Impact of the use of cost-effectiveness analysis (CEA) on patient access to innovative medicines for cancer and rare diseases: A comparison of CEA versus non-CEA markets

doi:10.21203/rs.3.rs-5312921/v1

Download PDF

Research Article

Impact of the use of cost-effectiveness analysis (CEA) on patient access to innovative medicines for cancer and rare diseases: A comparison of CEA versus non-CEA markets

https://doi.org/10.21203/rs.3.rs-5312921/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Although previous studies have examined approaches to health technology assessment (HTA) and different resulting levels of patient access across individual markets, data focusing on the use of cost-effectiveness analysis (CEA) versus other primary approaches to value assessment (e.g., based on relative clinical effectiveness) were previously lacking in the published literature. This study aimed to identify how the use of CEA impacted national-level patient access decisions for innovative medicines.

Methods

National-level patient access decisions in 10 markets with established systems (CEA: Australia, Canada, England, Scotland, Sweden; non-CEA: France, Germany, Italy, Spain, United States [US]) were assessed for all first indications of oncology and orphan drugs with initial European Medicines Agency regulatory approval between 2016 and 2019, and all subsequent indications approved to December 2021. Corresponding US, Australian, and Canadian approvals for the respective drugs were also included. To ensure consistency across markets, a patient access decision was defined as the first applicable national-level access outcome decision recommending some positive level of patient access (either full access or restricted access) or the final negative recommendation (no access).

Results

The sample included 129 unique drug/indication pairs (70 individual drugs). Overall, 862 patient access decisions were identified. The number of patient access decisions was markedly higher in non-CEA (n = 510) than CEA markets (n = 352). The total number and proportion of full patient access decisions was also higher in non-CEA (n = 335 [65.7%]) than CEA (n = 124, [35.2%]) markets, with restricted patient access decisions more common in CEA (n = 150 [42.6%]) than non-CEA (n = 113 [22.2%]) markets. The types of patient access decisions (i.e., no vs. full vs. restricted vs. no access) also varied across agencies, even within CEA and non-CEA markets.

Conclusions

National-level positive patient access decisions for innovative oncology and orphan medicines were more limited in markets that primarily use CEA in value assessments than in those that do not. Patient access restrictions beyond the regulatory label were also applied more often in CEA than non-CEA markets at the national level, recognizing that additional (and heterogeneous) barriers that impact ultimate patient access may occur downstream of the national decision in some markets.

Cost-effectiveness analysis

health technology assessment

patient access

oncology drugs

orphan drugs

value assessment

Approaches to determining patient access to new drugs vary around the world and, in many markets, regulatory approval of a drug is only the first step in securing its availability to patients. Drugs may then be subject to value or health technology assessment (HTA) to inform patient access decisions. The outcome of the value assessment or HTA process can inform decisions that result in access for all patients in accordance with the licensed indication, access for a subset of patients, or no patient access. Some markets also use the national HTA decision to inform subsequent pricing negotiations or downstream access decisions at different levels.

The approach to assessing the value of a new drug through HTA varies among markets. Some markets take a clinical effectiveness approach to assessing value, with price and budget impact addressed at a separate and subsequent stage. Others focus on a mixture of clinical benefit and cost, and some focus primarily on cost-effectiveness analysis (CEA).

Previous publications have reported that patient access decisions may vary due to market-specific processes and evidence requirements.^1–4 However, to our knowledge, the impact on patient access decisions of using CEA as the primary approach to national-level HTA, rather than other approaches, has not been directly assessed. The current study therefore aimed to address this question for innovative medicines in oncology and rare diseases.

Included markets

The impact of applying a CEA versus non-CEA (e.g., comparative clinical benefit) approach to national-level patient access decisions was assessed across 10 markets with established patient-access decision making processes. Analyses focused on national-level decisions as the heterogeneity of (potentially multilevel) downstream access decisions across markets limited the ability to directly examine any further effects on subsequent decisions.

The CEA markets were Australia (Pharmaceutical Benefits Advisory Committee [PBAC]), Canada (Canadian Agency for Drugs and Technologies in Health [CADTH]), England (National Institute for Health and Care Excellence [NICE], which also covers Wales and Northern Ireland), Scotland (Scottish Medicines Consortium [SMC]), and Sweden (Tandvårds-och läkemedelsförmånsverket [TLV; Dental and Pharmaceutical Benefits Agency]). The non-CEA markets were France (Haute Autorité de Santé [HAS; French National Authority for Health], Germany (Gemeinsamer Bundesausschuss [G-BA; Federal Joint Committee), Italy (Agenzia Italiana del Farmaco [AIFA; Italian Medicines Agency]), Spain (Agencia Española de Medicamentos y Productos Sanitarios [AEMPS; Spanish Agency of Medicines and Medical Devices]), and the United States (US; Food and Drug Administration [FDA]).

