Integrated budget impact model to estimate the impact of introducing selpercatinib as a tumor-agnostic treatment option for patients with RET-altered solid tumors in the US

Abstract

Objective

To estimate the potential budget impact on US third party payers (commercial or Medicare) associated with addition of selpercatinib as a tumor-agnostic treatment for patients with Rearranged during Transfection (RET)-altered solid tumors.

Methods

An integrated budget impact model (iBIM) with 3-year (Y) time horizon was developed for 19 RET-altered tumors. It is referred to as an integrated model because it is a single model that integrated results across multiple tumor types (as opposed to tumor-specific models developed traditionally). The model estimated eligible patient populations and included tumor-specific comparator treatments for each tumor type. Estimated annual total costs (2022USD, $) included costs of drug, administration, supportive care, and toxicity. For a one-million-member plan, the number of patients with RET-altered tumors eligible for treatment, incremental total costs, and incremental per-member per-month (PMPM) costs associated with introduction of selpercatinib treatment were estimated. Uncertainty associated with model parameters was assessed using various sensitivity analyses.

Results

Commercial perspective estimated 11.68 patients/million with RET-altered tumors as treatment-eligible annually, of which 7.59 (Y1), 8.17 (Y2), and 8.76 (Y3) patients would be selpercatinib-treated (based on forecasted market share). The associated incremental total and PMPM costs (commercial) were estimated to be: $873,099 and $0.073 (Y1), $2,160,525 and $0.180 (Y2), and $2,561,281 and $0.213 (Y3), respectively. The Medicare perspective estimated 55.82 patients/million with RET-altered tumors as treatment-eligible annually, of which 36.29 (Y1), 39.08 (Y2), and 41.87 (Y3) patients would be selpercatinib-treated. The associated incremental total and PMPM costs (Medicare) were estimated to be: $4,447,832 and $0.371 (Y1), $11,076,422 and $0.923 (Y2), and $12,637,458 and $1.053 (Y3), respectively. One-way sensitivity analyses across both perspectives identified drug costs, selpercatinib market share, incidence of RET, and treatment duration as significant drivers of incremental costs.

Conclusions

Three-year incremental PMPM cost estimates suggest a modest impact on payer-budgets associated with introduction of tumor-agnostic selpercatinib treatment.

Keywords:

• Budget impact
• commercial
• Medicare
• non-small cell lung cancer
• RET-altered tumors
• selpercatinib
• thyroid cancer
• tumor-agnostic

JEL CLASSIFICATION CODES:

• I10
• I1
• I
• I11

Introduction

Budget impact analyses (BIA) are used to estimate the affordability of new healthcare interventions versus existing standard of care (SoC). BIAs are primarily conducted from payers’ perspective and include cost measures relevant to them, and are used by the payers to inform their formulary and reimbursement related decision-making^1–3. Over the years, cancer treatment has shifted from a histology-based approach to a tumor-agnostic approach, which involves identifying molecular biomarkers and targeting oncogenic drivers, irrespective of the tumor type. This approach is evaluated in basket trials^4–6, making it important to understand the monetary impact of treatments on multiple tumor types simultaneously. As more trials evaluating tumor-agnostic treatments are anticipated⁷, traditional approaches to developing important payer resources (like BIA)^4–6 may have to be revisited to ensure that the methodology is aligned and able to accommodate the tumor-agnostic focused drug development. This will help estimate economic outcomes from a tumor-agnostic perspective for stakeholder considerations⁸.

Oncogenic alterations in Rearranged during Transfection (RET) genes are reported in multiple solid tumor types^9–12 and are associated with tumor proliferation, metastasis, and poor prognosis^13,14. The overall prevalence of RET-altered tumors is low (<2–3%)^15–18. Selpercatinib is a highly selective and potent RET kinase inhibitor (RET inhibitor, RETi), which was approved in September 2022 by the US FDA for RET fusion-positive solid tumors (i.e. tumor-agnostic) after being approved in May 2020 for RET-altered non-small cell lung cancer (NSCLC), thyroid cancer (TC), and medullary TC (MTC)^19–22. Prior to approval of selpercatinib, RET-altered tumors were commonly treated with anti-angiogenic therapy or immune checkpoint inhibitor alone or in combination with chemotherapy (NSCLC), multi-kinase inhibitors (MKIs; TC and MTC)^23–25, with limited data available for other RET-altered tumor types.

