Cost-effectiveness analysis of olaparib and
niraparib as maintenance therapy for women with
recurrent platinum-sensitive ovarian cancer
John Hang Leung, Hui-Chu Lang, Shyh-Yau Wang, Hsueh Fang Lo & Agnes LF
Chan
To cite this article: John Hang Leung, Hui-Chu Lang, Shyh-Yau Wang, Hsueh Fang Lo & Agnes
LF Chan (2021): Cost-effectiveness analysis of olaparib and niraparib as maintenance therapy for
women with recurrent platinum-sensitive ovarian cancer, Expert Review of Pharmacoeconomics &
Outcomes Research, DOI: 10.1080/14737167.2021.1954506
To link to this article: https://doi.org/10.1080/14737167.2021.1954506
Published online: 06 Aug 2021.
Submit your article to this journal
Article views: 48
View related articles
View Crossmark data
ORIGINAL RESEARCH
Cost-effectiveness analysis of olaparib and niraparib as maintenance therapy for
women with recurrent platinum-sensitive ovarian cancer
John Hang Leunga†
, Hui-Chu Langb†
, Shyh-Yau Wangc
, Hsueh Fang Loa and Agnes LF Chand
a
Department of Obstetrics and Gynecology, Ditmanson Medical Foundation Chiayi Christian Hospital, Chiayi, Taiwan; b
Institute of Hospital and
Health Care Administration, National Yang-Ming Chiao Tung University, Taipei, Taiwan; c
Department of Radiology, An-Nan Hospital, China Medical
University, Tainan, Taiwan; d
Department of Pharmacy, An-Nan Hospital, China Medical University, Tainan, Taiwan
ABSTRACT
Objective: We evaluated the cost-effectiveness of olaparib and niraparib as maintenance therapy for
patients with platinum-sensitive recurrent ovarian cancer.
Methods: A decision analysis model compared the costs and effectiveness of olaparib and niraparib
versus placebo for patients with or without germline BRCA mutations. Resource use and associated
costs were estimated from the 2020 National Health Insurance Administration reimbursement price list.
Clinical effectiveness was measured in progression-free survival per life-years (PFS-LY) based on the
results of clinical trials SOLO2/ENHOT-Ov21 and ENGOT-OV16/NOVA. The incremental cost-effectiveness
ratio (ICER) was estimated from a single-payer perspective.
Results: In the base case, olaparib was the more cost-effective treatment regimen. The ICERs for
olaparib and niraparib compared to placebo were NT$1,804,785 and NT$2,340,265 per PFS-LY, respec￾tively. Tornado analysis showed that PFS and the total resource use cost of niraparib regimen for
patients without gBRCA were the most sensitive parameters impacting the ICER. The ICERs for both
drugs in patients with a gBRCA mutation were lower than in patients without a gBRCA mutation.
Probabilistic sensitivity analysis indicated that olaparib was more cost-effective than niraparib at the
willingness-to-pay threshold of NT$2,602,404 per PFS life-year gained.
Conclusion: Olaparib was estimated to be less cost and more effective compared to niraparib as
maintenance therapy for patients with recurrent platinum-sensitive ovarian cancer.
ARTICLE HISTORY
Received 13 April 2021
Accepted 8 July 2021
KEYWORDS
Cost-effectiveness analysis;
recurrent platinum-sensitive
ovarian cancer; maintenance
therapy; niraparib; olaparib
1. Introduction
An increasing number of women in the US and Taiwan are
diagnosed with ovarian cancer (OC) each year [1,2].
Approximately 80% of patients are diagnosed at advanced
stage III or IV,and 80–85% with high risk of relapse after initial
treatment with surgery and platinum chemotherapy [3,4].
Researchers are continuing to investigate novel agents and
the molecular biology of ovarian cancer in order to decrease
the relapse rate and prolong remission. Bevacizumab has been
investigated in GOG 218(NCT00262847) and ICON 7
(NCT00483782) trials with evidence of benefit in overall survi￾val (OS) for patients with newly diagnosed stage IV ovarian
cancer and a subset of patients at high risk of progression,
respectively [5,6]. Furthermore, OCEANS trial (NCT00434642)
showed no evidence of benefit in OS in platinum-sensitive
recurrent OC. U.S. Food and Drug Administration (FDA)
approved its use with chemotherapy in maintenance therapy
for newly diagnosed advanced OC patients [7]. More recently,
three poly (ADP-ribose) polymerase (PARP) inhibitors, olaparib,
niraparib, and rucaparib, were approved by the FDA as main￾tenance therapy for patients with platinum-sensitive, relapsed
epithelial ovarian, fallopian tube, or primary peritoneal cancer
who had a complete or partial response to platinum-based
chemotherapy [8–10]. Olaparib has just been approved for
reimbursement by the Taiwan National Health Insurance
Administration (NHIA) as maintenance treatment for patients
with germline or somatic BRCA 1/2 mutation (gBRCA) [11].
