Cost–Utility Analysis of Palonosetron-Based Therapy in Preventing Emesis Among Breast Cancer Patients
We estimated the cost-utility of palonosetron-based therapy compared with generic ondansetron-based therapy throughout four cycles of anthracycline and cyclophosphamide for treating women with breast cancer. We developed a Markov model comparing six strategies in which ondansetron and palonosetron are combined with either dexamethasone alone, dexamethasone plus aprepitant following emesis, or dexamethasone plus aprepitant up front. Data on the effectiveness of antiemetics and emesis-related utility were obtained from published sources. Relative to the ondansetron-based two-drug therapy, the incremental cost–effectiveness ratios for the palonosetron-based regimens were $115,490/quality-adjusted life years (QALY) for the two-drug strategy, $199,375/QALY for the two-drug regimen plus aprepitant after emesis, and $200,526/QALY for the three-drug strategy. In sensitivity analysis, using the $100,000/QALY benchmark, the palonosetron-based two-drug strategy and the two-drug regimen plus aprepitant following emesis were shown to be cost-effective in 39% and 26% of the Monte Carlo simulations, respectively, and with changes in values for the effectiveness of antiemetics and the rate of hospitalization. The cost-utility of palonosetron-based therapy exceeds the $100,000/QALY threshold. Future research incorporating the price structure of all antiemetics following ondansetron's recent patent expiration is needed.
Markov Model Comparing Palo-Based Therapy vs Onda-Based Therapy for Prophylaxis of Chemotherapy-Induced Emesis in Breast Cancer Patients Receiving Four Cycles of AC-Based Chemotherapy (1) Onda (32 mg intravenously) + dex (8 mg intravenously) on day 1, followed by dex (4 mg orally twice a day) on days 2−5. (2) Onda (32 mg intravenously) + dex (8 mg intravenously) on day 1, followed by dex (4 mg orally twice a day) on days 2−5 and aprepitant in the subsequent cycles following the occurrence of emesis (ie, onda 16 mg orally + aprepitant 125 mg orally + dex 12 mg orally on day 1 followed by aprepitant 80 mg orally on days 2−3). (3) Palo (0.25 mg intravenously) + dex (8 mg intravenously) on day 1, followed by dex (4 mg orally twice a day) on days 2−5. (4) Palo (0.25 mg intravenously) + dex (8 mg intravenously) on day 1, followed by dex (4 mg orally twice a day) on days 2−5 and aprepitant in the subsequent cycles following the occurrence of emesis (ie, palo 0.25 mg intravenously + aprepitant 125 mg orally + dex 12 mg orally on day 1 followed by aprepitant 80 mg orally on days 2−3). (5) Onda (16 mg orally) + aprepitant (125 mg orally) + dex (12 mg orally) on day 1 followed by aprepitant (80 mg orally) on days 2−3. (6) Palo (0.25 mg intravenously) + aprepitant (125 mg orally) + dex (12 mg orally) on day 1 followed by aprepitant (80 mg orally) on days 2−3. Palo = palonosetron; onda = ondansetron; AC = anthracycline and cyclophosphamide; dex, dexamethasone
We modeled emesis-related outcomes and direct medical costs (from a third-party payer perspective within the context of the U.S. health-care system) over a total of four cycles of chemotherapy as patients receiving AC-based regimens usually undergo at least four cycles of AC.10 We performed all analyses using TreeAge Pro 2009 Suite (Decision Analysis; TreeAge Software, Williamstown, MA). The study was submitted to our institutional review board and was determined to be exempt from review.
Probability Data
Two-drug prophylactic regimens
We estimated the effectiveness of the 5-HT3 antagonists based on secondary analysis of the raw data from the randomized clinical trial (RCT) directly comparing onda and palo when used alone for prevention of emesis associated with MEC, including 90 breast cancer patients from the palo 0.25-mg arm and 82 from the onda 32-mg arm who received AC-based chemotherapy (Table 1).5 Effectiveness estimates for palo 0.25 mg were augmented by data on 117 breast cancer patients on AC-based chemotherapy participating in a multicenter RCT comparing palo with dolasetron (Table 1).4 We assumed that dex adds the same relative benefit to either first- or second-generation 5-HT3 antagonists and obtained the expected additional benefit of dex in preventing acute emesis based on the results of an RCT comparing a single-dose of granisetron in combination with dex vs granisetron given alone to patients undergoing MEC (Table 2).11 Since in the aforementioned study dex was only given on day 1 of chemotherapy, the estimated additional benefit of adding dex to a 5-HT3 inhibitor on the delayed period was obtained from another RCT; this study, conducted by the Italian Group for Antiemetic Research, compared dex alone, dex plus onda, or placebo on days 2−5 of MEC.12
| MODEL PARAMETERS | BASE-CASE VALUES (RANGES) | DATA SOURCES |
|---|---|---|
| Probability of acute emesis control on cycle 1 of AC: | ||
| Onda-based two-drug strategyc | 0.84 (0.74−0.93) | Gralla et al,a The Italian Group[5] and [11] |