In this study, an integrated budget impact model (iBIM) was developed with an aim to estimate the number of patients diagnosed with incident RET-altered solid tumors eligible for treatment with RETi in the US, and to determine the potential budget impact on US third party payers’ formulary budgets associated with the addition of selpercatinib as a tumor-agnostic treatment for patients with these tumor types. The model was referred to as integrated as it is a single model that integrated results across multiple tumor types or indications (as opposed to indication-specific or tumor-specific models developed traditionally).

Methods

Model structure and perspective

This iBIM was developed using Microsoft Excel®, and in accordance with the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) good practice guidelines for conducting BIA¹.

The model included 19 RET-altered tumor types and was developed from the US third party payer perspectives including commercial and Medicare. The model was built using a 3-year (Y) time horizon that allowed determining the impact of varying treatment mix inclusive and exclusive of selpercatinib on payers’ budget. The same structure and approach were applied to both perspectives, with the primary difference being cost and input data used in estimation of the patient population. The model estimated differences in costs between standard of care (SoC) treatment mix in the absence of selpercatinib (without selpercatinib) and the treatment mix with selpercatinib as an available treatment option (with selpercatinib) for all tumor types. The model structure is presented in Figure 1. Key model assumptions and/or limitations are presented in Supplementary Table S1.

Figure 1. iBIM framework.

* NSCLC, pancreatic cancer, CRC, MTC, TC, salivary gland cancer, cholangiocarcinoma, gastric cancer, small intestine cancer, ovarian cancer, breast cancer, prostate cancer, xanthogranuloma, melanoma skin cancer, non-melanoma skin cancer, sarcoma, pulmonary carcinosarcoma, carcinoid, carcinoma of unknown primary.

Abbreviations: CRC, colorectal cancer; NSCLC, non-small cell lung cancer; MTC, medullary thyroid cancer; TC, thyroid cancer.

Base-case analysis

Patient population

Both commercial and Medicare perspectives were assumed to have one-million plan members. Assumptions and calculations for number of eligible patients for the treatment for each RET-altered tumor type was performed using the base-case population from each plan, disease stage, histology, testing rate, incidence of RET-alteration, proportion of patients eligible for treatment, proportion of patients receiving first-line (1 L) therapy, and proportion of patients receiving second-line or beyond (2 L+) therapy similar to the methodology, as published previously²⁶. As an example, Supplementary Table S2 presents detailed attrition demonstrating the estimation of patient population with advanced or metastatic RET fusion-positive NSCLC and Supplementary Table S3 presents the detailed inputs used similarly for each tumor type included in the model.

The 19 RET-altered tumor types included in the iBIM were NSCLC, MTC, TC, salivary gland cancer, pancreatic cancer, colorectal cancer (CRC), cholangiocarcinoma (CCA), gastric cancer, small intestine cancer, ovarian cancer, breast cancer, prostate cancer, xanthogranuloma, melanoma skin cancer, non-melanoma skin cancer, sarcoma, pulmonary carcinosarcoma, carcinoid tumors, and carcinoma of unknown primary. Consistent with the US FDA-approved label, NSCLC, MTC, and TC tumors accounted for both first-line (1 L) and previously (2 L+) treated settings while the remaining tumor types only accounted for 2 L + setting in the model²⁷. The tumor types included in this iBIM were based on RET-altered tumor types included in the phase 1/2 LIBRETTO-001²⁰ and ARROW²⁸ studies as well as the published literature indicating the presence of oncogenic RET-alterations in multiple solid tumor types^15–18.

Treatments

For each tumor type, the most frequently used treatment regimen in the absence of selective RETi was identified using the IPSOS Oncology Data Monitor²⁹ and included in the model as the SoC for the tumor type (defined as the SoC comparator representative of treatment options available for the disease). Additionally, pralsetinib was included for tumor types where it was approved for use in the US at the time of model development^30,31. These treatments comprised SoC options in the “without selpercatinib” mix, while selpercatinib was included as an additional treatment option in the “with selpercatinib” mix. Please refer to Supplementary Tables S4 and S5 for details on tumor-specific treatments included in the model.

