Breast Cancer Research and Treatment

, Volume 132, Issue 2, pp 747–751

A cost-benefit analysis of bevacizumab in combination with paclitaxel in the first-line treatment of patients with metastatic breast cancer

Authors

  • Alberto J. Montero
    • Division of Hematology/OncologyUniversity of Miami Sylvester Comprehensive Cancer Center
  • Kiran Avancha
    • University of Miami Miller School of Medicine Office of Research
  • Stefan Glück
    • Division of Hematology/OncologyUniversity of Miami Sylvester Comprehensive Cancer Center
    • Johns Hopkins Singapore International Medical Centre and Johns Hopkins University School of Medicine
Brief Report

DOI: 10.1007/s10549-011-1919-y

Cite this article as:
Montero, A.J., Avancha, K., Glück, S. et al. Breast Cancer Res Treat (2012) 132: 747. doi:10.1007/s10549-011-1919-y

Abstract

Bevacizumab in combination with chemotherapy increases progression-free survival (PFS), but not overall survival when compared to chemotherapy alone in the treatment of metastatic breast cancer (MBC). Recently in November, 2011 the Food and drug administration revoked approval of bevacizumab in combination with paclitaxel for the treatment of MBC. The European Medicines Agency, in contrast, maintained its approval of bevacizumab in MBC. While neither agency considers health economics in their decision-making process, one of the greatest challenges in oncology practice today is to reconcile hard-won small incremental clinical benefits with exponentially rising costs. To inform policy-makers in the US, this study aimed to assess the cost-effectiveness of bevacizumab/paclitaxel in MBC, from a payer perspective. We created a decision analytical model using efficacy and adverse events data from the ECOG 2100 trial. Health utilities were derived from available literature. Costs were obtained from the Center for Medicare Services Drug Payment Table and Physician Fee Schedule and are represented in 2010 US dollars. Quality-adjusted life-years (QALY) and incremental cost-effectiveness ratio (ICER) were calculated. Sensitivity analyses were performed. Bevacizumab added 0.49 years of PFS and 0.135 QALY with an incremental cost of $100,300, and therefore a cost of $204,000 per year of PFS gained and an ICER of $745,000 per QALY. The main drivers of the model were drug acquisition cost, PFS, and health utility values. Using a threshold of $150,000/QALY, drug price would have to be reduced by nearly 80% or alternatively PFS increased by 10 months to make bevacizumab cost-effective. The results of the model were robust in sensitivity analyses. Bevacizumab plus paclitaxel is not cost-effective in treating MBC. Value-based pricing and the development of biomarkers to improve patient selection are needed to better define the role of the drug in this population.

Keywords

Breast cancerCost-benefit analysisPharmacoeconomicsBevacizumabEconomicsMetastatic breast cancerQALY

Introduction

The advent of molecularly targeted agents has directly contributed to improving clinical outcomes in metastatic breast cancer. Over the last two decades 5-year overall survival rates in patients with stage 4 breast cancer in the US have improved to approximately 23% [1]. However, these improvements come at a cost particularly when we consider the cost of monoclonal antibodies such as trastuzumab and bevacizumab which are at present among the most expensive class of molecularly targeted drugs. Consequently, breast cancer is estimated to be the most expensive cancer to treat in the U.S., costing an estimated $16.5 billion in 2010, with an unsustainable rate of increase [2]. While the cost of breast cancer care is increasing in virtually all high-income countries, off-label use of expensive medications that provide either only a very marginal or no benefit is debatable [2].

