Olaparib plus bevacizumab as a first-line maintenance treatment for patients with advanced ovarian cancer by molecular status: an updated PAOLA-1 based cost-effectiveness analysis

Abstract

Objective

The PAOLA-1 trial (NCT02477644) reported final survival benefit associated with olaparib plus bevacizumab maintenance treatment of patients with advanced ovarian cancer (AOC) based on molecular status. Our aimed to compare the cost-effectiveness of olaparib plus bevacizumab for overall patients, patients with a breast cancer susceptibility genes (BRCA) mutation, homologous recombination deficiency (HRD), or HRD without BRCA mutations AOC from the context of the American healthcare system.

Methods

Analysis of health outcomes in life-years (LYs), quality-adjusted life-years (QALYs), and the incremental cost-effectiveness ratio (ICER) in various molecular status-based AOC patient at a $150,000/QALY of willingness-to-pay was performed using a state-transitioned Markov model with a 20-year time horizon. Meanwhile, sensitivity analyses assessments were also used to gauge the model’s stability.

Results

The ICERs of olaparib plus bevacizumab versus bevacizumab alone were $487,428 ($374,758), $249,579 ($191,649), $258,859 ($198,739), and $270,736 ($206,640) per QALY (LY) in the overall patients, patients with BRCA mutations, patients with HRD, and patients with HRD without BRCA mutations AOC, respectively, which indicated that The ICERs was higher than $150,000/QALY in the US. Progression-free survival (PFS) value and olaparib cost emerged as the primary influencing factors of these findings in the sensitivity analysis.

Conclusion

At current cost levels, olaparib plus bevacizumab treatment is not a cost-effective treatment for patients with AOC regardless of their molecular status in the US. However, this maintenance treatment may be more favorable health advantages for patients with BRAC mutations AOC.

Synopsis

The Markov model was first established based on the updated PAOLA-1 trial. We evaluated the cost-effectiveness of olaparib plus bevacizumab (OB) for advanced ovarian cancer (AOC). The incremental cost-effectiveness ratio of OB versus bevacizumab was $487,428/quality-adjusted life-year for overall patients with AOC. OB is not cost-effective for AOC regardless of their molecular status in the US.

INTRODUCTION

Ovarian cancer (OC) is the fifth most lethal and the eleventh most common cancer in American women with forecasts estimating that there will be 19,880 and 12,810 OC diagnoses and deaths in 2022 alone [1]. Impairment of double-stranded DNA repair as a result of homologous recombination deficiency (HRD) owing to the inactivation of breast cancer susceptibility genes breast cancer susceptibility genes 1 and 2 (BRCA1 and BRCA2) is a prevalent oncogenic driver in OC patients [2]. Women harboring these BRCA gene mutations are at an elevated risk of OC, with this being the risk factor associated with the highest odds of OC development [3,4]. Epithelial OC is the most common type of ovarian cancer, with approximately 70% of cases diagnosed at advanced stages and a poor prognosis, with a 5-year survival rate of less than 10% [3].

The standard first-line treatment for newly diagnosed advanced ovarian cancer (AOC) patients over the last decade has been surgical tumor cytoreduction followed by an adjuvant chemotherapy regimen composed of platinum and nonplatinum (taxane-based) drugs [5]. In the large phase III ICON7 (ISRCTN91273375) and GOG-0218 (NCT00262847) trials, individuals with AOC who received surgical tumor cytoreduction and chemotherapy had a median progression-free survival (PFS) of 10 to 17 months, with the vast majority of them having tumor recurrence [6,7]. Therefore, we need innovative drugs or therapeutic strategies, irrespective of surgical or molecular status, that have a significant clinical response.