Niraparib did not obtain approval, though it may have similar
effectiveness as olaparib. Patients treated with niraparib need
to pay out of pocket. In a randomized, double-blind, phase III
trial (ENGOT-OV16/NOVA), patients with gBRCA treated with
niraparib maintenance therapy had a median progression-free
survival (PFS) of 21.0 months versus 5.5 months with placebo
(HR = 0.27, 95% CI = 0.17–0.41) and 9.3 months versus
3.9 months for niraparib and placebo in non-gBRCA patients,
respectively(HR = 0.45, 95% CI = 0.34–0.61) [12]. An interna￾tional, multicenter, double-blind, randomized, placebo￾controlled phase 3 trial (SOLO-2 /ENGOT-Ov21,
NCT01874353) reported that patients with gBRCA treated
with olaparib maintenance therapy had a significant longer
median PFS (19.1 months) than the placebo group (5.5 months;
HR = 0.30, 95% CI = 0.22–0.41) [13]. Of patients with wild-type
BRCA,the median PFS was significantly longer in the olaparib
CONTACT Agnes LF Chan [email protected] Department of Pharmacy, An-Nan Hospital, China Medical University. No. 66, Sec. 2, Changhe Rd., Annan
Dist., Tainan, Taiwan; Hsueh Fang Lo [email protected] Department of Obstetrics and Gynecology, Ditmanson Medical Foundation Chiayi Christian
Hospital, No.539, Zhongxiao Rd., East Dist., Chiayi, Taiwan †
Co-first authors.
EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH

https://doi.org/10.1080/14737167.2021.1954506

© 2021 Informa UK Limited, trading as Taylor & Francis Group
group than in the placebo group (7.4 months vs.
5.5 months) [14].
Cost-effectiveness analysis (CEA) is a well-established and gen￾erally used analytical method of assessing the costs and clinical
effectiveness of drugs with higher therapeutic value. CEA is one of
the pharmacoeconomic assessment methods used by the Health
Technology Assessment (HTA) division of Taiwan’s Center for Drug
Evaluation to allocate the limited healthcare resources.This is parti￾cularly important given the country’s increasing healthcare expen￾ditures [15]. In view of the clinical benefit of PARP inhibitors as
maintenance therapy, we have much interest in their economic
value in treating patients with platinum-sensitive recurrent ovarian
cancer. Recently published CEAs have suggested that routine use of
PARP inhibitors in the recurrent stage is not cost-effective from
a healthcare perspective [16–18].Therefore, we focus on the cost￾effectiveness (CE) of the newly approved olaparib and niraparib
compared to placebo as maintenance therapies in platinum￾sensitive recurrent ovarian cancer patients from a healthcare
perspective.
2. Methods
2.1. Model structure
We created a decision analysis model to compare the CE of three
maintenance treatment strategies in platinum-sensitive recurrent
ovarian cancer patients with or without gBRCA who had previously
been treated with platinum–based chemotherapy:(1) test for
gBRCA mutation, followed by target olaparib use; (2)test for
gBRCA mutation, followed by target niraparib use; (3) placebo
(observation). Model inputs for the cost of resource use were
retrieved from the NHIA reimbursement price list reported in
2020. Clinical outcomes were assessed using PFS life-years based
on the published results of the SOLO-2 /ENGOT-Ov21 trial for
olaparib and ENGOT-OV16/NOVA trial for niraparib [12,13].The 24-
month time horizon was chosen based on survival data from the
two clinical trials. All costs of resource use were discounted at a real
annual rate of 3% to adjust for the current relative value of the
Taiwan dollar. We assumed a willingness-to-pay (WTP) threshold of
NT$2,602,404 (US$92,943;US$1 = NT$28) per quality-adjusted life￾year (QALY) gained. This WTP threshold was suggested by the
World Health Organization [19,20]. The model was constructed
from the healthcare perspective using TreeAge Pro Suite 2020
software (R1.0 Release; TreeAge Inc., Williamstown, MA).
2.2. Treatment strategy
The treatment regimens for hypothetical patients were
derived from two clinical trials [12,13].According to expert
opinions, the dose and schedule, PFS, and adverse drug events
were assumed to be the same as those in the trials.Patients in
the olaparib group received 300 mg olaparib as a 150 mg
tablet twice daily. In the nirapirab group, patients received
300 mg niraparib once daily. Both groups were matched with
placebo tablets until disease progression. Each treatment cycle
was 28 days per year. Grade 3 or above adverse events (SAE)
could be managed by treatment interruption and dose
reduction.