An excellent example is bevacizumab, a monoclonal antibody that targets circulating vascular endothelial growth factor-A (VEGF), thereby interfering with the process of tumor angiogenesis by preventing this ligand from interacting with its receptor [3]. Bevacizumab was the first anti-angiogenic drug approved for the treatment of cancer with initial approval in the setting of advanced colorectal cancer and later in lung cancer in combination with chemotherapy [4, 5]. The combination of bevacizumab and weekly paclitaxel, after the results of the E2100 randomized phase 3 trial demonstrated a near doubling in progression-free survival (PFS) with the combination over paclitaxel alone [11.4 months vs. 5.9 months; hazard ratio 0.6; P < 0.001], was initially granted “accelerated” approval by the US Food and Drug Administration (FDA) in 2008 as first-line treatment of HER2-negative metastatic breast cancer (MBC). The FDA then recently revoked the MBC indication from bevacizumab, on November 2011, citing safety concerns and the failure of subsequent phase 3 studies to show a similar magnitude of PFS benefit as E2100, and no significant improvement in OS [6]. The European Medicines Agency (EMA), however, interpreted the bevacizumab data differently than the FDA, and bevacizumab continues to be an approved option for MBC. In fact, the EMA’s Committee for medicinal products for human use (CHMP) has approved extending the indication of bevacizumab in MBC, and also permitting it to be combined with capecitabine in the first-line setting when a taxane/anthracycline combination cannot be used [3].

While neither the FDA nor the EMA considers health economics in their decision-making process, one of the greatest challenges in the practice of oncology today is to reconcile hard-won small incremental clinical benefits with exponentially rising costs. This study aimed to assess the cost-effectiveness of bevacizumab plus solvent-based paclitaxel in MBC, from a payer perspective and to better inform US policymakers. We created a decision analytical model using efficacy and adverse events data from the ECOG 2100 trial, because, thus far has the longest reported prolongation of PFS, in absolute terms, with the combination of bevacizumab with cytotoxic chemotherapy in the MBC setting.

Methods

Overall model

To inform policy-makers, we aimed to assess the cost-effectiveness of bevacizumab in combination with paclitaxel to paclitaxel alone in the treatment of patients with metastatic breast cancer from a payer perspective in the United States. Using Excel 2007 software (Microsoft corp., Redmond, WA, USA), we developed the decision-analytic model depicted in Fig. 1. Patients with advanced breast cancer may receive paclitaxel alone or with bevacizumab in the first-line setting. Upon progression there is no cross-over to bevacizumab for patients who received paclitaxel alone. The model assumes that second-line therapy and beyond, as well as best supportive care (BSC) would be similar in both arms and therefore costs would be evenly distributed and would not bear importance in the incremental costs and benefits incurred to the bevacizumab arm.
https://static-content.springer.com/image/art%3A10.1007%2Fs10549-011-1919-y/MediaObjects/10549_2011_1919_Fig1_HTML.gif
Fig. 1

Tree-diagram shows the model assessing the costeffectiveness of bevacizumab in combination with paclitaxel in the treatment of patients with advanced breast cancer

Model inputs

Costs were obtained from the Center for Medicare Services Drug Payment Table and Physician Fee Schedule and are represented in 2010 US dollars. Base case patient had a height of 65 inches and a weight of 70 kg and therefore a BSA of 1.79 m2 (using Mosteller). We included costs for drug acquisition, laboratory tests, physician, and administration fees. We decided not to include the cost of treating adverse events as these were considered minor compared to medication acquisition costs. No discounting was used due to the short-time horizon.

Time in each treatment state was based on data from ECOG 2100. Length of time on first-line treatment was derived from median progression-free survival time. Health utilities were derived from the work by Lloyd et al. [7]. Baseline and incremental and decrement values were as follows: baseline 0.715, response (+0.075), progression (−0.272), febrile neutropenia (−0.15), diarrhea or vomit (−0.103), fatigue (−0.115), hair loss (−0.114), neuropathy (−0.155). Utility value for second-line therapy and beyond (before terminal state) was 0.45. Terminal state was considered to be the last 6 months and carried a utility value of 0.19. Incidence of adverse events were as follows; paclitaxel alone: alopecia, 79%; febrile neutropenia, 0.6%; diarrhea + vomiting, 2.7%; fatigue, 9.1%; neuropathy, 21.6%; for bevacizumab in combination with paclitaxel they were: alopecia, 80%; febrile neutropenia, 0.8%; diarrhea + vomiting, 2.7%; fatigue, 9.1%; neuropathy, 23%. Response rates were 36.9 and 21.2% for the combination and paclitaxel alone, respectively. The average health utility for first-line treatment was 0.59 with paclitaxel and 0.6 with the combination.

