
0.4321
$81,121
–
–
Death
Costs are represented as per patient for each 6 months.
Salvage
Transplant Refractory Subsequent Remissions
Death
compared with routine clinical observation of patients in their second remission from FL over 5 years (Figure 1). To construct the model, we first identified all possible health states that patients in second complete remission could enter throughout the rest of their lives. We termed these states as follows: (1) disease-free survivor; (2) undergoing salvage treatment; (3) subsequent remissions; (4) refractory disease; (5) transplantation; and (6) death. We constructed a graphical model (Figure 1) to demonstrate the conceivable pathways patients can transition from one health state to another. Finally, we reviewed the literature to derive the probabilities that each health state transition will occur, the quality of life that is associated with being in each health state, and the costs expected while residing in that health state. Transition Probabilities Most patients with FL in the United States receive rituximab as part of their induction therapy, and we accordingly focused our analysis on the cohort of patients who received R-CHOP (rituximab with cyclophosphamide/doxorubicin/vincristine/ prednisone) and were then randomized to adjuvant rituximab versus observation.4 In this arm of the trial, patients who received R-CHOP and obtained a complete remission were randomized to receive 375 mg/m2 of rituximab every 3 months for 2 years or observation. We used the observed 4-year relapse-free survival (RFS) rates of 60% and 32%, respectively, as reported from the trial. The rate of relapse was calculated for each 6-month cycle of the model and extended for the full 5 years. Non–cancerrelated death rates were expected to be quite different for the previously reported clinical trial group and the age group selected for the current modeling study. We incorporated this
aspect in the model by estimating the age-specific death rates for each year and deducting the proportion of patients who would be expected to die from other conditions from the population remaining in the model.6 As no published data suggests what proportion of patients will elect to receive additional salvage chemotherapy upon subsequent relapse, we used expert opinion to estimate these probabilities. Patients who do receive additional salvage chemotherapy might then go on subsequent remissions. Although numerous options might be considered for patients attempting to have and maintain a third remission, we focused our attention on combination chemotherapy, radioimmunotherapy, and autologous BMT (ABMT). Median DFS for radioimmunotherapy,7 multiagent chemotherapy,8 and ABMT9 were previously reported. Some patients will elect to consolidate their subsequent remission and undergo an ABMT. Again, although data are limited, a Dutch retrospective review suggested 8% of their patients with FL received an ABMT during their treatment course.10 Sensitivity Analyses One-way and 2-way sensitivity analyses were conducted to examine the effect of uncertainty in our model parameters on the estimates. Sensitivity analyses for rates estimated from time-to-event curves were conducted using data points 10% before and after the original point estimate. For example, if the observed survival at 60 months was the original point estimate, we used survival at 54 and 66 months for the sensitivity analyses. For other proportions and continuous variables, we used ± 20% of the expected value. Finally, because we consider ABMT to be a unique decision point in the model, we assessed the effect of having a more durable third remission, as might be expected with ABMT, and less durable, as might be expected with radioimmunotherapy. Utilities A health state utility reflects the value that patients would assign to living in a specific state of health. Such values have been identified for patients with cancer by the use of validated survey instruments. The results are scored on a range of “0” for death (or health worse than death) to “1” for perfect health. We used utilities that have been previously reported for each of the health states in our model (Table 1).11-13
Clinical Lymphoma & Myeloma June 2008
•
167
Cost-Effectiveness of Rituximab in FL Figure 3 Overall Survival
100
100
80
80 Overall Survival (%)
Disease-Free Survival (%)
Figure 2 Disease-Free Survival
60
40
60
40
20
20 Observation Extended Adjuvant Rituximab
0
2
Observation Extended Adjuvant Rituximab
4
6
0
Years
Cost Costs were analyzed from the perspective of the US healthcare system. Previous studies have reported cost of care for patients with lymphoma during salvage chemotherapy,14 transplantation,15,16 and refractory disease states.14 These costs were inflated by the Medical Care Consumer Price Index to reflect 2006 medical care costs and used in the model. Costs for patients in event-free and subsequent remission health states were estimated by adding the costs of full-body computed tomography scans (Current Procedural Technology [CPT] codes 71260, 74160, and 72193), 2 physician visits (CPT 99214), and routine blood evaluations during those visits (CPT 85021 and 80012).17 Costs for adjuvant rituximab include the costs of routine observation plus the cost of the time in the infusion chair,18 chemotherapy nurse time,19 pharmacist preparation,20 and rituximab acquisition. The total costs for each health state are summarized in Table 1. Adjustments Historical cost estimates were adjusted for inflation using the US healthcare services price index available at www.data.bls. gov/cgi-bin/surveymost. Future costs and future quality-adjusted life-years (QALYs) were discounted at a rate of 3% annually.