Market shares (based on internal real-world data and/or forecasting data) were specified for each SoC treatment in the “without selpercatinib” mix at 100% or 75% (if pralsetinib was an added SoC in the mix), which were reduced by the annual forecasted market share of selpercatinib in the “with selpercatinib” mix. The assumed market share for selpercatinib was 65%, 70%, and 75% in years 1 (Y1), 2 (Y2), and 3 (Y3), respectively. Pralsetinib market share was assumed to be constant at 25% across all years in “with and without selpercatinib” mixes. These values were the same for commercial and Medicare perspectives in the model.

Cost inputs

The iBIM used several comprehensive medical resources to estimate the costs of each treatment including drug acquisition costs, administration costs, treatment-related costs (monitoring, supportive care, and its administration), and adverse event (AE) related costs. The costs were reported for individual years and were not discounted, consistent with best practice guidance¹. All costs were inflated to 2022 USD using the half-year 2022 conversion factor for the medical care component of the consumer price index (CPI) that was available at the time of model development³².

Drug costs

The drug acquisition and relevant administration costs (Supplementary Tables S4 and S5) for each treatment were calculated based on the recommended drug dosing and the average patient bodyweight for each tumor type, as well as the published median duration of treatment³³. The wholesale acquisition cost (WAC) of drugs sourced from the REDBOOK online database was used for commercial perspective while the Medicare perspective used the payment allowance limits for Medicare Part B drugs as available, and WAC costs where payment allowance limits were not available³⁴. For intravenous treatment, infusion costs were retrieved from the Centers for Medicare and Medicaid Services (CMS) physician fee schedule for Medicare perspective and were inflated using a multiplier (×1.65) for commercial perspective³⁵ (Supplementary Table S1).

Treatment-related costs

Treatment-related costs included monitoring costs (lab fee and physician fee), premedication costs/supportive care drug costs, and their administration costs. Treatment-related costs for each regimen were derived primarily from prescribing information and recommended treatment guidelines (Supplementary Tables S4 and S5).

AE rates and related costs

Grade 3 or higher AEs associated with treatment were included in the model if rates were ≥5% based on published clinical trials related to each treatment included. Costs associated with managing AEs, with an assumption that grade ≥3 AE required hospitalization, were sourced from the Healthcare Cost and Utilization Project database³⁶. It was also assumed that rates of AEs and the dosing regimen were the same for commercial and Medicare perspectives. Supplementary Tables S6–S8 present the AE-related costs for treatment by tumor types included in the iBIM.

Analyses

Annual total estimated patient populations diagnosed with incident RET-altered tumors eligible to receive treatment and those treated with selpercatinib (based on forecasted market share) were presented separately for one-million member commercial and Medicare plan perspectives. Similarly, total costs (for “without selpercatinib” and “with selpercatinib”) and the per-member per-month (PMPM) costs were estimated and reported for each of the 3 years. The PMPM cost was calculated as the total cost for a selected year divided by the plan population divided by 12 months. The incremental costs (“with selpercatinib” minus “without selpercatinib”) representing the budget impact associated with introduction of selpercatinib as an additional treatment option in Y1, Y2, and Y3 were also reported individually for 3 years. Incremental total costs were also disaggregated into costs of drug acquisition, drug administration/Tx infusion, treatment-related, and AE-related. All costs were reported for the integrated tumor-agnostic population as well as reported separately for individual tumor types.

Sensitivity and scenario analyses

One-way sensitivity analyses (OWSA)

Base-case model parameters were varied by ±10% to assess uncertainty in estimated incremental costs for both perspectives across 3 years. The ten most influential parameters were presented in tornado graphs. This OWSA was performed for the integrated model including all tumor types, however, functionality was built in the model to conduct OWSA for a single or a selected list of tumor types as well (data not shown).