Cost-effectiveness evaluation, sensitivity analysis, and scenarios

Quality-adjusted life-years (QALY) and incremental cost-effectiveness ratio (ICER) were calculated. The primary endpoint was the ICER for bevacizumab in combination with paclitaxel versus paclitaxel alone. We also calculated the cost per progression-free survival year added as the main endpoint in E2100 was progression-free survival.

We performed one-way sensitivity analysis to test the robustness of the model. We also performed several other analysis including different scenarios and assumptions, such as using clinical outcomes from other trials and with bevacizumab in combination with other chemotherapeutic drugs, as well as a comparison of gemcitabine in addition to paclitaxel versus paclitaxel alone [8, 9]. We also aimed to evaluate at what cost and/or benefits the ICER would be within acceptable thresholds.

Results

Table 1 summarizes the value and data source for each of the model inputs. Bevacizumab was not cost-effective using commonly accepted thresholds for willingness-to-pay. The addition of bevacizumab to weekly paclitaxel, added 0.49 years of PFS and 0.135 QALY, and was associated with an incremental cost of $100,300. The overall cost of paclitaxel plus bevacizumab was therefore $204,000 per year of PFS gained, with an incremental cost-effectiveness ratio (ICER) of $745,000 per QALY (Table 2). The main cost element was the cost of bevacizumab. One-way sensitivity analyses were performed to evaluate the potential impact of changes in model inputs. The primary drivers of this cost-effectiveness model were: cost of treatment with bevacizumab, PFS, and health utility values.
Table 1

Values of input variables

Input

Cost (2010 USD)

 

Source

Monthly paclitaxel cost

$404.10

 

CMS

Monthly paclitaxel/bevacizumab cost

$8,707.50

  

Time in each treatment state (months)

Paclitaxel

Paclitaxel/Bevacizumab

Source

Total survival time

25.2

26.7

ECOG 2100

Time in treatment state under first-line treatment

5.9

11.8

 

Health Utilities by treatment state

 

Source

First-line treatment: Paclitaxel

0.59

Derived from Lloyd et al. [7], adjusted for specific adverse events and response rate with each treatment

First-line treatment: Paclitaxel/Bevacizumab

0.60

Second-line treatment and beyond

0.45

Terminal State

0.19

Table 2

Results of the bevacizumab cost-effectiveness model

Systemic therapy

Cost (2010 USD)

Incremental cost

QALYs

Incremental QALYs

ICER

Paclitaxel

$2,384.28

 

0.855

  

Paclitxel/Bevacizumab

$102,748.10

$100,363.82

1.02

0.135

$745,000/QALY

Overall survival in the bevacizumab plus paclitaxel arm would have to improve by an added 13.3 months, i.e., it would have to reach 40 months, to generate an ICER below $160,000 (Doubling median OS to 53.4 months would lead to an ICER of $88,425). Progression-free-survival would have to improve by an added 10 months (i.e., to 21.8 months) to generate an ICER below $160,000. Doubling PFS to 23.6 months would generate an ICER of $138,700. The cost of treatment with bevacizumab would have to be reduced by nearly 80% to generate an ICER below $160,000.

In other scenario analyses assessing bevacizumab in combination with docetaxel or capecitabine versus the respective single agent alone (utilizing data from AVADO and RIBBOn1) the anti-angiogenic agent was not found to be cost effective either. Bevacizumab and docetaxel generated 0.0375 QALY with an ICER of $1,937,000 over docetaxel alone (Based on AVADO). Bevacizumab in addition to capecitabine led to an increase of 0.168 QALY and an ICER of 425,000/QALY over capecitabine alone. In other words, bevacizumab was found to be even less cost-effective with docetaxel and did not reach accepted thresholds with capecitabine.