2.3. Resource utilization and costs
The resource use costs included drug costs, laboratory tests
for monitoring, diagnostic imaging (e.g. MRI), physician visits,
pharmacy dispensations, administration, and severe adverse
events (SAEs) (Table 1).The costs of drugs and monitoring
costs were calculated based on a 28-day cycle and the
resource use rates described in the two clinical trials. Dose
adjustments were used to reflect treatment of severe AEs that
occurred. The drug price for the niraparib tablet was assumed
to be the 2017 RED BOOK wholesale price in the USA [21] and
the discount percentage was estimated based on the reim￾bursement price of olaparib. Based on expert opinion, we
only included severe grade 3 or greater AEs because they
may require active treatment and hospitalization, which
increases the costs. The management costs for SAE were
obtained from the literature and multiplied by the SAE inci￾dence rates to estimate the total costs associated with AEs by
treatment [22](Table 1).
Table 1. Direct medical cost imputed in decision tree model.
Parameters Base-Case Value Total regimen cost, NTD Source
Drug Cost Unit cost, NTD Olaparib Naparib
Olaparib, 150 mg/ tablet 1600 (USD57.00) 358,400 NHI reimbursement price for no dose reduction
Olaparib, 150 mg/ tablet 1600 (USD57.00) 2,016,000 NHI reimbursement price for dose reduction due to SAEs
Niraparib, 100 mg/table 1921 (USD 69.00) 322,728 Assumed reimbursement price for no dose reduction
Niraparib, 100 mg/table 1921 (USD 69.00) 1,828,792 Assumed reimbursement price for dose reduction due to SAEs
Administration cost
Physician OPD visit 356 (USD 13.00) 7476 8900 NHI reimbursement price
Dispensing fee 69 (USD 2.5) 897 966 NHI reimbursement price
Follow up monitoring cost
CA 125 test 400 (USD 14.3) 8400 10,000 NHI reimbursement price
Complete blood count 200 (USD 7.00) 4200 5400 NHI reimbursement price
MRI with contrast 13,945 (USD 498) 125,505 153,395 NHI reimbursement price
Managing grade 3/4 adverse events
Anemia 107,772 143,528 Reference 19
Neutropenia 24,052 94,276 Reference 19
Thrombocytopenia 6356 214,814 Reference 19
Remarks:MRI, Magnetic resonance imaging; OPD, out-patient department; CA, cancer antigen, Cost 1 USD = NTD 28.
2 J. H. LEUNG ET AL.
2.4. Utilities
Utility data for nirapirab and olaparib were obtained from the
ENGOT-OV16/NOVA trial and the health-related quality of life
study of SOLO-2 /ENGOT-Ov21, respectively [12,13].
2.5. Effectiveness estimates
The primary endpoints of the clinical trials were PFS. In our
study, PFS life-years were used as the effectiveness out￾come for the base-case scenario. Effectiveness estimates
were obtained from the SOLO-2/ENGOT-Ov21 trial and
ENGOT-OV16/NOVA trial. In the SOLO-2/ENGOT-Ov21 trial,
the proportion of patients with gBRCA in the olaparib
group and placebo group was 97%. Median PFS with ola￾parib maintenance therapy for gBRCA and non-gBRCA was
19.10 months and 7.4 months, respectively [13]. In the
ENGOT-OV16/NOVA trial, the proportion of patients with
gBRCA mutations in the niraparib group and placebo
group was 98.6% and 93.9%, respectively. In the gBRCA
cohort, the median PFS was longer with niraparib versus
placebo (21.0 months vs. 5.5 months). For patients in the
homologous recombination deficiency (HRD) non-gBRCA
cohort, the median PFS with niraparib versus placebo was
12.9 months and 3.8 months, respectively [12].
2.6. Sensitivity analyses
Sensitivity analyses were performed to validate the robust￾ness of our model. The parameters were hypothesized to
change over a range of ± 30% relative to the value used
for the base case (Table 2). A tornado diagram was used to
determine which parameters had the greatest impact on
the incremental cost-effectiveness ratio (ICER). We then
performed one-way sensitivity analysis by varying one key
parameter, identified by tornado diagram, at a time within
the assumed range as mentioned above. In addition,
the acquisition cost of niraparib and the total cost of main￾tenance regimen were reduced by 16.7% and 30% to test at
what cost of niraparib would be cost-effective compared to
olpaparib and placebo at WTP threshold of NT$ 2,602,404.
A probabilistic sensitivity analysis (PSA) was performed
by varying all parameters simultaneously with 1000 Monte￾Carlo simulations, each run with randomly selected sam￾ples from the distributions of model inputs. Each para￾meter was assigned a distribution based on its
characteristics in the PSA. Beta distributions were used
for effectiveness and probabilities in clinical health-state.