Bevacizumab, a monoclonal anti-vascular endothelial growth factor A (anti-VEGF-A) antibody, combined with chemotherapy and then sustained bevacizumab maintenance therapy has shown substantial clinical benefit for patients with AOC of all stages of the disease and with no evidence of disease progression after surgery and has subsequently become the standard choice for patients with newly diagnosed AOC [6,7,8,9,10,11]. Approximately 50% of all OC patients exhibit HRD-positive tumors, the majority of which are driven by BRCA mutations [12]. Therefore, there is a renewed focus on the early identification of disease-related biomarkers and molecular status indicators for use in deciding upon the most effective therapy approaches. The oral poly (ADP-ribose) polymerase (PARP) inhibitor olaparib shows promise as a cost-effective and clinically-effective treatment for AOC in its development as a first-line medication [13]. In the phase III study PAOLA-1 (NCT02477644), olaparib plus bevacizumab were superior to bevacizumab for boosting the PFS of overall patients with AOC (median 22.1 months vs. 16.6 months, hazard ratio [HR]=0.59; 95% confidence interval [CI]=0.49–0.72), patients with HRD tumor (46.8 vs. 17.6, HR=0.41; 95% CI=0.32–0.54), patients with a tumor BRCA mutation (37.2 vs. 21.7, HR=0.31; 95% CI=0.20–0.47), and Patients with HRD tumors without a BRCA mutation (28.1 vs. 16.6, HR=0.43; 95% CI=0.28–0.66) [14,15]. In addition, olaparib plus bevacizumab significantly prolonged overall survival (OS) compared with bevacizumab in patients with HRD tumor (75.2 vs. 57.3, HR=0.62; 95% CI=0.45–0.85) and patients with a tumor BRCA mutation (75.2 vs. 66.9, HR=0.60; 95% CI=0.39–0.93) [15]. However, the prolongation of OS was not statistically significant in patients with AOC (56.5 vs. 51.6, HR=0.92; 95% CI=0.76–1.12), and patients with HRD tumors without a BRCA mutation (NR vs. 52, HR=0.71; 95% CI=0.45–1.13) [15]. Due to the clinical relevance of these results, on May 8, 2020, the Food and Drug Administration (FDA) approved the combination therapy of olaparib plus bevacizumab for the treatment of patients with advanced epithelial ovarian, fallopian tube, or primary peritoneal carcinoma who have tumors that have tested positive for HRD and who have either a partial or complete response to first-line platinum-based chemotherapy [16].

The clinical application of olaparib and bevacizumab must be guided by the relative efficacy and economic value of such treatment, underscoring a need for health economic analyses focused on survival benefits of patients with AOC. Accurate patient selection based on early biomarkers to maximize efficacy must also be tempered by strong evidence of cost-effectiveness to ensure that oncologists provide patients with the most optimal therapeutic regimen for their individual situations. In an effort to address these issues, the cost-effectiveness of olaparib plus bevacizumab vs single-agent bevacizumab for first-line maintenance therapy in patients with AOC in selected molecular status subgroups was investigated in this research.

MATERIALS AND METHODS

1. Clinical data inputs

Patients’ baseline data from the PAOLA-1 trial were employed in a Markov model to create a hypothetical patient cohort, ultimately including 806 patients with AOC [14,15]. Of the 537 and 269 patients respectively assigned at random to undergo olaparib plus bevacizumab and bevacizumab monotherapy, 255 (47.5%) and 132 (49.1%) were HRD-positive cohort, 161 (30.0%) and 80 (29.7%) were BRCA mutations cohort, and 97 (18.1%) and 55 (20.5%) were HRD-positive without BRCA mutations cohort, respectively [14,15]. Based on the PAOLA-1 trials, olaparib was administered orally at a dose of 300 mg twice daily for a total duration of up to 2 years, and bevacizumab was administered intravenously at a dose of 15 mg/Kg every 3 weeks for a total duration of up to 15 months [14]. The average age, body weight, body surface area, and serum creatinine levels of these patients were 60 years, 70 kg, 1.84 m², and 1 mg/dL, respectively (Table 1) [14,17,18]. Imaging studies using computed tomography (CT) were performed every 24 weeks to monitor tumor progression and assess patient progress [14,15]. The assigned first-line maintenance treatment regimens were maintained until progressive disease (PD) or unacceptable adverse events (AEs) were observed, after which carboplatin plus paclitaxel was administered to 260 (48.4%) and 164 (61.0%) patients in the olaparib plus bevacizumab and bevacizumab monotherapy groups, respectively (Table 1), as per study guidelines [19,20]. The remaining patients were provided with the best supportive care (BSC) until death occurred, and terminal care was provided in all death cases. For further details regarding drug dosages, administration methods, and price per unit, see Table S1 of supplementary materials. This inquiry was guided according to the checklist of the reporting standards regarding the consolidated health economic evaluation (CHEERS) (Table S2).