Alternative comparator costing

An additional sensitivity analysis was conducted using an alternative approach for costing SoC, in which average costs associated with treating each disease (in 1 L or 2 L + settings as applicable) were obtained from published literature instead of micro-costing only one treatment selected to be the representative of the disease (in base-case approach) for each tumor type included in the model. In the absence of published literature for a few rare tumor types, an estimated cost of treating each disease from the base-case was used. Supplementary Table S9 presents the average SoC costs used in this sensitivity analyses and the assumptions made for each tumor type.

Scenario analyses

In addition to the base-case analysis, which assumed a 100% RET testing rate and included 19 RET-altered tumors, we also conducted analyses for two different scenarios. In scenario 1 (50% RET testing scenario), the rate of RET testing was assumed as 50%, to estimate the potential budget impact to payers that is reflective of contemporary real-world evidence, which demonstrates that not all patients with cancer currently receive biomarker testing³⁷. In scenario 2 (gastrointestinal tumor-specific scenario, GI-specific), the integrated model was restricted to five GI-specific tumor types (pancreatic cancer, CRC, CCA, gastric cancer, and small intestinal cancer) to provide the estimated size of the eligible patient population and associated budgetary impacts in GI-setting. The GI-specific scenario was conducted given the higher prevalence of these diseases observed in the non-lung/non-thyroid cohort of patients treated with selpercatinib in the LIBRETTO-001 trial as well as the high unmet need reported in this population, and the ARROW trial for pralsetinib^22,38,39. These scenario analyses followed the same costing approach as described in the base-case, unless specified otherwise.

Results

Base-case

Commercial perspective (1-million-member plan)

Overall, 11.68 patients per million were estimated to have an incident diagnosis of RET-altered advanced or metastatic cancer and considered eligible for treatment annually, of which 7.59 (Y1) to 8.76 (Y3) patients were estimated to be treated with selpercatinib (Table 1).

Table 1. Estimated patient population per million.

	Estimated # of patients per million
	Commercial				Medicare
		Treated with selpercatinib in the model				Treated with selpercatinib in the model
Tumor type	With RET-altered tumors	Y1	Y2	Y3	With RET-altered tumors	Y1	Y2	Y3
NSCLC (1 L and 2 L+)	5.84	3.79	4.09	4.38	34.37	22.34	24.06	25.78
MTC (1 L and 2 L+)	1.16	0.76	0.81	0.88	2.73	1.78	1.92	2.05
TC (1 L and 2 L+)	1.34	0.87	0.94	1	3.51	2.28	2.45	2.63
Pancreatic	0.27	0.18	0.19	0.2	1.5	0.97	1.05	1.12
CRC	1.58	1.02	1.1	1.18	6.57	4.27	4.6	4.93
Salivary	0.09	0.06	0.06	0.06	0.38	0.25	0.26	0.28
CCA	0.01	0	0	0	0.03	0.02	0.02	0.03
Gastric	0.05	0.03	0.04	0.04	0.22	0.15	0.16	0.17
Small intestine	0.04	0.02	0.03	0.03	0.17	0.11	0.12	0.12
Ovarian	0.32	0.21	0.23	0.24	1.32	0.86	0.93	0.99
Breast	0.06	0.04	0.04	0.04	0.2	0.13	0.14	0.15
Prostate	0.14	0.09	0.1	0.1	0.66	0.43	0.46	0.5
Xanthogranuloma	0.01	0.01	0.01	0.01	0.01	0.01	0.01	0.01
Melanoma skin	0.01	0	0	0	0.03	0.02	0.02	0.02
Non-melanoma skin	0.02	0.01	0.01	0.01	0.26	0.17	0.18	0.19
Sarcoma	0.03	0.02	0.02	0.02	0.08	0.05	0.06	0.06
Pulmonary Carcinosarcoma	0.37	0.24	0.26	0.28	2.1	1.36	1.47	1.57
Carcinoid	0.27	0.18	0.19	0.2	1.2	0.78	0.84	0.9
Carcinoma of unknown primary	0.09	0.06	0.06	0.06	0.48	0.31	0.34	0.36
Total	11.68	7.59	8.17	8.76	55.82	36.29	39.08	41.87

Patients included are per 1-million-members. 1L and 2 L + are indicated for NSCLC, MTC, and TC. For all other tumor types, the treatment was 2 L+.