These distributions were constrained to the interval
between 0 and 1 [23,24]. Distribution parameters alpha
and beta were calculated based on the mean and standard
deviation of our estimates and the default formulas pro￾vided in the TreeAge Pro Suite 2020 software. PERT dis￾tributions were used for costs with greater influence on
the ICER because it generated a distribution that may
better reflect the realistic probability distribution. The
PERT distribution can provide a close fit to the normal
distribution based on the values provided [25,26].
Therefore, PERT distribution parameters were based on
the maximum and minimum of our estimates, with the
default formulas provided in TreeAge Pro Suite 2020 soft￾ware. CE acceptability curves were obtained by plotting
the probabilities of each treatment strategy against WTP
over 1,000 iterations in the model.
The scenario sensitivity analysis simulated quality￾adjusted PFS life-years (QA-PFS) as an effectiveness out￾come. QA-PFS were calculated by multiplying the PFS life￾years with health-state utility values (HSUVs), which were
adjusted by disutility to reflect the influence of grade 3 or
4 SAEs on the quality of life in these patients. PFS was
adjusted downward by QA-PFS.
2.7. Cost-effectiveness analysis
Costs were defined as the total resource use costs in each
regimen. The effectiveness was defined as ICER, repre￾sented by the incremental costs associated with one incre￾mental unit of effectiveness (PFS life-years) gain in the
base case scenario. Incremental costs and effectiveness
were estimated by the difference between the two treat￾ment regimens being compared. The ICER was calculated
by the incremental cost divided by the incremental
effectiveness.
1.75 1.225–2.275 Beta
PFS for Non-gBRCA 0.775 0.543–1.008 Beta
Placebo PFS for gBRCA 0.46 0.332–0.598 Beta
PFS for Non-gBRCA 0.32 0.224–0.416 Beta
Probabilities, %
Rate of gBRCA mutation 0.20 0.14–0.26 Beta
Probabilities of dose reduction
Olaparib 0.25 0.175–0.325 Beta
Niraparib 0.665 0.466–0.865 Beta
Probabilities of severe grade 3–4 adverse events
Olaparib
Anemia 19.00 13.3–24.7 Beta
Neutropenia 5.00 3.5–6.5 Beta
Niraparib
Anemia 25.30 17.7–32.9 Beta
Neutropenia 19.60 13.7–25.5 Beta
thrombocytopenia 33.80 23.7–43.9 Beta
Utility Values
Niraparib gBRCA 0.840 0.588–1.092 Beta
Non-gBRCA 0.835 0.626 − 1.044 Beta
Olaparib gBRCA 0.80 0.56–1.04 Beta
Non-gBRCA 0.80 0.56–1.04 Beta
Abbreviations: PFS-LY, progression free survival-life years; gBRCA, germline BRCA
Costs are in 2020 (1USD = NTD 28)
EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 3
3. Results
3.1. Base case
Results of the baseline CEA are shown in (Table 3). ICER was
used to measure the CE of these regimens. For olaparib, the
drug cost of total cycles for patients without a dose reduction
was NT$358,400 (US$12,800), and for patients with a dose
reduction was NT$2,016,000 (US$72,000). For niraparib, the
drug cost of total cycles for patients without a dose reduction
was NT$322,728 (US$11,526), and for patients with a dose
reduction was NT$1,828,792 (US$65,314).The resource use
costs for the treatment of patients with severe AEs were NT
$138,180 (US$4935) and NT$452,620 (US$16,165) for olaparib
and niraparib, respectively.The costs for disease monitoring
and management of SAEs are given in (Table 1).
In the base case model, niraparib (NT$1,604,913) was a more
costly regimen than olaparib (NT$992,226) and placebo (NT
$157,693; Table 3A). The total effectiveness (PFS life-years) was
favorable with niraparib (0.97),followed by olaparib (0.81) and
placebo (0.35).For all patients, the ICERs were NT$1,804,785(US
$64,457) per PFS life-year with olaparib and NT$2,340,265 (US
$83,581) per PFS life-year with niraparib.The ICERs for both
regimens were below the WTP threshold, however, olaparib
was more cost-effective than niraparib (Table 3A) .
We also examined the impact of the presence of absence of
BRCA mutation on the ICERs. The PFS time was longer in the
gBRCA mutation group than the non-gBRCA multation group
for both olaparib and niraparib. Therefore, the ICERs were
lower than the assumed WTP threshold of NT$2,602,404 in
the gBRCA mutation patients (NT$737,219.5 for olaparib and
NT$1,120,139 for niraparib per PFS life-year;Table 3B) .In
patients with non-gBRCA mutation, the ICERs were higher
than the assumed WTP threshold of NT$2,602,404 (NT
$2,828,924 for olaparib and NT$3,216,044 for niraparib per
PFS life-year; Table 3C). These results indicated that olaparib
and niraparib may be cost-effective in the treatment of
patients with gBRCA mutation. However, for non-gBRCA
patients, neither regimen was cost-effective.