Table 1. Model parameters: key clinical and health preference data.

Parameters			Baseline value	Range		Reference	Distribution
				Minimum	Maximum
Clinical data
	Weibull survival model for OS of olaparib plus bevacizumab
		Overall patients	Scale=0.0032233, Shape=1.3412093	-	-	[14,15]	-
		Patients with a tumor BRCA mutation	Scale=0.0008312, Shape=1.4656368	-	-		-
		Patients with HRD tumors	Scale=0.0013868, Shape=1.4043253	-	-		-
		Patients with HRD tumors without a BRCA mutation	Scale=0.0017414, Shape=1.4345009	-	-		-
	Weibull survival model for PFS of olaparib plus bevacizumab
		Overall patients	Scale=0.0088569, Shape=1.3634685	-	-	[14,15]	-
		Patients with a tumor BRCA mutation	Scale=0.0013118, Shape=1.6451884	-	-		-
		Patients with HRD tumors	Scale=0.004515, Shape=1.359471	-	-		-
		Patients with HRD tumors without a BRCA mutation	Scale=0.0030826, Shape=1.6428677	-	-		-
	Weibull survival model for OS of bevacizumab
		Overall patients	Scale=0.0025282, Shape=1.427602	-	-	[14,15]	-
		Patients with a tumor BRCA mutation	Scale=0.0016336, Shape=1.4217268	-	-		-
		Patients with HRD tumors	Scale=0.0014855, Shape=1.500712	-	-		-
		Patients with HRD tumors without a BRCA mutation	Scale=0.0009159, Shape=1.6748696	-	-		-
	Weibull survival model for PFS of bevacizumab
		Overall patients	Scale=0.013731, Shape=1.38569	-	-	[14,15]	-
		Patients with a tumor BRCA mutation	Scale=0.0054001, Shape=1.5892609	-	-		-
		Patients with HRD tumors	Scale=0.014216, Shape=1.316224	-	-		-
		Patients with HRD tumors without a BRCA mutation	Scale=0.007928, Shape=1.592306	-	-		-
Risk for main AEs in olaparib plus bevacizumab group
	Risk of fatigue		0.050	0.040	0.060	[14]	Beta
	Risk of neutropenia		0.060	0.048	0.072	[14]	Beta
	Risk of lymphopenia		0.070	0.056	0.084	[14]	Beta
	Risk of anemia		0.170	0.136	0.204	[14]	Beta
	Risk of hypertension		0.190	0.152	0.228	[14]	Beta
Risk for main AEs in bevacizumab group
	Risk of hypertension		0.300	0.240	0.360	[14]	Beta
Proportion of receiving active second-line treatment
	Olaparib plus bevacizumab		0.484	0.387	0.581	[19]	Beta
	Bevacizumab		0.610	0.488	0.732	[19]	Beta
Utility and disutility
	Utility of PFS		0.779	0.623	0.935	[24]	Beta
	Utility of PD		0.753	0.602	0.904	[24]	Beta
	Disutility of leukopenia		0.090	0.072	0.108	[21]	Beta
	Disutility of fatigue		0.170	0.136	0.204	[25]	Beta
	Disutility of neutropenia		0	-	-	[25]	-
	Disutility of anemia		0	-	-	[25]	-
	Disutility of hypertension		0	-	-	[25]	-
Body weight (kg)			70	56	84	[17,18]	Normal
Body surface area (m2)			1.84	1.47	2.21	[17,18]	Normal
Discount rate			0.05	0.04	0.06	[22,23]	Uniform

AE, adverse event; BRCA, breast cancer susceptibility genes; HRD, homologous recombination deficiency; OS, overall survival; PD, progressive disease; PFS, progression-free survival.