Abbreviations: 1L, first-line; 2 L+, second-line and beyond; CCA, cholangiocarcinoma; CRC, colorectal cancer; NSCLC, non-small cell lung cancer; MTC, medullary thyroid cancer; RET, REarranged during Transfection; TC, thyroid cancer.

The incremental total costs of adding selpercatinib as a tumor-agnostic treatment option for patients with RET-altered tumors were estimated to be $873,099 (Y1), $2,160,525 (Y2), and $2,561,281 (Y3) (Figure 2A), with corresponding incremental PMPM costs estimated at $0.073 (Y1), $0.180 (Y2), and $0.213 (Y3) (Figure 2B). For the included tumor types, the estimated incremental total costs ranged from −$3,395 to $252,537 (Y1), $372 to 922,071 (Y2), and $398 to $1,017,597 (Y3); and the estimated corresponding incremental PMPM costs ranged from $0.000 to $0.021 (Y1), $0.000 to $0.077 (Y2), and $0.000 to $0.085 (Y3) (Supplementary Tables S10–S12), including disaggregated incremental costs.

Figure 2. Incremental costs for commercial and medicare perspectives – all tumor types, base-case.

Medicare perspective (1-million-member plan)

Overall, 55.82 patients per million were estimated to have an incident diagnosis of RET-altered advanced or metastatic cancer and considered eligible for treatment annually, of which 36.29 (Y1) to 41.87 (Y3) patients were estimated to be treated with selpercatinib (Table 1).

The incremental total costs of adding selpercatinib as a tumor-agnostic treatment option for patients with RET-altered tumors were estimated to be $4,447,832 (Y1), $11,076,422 (Y2), and $12,637,458 (Y3) (Figure 2A), with corresponding incremental PMPM costs of $0.371 (Y1), $0.923 (Y2), and $1.053 (Y3) (Figure 2B). For the included tumor types, the estimated incremental total costs ranged from –$7,926 to $1,490,016 (Y1), $1,205 to $5,200,859 (Y2), and $1,291 to $5,746,766 (Y3); and the estimated corresponding incremental PMPM costs ranged from $−0.001 to $0.124 (Y1), $0.000 to $0.433 (Y2), and $0.000 to $0.479 (Y3) (Supplementary Tables S13–S15), including disaggregated incremental costs.

Sensitivity and scenario analyses

OWSA

In the commercial perspective, the drug acquisition cost of selpercatinib was the most important driver of incremental costs across all 3 years, with duration of selpercatinib or comparator treatment, comparator acquisition cost, selpercatinib market share, and the incidence of RET being the other most influential drivers of incremental costs (Figure 3, Supplementary Figures S1 and S2). These observations were similar in the Medicare perspective with a few notable differences. For example, selpercatinib market share was not one of the prominent drivers of incremental cost in Y1. In addition, for Y2 and Y3 of Medicare perspective, the percent of patients tested for RET and the number of 1 L a/mNSCLC patients with RET fusion were additional drivers of incremental cost (Figure 4, Supplementary Figures S3 and S4).

Figure 3. OWSA, base-case – commercial perspective (Y1).

Abbreviations: 1L, first-line; 2L+, second-line and beyond; CRC, colorectal cancer; NSCLC, non-small cell lung cancer; RET, REarranged during Transfection; Y1, year 1.

Figure 4. OWSA, base-case – Medicare perspective (Y1).

Abbreviations: 1L, first-line; 2L+, second-line and beyond; CRC, colorectal cancer; NSCLC, non-small cell lung cancer; RET, REarranged during Transfection; Y1, year 1.

Alternative comparator costing

The results for the alternative comparator costing analysis were mostly similar to the results of base-case analysis. Please refer to Supplementary Figures S5A and B and Supplementary Tables S16–S21 for detailed results.

Scenario analysis

For scenario 1, decreasing the rate of testing from 100% to 50% decreased the incremental total cost and incremental PMPM cost to half of the values in the base-case (Figure 2 and Table 2).

Table 2. Scenario analyses.