3.2. Sensitivity analyses
The tornado diagram showed that the ICERs were most sensi￾tive to the estimates of median PFS for the regimens and the
total cost of niraparib for treating patients with non-gBRCA
multation (Figure 1). In one-way sensitivity analyses, the ICERs
changed substantially when PFS and total cost of niraparib for
patients with non-gBRCA multation were varied within the
appropriate hypothesized range. The results were robust gen￾erally. The ICERs for olaparib remained below the acceptable
WTP threshold. However, the ICERs for niraparib were compar￾able to olaparib if the acquisition cost and the total cost of
niraparib were reduced by 16.7% and 30% (Table 4).
The PSA generated the CE acceptability curve of the three
regimens (olaparib, niraparib, placebo); the CE acceptability
curve with the higher percentage is considered to be more
cost-effective. Thus, olaparib was about 100% cost-effective at
WTP of NT$2,602,404. At WTP of less than NT$ 1,487,088 per
PFS life-years, neither of the olaparib and niraparib regimens
were cost-effective compared with placebo (Figure 2).
In the scenario sensitivity analysis, we implemented the
adjusted QA-PFS as the effectiveness outcome. The incremen￾tal QA-PFS was 0.28 for olaparib and 0.32 for niraparib.
Compared to common baseline placebo, the ICERs for both
regimens were higher than the acceptable WTP threshold (NT
$ 2,965,645 for olaparib; NT$4,594,348) (Table 5), meaning that
both regimens were not cost-effective because the cost of
gaining one additional QA-PFS was very expensive.
4. Discussion
Costs of cancer treatment have continuously escalated with
the launch of new novel target agents in the healthcare
market worldwide. It has become a substantial financial bur￾den to both patients and healthcare payers [26]. It is likely
impossible for the NHIA healthcare program to pay for all new
novel target drugs within an acceptable CE threshold of less
than NT$2,602,404 (three times a country’s per-capita gross
domestic product).
Our analysis indicates that olaparib is likely to be more cost￾effective than niraparib as a maintenance therapy in patients
with platinum-sensitive recurrent ovarian cancer, particularly for
patients with gBRCA mutation. Our results were not consistent
with the previous CEAs performed by Smith et al. and Zhong
et al. [17,27]. They reported an ICER of 258,864 USD per PFS life￾year and 226,000 USD per PFS life-year gain for olaparib main￾tenance therapy compared to placebo in patients with gBRCA
Table 3. Results of cost-effectiveness of olaparib and niraparib.
Treatment strategy Total Cost (NTD)
Incr Cost
compared to placebo
Total Effectiveness
(PFS Life-Year)
Incr eff (as PFS Life-Year
(compared with placebo)
ICER
(Incr cost per Incr eff)
(compared with placebo)
A.Base- case cost-effectiveness in platinum-sensitive recurrent ovarian cancer patients
Placebo 157,693 0.35
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,604,913 1,447,220 0.97 0.62 2,340,265
B. Base- case cost-effectiveness in patients with gBRCA mutation
Placebo 157,693 0.46
Olaparib 992,226 834,533 1.59 1.13 737,219.5
Niraparib 1,604,913 1,447,220 1.75 1.29 1,120,139
C.Base- case cost-effectiveness in patients with non-gBRCA mutation
Placebo 157,693 0.33
Olaparib 992,226 834,533 0.62 0.29 2,828,924
Niraparib 1,604,913 1,447,220 0.78 0.45 3,216,044
Remarks: Incr = incremental; eff = effectiveness;PFS = progression free survival; gBRCA = germline; ICER = incremental cost-effectiveness ratio
4 J. H. LEUNG ET AL.
mutation. The ICER in our CEA study was NT$1,804,785 (US
$64,457) per PFS life-year for olaparib maintenance therapy,
which is lower than that reported by Smith et al. and Zhong
et al. The difference may be due to the lower acquisition drug
cost, which was reduced by our NHIA annually. However, if the
acquisition costs and the total cost of niraparib regimen were
assumed to reduce by 16.7% and 30%, niraparib may be con￾sidered a cost-effective maintenance therapy (NT$
1,707,096 per PFS-LY) for patients with severe AE and non￾gBRCA mutation (Table 4). This result was in line with the
current published study conducted by Dottino JA. They con￾cluded that niraparib may be considered a cost-effective main￾tenance therapy by lowering its supply cost [28] .