2. Construct model

A comprehensive Markov model was developed to evaluate the cost-effectiveness of olaparib plus bevacizumab compared with bevacizumab alone as first-line maintenance treatment for AOC from a US societal perspective with the TreeAge Software (TreeAge Pro 2021-, available at: https://www.treeage.com). This state-transition model combined effectiveness and total costs for hypothetical AOC patient groups, and it accounted for three mutually exclusive health states (PFS, PD, and death). At model initiation, all patients were in the PFS state, and at the end of each cycle, they could transition to a PD or death state or could maintain their current state. (Fig. S1). A model cycle of 3 weeks and a 20-year time horizon within which over 99% of patients were expected to reach the death state was established based on clinical trial drug administration, follow-up periods, and survival data.

Transition probabilities were estimated using PAOLA-1 trial data. As precise baseline patient data for this trial were not available, the survival data for individual groups were extracted with GetData Graph Digitizer (version 2.26, available at: http://www.getdata-graph-digitizer.com/index.php) from Kaplan-Meier (KM) curves. The exponential, log-logistic, log-normal, Gompertz, and Weibull distributions were then selected to fit the olaparib plus bevacizumab and bevacizumab groups survival curves by estimating prediction errors as per the Akaike information criterion (AIC) and Bayesian information criterion (BIC) (Fig. S2, Table S3). Of these, the Weibull distribution was ultimately found to most effectively reconstruct the individual patient data, with the γ (scale) and λ (shape) parameter distributions being calculated using R (version 4.1.1, available at: http://www.rproject.org) (Table 1) [17].

Overall costs, life-years (LYs), quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratio (ICER) values were all considered important research outcomes. ICER refers to the incremental cost required to achieve the incremental effect, which compares multiple interventions by comparing the ratio of cost differences and effective output differences between different interventions. ICER=∆C/∆E (∆C is the incremental cost, ∆E is the incremental effect). The calculated ICER result should also be compared with the patient’s willingness-to-pay (WTP) to see whether the intervention is cost-effective. For US payers, the model’s WTP threshold was set at $150,000/QALY [17,21]. An annual discount rate of 5% for healthcare expenses and benefits was additionally implemented [22,23].

3. Utility and cost input

Utilities were used to obtain QALYs by discount LYs. The PAOLA-1 trial collected the global health status-quality of life dimension of the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire (EORTC QLQ-C30), a questionnaire used to determine the quality of life (QoL) from randomization to secondary progression in patients with different disease domains. We assigned each health state different utility weights which range from 0 to 1 (standing for individual’s preference in health condition, death to perfect health) according to published literature. An average healthy utility of 0.779 and 0.753 for PFS and PD status, respectively, was established in US populations [24]. Grade 3-4 treatment-related adverse events (AEs) with a disutility value ≥5% were additionally taken into consideration [21,25].

Direct medical costs in this study included the costs of drugs, laboratory tests, tumor imaging, laboratory tests, HRD testing, germline BRCA testing, administration, AE-related treatments, BSC, and terminal care were calculated as direct medical costs in our study (Table 2). Drug prices were determined using the drug price inquiry website and the Centers for Medicare & Medicaid Services [26,27]. All other direct medical costs were determined based on published reports [25,28,29,30]. Grade 3–4 treatment-related AEs with a disutility value ≥5% were additionally taken into consideration [31]. All medical costs were adjusted by the consumer price index and presented in 2022 US dollars.

Table 2. Cost estimates (US $).