Results Disease-free and OS estimates for patients with FL aged 65-70 years in their second remission were calculated for patients treated with extended adjuvant rituximab and observation (Figures 2 and 3). After 5 years, DFS is expected to be 47% and 22%, and OS is estimated to be 73% and 61% for extended adjuvant rituximab and observation, respectively, during the second remission. The discounted incremental cost-effectiveness ratio (ICER) for the addition of adjuvant rituximab is estimated to be $19,522/QALY gained. The unadjusted ICER is $16,586/QALY gained. One-way and 2-way sensitivity analyses were completed to assess the stability of the model. All variables were assessed as described in the methods section. We repeated the model
168
•
Clinical Lymphoma & Myeloma June 2008
2
4
6
Years
using the actual RFS rates reported 10% before and after 4 years since randomization. At the later time point, the difference in OS is smaller. The adjuvant and observation groups are expected to have 39% and 32% survival rates, respectively, under these assumptions. In contrast, the original model assumed 60% and 32% DFS after 4 years of follow-up. This smaller benefit in DFS translated to an adjusted ICER of $181,105, which is higher than values generally accepted as cost-effective in the United States. The lowest (most costeffective regimen) ICER, $6,991, was predicted if the cost of the rituximab was decreased by 20%. Another important variable was the proportion of patients choosing further aggressive salvage therapy after their second relapse (Table 2). Because previously published data is not available, we relied upon expert opinion to estimate that 80% of patients in this age range would choose and be eligible for further aggressive treatments. However, if 96% of patients undergo further therapy, the ICER increases to $32,381, while if only 64% are able to take further aggressive therapy, the ICER declines to $12,576. Finally, the durability of the third remission also affects the ICER. Patients with longer subsequent remissions, as might be expected with ABMT, have an increased ICER to $32,066, while patients with shorter subsequent remissions, as might be expected with radioimmunotherapy, have a decreased ICER of $11,093. A lower DFS rate and more patients choosing further aggressive treatments resulted in the most expensive 2-way analysis, with an ICER of $258,833/QALY. Conversely, if 20% less was spent for the rituximab and subsequent remissions are less durable, then the model becomes dominant, and adjuvant treatment is both less expensive and improves health.
Discussion Rationally incorporating the increasing number of active agents into the care of patients with FL is an ongoing challenge. The long natural history and the likelihood that patients will have subsequent remissions make it difficult to conclusively
John W. Hayslip, Kit N. Simpson Table 2 Key Sensitivity Analysis Results Model Estimate
Advantageous Sensitivity Estimate
Lower ICER
Deleterious Sensitivity Estimate
Higher ICER
Rituximab-Treated Disease-Free Survival
60%
60%
$19,522
39%
$181,105
Patients Attempting Salvage Chemotherapy at Second Relapse
80%
64%
$12,576
96%
$32,381
5%
6%
$19,483
4%
$19,562
Sensitivity Measure
Outcomes of Treatment After Second Relapse Death Proceed to ABMT
8%
9.6%
$18,842
6.4%
$20,204
Have third remission
89%
71.2%
$12,952
95%
$21,949
Refractory patients attempt another salvage
50%
40%
$19,209
60%
$23,974
Event-free survivor
0.73
0.876
$14,847
0.584
$28,494
Salvage chemotherapy
0.66
0.528
$18,184
0.792
$21,071
Transplantation
0.65
0.52
$19,414
0.78
$19,630
Palliative
0.43
0.344
$18,847
0.516
$20,246
$9088
$7270
$6991
$10,906
$32,052
Quality of Life (Health Utility)
Cost Acquisition 2 treatments of rituximab Salvage chemotherapy care
$27,233
$32,680
$16,487
$21,556
$22,556
Transplantation
$89,916
$107,899
$18,755
$71,932
$20,288
Refractory disease care
$81,121
$97,345
$12,766
$64,897
$26,277
identify optimal treatment algorithms that include multiple lines of therapy. Therefore, the importance of surrogate endpoints, such as time to progression (TTP), is difficult to comprehend and translate into meaningful data for guiding patient therapies. Additionally, differences in patterns of care across countries and between patients with different characteristics obscure the relevance of definitive endpoints such as OS from clinical trials. In the absence of pertinent data, decision analysis modeling, as done in this study, can provide valuable insights and offer our best opportunity to extrapolate available data under a clearly identified set of assumptions to fill the current need for information that can help inform clinical decisions. Our results suggest that the use of extended adjuvant rituximab, 375 mg/m2 intravenously every 3 months for 2 years, offers a clinical benefit to patients aged 65-70 years who have a second remission of FL at a reasonable cost. Admittedly, the model is moderately unstable and is particularly sensitive to variations in the assumed relapse rates in patients treated with rituximab, and the actual ICER might be worse than predicted. However, although the ICER might be less favorable, no deleterious health effect is expected. The instability in the ICER supports the need for ongoing updates of survival rates for patients treated with extended adjuvant rituximab. If the benefit of adjuvant rituximab is durable, then the ICER will be even better than predicted. On the contrary, if the benefit is less durable and the relapse rates begin to equalize after 5 years, then the ICER will increase toward the more costly values reported in our sensitivity analysis. Additionally, US patients might be more likely to receive R-CHOP as first-line therapy of FL and would therefore not