Our study indicated that gBRCA status might also influence
ICERs. Olariparib or niraparib maintenance therapy was cost￾effective for patients with gBRCA mutation (NT$737,219.5 (US
$26,329) for olaparib and NT$1,120,139 (US$40,005) for niraparib
per PFS life-year)(Table 3B). It is lower than that reported by
Zhong et al. They reported the ICER for olaparib and niraparib
maintenance therapy compared to placebo in patients with
gBRCA mutation were 226,000 USD per PFS life-year gain and
197,000 USD per PFS life-year gain, respectively. The difference
could be due to the drug cost, costs of managing SAEs, and the
treatment cycle. Dose reducton induced by SAE might also be
another factor impacted the ICERs. The drug cost for olaparib
reimbursed by our healthcare payer for 24 month treatment was
NT$358,400 (US$12,800) for patients without a dose reduction
and NT$2,016,000 (US$72,000) for patients with a dose reduc￾tion. For niraparib, the drug cost of 24 months treatment for
patients without a dose reduction was NT$322,728 (US$11,526),
and for patients with a dose reduction was NT$1,828,792 (US
$65,314). (Table 1). Therefore, the reimbursement cost for both
drugs are lower than that paid by healthcare payers in the United
States. Our results also suggest that olaparib or niraparib main￾tenance therapy is not cost-effective for patients with non￾gBRCA1/2 mutation. The ICERs were unfavorable for both regi￾mens, possibly because less PFS benefit was shown in non￾gBRCA patients without evidence of tumor HRD. In contrast,
the greatest PFS benefit of olaparib or niraparib was shown in
patients with gBRCA1/2.
The main strength of our study was that the drug price for
olaparib and resource use in the model were derived from the
NHIA reimbursement price list, which is an open and transpar￾ent price list. In addition, the WTP threshold is commonly used
by the healthcare payer in their evaluation process [29].
There are several limitations of this study. First, we estimated
the management costs of severe adverse hematological events
based on the literature for all cancers, which may not accurately
reflect the cost of severe hematological AEs in ovarian cancer
patients [22]. The cost may be over-estimated. Secondly, the
cost of the gBRCA test was not included in this study because
it is an out-of-pocket charge for patients. The total cost of
resource use may be underestimated. Third, assumptions
about the HSUVs in our study were based on the NOVO trial
and SOLO2 extension study to evaluate the health-related qual￾ity of life of patients with gBRCA [11,30]. In the NOVO trial, the
FACT-O questionnaire and European Quality of Life
5-Dimensions questionnaire (EQ-5D-5 L)were used for patients
with and without gBRCA. The results of the two scales were
reported separately and in detail. However, in the SOLO2 trial,
they also used FACT-O and EQ-5D-5 L to evaluate the health￾related quality of life but only focused on patients with gBRCA,
supplying no data for non-gBRCA and placebo. The results only
provided a mean utility value. Based on the inconsistent data
from these trials, we assessed the impact of QA-PFS in the
sensitivity analysis. Finally, PFS was used as the outcome mea￾sure in our study because no overall survival data were available
for olaparib and niraparib as maintenance therapy in the SOLO2
and NOVO trials.The effectiveness may be under-estimated.
However, the sensitivity analysis was robust.
Figure 1. Tornado diagram for key parameters impact the ICERs.
EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 5
5. Conclusion
Our study demonstrated that olaparib and niraparib had ICERs
below NT$2,602,404 per PFS-LYS as compared with placebo.
Olaparib or niraparib maintenance therapy in patients with
gBRCA mutations is more cost-effective than in patients with￾out gBRCA mutations. If QA-PFS is used as an outcome mea￾sure and implemented into the model, neither olaparib nor
niraparib is cost-effective. In conclusion, olaparib is more cost￾effective than niraparib in using the acceptable WTP threshold
of NT$2,602,404 per PFS-LYS from the Taiwan NHAI perspec￾tive. However, we need further mature overall survival data to
support these results.
Notes on contributors
AC and JHL conceived of the presented idea and built the model. LHC
analyzed the data. AC wrote the manuscript in consultation with LHF and
WSY. All authors discussed the results and contributed to the final manuscript.
Table 4. One- way sensitivity analysis for important parameters.
Parameters Base-case Value Strategy Cost Inc cost Eff (PFS-LY)
Inc Eff
(PFS-LY)
ICER
(PFS-LY)
Costs for Niraparib,
Non-gBRCA, dose reduction
1,947,229 1,363,060
(−30%)
Placebo 157,693 0.00 0.35 0.00 0.00
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,294,135 301,909 0.97 0.16 1,935,315
2,531,398 (+30%) Placebo 157,693 0.00 0.35 0.00 0.00
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,915,691 923,465 0.97 0.156 5,919,647
Costs for Niraparib,
Non-gBRCA, dose reduction
1,701,861
Drug cost (−16.7%)
1,155,563(−30%) Placebo 157,693 0.00 0.35 0.00 0.00
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,126,798 134,573 0.97 0.16 1,707,096
2,146,047 (+30%) Placebo 157,693 0.00 0.35 0.00 0.00
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,653,736 661,510 0.97 0.16 4,240,449
PFS, year
Niraparib, non-gBRCA,
no dose reduction
0.775 0.543 (−30%) Placebo 157,693 0.00 0.35 0.00 0.00
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,604,913 612,687 0.91 0.09 6,530,175