# Patients (per million), Commercial		# Patients (per million), Medicare		Incremental cost ($)	Commercial			Medicare
With RET-altered tumors	Selpercatinib treated	With RET-altered tumors	Selpercatinib treated		Y1	Y2	Y3	Y1	Y2	Y3
Scenario 1: 50% RET testing
5.84	3.79 (Y1); 4.09 (Y2); 4.38 (Y3)	27.91	18.14 (Y1); 19.54 (Y2); 20.93 (Y3)	PMPM	0.036	0.09	0.107	0.185	0.462	0.527
				Total Cost	436,550	1,080,262	1,280,640	2,223,916	5,538,211	6,318,729
Scenario 2: GI tumor-specific
1.94	1.26 (Y1); 1.36 (Y2); 1.46 (Y3)	8.49	5.52 (Y1); 5.94 (Y2); 6.37 (Y3)	PMPM	0.013	0.014	0.015	0.062	0.067	0.072
				Total Cost	156,492	168,530	180,568	748,268	805,827	863,386

All values are presented in terms of 2022$. Incremental Cost: cost with selpercatinib – cost without selpercatinib.

GI-tumors included: pancreatic cancer, CRC, CCA, gastric cancer, and small intestine cancer.

Abbreviations: GI, gastrointestinal; PMPM, per-member per-month; RET, REarranged during Transfection; Y, year.

For scenario 2, the incremental total costs and incremental PMPM costs for GI-tumors accounted for approximately 18%, 8%, and 7% of the incremental PMPM for a tumor-agnostic population (as reported in base-case) in Y1, Y2, and Y3 in the commercial perspective, respectively. From the Medicare perspective, GI-tumors accounted for approximately 17%, 7%, and 7% of the incremental PMPM for a tumor-agnostic population (as reported in base-case) in Y1, Y2, and Y3, respectively (Figure 2 and Table 2).

Discussion

To the best of our knowledge, this is the first BIM that developed an approach to estimate the economic impact of a new treatment approved for tumor-agnostic use in the US. In this iBIM, we estimated the annual number of patients diagnosed with incident RET-altered tumors eligible to receive treatment and reported a modest incremental cost of introducing selpercatinib to a third party payers’ budget in the US for treatment of patients with RET-altered tumors. Most influential parameters affecting the incremental costs included cost of drug acquisition (selpercatinib or comparator SoC), share of market of selpercatinib, incidence of RET-alteration, and duration of selpercatinib treatment. These notable findings were consistent across both commercial and Medicare perspectives evaluated in this study.

The estimated total annual number of patients diagnosed with RET-altered tumors was 11.68 (commercial) and 55.82 (Medicare) per million, which ranged from 0.01 to 5.84 (commercial) and 0.03 to 34.37 (Medicare) patients per million for different tumor types. The estimated number of patients was expected to be low given the known incidence of RET-alterations across tumor types^{16–18,40–43}, which in turn might be a function of low rates of biomarker testing in the real-world. Duff et al. previously estimated a patient population with RET-altered lung (2–4.6 patients/million) and thyroid cancers (0.5–0.7 patients/million) assuming an increased rate of testing for RET (30–70%) over 3 years in the US⁴⁴. Despite using lower and contemporary rates of incidence of RET-alterations in lung and thyroid cancers in this study versus the analysis by Duff et al. the estimated patient population numbers for the same tumor types were higher in the current base-case analysis because the RET testing rate was assumed at 100% to avoid potential underestimation of impact to payers.

To our knowledge, no other study has published a comprehensive estimation of eligible patient population diagnosed with RET-altered tumors in the US and this is one of the key strengths of this current analysis. As molecular testing increases over time, the incidence of RET may change, potentially affecting the numbers presented in this analysis. Currently, however, the low testing rates for identification of biomarkers translates into lower use of appropriate treatments, which is also described in literature where providers reported sub-optimal knowledge of screening for RET-alterations that resulted in initiating a sub-optimal treatment plan⁴⁵.