be candidates to receive R-CHOP for their first recurrence. Although a separate trial supported a benefit for adjuvant rituximab following a fludarabine-based regimen for patients with relapsed FL and mantle cell lymphoma (MCL),21 it is not known how variations in treatment sequence will affect the potential benefit of extended adjuvant rituximab. Besides the durability of rituximab effect, the sensitivity analyses highlight the importance of other key factors in the model. Decreases and increases in the cost of rituximab by just 20% can profoundly change the ICER to $6991 and $32,052 respectively, indicating the importance of pricing policies for the drug. Patients who are unwilling or unable to undergo further treatment at relapse will have an expected greater survival benefit from extended adjuvant rituximab. Conversely, patients who might be expected to undergo aggressive treatment after a second relapse might gain less absolute survival benefit from extended adjuvant rituximab, and their treatment is expected to be less cost-effective. Other authors have reported results that further corroborate the improvement in DFS after extended rituximab administration. Rituximab given for 4 consecutive weeks at 3 and 9 months after having a salvage remission was associated with significantly longer TTP in patients with relapsed FL and MCL. Further, rituximab given for 4 consecutive weeks every 6 months for 2 years has reported a superior progression-free survival and OS than observation after induction with cyclophosphamide, vincristine, and prednisone.22 These reports and others contribute to the growing body of data that supports a health benefit for the extended use of rituximab in the adjuvant setting for patients with indolent lymphomas. We expect the next
Clinical Lymphoma & Myeloma June 2008
•
169
Cost-Effectiveness of Rituximab in FL generation of studies will have to assess if extended rituximab use in previous lines of therapy will abrogate the benefit in later lines of therapy.
Conclusion Oncologists and patients with cancer are challenged to quickly assimilate the knowledge and experience gained from clinical trials into the treatment decisions that must be made today. From this study, we conclude that extended adjuvant rituximab for US patients aged 65-70 years with FL who have a second remission is expected to increase QALYs at an acceptable cost to the US healthcare system. The current analysis is based upon the best available data and attempts to model the expected outcomes of a new treatment strategy upon a patient population that has not yet been studied. Further follow-up is critical to assess the durability of the improved DFS after the extended rituximab portion of the therapy is completed because this will allow us to more accurately and confidently predict the costs and benefits of this promising new treatment strategy.
References 1. Groves FD, Linet MS, Travis LB, et al. Cancer surveillance series: nonHodgkin's lymphoma incidence by histologic subtype in the United States from 1978 through 1995. J Natl Cancer Inst 2000; 92:1240-51. 2. Swenson WT, Wooldridge JE, Lynch CF, et al. Improved survival of follicular lymphoma patients in the United States. J Clin Oncol 2005; 23:5019-26. 3. Zelenetz AD, Advani RH, Buadi F, et al. Non-Hodgkin’s lymphoma. Clinical practice guidelines in oncology. J Natl Compr Canc Netw 2006; 4:258-310. 4. van Oers MH, Klasa R, Marcus RE, et al. Rituximab maintenance improves clinical outcome of relapsed/resistant follicular non-Hodgkin lymphoma in patients both with and without rituximab during induction: results of a prospective randomized phase 3 intergroup trial. Blood 2006; 108:3295-301. 5. Morton LM, Wang SS, Devesa SS, et al. Lymphoma incidence patterns by WHO subtype in the United States, 1992-2001. Blood 2006; 107:265-76. 6. Arias E. United States Life Tables, 2003. Natl Vital Stat Rep 2006; 54:1-40. 7. Fisher RI, Kaminski MS, Wahl RL, et al. Tositumomab and iodine-131 tositumomab produces durable complete remissions in a subset of heavily pretreated patients with low-grade and transformed non-Hodgkin’s lymphomas. J Clin Oncol 2005; 23:7565-73. 8. Hertzberg MS, Crombie C, Benson W, et al. Outpatient fractionated ifosfamide, carboplatin and etoposide as salvage therapy in relapsed and refractory non-Hodgkin’s and Hodgkin’s lymphoma. Ann Oncol 2006; 17(suppl 4):iv25-30.