1.008 (+30%) Placebo 157,693 0.00 0.35 0.00 0.00
Olaparib 992,226 834,533 0.81 0.46 1,804,785
Niraparib 1,604,913 612,687 1.03 0.22 2,804,779
PFS, year
Niraparib, gBRCA,
6 J. H. LEUNG ET AL.
Declaration of interest
The authors have no relevant affiliations or financial involvement with any
organization or entity with a financial interest in or financial conflict with
the subject matter or materials discussed in the manuscript. This includes
employment, consultancies, honoraria, stock ownership or options, expert
testimony, grants or patents received or pending, or royalties.
Consent for publication
Not applicable.
Reviewers Disclosure
Peer reviewers on this manuscript have no relevant financial relationships
or otherwise to disclose.
Ethics approval and consent to participate
This work did not require any written patient consent and the local ethics
approval.
Availability of data and materials
The datasets supporting the conclusions of this article are included within
the article.
Funding
This paper was not funded.
References
Papers of special note have been highlighted as either of interest (•) or of
considerable interest (••) to readers.
1. American Cancer Society. cited 2020 Dec. https://www.cancer.org/
cancer/ovarian-cancer/about/key-statistics.html
2. Taiwan Cancer registry. 2019.Cancer registry annual report, 2018
TAIWAN
3. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer
J Clin. 2020;70(1):7–30. Accessed January 2021
4. Herzog TJ, Armstrong DK. First-line chemotherapy for advanced
(stageIII or IV) epithelial ovarian, fallopian tubal, and peritoneal
cancer; Vora SR,ed. UpToDate; 2021 Feb 18; cited [2021 Mar 3].
Available from: https://www.uptodate.com/contents/first-line￾chemotherapy-for-advanced-stage-iii-or-iv-epithelial-ovarian￾fallopian-tubal-and-peritoneal-cancer?source=see_
link#H760206841
5. Tewari KS, Burger RA, Enserro D, et al. Final overall survival of
a randomized trial of bevacizumab for primary treatment of
ovarian. Cancer. 2019;37:2317–2328.
6. Oza AM, Cook AD, Pfisterer J, et al. Standard chemotherapy with or
without bevacizumab for women with newly diagnosed ovarian
cancer (ICON7): overall survival results of a phase 3 randomised
trial. Lancet Oncol. 2015;16(8):928–936.
7. Wang Q, Peng H, Qi X, et al. Targeted therapies in gynecologica
cancers: a comprehensive review of clinical evidence. Signal
Transduct Target Ther. 2020;5(1):137.
•• The article reporting on approval for three PARPis.
8. Research C for DE and Approved Drugs-Niraparib (ZEJULA). 2018.
[cited 2018 Jul 13]. https://www.fda.gov/drugs/informationon
drugs/approveddrugs/ucm548487.htm
9. .Research C for DE and Approved drugs-FDA approves olaparibtab￾lets for maintenance treatment in ovarian cancer. 2018 [cited 2021
Mar 13]. https://www.fda.gov/Drugs/InformationOnDrugs/
ApprovedDrugs/ucm572143.htm
Figure 2. Probability sensitivity analysis for parameter of PFS-life years. The curves represent the percentage of cost-effective iteration of each regimen at different
WTP thersholds. The regiments with higher percentage is considered to be more cost-effective at WTP value. When WTP exceeded NT$1,487,088 per PFS life years,
olaparib is more cost-effective regimen.
Table 5. Cost-effectiveness analysis of scenario sensitivity analysis.
Treatment strategy Total Cost (NTD)
Incr Cost
compared to placebo
Total Effectiveness
(QA-PFS Life-Year)
Incr eff (as QA-PFS Life-Year
(compared with placebo)
ICER
(Incr cost per Incr eff)
(compared with placebo)
Placebo 157,693 0.28
Olaparib 992,226 834,533 0.56 0.28 2,965,645
Niraparib 1,604,913 1,447,220 0.60 0.32 4,594,348
EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 7
10. Research C for DE and. Approved drugs-Rucaparib. 2018 [cited
2021 Mar 13]. https://www.fda.gov/Drugs/InformationOnDrugs/
ApprovedDrugs/ucm533891.htm
11. National Health insurance Administration. Ministry of health and
welfare.national health insurance drug payment items and pay￾ment standards; [cited 2021 Mar 3]. https://www.nhi.gov.tw/
Content_List.aspx?n=E70D4F1BD029DC37&topn=
3FC7D09599D25979.
• The pivotal clinical trial that was the main clinical source of the
cost-effectiveness analysis.
12. Mirza MR, Monk BJ, Herrstedt J, et al. Niraparib maintenance ther￾apy in platinum-sensitive,recurrent ovarian cancer. N Engl J Med.