Parameters		Baseline value	Range		Reference	Distribution
			Minimum	Maximum
Drug cost, $/per cycle
	Olaparib	3,657	2,926	4,388	[26]	Gamma
	Bevacizumab	7,326	5,861	8,791	[27]	Gamma
	Carboplatin	23	18	28	[27]	Gamma
	Paclitaxel	35	28	42	[27]	Gamma
Cost of AEs
	Bevacizumab	76	61	91	[25]	Gamma
	Olaparib plus bevacizumab	291	233	349	[25,28]	Gamma
Laboratory per cycle		4	3	5	[26]	Gamma
Tumor imaging per cycle		105	84	126	[25]	Gamma
Administration per cycle		124	99	149	[25]	Gamma
Germline BRCA testing per patient		2,901	2,321	3,481	[29]	Gamma
HRD test per patient		4,682	3,746	5,618	[29]	Gamma
Best supportive care per cycle		4,143	3,314	4,972	[30]	Gamma
Terminal care per patient		85,904	68,723	103,085	[25]	Gamma

AE, adverse event; BRCA, breast cancer susceptibility genes; HRD, homologous recombination deficiency.

4. Sensitivity analyses

Numerous sensitivity evaluations were performed to evaluate the stability of the model. One-way sensitivity analysis included the extreme values of each model parameter being tested to see what influence they had on the conclusions. Tornado diagrams were used to present the results of these analyses, with all parameters varying by ±20% [21]. A probabilistic sensitivity analysis was additionally conducted in which key parameters were varied randomly within the distribution range with 10,000 Monte Carlo simulations, all parameters randomly sampling from the distributions as recommended based on parameter types [21]. Both utility values and the incidence of AEs fit β-distribution (Beta) and costs for γ-distribution (Gamma). The results of this analysis were presented using scatter plots and acceptability curves.

5. Ethics statement

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors, it does not require the approval of the independent ethics committee.

RESULTS

1. Base-case analysis

Over a 20-year time horizon, the higher cost of olaparib plus bevacizumab was highly associated with better health advantages (LYs and QALYs) than bevacizumab monotherapy. Olaparib plus bevacizumab produced 2.769, 4.433, 3.850, and 3.129 QALYs (3.633, 5.812, 5.045, and 4.107 LYs) and bevacizumab monotherapy gained 2.526, 3.322, 2.869, and 2.486 QALYs (3.317, 4.367, 3.767, and 3.265 LYs) for overall patients, patients with a BRCA mutation, patients with HRD, and patients with HRD without a BRCA mutation AOC, respectively. The cost of bevacizumab monotherapy was calculated to be $328,717, $328,717, $379,412, and $329,147 whereas for the olaparib plus bevacizumab therapy, it was estimated as $447,049, $713,630, $633,406, and $502,979, respectively for the aforementioned groups. The ICERs of olaparib plus bevacizumab versus bevacizumab alone were $487,428 ($374,758), $249,579 ($191,649), $258,859 ($198,739), and $270,736 ($206,640) per QALY (LY), respectively (Table 3). Our results suggest that olaparib plus bevacizumab is not an optimal strategy as a first-line maintenance treatment for patients with AOC, regardless of molecular status.

Table 3. Results of the base-case analysis.

Treatment		Total cost $	LYs	ICER $/LY*	QALYs	ICER $/QALY†
Overall patients
	Bevacizumab	328,717	3.317	NA	2.526	NA
	Olaparib plus bevacizumab	447,049	3.633	374,758	2.769	487,428
Patients with BRCA mutations
	Bevacizumab	436,564	4.367	NA	3.322	NA
	Olaparib plus bevacizumab	713,630	5.812	191,649	4.433	249,579
Patients with HRD
	Bevacizumab	379,412	3.767	NA	2.869	NA
	Olaparib plus bevacizumab	633,406	5.045	198,739	3.850	258,859
Patients with HRD without BRCA mutations
	Bevacizumab	329,147	3.265	NA	2.486	NA
	Olaparib plus bevacizumab	502,979	4.107	206,640	3.129	270,736

BRCA, breast cancer susceptibility genes; HRD, homologous recombination deficiency; ICER, incremental cost-effectiveness ratio; LY, life-year; NA, not applicable; QALY, quality-adjusted life-year.