170
•
Clinical Lymphoma & Myeloma June 2008
9. Brice P, Simon D, Bouabdallah R, et al. High-dose therapy with autologous stem-cell transplantation (ASCT) after first progression prolonged survival of follicular lymphoma patients included in the prospective GELF 86 protocol. Ann Oncol 2000; 11:1585-90. 10. van Agthoven M, Kramer MH, Sonneveld P, et al. Cost analysis of common treatment options for indolent follicular non-Hodgkin’s lymphoma. Haematologica 2005; 90:1422-32. 11. Doorduyn JK, Buijit I, Uyl-de Groot CA, et al. Quality of life (QOL) in elderly patients with aggressive non-hodgkin’s lymphoma (NHL) treated with CHOP. Blood 2001; 98:430a (Abstract 1803). 12. Doorduijn J, Buijt I, Holt B, et al. Self-reported quality of life in elderly patients with aggressive non-Hodgkin's lymphoma treated with CHOP chemotherapy. Eur J Haematol 2005; 75:116-23. 13. van Agthoven M, Segeren CM, Buijt I, et al. A cost-utility analysis comparing intensive chemotherapy alone to intensive chemotherapy followed by myeloablative chemotherapy with autologous stem-cell rescue in newly diagnosed patients with stage II/III multiple myeloma; a prospective randomised phase III study. Eur J Cancer 2004; 40:1159-69. 14. Kutikova L, Bowman L, Chang S, et al. Medical costs associated with non-Hodgkin’s lymphoma in the United States during the first two years of treatment. Leuk Lymphoma 2006; 47:1535-44. 15. Meehan KR, Areman EM, Ericson SG, et al. Mobilization, collection, and processing of autologous peripheral blood stem cells: development of a clinical process with associated costs. J Hematother Stem Cell Res 2000; 9:767-71. 16. Lee SJ, Klar N, Weeks JC, et al. Predicting costs of stem-cell transplantation. J Clin Oncol 2000; 18:64-71. 17. Medical Fees in the United States 2004. Los Angeles, CA: Practice Management Information Corporation; 2004. 18. Simpson KN, Coughlin CM, Eron J, et al. Idiopathic thrombocytopenia purpura: treatment patterns and an analysis of cost associated with intravenous immunoglobulin and anti-D therapy. Semin Hematol 1998; 35(suppl 1):58-64. 19. Lamkin L, Rosiak J, Buerhaus P, et al. Oncology Nursing Society Workforce Survey. Part I: perceptions of the nursing workforce environment and adequacy of nurse staffing in outpatient and inpatient oncology settings. Oncol Nurs Forum 2001; 28:1545-52. 20. Brixner DI, Oderda GM, Nickman NA, et al. Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices. J Natl Compr Canc Netw 2006; 4:197-208. 21. Forstpointner R, Unterhalt M, Dreyling M, et al. Maintenance therapy with rituximab leads to a significant prolongation of response duration after salvage therapy with a combination of rituximab, fludarabine, cyclophosphamide, and mitoxantrone (R-FCM) in patients with recurring and refractory follicular and mantle cell lymphomas: results of a prospective randomized study of the German Low Grade Lymphoma Study Group (GLSG). Blood 2006; 108:4003-8. 22. Hochster HS, Weller E, Gascoyne RD, et al. Maintenance rituximab after CVP results in superior clinical outcome in advanced follicular lymphoma (FL): results of the E1496 phase III trial from the Eastern Cooperative Oncology Group and the Cancer and Leukemia Group B. Blood 2005; 106:106a (Abstract 349).
Abstract Purpose: A recent Intergroup trial showed that receiving adjuvant rituximab after having a second remission from follicular lymphoma (FL) improved progression-free and overall survival (OS). The current study was conducted to determine the incremental cost-effectiveness ratio of this strategy in the United States. Patients and Methods: We constructed a transition state model to estimate the incremental cost-effectiveness ratio of extended adjuvant rituximab for 2 years for patients having a second FL remission. Event-free and OS rates were estimated from the recently published Intergroup trial. These were adjusted to reflect the contribution of non-cancer–specific mortality for patients aged 65-70 years, a more commonly affected age group than in the Intergroup trial, which had a median age of 54 years. Previously reported quality of life and cost estimates were obtained from peer-reviewed sources. Results: Five years after a second induction with R-CHOP (rituximab with cyclophosphamide/doxorubicin/vincristine/ prednisone), disease-free survival is expected to be 47% and 22%, and the OS rates are estimated to be 73% and 61% for extended adjuvant rituximab and observation, respectively, during the second remission. The discounted incremental cost-effectiveness ratio for the addition of adjuvant rituximab is estimated to be $19,522 per quality-adjusted life-years gained. Conclusion: Extended adjuvant rituximab offers a clinical benefit to patients aged 65-70 years who have a second remission from FL at a cost generally acceptable to the US healthcare system.