2016;375(22):2154–2164.
13. Pujade-Lauraine E, Ledermann JA, Selle F, et al. Olaparib tablets as
maintenance therapy in patients with platinum-sensitive, relapsed
ovarian cancer and a BRCA1/2 mutation (SOLO2/ENGOT-Ov21):
a doubleblind, randomised, placebo-controlled, phase 3 trial.
Lancet Oncol. 2017;18(9):1274–1284.
14. Ledermann J, Harter P, Gourley C, et al. Olaparib maintenance
therapy in patients with platinum-sensitive relapsed serous ovarian
cancer: a preplanned retrospective analysis of outcomes by BRCA
status in a randomised phase 2 trial. Lancet Oncol. 2014;15(8):
852–861.
•• An interesting article on cost-Effectiveness analysis of
Niraparib and Olaparib as Maintenance Therapy for Patients
with Platinum-Sensitive Recurrent Ovarian Cancer.
15. Center for Drug Evaluation. Taiwan. [cited 2021 Mar 3]. https://
www.cde.org.tw/eng/HTA/business
16. Secord AA, Barnett JC, Ledermann JA, et al. Cost-effectiveness of
BRCA1 and BRCA2 mutation testingto target PARP inhibitor use in
platinum-sensitiverecurrent ovarian cancer. Int J Gynecol Cancer.
2013;23(5):846–852.
17. Zhong L, Tran AT, Tomasino T, et al. Cost-effectiveness of niraparib
and olaparib as maintenance therapy for patients with
platinum-sensitive recurrent ovarian cancer.J. Manage Care Spec
Pharm. 2018;24(12):1219–1228.
18. Cheng LJ, Wrong G, Chay WY, et al. Cost-effectiveness of ola￾parib maintenance therapy when used with and without
restriction by BRCA1/2 mutation status for platinum-sensitive
relapsed ovarian cancer. Expert Rev Pharmacoecon Outcomes
Res. 2021;21(3):441–448.
19. Murray CJL, Evans DB, Acharya A, et al. Development of WHO
guidelines on generalized cost-effectiveness analysis. Health Econ.
2000;9(3):235–251.
20. International Monetary Fund. World economic outlook (April 2019)
GDP current price.[cited 2018 Aug 12]. http://www.imf.org/exter
nal/datamapper/NGDPD@WEO/OEMDC/ADVEC/WEOWORLD25
21. Truven Health Analytics, IBM Watson. Micromedex RED BOOK.data￾base; RED BOOK. 2019. [cited 2020 Sept 28]. Available from: http://
truvenhealth.com/Products/Micromedex/Product-Suites/Clinical￾Knowledge/RED-BOOK
22. Wong W, Yim YM, Kim A, et al. Assessment of costs associated with
adverse events in patients with cancer. PloS One. 2018;13(4):
e0196007–e0196007.
23. Gray A, Clarke PM, Wolstenholme JL, et al. Applied methods of
cost- effectiveness analysis in health care. Oxford, New York.:
Oxford University Press; 2010.
24. Sculpher M. The use of a probabilistic sensitivity analysis for deci￾sion making: the example of drug eluting stents. Centre for Health
Economics University of York, uk. 2008 Aug 17.
25. Overview of the beta-PERT distribution for modeling expert data.
[cited 2021 Mar 21]. https://www.riskamp.com/beta-pert
26. Mariotto AB, Yabroff KR, Shao Y, et al. Projections of the cost of
cancer care in the United States: 2010–2020.J. Natl Cancer Inst.
2011;103(2):117–128.
27. Smith HJ, Walters Haygood CL, Arend RC, et al. PARP inhibitor
maintenance therapy for patients with platinum-sensitive recurrent
ovarian cancer: a cost-effectiveness analysis. Gynecol Oncol.
2015;139(1):59–62.
28. Dottino JA, Moss HA, Lu KH, et al. U.S. food and drug
dministration-approved poly (ADP-Ribose) polymerase inhibitor
maintenance therapy for recurrent ovarian cancer: a
cost-effectiveness analysis. Obstet Gynecol. 2019;133(4):795–802.
29. Shiroiwa T, Sung YK, Fukuda T, et al. International survey on
willingness-to-pay (WTP) for one additional QALY gained: what is
the threshold of cost effectiveness? Health Econ. 2010;19(4):422–437.
30. Friedlander M, Gebski V, Gibbs E, et al. Health-related quality of life
and patient-centred outcomes with olaparib maintenance after
chemotherapy in patients with platinum-sensitive, relapsed ovarian MK-4827
cancer and a BRCA1/2 mutation (SOLO2/ENGOT Ov-21): a
placebo-controlled, phase 3 randomised trial. Lancet Oncol.
2018;19(8):1126–1134.
8 J. H. LEUNG ET AL.