^*Compared to olaparib plus bevacizumab ($/LY).

^†Compared to olaparib plus bevacizumab ($/QALY).

2. Sensitivity analysis

According to the results of the one-way sensitivity analysis, the utility of the PFS (varying from 0.6232 to 0.9348, with the ICER ranging from $386,005/QALY to $661,144/QALY) and the cost of olaparib (varying from $2,926 to $4,388 each cycle, with the ICER ranging from $429,464/QALY to $545,391/QALY) had the most significant impact on the results of ICERs (Fig. 1). The cost of follow-up, the cost of second-line treatment, and the cost of AE-related treatment in the bevacizumab group had little impact on the results of the model (Fig. 1).

Fig. 1. The one-way sensitivity analyses for olaparib plus bevacizumab strategy compared to bevacizumab strategy.

BRCA, breast cancer susceptibility genes; HRD, homologous recombination deficiency; ICER, incremental cost-effectiveness ratio; OB, olaparib plus bevacizumab; PD, progressive disease; PFS, progression-free survival; QALY, quality-adjusted life-year; WTP, willingness-to-pay.

In the probabilistic sensitivity analyses, the cost-effectiveness acceptability curve revealed that the cost-effectiveness of olaparib plus bevacizumab increases with the increase of the WTP threshold. If the WTP threshold increased to approximately $420,000, $191,000, $230,000, and $255,000 per QALY, there was a 50% chance that olaparib plus bevacizumab was cost-effective compared with bevacizumab alone for overall patients, patients with a BRCA mutation, patients with HRD, and patients with HRD without a BRCA mutation AOC, respectively (Fig. 2). In summary, the proportions of olaparib plus bevacizumab being cost-effective compared with bevacizumab monotherapy for AOC, regardless of molecular status, at the WTP thresholds of $150,000 per QALY in the US were 0% (Fig. 2, Fig. S3).

Fig. 2. The cost-effectiveness acceptability curves for olaparib plus bevacizumab strategy compared to bevacizumab strategy. (A) Overall patients, (B) Patients with HRD tumors, (C) Patients with a tumor BRCA mutation, and (D) Patients with HRD tumors without a BRCA mutation.

BRCA, breast cancer susceptibility genes; HRD, homologous recombination deficiency; QALY, quality-adjusted life-year.

DISCUSSION

In the US, OC-related healthcare expenditures in 2020 were approximately %6.4 billion, and national medical service and prescription drug costs are respectively forecast to rise by 34% and 17% by 2030 [32,33]. These rising healthcare costs emphasize the importance of value-based oncology. The advent of olaparib and other PARP inhibitors such as niraparib and rucaparib has led to rapid changes in the AOC treatment landscape, drawing significant interest from many sectors. The SOLO1 trial demonstrated that maintenance olaparib monotherapy agent in newly diagnosed AOC patients was associated with clinical improvements and a median PFS of 56 months [34]. Muston et al. [35] then analyzed the associated cost-effectiveness and found that olaparib had incremental costs of $51,986 per QALY, respectively. That is, olaparib as a first-line maintenance treatment for patients with AOC was cost-effective compared with routine monitoring. Further studies revealed that the addition of antiangiogenic agents to PARP inhibitors resulted in longer PFS than PARP inhibitors alone [14,19,36]. Among them, updated final OS data for olaparib plus bevacizumab, recently reported by PAOLA-1, provide exciting evidence for first-line maintenance treatment for patients with AOC. Therefore, the cost-effectiveness of the olaparib plus bevacizumab strategy needs to be revised. The current study was a data-based cost-effectiveness analysis of olaparib plus bevacizumab versus bevacizumab monotherapy as first-line maintenance treatment for patients with AOC from the perspective of US society, taking into account molecular status.