The addition of selpercatinib as a tumor-agnostic treatment for patients with RET-altered tumors in this analysis was shown to be associated with an increase of 7–21 cents (commercial, percent change: 50% [Y1], 100% [Y2], and 118% [Y3]) or 37–105 cents (Medicare, percent change: 56% [Y1], 117% [Y2], and 137% [Y3]) on a PMPM basis over Y1 to Y3 in the US. This assumed the rate of testing for RET at 100%, however, these estimates were reduced to 4–11 cents (commercial; percent change: 51% [Y1], 101% [Y2], and 78% [Y3]) or 18–53 cents (Medicare; percent change: 56% [Y1], 118% [Y2], and 134% [Y3]) over Y1 to Y3 when the rate of testing for RET was reduced to 50% (scenario analysis 1). The estimates from scenario analysis 1 may better reflect the current budget impact to payers given that rates of biomarker testing in the US are approximately 50–60%^23,24. Focusing on GI-specific cancers alone (scenario analyses 2), these numbers ranged from 1.3–1.5 cents (commercial, percent change: 118% [Y1], 127% [Y2], and 136% [Y3]) and 6–7 cents (Medicare, percent change: 175% [Y1], 186% [Y2], and 200% [Y3]) on a PMPM basis over Y1 to Y3, suggesting a modest impact to address a high unmet need. Collectively, these findings indicate that including selpercatinib as an additional treatment option for patients with RET-altered tumors resulted in a modest increase in cost burden for payers. The increase in incremental PMPM costs in this iBIM is attributed to the assumed increase in the market share of selpercatinib over 3 years, and these PMPM costs were higher for the Medicare than observed for the commercial perspective due to the greater number of patients diagnosed with RET-altered tumors in those aged 65 years or more. This was consistently observed in both base-case and scenario analyses.

Non-selective inhibitors of RET (such as cabozantinib, vandetanib, and lenvatinib) have demonstrated lower efficacy in few RET-altered cancers compared with selpercatinib, with high incidence of AEs^46,47. Furthermore, the clinical benefit of MKIs in the RET-altered cancers has generally been poor, with modest benefit in RET-mutant MTC^48,49. In the LIBRETTO-431 (vs. platinum-based chemotherapy +/- pembrolizumab) and LIBRETTO-531 (vs. MKI – cabozantinib or vandetanib) phase 3 randomized controlled trials, selpercatinib demonstrated significant and clinically meaningful improvements in progression-free survival in patients with RET-altered NSCLC and MTC in the first-line setting, respectively^50,51. Additionally, in the LIBRETTO-001 trial, selpercatinib showed clinically meaningful activity in various RET-altered non-lung/non-thyroid solid tumors and was reported as well-tolerated²². The published clinical efficacy evidence of selpercatinib and the modest economic impact of selpercatinib observed in this analysis together indicate the added value of selpercatinib as a tumor-agnostic treatment option for patients with RET-altered tumors. The highly selective nature of selpercatinib and these data also suggest making biomarker testing available to a broader patient population, which will help in initiating the appropriate biomarker-driven targeted therapy in eligible patients wherever available. A collaborative effort between regulatory bodies, industry, laboratories, and academia is important to have a consensus on factors such as patient heterogeneity and interpretation of NGS results⁵². This is required to create a unified approach to testing, which includes waiting for test results (in most cases, as applicable) and subsequently initiating appropriate concordant targeted therapy for eligible patients.

Strengths and limitations

This iBIM has several strengths. The model was flexible to integrate data across 19 different tumor types for a wholistic view as well as to view results independently for each tumor type as desired. The model allowed for inclusion of various inputs for drug acquisition costs (WAC or AWP – data not shown), multiple costing approaches, and study perspective (commercial and Medicare). The model also allows users to conduct sensitivity analyses at the global/integrated level across all tumor types as well as for a specific tumor type of choice. Overall, the present iBIM provides monetary impact to payers associated with use of a single new drug across all plausible indications, providing them with more practical information versus individual disease-specific indications as reported traditionally. The scenario analysis of 50% testing rate for RET reduced the incremental total and incremental PMPM costs to 50% of the base-case values, which may be the actual incremental values in the real-world setting where all patients are not tested for biomarkers. The scenario analysis showed GI-tumors form approximately 17% of the RET-altered tumor cases, making it a large subgroup after NSCLC, MTC, and TC. It is expected that this wholistic information will be helpful to payers in formulary decision-making. Additionally, this iBIM serves as a proof of concept for future economic models evaluating the impact of treatments approved for tumor-agnostic use (or potentially for many indications) developed using innovative trial designs such as basket trials. A systematic review of oncology study protocols demonstrated that the number of master trials, including basket trials, have increased in the last 5 years and are expected to increase in the future. This is due to the potential efficiency they offer in drug development and approval process⁵³.