| Palo-based two-drug strategyc | 0.87 (0.81−0.94) | Eisenberg et al,a Gralla et al,a The Italian Group[4], [5] and [11] |
| Onda-based three-drug strategyd | 0.88 (0.85−0.91) | Warr et al7 |
| Palo-based three-drug strategyd | 0.96 (0.89−0.99) | Grote et al, Grunberg et al[40] and [41] |
| Probability of delayed emesis control following control of acute emesis on cycle 1 of ACc: | ||
| Onda-based two-drug strategyd | 0.75 (0.62–0.85) | The Italian Group12 |
| Palo-based two-drug strategyc | 0.85 (0.78–0.91) | Eisenberg et al,a Gralla et al,a The Italian Group[4], [5] and [12] |
| Onda-based three-drug strategyd | 0.86 (0.82–0.90) | Warr et al7 |
| Palo-based three-drug strategyc | 0.96 (0.91–0.97) | Eisenberg et al,a Gralla et al,a Warr et al[4], [5] and [7] |
| Probability of delayed emesis control following acute emesis on cycle 1 of ACc: | ||
| Onda-based two-drug strategyc | 0.46 (0.31–0.62) | Gralla et al,a The Italian Group[5] and [12] |
| Palo-based two-drug strategyc | 0.44 (0.27–0.59) | Eisenberg et al,a Gralla et al,a The Italian Group[4], [5] and [12] |
| Onda-based three-drug strategyd | 0.44 (0.29–0.57) | Warr et al7 |
| Palo-based three-drug strategyc | 0.51 (0.41–0.67) | Eisenberg et al,a Gralla et al,a Warr et al[4], [5] and [7] |
| Relative probability of emesis control in subsequent cycles of ACc: | ||
| Two-drug therapy | 0.987 (0.970–1.0) | Herrstedt et al14e |
| Three-drug therapy | 1.013 (1.0–1.030) | Herrstedt et al14e |
| Probability of hospitalization (among patients who develop emesis) per cycle of ACd: | ||
| Onda-based regimens | 0.0035 (0.0001−0.019) | Data from Medstat MarketScan16 |
| Palo-based regimens | 0.0017 (0.00004−0.0089) | Data from Medstat MarketScan, Haislip et al[16] and [19]b |
| Probability of office visit use (among patients who develop emesis) per cycle of ACd: | ||
| Onda-based regimens | 0.10 (0.07−0.14) | Data from Medstat MarketScan16 |
| Palo-based regimens | 0.05 (0.03−0.07) | Data from Medstat MarketScan, Haislip et al[16] and [19]b |
| Probability of rescue medicine utilization use (among patients who develop emesis) per cycle of ACd: | ||
| Onda-based regimens | 0.61 (0.46−0.75) | Gralla et al5a |
| Palo-based regimens | 0.56 (0.45−0.66) | Eisenberg et al, Gralla et al[4] and [5]a |
| Utility weights for emesis per cycle of ACf: | ||
| Acute and delayed emesis | 0.15 (0.10−0.20) | Sun et al20 |
| Acute emesis and no delayed emesis | 0.76 (0.70−0.83) | Sun et al20 |
| No acute emesis and delayed emesis | 0.20 (0.14−0.26) | Sun et al20 |
| No acute and no delayed emesis | 0.92 (0.86−0.99) | Sun et al20 |
AC = anthracycline and cyclophosphamide; onda = ondansetron; palo = palonosetron.
a Included in the analysis was the subset of women with breast cancer receiving AC-based chemotherapy.b We obtained an estimate of emesis-related hospitalization and office visit utilization based on data from Medstat MarketScan, HPM subset (Medstat Group, Inc., Ann Arbor, MI) on 707 breast cancer patients who received the first cycle of AC-based chemotherapy from 1996 to 2002 and either were admitted to the hospital or had an office visit for treatment of vomiting or dehydration. Since palo was only introduced into the U.S. market in 2003, we assumed that all these breast cancer patients received onda-based antiemetic prophylaxis. As a result, we estimated the differential rate of health-care resource utilization based on Haislip et al's19 reported differential incidence of extreme events associated with chemotherapy-induced nausea and vomiting experienced by community-based breast cancer patients who received either onda or palo for emesis prophylaxis following the first cycle of chemotherapy.c Of note is that there are two different methods for applying the benefit of adding dex and/or aprepitant to a 5-HT3 antagonist: (1) rate of emesis with 5-HT3* relative risk of emesis by adding dex and/or aprepitant and (2) rate of emesis control with 5-HT3 * relative risk of emesis control by adding dex and/or aprepitant. These produce substantially different results, with the former method skewing the results toward the least effective 5-HT3 and the latter skewing it toward the most effective one. As a result, we estimated the probability of emesis by averaging the results obtained using the two different methods. Of note is that the ranges for these effectiveness estimates were obtained by applying the two different methods to the lower and upper bounds of the 95% confidence intervals derived from the clinical trials comparing the 5-HT3 antagonists when used alone.d Ranges were obtained by constructing 95% confidence intervals for observed proportions using the normal approximation to the binomial distribution.e Ranges are based on the minimum and maximum values observed in Herrstedt et al's14 clinical trial of multicycle chemotherapy.f Ranges are based on the estimate's actual 95% confidence intervals obtained from Sun et al's20 data.