Clinical Lymphoma & Myeloma, Vol. 8, No. 3, 166-170, 2008; DOI: 10.3816/CLM.2008.n.020 Key words: CHOP, Health state utility, Indolent lymphoma, Salvage therapy
Introduction Follicular lymphoma (FL) is the most common indolent non-Hodgkin lymphoma. Follicular lymphoma has an average annual incidence of 2.6 per 100,000 people, afflicting an estimated 8000 new Americans in 2007.1 Many patients will have a remission with their initial treatment, but eventual relapse is expected as part of the natural history of FL. Although many treatment options exist, the median survival of patients diagnosed with FL is 7.8 years.2 Despite the numerous available treatment options for patients with FL, there is little consensus regarding the optimal treatment algorithm for patients as they progress through the various stages of their disease course. Indeed, the National Comprehensive Cancer Network practice guidelines endorse approaches ranging from observation to multiagent 1Markey Cancer Center, University of Kentucky, Lexington 2Center for Health Economic and Policy Studies, College of
chemotherapy for the initial treatment of FL and include autologous and allogeneic bone marrow transplantation (BMT) as reasonable second-line treatments.3 The heterogeneity of these recommendations reflects the variability of the disease and our limited understanding of how to optimally use our current treatments to maximize health while minimizing illness and expense. Recently, a new strategy of extended adjuvant rituximab given to patients with FL in their second remission was reported to improve disease-free survival (DFS) and overall survival (OS) compared with standard observation.4 Patients, physicians, and payers will have to struggle to incorporate these new findings as they consider the myriad options for treatment of patients with FL. To aid in these complex decisions, we constructed a transition state model to estimate the cost and patient benefit of extended adjuvant rituximab therapy during second remission for patients with FL aged 65-70 years in the United States.
Health Professions,
Medical University of South Carolina, Charleston
Patients and Methods
Submitted: Mar 27, 2008; Accepted: Apr 7, 2008
We restricted this study to US patients aged 65-70 years (the age range with the most cases of FL5) in their second remission from FL. A transition state model was created to calculate the incremental cost-effectiveness ratio of adjuvant rituximab
Address for correspondence: John W. Hayslip, MD, MSCR, Markey Cancer Center, University of Kentucky, 800 Rose St, CC450, Lexington, KY 40536 Fax: 859-257-7715; e-mail: [email protected]
Electronic forwarding or copying is a violation of US and International Copyright Laws. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by CIG Media Group, LP, ISSN #1557-9190, provided the appropriate fee is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA 978-750-8400.
166
•
Clinical Lymphoma & Myeloma June 2008
Figure 1 Transition State Model Describing Patients with Follicular Lymphoma Who Successfully Have a Second Remission Disease-Free Survivor in Second Remission
Relapse
Table 1 Estimated Utilities and Costs for Health States Health State
Utility
Cost
Disease Free
0.7320
$3233
Salvage
0.6621
$27,233
Subsequent Remissions
0.7320
$3233
Transplantation
0.6522
$89,916
Refractory
0.4321
$81,121
–
–
Death
Costs are represented as per patient for each 6 months.
Salvage
Transplant Refractory Subsequent Remissions
Death
compared with routine clinical observation of patients in their second remission from FL over 5 years (Figure 1). To construct the model, we first identified all possible health states that patients in second complete remission could enter throughout the rest of their lives. We termed these states as follows: (1) disease-free survivor; (2) undergoing salvage treatment; (3) subsequent remissions; (4) refractory disease; (5) transplantation; and (6) death. We constructed a graphical model (Figure 1) to demonstrate the conceivable pathways patients can transition from one health state to another. Finally, we reviewed the literature to derive the probabilities that each health state transition will occur, the quality of life that is associated with being in each health state, and the costs expected while residing in that health state. Transition Probabilities Most patients with FL in the United States receive rituximab as part of their induction therapy, and we accordingly focused our analysis on the cohort of patients who received R-CHOP (rituximab with cyclophosphamide/doxorubicin/vincristine/ prednisone) and were then randomized to adjuvant rituximab versus observation.4 In this arm of the trial, patients who received R-CHOP and obtained a complete remission were randomized to receive 375 mg/m2 of rituximab every 3 months for 2 years or observation. We used the observed 4-year relapse-free survival (RFS) rates of 60% and 32%, respectively, as reported from the trial. The rate of relapse was calculated for each 6-month cycle of the model and extended for the full 5 years. Non–cancerrelated death rates were expected to be quite different for the previously reported clinical trial group and the age group selected for the current modeling study. We incorporated this