Discussion

Our results confirm that even assuming the most optimistic clinical data from E2100, the combination of bevacizumab with weekly paclitaxel is not cost effective as first-line therapy in MBC. Using a threshold of $150,000/QALY, as a definition of what constitutes a cost-effective therapy, the ICER of $745,000/QALY found in our model with paclitaxel/bevacizumab, means that the overall costs associated with bevacizumab would have to be reduced by nearly 80%. Alternatively PFS would have to be increased by an additional 10 months, i.e., a median PFS of 21.4 months, in order to make bevacizumab plus chemotherapy cost-effective in the setting of MBC.

One other evaluation has been published to date with an assessment of the cost-effectiveness of bevacizumab added to paclitaxel from a Swiss health care system perspective [10]. The authors utilized a Markov model, and also utilizing clinical trial results from ECOG 2100. In this study, it was shown that use of bevacizumab in the treatment of MBC was associated with an additional cost of EUR 40,369 and generated an increment of 0.22 QALY and an ICER of EUR 189,427/QALY. In their probabilistic sensitivity analysis, the willingness-to-pay threshold of EUR 60,000 was never reached. These results, like ours, provide additional evidence that the combination of bevacizumab with chemotherapy is not cost-effective at currently accepted thresholds in the treatment of metastatic breast cancer [11].

The current growth of health care expenditures in the United States remains on an unsustainable trajectory [2]. MBC will remain a significant economic burden in the US based on major demographic shifts, i.e., an aging population and fact that incidence of breast cancer increases with age. Even under the most optimistic assumptions, using only the E2100 dataset which showed the greatest magnitude of benefit in terms of delaying median PFS, and combination with weekly paclitaxel an inexpensive chemotherapeutic agent, the use of bevacizumab as first-line therapy in MBC is not cost effective. This dilemma is not unique to bevacizumab of course, but is a fact of cancer care in the twenty-first century [2]. In fact, other recently FDA approved therapies for the treatment of metastatic solid tumors such as sipuleucel-T (metastatic prostate cancer) and ipilimumab (metastatic melanoma) are even more expensive than bevacizumab providing what many would argue are similar marginal benefits from cost-benefit perspective.

However, bevacizumab could be utilized in a more cost-effective manner in the setting of MBC. On the efficacy side of the cost equation, identification of robust predictive biomarkers would increase the overall efficacy of bevacizumab in MBC by treating only those patients most likely to respond. Similarly, development of other drugs that are more effective in inhibition of tumor angiogenesis would also have the potential of increasing the overall anti-tumor effect of this therapeutic approach. On the cost side of the equation, technical advances that reduce the production cost of monoclonal antibodies would have a favorable impact on the cost-effectiveness of bevacizumab. For example, development of less expensive generic and biosimilar antibodies is already under way [12]. Development of other less expensive alternatives to antibodies may represent a technological innovation that may bring the cost down of targeted anti-angiogenic therapy in treating MBC.

There are several limitations of this study. First, our model likely underestimates the overall cost of bevacizumab in MBC because we did not take into account the cost of treating bevacizumab related grade 3–4 severe adverse toxicities, such as hypertension, proteinuria, bowel perforation, bleeding, and thromboembolic events. While they are uncommon, these complications known to be attributed to bevacizumab would clearly increase the costs associated with this drug above what is quantified in our model.

In conclusion, bevacizumab plus weekly paclitaxel is not cost-effective in the treatment of metastatic breast cancer at an overall cost of $204,000 per year of PFS gained, with an incremental cost-effectiveness ratio (ICER) of $745,000 per QALY. In order to meet currently accepted definitions of cost-effectiveness, the overall costs of bevacizumab would have to be decreased by approximately 80% or the overall median PFS observed in E2100 would have to be increased twofold. From an economic perspective, the removal of the FDA indication for MBC is entirely rational. Until predictive biomarkers are developed that can accurately identify the subset of patients that derive benefit from bevacizumab therapy, from a health economics perspective, it should not be considered as part of standard of care for management of MBC, and should not be used on an off-label basis in the U.S.

Conflicts of interest

Gilberto Lopes has received honorarium and grant funding from Roche. Stefan Gluck has received research funding from Genentech. The other authors have no conflict of interest to declare.

Copyright information

© Springer Science+Business Media, LLC. 2011