There are also a few limitations of this model. The assumptions of static populations, incident rates, and market share are approximations that simplify the use of this iBIM; however, these are modifiable inputs in the model and could be changed as appropriate for a stakeholder. The number of patients in each population of interest or the relative costs of treatment were not expected to change substantially over a 3-year period. The costs of drug treatment were based on WAC pricing from RED BOOK, and since they do not include any discounts, deductibles, or co-payments, they may not reflect the final cost to the budget holder. The AEs included in the model were only grade ≥3 that were assumed to incur hospitalization costs and were assumed to be independent when there may have been either additional costs due to their severity or savings due to concurrent hospitalizations. The model did not account for adherence to treatment (and associated drug wastage). Treatment duration and the rates of AEs were obtained from primary clinical trials of treatments and were not based on meta-analyses; these inputs may or may not differ from common clinical practice. Finally, while pralsetinib was included as an additional comparator in the model for patients with MTC per approved label and national treatment guidelines available during the development of the model, this indication was recently withdrawn⁵⁴, however, it does not have a substantial impact on the results presented here.

Conclusions

The integrated model estimated approximately 12 (commercial) or 56 (Medicare) patients per million per year that were diagnosed with RET-altered tumors and would be eligible for treatment with RET inhibitors. The 3-year incremental PMPM cost estimates suggest modest impact on payer-budgets associated with the introduction of selpercatinib as a tumor-agnostic treatment. Among various factors, selpercatinib acquisition cost was the most significant driver of incremental costs. Lastly, the methodology described in this iBIM serves as a proof of concept for future economic models evaluating treatments approved for tumor-agnostic use.

Transparency

Declaration of funding

Medical Decision Modeling Inc. received funding from Eli Lilly and Company to conduct the study.

Declaration of financial/other relationships

NRB, AMG, and AAA are employees and minor stockholders at Eli Lilly and Company. JM and LS are employees of Medical Decision Modeling Inc., which was contracted by Eli Lilly and Company to perform the analysis.

Author contributions

Conception of work: NRB, JM, AAA LS. Design of work: NRB, AMG, JM, LS. Acquisition of data: JM, LS. Data Analysis: JM, LS. Data interpretation: NRB, AMG, JM, AAA, LS. Manuscript writing and critical revision: NRB, AMG, JM, AAA, LS. Final approval: NRB, AMG, JM, AAA, LS.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Ethics approval

This iBIM was exempt from institutional review board approval as it did not include patients directly.

Consent to participate

Due to the nature of the study, consent to participate from patients was not required.

Consent to publication

The manuscript does not contain any individual person’s data in any form. Therefore, consent for publication was not required.

Previous presentation

The abstract and poster were presented at the International Society for Pharmacoeconomics and Outcomes Research (ISPOR); Boston, MA; 7–10 May 2023 (Poster ID EE507).

Acknowledgements

The authors would like to acknowledge Frank Cinfio, a former employee of Medical Decision Modeling Inc., for his contributions in model development. The authors would also like to acknowledge Karan Sharma from Eli Lilly Services India Private Limited for his medical writing and editorial assistance during the preparation of this manuscript.

Data sharing statement

The integrated budget impact model could be available to health care decision makers, healthcare organizations, and third party payers by contacting The Lilly Answers Center at 1-800-LillyRx (1-800-545-5979).

Disclosure statement

Naleen Raj Bhandari, Adrienne M. Gilligan, and Amine Ale-Ali are employees and minor stockholders at Eli Lilly and Company. Julie Myers and Lee Smolen are employees of Medical Decision Modeling Inc., which was contracted by Eli Lilly and Company to perform the analysis.