aspect in the model by estimating the age-specific death rates for each year and deducting the proportion of patients who would be expected to die from other conditions from the population remaining in the model.6 As no published data suggests what proportion of patients will elect to receive additional salvage chemotherapy upon subsequent relapse, we used expert opinion to estimate these probabilities. Patients who do receive additional salvage chemotherapy might then go on subsequent remissions. Although numerous options might be considered for patients attempting to have and maintain a third remission, we focused our attention on combination chemotherapy, radioimmunotherapy, and autologous BMT (ABMT). Median DFS for radioimmunotherapy,7 multiagent chemotherapy,8 and ABMT9 were previously reported. Some patients will elect to consolidate their subsequent remission and undergo an ABMT. Again, although data are limited, a Dutch retrospective review suggested 8% of their patients with FL received an ABMT during their treatment course.10 Sensitivity Analyses One-way and 2-way sensitivity analyses were conducted to examine the effect of uncertainty in our model parameters on the estimates. Sensitivity analyses for rates estimated from time-to-event curves were conducted using data points 10% before and after the original point estimate. For example, if the observed survival at 60 months was the original point estimate, we used survival at 54 and 66 months for the sensitivity analyses. For other proportions and continuous variables, we used ± 20% of the expected value. Finally, because we consider ABMT to be a unique decision point in the model, we assessed the effect of having a more durable third remission, as might be expected with ABMT, and less durable, as might be expected with radioimmunotherapy. Utilities A health state utility reflects the value that patients would assign to living in a specific state of health. Such values have been identified for patients with cancer by the use of validated survey instruments. The results are scored on a range of “0” for death (or health worse than death) to “1” for perfect health. We used utilities that have been previously reported for each of the health states in our model (Table 1).11-13
Clinical Lymphoma & Myeloma June 2008
•
167
Cost-Effectiveness of Rituximab in FL Figure 3 Overall Survival
100
100
80
80 Overall Survival (%)
Disease-Free Survival (%)
Figure 2 Disease-Free Survival
60
40
60
40
20
20 Observation Extended Adjuvant Rituximab
0
2
Observation Extended Adjuvant Rituximab
4
6
0
Years
Cost Costs were analyzed from the perspective of the US healthcare system. Previous studies have reported cost of care for patients with lymphoma during salvage chemotherapy,14 transplantation,15,16 and refractory disease states.14 These costs were inflated by the Medical Care Consumer Price Index to reflect 2006 medical care costs and used in the model. Costs for patients in event-free and subsequent remission health states were estimated by adding the costs of full-body computed tomography scans (Current Procedural Technology [CPT] codes 71260, 74160, and 72193), 2 physician visits (CPT 99214), and routine blood evaluations during those visits (CPT 85021 and 80012).17 Costs for adjuvant rituximab include the costs of routine observation plus the cost of the time in the infusion chair,18 chemotherapy nurse time,19 pharmacist preparation,20 and rituximab acquisition. The total costs for each health state are summarized in Table 1. Adjustments Historical cost estimates were adjusted for inflation using the US healthcare services price index available at www.data.bls. gov/cgi-bin/surveymost. Future costs and future quality-adjusted life-years (QALYs) were discounted at a rate of 3% annually.
Results Disease-free and OS estimates for patients with FL aged 65-70 years in their second remission were calculated for patients treated with extended adjuvant rituximab and observation (Figures 2 and 3). After 5 years, DFS is expected to be 47% and 22%, and OS is estimated to be 73% and 61% for extended adjuvant rituximab and observation, respectively, during the second remission. The discounted incremental cost-effectiveness ratio (ICER) for the addition of adjuvant rituximab is estimated to be $19,522/QALY gained. The unadjusted ICER is $16,586/QALY gained. One-way and 2-way sensitivity analyses were completed to assess the stability of the model. All variables were assessed as described in the methods section. We repeated the model
168
•
Clinical Lymphoma & Myeloma June 2008
2
4
6
Years
using the actual RFS rates reported 10% before and after 4 years since randomization. At the later time point, the difference in OS is smaller. The adjuvant and observation groups are expected to have 39% and 32% survival rates, respectively, under these assumptions. In contrast, the original model assumed 60% and 32% DFS after 4 years of follow-up. This smaller benefit in DFS translated to an adjusted ICER of $181,105, which is higher than values generally accepted as cost-effective in the United States. The lowest (most costeffective regimen) ICER, $6,991, was predicted if the cost of the rituximab was decreased by 20%. Another important variable was the proportion of patients choosing further aggressive salvage therapy after their second relapse (Table 2). Because previously published data is not available, we relied upon expert opinion to estimate that 80% of patients in this age range would choose and be eligible for further aggressive treatments. However, if 96% of patients undergo further therapy, the ICER increases to $32,381, while if only 64% are able to take further aggressive therapy, the ICER declines to $12,576. Finally, the durability of the third remission also affects the ICER. Patients with longer subsequent remissions, as might be expected with ABMT, have an increased ICER to $32,066, while patients with shorter subsequent remissions, as might be expected with radioimmunotherapy, have a decreased ICER of $11,093. A lower DFS rate and more patients choosing further aggressive treatments resulted in the most expensive 2-way analysis, with an ICER of $258,833/QALY. Conversely, if 20% less was spent for the rituximab and subsequent remissions are less durable, then the model becomes dominant, and adjuvant treatment is both less expensive and improves health.