These decision analysis model results revealed that in the overall AOC patient population, olaparib plus bevacizumab was less cost-effective than bevacizumab alone, with an ICER of $487,428/QALY, significantly above the WTP of $150,000/QALY in the US, thus indicating that this combination treatment regimen is not cost-effective relative to bevacizumab monotherapy as a first-line maintenance therapy in this patient population. The additional costs associated with olaparib plus bevacizumab mainly came from the cost of drugs and the management cost of treatment-related AEs, which means that reducing the cost of treatment strategies and the occurrence of AEs must be considered to increase survival. One-way sensitivity analyses revealed that changes in each parameter did not impact study conclusions, thus confirming model robustness. Probabilistic sensitivity analyses also indicated that olaparib plus bevacizumab had a 0% chance of being cost-effective relative to bevacizumab alone. Based on the PAOLA-1 trial results, the OS of patients in the combination treatment group was just 4.9 months longer than that of patients in the olaparib plus bevacizumab group, and this difference was not significant [15]. The combination treatment costs in this setting were higher while the clinical efficacy was not as significant, thus potentially explaining why olaparib plus bevacizumab was not cost-effective relative to bevacizumab monotherapy. This may suggest a lack of balance between the cost and efficacy of first-line maintenance combination therapy, emphasizing the need for future price adjustments in order to achieve a wider range of acceptable standards.

PARP inhibitors have recently been shown to significantly improve the quality of life and survival of many OC patients. As not all patients attain benefit from PARP inhibitor treatment, however, it is essential that those patients who are best suited to this treatment strategy be reliably identified. The growing prevalence of biomarker testing offers an opportunity to provide cancer patients with appropriate targeted therapies. An understanding of the relative cost-effectiveness of olaparib plus bevacizumab in AOC patients with particular HRD and BRCA statuses can thus guide decision-making efforts for both healthcare providers and payers. The model used in this study computed ICERs of olaparib plus bevacizumab versus bevacizumab alone were $249,579, $258,859, and $270,736 per QALY for patients with BRCA mutations, patients with HRD, and patients with HRD without BRCA mutations AOC, respectively. While not cost-effective, the combined olaparib plus bevacizumab regimen was associated with more favorable health benefits in patients with BRCA mutations and HRD-positive AOC, in line with several prior reports [37,38,39,40]. In two recent retrospective analyses of 33 and 42 patients in France and China harboring BRCA mutations or HRD-positive AOC, respectively, designed to assess prognostic outcomes associated with PARP inhibitor treatment, the median PFS and OS of patients with BRCA mutations were 20.9 vs. 37.7 months (p=0.210) and 151.2 vs. 122.5 months (p=0.520), while HRD status was an independent predictor of PFS (HR=0.67; 95% CI=0.49–0.92; p=0.010) [37,38]. Two other meta-analyses including 5,005 and 3,070 patients with OC revealed significant PARP inhibitor-related improvements in the PFS of patients with BRCA mutations OC (HR=0.29; 95% CI=0.24–0.34 and 0.34; 0.28–0.41) and HRD-positive OC (HR=0.40; 95% CI=0.32–0.48 and 0.39; 0.29–0.53) [39,40]. As these new combination treatment regimens are associated with high costs, alternative treatments should be considered as appropriate for patients with AOC based on their molecular status, with the early detection of these prognostic biomarkers being vital to achieving optimal patient outcomes.

As biomarker testing becomes more common, patients will increasingly choose to receive targeted therapies. Therefore, payers and healthcare decision-makers need to consider the cost of primary therapy and biomarker testing such as HRD or BRCA, and the cost-effectiveness between performing PARP and targeted therapy. This is because olaparib combined with bevacizumab has a stronger health benefit than bevacizumab alone. Olaparib combined with bevacizumab produced significant clinical benefits over the life cycle of bevacizumab, with benefits concentrated in LYs. Although treatment of olaparib is limited to 2 years, many patients still benefit after that. After bevacizumab monotherapy progressed, the upfront drug costs of olaparib were offset by long-term patient benefits and subsequent treatment due to the use of PARP [24].