Discussion Rationally incorporating the increasing number of active agents into the care of patients with FL is an ongoing challenge. The long natural history and the likelihood that patients will have subsequent remissions make it difficult to conclusively
John W. Hayslip, Kit N. Simpson Table 2 Key Sensitivity Analysis Results Model Estimate
Advantageous Sensitivity Estimate
Lower ICER
Deleterious Sensitivity Estimate
Higher ICER
Rituximab-Treated Disease-Free Survival
60%
60%
$19,522
39%
$181,105
Patients Attempting Salvage Chemotherapy at Second Relapse
80%
64%
$12,576
96%
$32,381
5%
6%
$19,483
4%
$19,562
Sensitivity Measure
Outcomes of Treatment After Second Relapse Death Proceed to ABMT
8%
9.6%
$18,842
6.4%
$20,204
Have third remission
89%
71.2%
$12,952
95%
$21,949
Refractory patients attempt another salvage
50%
40%
$19,209
60%
$23,974
Event-free survivor
0.73
0.876
$14,847
0.584
$28,494
Salvage chemotherapy
0.66
0.528
$18,184
0.792
$21,071
Transplantation
0.65
0.52
$19,414
0.78
$19,630
Palliative
0.43
0.344
$18,847
0.516
$20,246
$9088
$7270
$6991
$10,906
$32,052
Quality of Life (Health Utility)
Cost Acquisition 2 treatments of rituximab Salvage chemotherapy care
$27,233
$32,680
$16,487
$21,556
$22,556
Transplantation
$89,916
$107,899
$18,755
$71,932
$20,288
Refractory disease care
$81,121
$97,345
$12,766
$64,897
$26,277
identify optimal treatment algorithms that include multiple lines of therapy. Therefore, the importance of surrogate endpoints, such as time to progression (TTP), is difficult to comprehend and translate into meaningful data for guiding patient therapies. Additionally, differences in patterns of care across countries and between patients with different characteristics obscure the relevance of definitive endpoints such as OS from clinical trials. In the absence of pertinent data, decision analysis modeling, as done in this study, can provide valuable insights and offer our best opportunity to extrapolate available data under a clearly identified set of assumptions to fill the current need for information that can help inform clinical decisions. Our results suggest that the use of extended adjuvant rituximab, 375 mg/m2 intravenously every 3 months for 2 years, offers a clinical benefit to patients aged 65-70 years who have a second remission of FL at a reasonable cost. Admittedly, the model is moderately unstable and is particularly sensitive to variations in the assumed relapse rates in patients treated with rituximab, and the actual ICER might be worse than predicted. However, although the ICER might be less favorable, no deleterious health effect is expected. The instability in the ICER supports the need for ongoing updates of survival rates for patients treated with extended adjuvant rituximab. If the benefit of adjuvant rituximab is durable, then the ICER will be even better than predicted. On the contrary, if the benefit is less durable and the relapse rates begin to equalize after 5 years, then the ICER will increase toward the more costly values reported in our sensitivity analysis. Additionally, US patients might be more likely to receive R-CHOP as first-line therapy of FL and would therefore not
be candidates to receive R-CHOP for their first recurrence. Although a separate trial supported a benefit for adjuvant rituximab following a fludarabine-based regimen for patients with relapsed FL and mantle cell lymphoma (MCL),21 it is not known how variations in treatment sequence will affect the potential benefit of extended adjuvant rituximab. Besides the durability of rituximab effect, the sensitivity analyses highlight the importance of other key factors in the model. Decreases and increases in the cost of rituximab by just 20% can profoundly change the ICER to $6991 and $32,052 respectively, indicating the importance of pricing policies for the drug. Patients who are unwilling or unable to undergo further treatment at relapse will have an expected greater survival benefit from extended adjuvant rituximab. Conversely, patients who might be expected to undergo aggressive treatment after a second relapse might gain less absolute survival benefit from extended adjuvant rituximab, and their treatment is expected to be less cost-effective. Other authors have reported results that further corroborate the improvement in DFS after extended rituximab administration. Rituximab given for 4 consecutive weeks at 3 and 9 months after having a salvage remission was associated with significantly longer TTP in patients with relapsed FL and MCL. Further, rituximab given for 4 consecutive weeks every 6 months for 2 years has reported a superior progression-free survival and OS than observation after induction with cyclophosphamide, vincristine, and prednisone.22 These reports and others contribute to the growing body of data that supports a health benefit for the extended use of rituximab in the adjuvant setting for patients with indolent lymphomas. We expect the next
Clinical Lymphoma & Myeloma June 2008
•
169
Cost-Effectiveness of Rituximab in FL generation of studies will have to assess if extended rituximab use in previous lines of therapy will abrogate the benefit in later lines of therapy.
Conclusion Oncologists and patients with cancer are challenged to quickly assimilate the knowledge and experience gained from clinical trials into the treatment decisions that must be made today. From this study, we conclude that extended adjuvant rituximab for US patients aged 65-70 years with FL who have a second remission is expected to increase QALYs at an acceptable cost to the US healthcare system. The current analysis is based upon the best available data and attempts to model the expected outcomes of a new treatment strategy upon a patient population that has not yet been studied. Further follow-up is critical to assess the durability of the improved DFS after the extended rituximab portion of the therapy is completed because this will allow us to more accurately and confidently predict the costs and benefits of this promising new treatment strategy.
References 1. Groves FD, Linet MS, Travis LB, et al. Cancer surveillance series: nonHodgkin's lymphoma incidence by histologic subtype in the United States from 1978 through 1995. J Natl Cancer Inst 2000; 92:1240-51. 2. Swenson WT, Wooldridge JE, Lynch CF, et al. Improved survival of follicular lymphoma patients in the United States. J Clin Oncol 2005; 23:5019-26. 3. Zelenetz AD, Advani RH, Buadi F, et al. Non-Hodgkin’s lymphoma. Clinical practice guidelines in oncology. J Natl Compr Canc Netw 2006; 4:258-310. 4. van Oers MH, Klasa R, Marcus RE, et al. Rituximab maintenance improves clinical outcome of relapsed/resistant follicular non-Hodgkin lymphoma in patients both with and without rituximab during induction: results of a prospective randomized phase 3 intergroup trial. Blood 2006; 108:3295-301. 5. Morton LM, Wang SS, Devesa SS, et al. Lymphoma incidence patterns by WHO subtype in the United States, 1992-2001. Blood 2006; 107:265-76. 6. Arias E. United States Life Tables, 2003. Natl Vital Stat Rep 2006; 54:1-40. 7. Fisher RI, Kaminski MS, Wahl RL, et al. Tositumomab and iodine-131 tositumomab produces durable complete remissions in a subset of heavily pretreated patients with low-grade and transformed non-Hodgkin’s lymphomas. J Clin Oncol 2005; 23:7565-73. 8. Hertzberg MS, Crombie C, Benson W, et al. Outpatient fractionated ifosfamide, carboplatin and etoposide as salvage therapy in relapsed and refractory non-Hodgkin’s and Hodgkin’s lymphoma. Ann Oncol 2006; 17(suppl 4):iv25-30.
170
•
Clinical Lymphoma & Myeloma June 2008
9. Brice P, Simon D, Bouabdallah R, et al. High-dose therapy with autologous stem-cell transplantation (ASCT) after first progression prolonged survival of follicular lymphoma patients included in the prospective GELF 86 protocol. Ann Oncol 2000; 11:1585-90. 10. van Agthoven M, Kramer MH, Sonneveld P, et al. Cost analysis of common treatment options for indolent follicular non-Hodgkin’s lymphoma. Haematologica 2005; 90:1422-32. 11. Doorduyn JK, Buijit I, Uyl-de Groot CA, et al. Quality of life (QOL) in elderly patients with aggressive non-hodgkin’s lymphoma (NHL) treated with CHOP. Blood 2001; 98:430a (Abstract 1803). 12. Doorduijn J, Buijt I, Holt B, et al. Self-reported quality of life in elderly patients with aggressive non-Hodgkin's lymphoma treated with CHOP chemotherapy. Eur J Haematol 2005; 75:116-23. 13. van Agthoven M, Segeren CM, Buijt I, et al. A cost-utility analysis comparing intensive chemotherapy alone to intensive chemotherapy followed by myeloablative chemotherapy with autologous stem-cell rescue in newly diagnosed patients with stage II/III multiple myeloma; a prospective randomised phase III study. Eur J Cancer 2004; 40:1159-69. 14. Kutikova L, Bowman L, Chang S, et al. Medical costs associated with non-Hodgkin’s lymphoma in the United States during the first two years of treatment. Leuk Lymphoma 2006; 47:1535-44. 15. Meehan KR, Areman EM, Ericson SG, et al. Mobilization, collection, and processing of autologous peripheral blood stem cells: development of a clinical process with associated costs. J Hematother Stem Cell Res 2000; 9:767-71. 16. Lee SJ, Klar N, Weeks JC, et al. Predicting costs of stem-cell transplantation. J Clin Oncol 2000; 18:64-71. 17. Medical Fees in the United States 2004. Los Angeles, CA: Practice Management Information Corporation; 2004. 18. Simpson KN, Coughlin CM, Eron J, et al. Idiopathic thrombocytopenia purpura: treatment patterns and an analysis of cost associated with intravenous immunoglobulin and anti-D therapy. Semin Hematol 1998; 35(suppl 1):58-64. 19. Lamkin L, Rosiak J, Buerhaus P, et al. Oncology Nursing Society Workforce Survey. Part I: perceptions of the nursing workforce environment and adequacy of nurse staffing in outpatient and inpatient oncology settings. Oncol Nurs Forum 2001; 28:1545-52. 20. Brixner DI, Oderda GM, Nickman NA, et al. Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices. J Natl Compr Canc Netw 2006; 4:197-208. 21. Forstpointner R, Unterhalt M, Dreyling M, et al. Maintenance therapy with rituximab leads to a significant prolongation of response duration after salvage therapy with a combination of rituximab, fludarabine, cyclophosphamide, and mitoxantrone (R-FCM) in patients with recurring and refractory follicular and mantle cell lymphomas: results of a prospective randomized study of the German Low Grade Lymphoma Study Group (GLSG). Blood 2006; 108:4003-8. 22. Hochster HS, Weller E, Gascoyne RD, et al. Maintenance rituximab after CVP results in superior clinical outcome in advanced follicular lymphoma (FL): results of the E1496 phase III trial from the Eastern Cooperative Oncology Group and the Cancer and Leukemia Group B. Blood 2005; 106:106a (Abstract 349).