One recent publication analyzed the cost-effectiveness of olaparib plus bevacizumab for the maintenance treatment of AOC [24]. Compared with bevacizumab monotherapy, olaparib plus bevacizumab presented an ICER of $56,863/QALY, which indicated that olaparib plus bevacizumab was cost-effective in the US [24]. Compared with our research, there were some shortcomings. First, they only analyzed the cost-effectiveness of OC in the treatment of the HRD-positive AOC subgroup. Because only 48% of the patients in this study were likely to be HRD-positive, single subgroup analysis was significantly limited. Second, they used data from the PAOLA-1 trial published in 2019 to extrapolate survival and final survival outcomes have been published. Finally, utility value was a crucial input parameter for cost-effectiveness. They do not consider the disutility value brought by serious AEs, which was the key point of the current cost-effectiveness analysis. However, our research has several important advantages worth noting. For one, these analyses were based on the synthesis of the most up-to-date data reported from the PAOLA-1 trial, including QoL, molecular status analysis, and final OS data that were published in separate reports in 2022 [14,15]. Long-term outcome data from this trial helped confirm the robustness of the models developed herein. In addition, all of the medical cost analyses conducted in this study were adjusted for the most recent US data in 2022, allowing for the minimization of any effects of varying medical costs on study outcomes. Lastly, two treatment settings and death-related costs across three different molecular status-based AOC patient subpopulations, potentially providing valuable real-world clinical guidance.

It is important to note that there are caveats to this research as well. Firstly, the model presented here relied only on mortality data from the phase III PAOLA-1 trial, since this was the only study to date comparing the safety and effectiveness of olaparib plus bevacizumab against bevacizumab monotherapy as a first-line maintenance therapy for patients with AOC by molecular status. As such, any biases inherent in that trial will be reflected in the present results. Second, only grade 3 or higher AEs were taken into consideration when performing these calculations given the less pronounced impact of grade 1–2 AEs, potentially contributing to the underestimation of the overall costs. However, one-way sensitivity analysis results suggest that these factors had no significant impact on our base-case outcome. Third, our model was based on the assumption that all patients underwent second-line therapy and BSC following initial disease progression, with no consideration for the potential of patients to continue treatment following the second progression, as no clear evidence or standard guidelines were available relating to these assumptions. One-way sensitivity analyses indicated that second-line treatment was not the main factor that influenced the conclusions of this study, thus having little impact on our results. Lastly, the PAOLA-1 trial was a multi-center study that included multiple countries and patients of various ethnicities, with the selected treatment plan being adjusted based upon the specific circumstances of a given patient, particularly during follow-up. As such, additional clinical trials will be needed to focus on specific study populations, follow-up treatments, and other factors with the potential to impact study outcomes.

In summary, the present results based on updated PAOLA-1 trial results and current drug prices suggest that combined treatment with olaparib plus bevacizumab is not likely to be a cost-effective alternative to single-agent bevacizumab treatment as a first-line maintenance therapy option for patients with AOC irrespective of their molecular status. However, this combined maintenance regimen may be associated with better health outcomes for patients with harboring BRCA mutations AOC. Early testing for biomarkers or molecular status can aid in identifying individuals likely to attain benefits from a given therapeutic approach, providing an opportunity for the individualized optimization of cancer treatment. The results of the present study will aid in the selection of the treatment options that are most cost-effective in these specific patient subpopulations.

SUPPLEMENTARY MATERIALS

Table S1

Drug dose and cost

Table S2

CHEERS checklist

Table S3

Summary of statistical goodness-of-fit of Kaplan-Meier curve

Fig. S1

Model structure.

Fig. S2

Kaplan-Meier curve fitting and extrapolation.

Fig. S3

Probability sensitivity analysis scatter plot.