Sample selection

All oncology and orphan drugs with approval for their first indication from the European Medicines Agency (EMA) between January 1, 2016, and December 31, 2019, were included in the analysis. Subsequent indications of the identified medicines were also included if they received EMA approval prior to December 31, 2021. National-level patient access was determined at the indication level as each drug/indication was considered as a distinct entity in the analysis; patient access decisions were only reviewed for those drugs that had received regulatory approval for that indication in each market and so were potentially available to physicians and patients. Orphan drug classification was assigned to all indications for a drug based on the EMA designation of the initial indication. The time period used in this study allowed the collection of recent data over a duration sufficient to obtain a representative sample of HTAs (given the procedural time associated with conducting HTAs following regulatory approval). For drugs with regulatory approval prior to 2021 it was assumed that there was sufficient time for an HTA assessment to have been conducted, whereas for drugs approved in 2021 it was assumed that HTA assessments were unlikely to have been completed.⁴ Generic drugs, diagnostics, devices, biosimilars, and hybrid products were excluded.

For all drugs included in the analysis, the regulatory approvals from the US FDA, Australian Therapeutic Goods Administration (TGA), and Health Canada were identified up to December 2021; any additional indications that did not have an existing EMA approval but were approved by the other regulatory agencies during that period were added to the dataset. As of January 2021, a separate regulatory approval by the Medicines and Healthcare Products Regulatory Agency (MHRA) in the United Kingdom (UK) would have been possible; however, all identified drug indications used in the UK were approved via the EMA.

For a drug indication to be included in the analysis, at least one HTA or patient access decision before the end of 2021 had to be identified.

Identification of patient access decisions

For each drug indication, the associated published HTA outcome and patient access decision was identified for each market and linked to the relevant regulatory approval, focusing on access decisions at a national level. Of note, some drugs received regulatory approval for more than one indication within the same (combined) assessment but were reviewed by some HTA agencies separately. In our analysis, these approvals were split into the separate drug indications. The level of patient access was based on the comparison between the patient population specified in the regulatory label and that in the corresponding patient access decision; decisions were classified in terms of national-level access for patients as full access relative to the regulatory label, restricted access, no access, or still in development, as described in Table 1.

Table 1

Types of patient access decision outcomes
Patient access decision outcome	Description	Market-specific details
		CEA markets					Non-CEA markets
		Australia (PBAC)	Canada (CADTH)	England (NICE)	Scotland (SMC)	Sweden (TLV)	France (HAS)	Germany (G-BA)	Italy (AIFA)	Spain (AEMPS)	US
Full access	Unrestricted patient access in accordance with approved regulatory indication	Recommended for reimbursement	Recommended for reimbursement	Recommended for NHS coverage	Recommended for NHS coverage	Recommended for reimbursement in population aligned with EMA label	Reimbursed population	All EMA approved drugs / indications	Class A (reimbursed by SSN) or Class H (reimbursed for hospital / specialist use)	TPR recommends position in line with EMA label / BIFIMED recommendation to finance	Assumes all drugs are available to all within FDA label indications
Restricted access	Restricted patient access compared with approved regulatory indication (e.g., sub-population)	Recommended for reimbursement with restrictions	Recommended for reimbursement with restrictions	Recommended for NHS coverage with restrictions	Recommended for NHS coverage with restrictions	Recommended for reimbursement with restrictions	Reimbursed with restrictions	NA	Reimbursed with restrictions (including where a register / prescription form is in place in that indication)	TPR recommends narrower position / BIFIMED restriction on financing	NA
No access	No patient access for approved regulatory indication	Not recommended for reimbursement	Not recommended for reimbursement	Not recommended for NHS coverage	Not recommended for NHS coverage	Not recommended for reimbursement	Not recommended for reimbursement	NA	Class C (not reimbursed by SSN) Class C (nn) (non-negotiated / assessed	Not recommended in TPR / BIFIMED non-financing decision	NA
In development	Assessment is in progress	NA (information unavailable^a)	Any draft or expected decisions	Any draft or expected decisions	Any draft or expected decisions	NA (information unavailable^a)	Any draft or expected decisions	NA	NA (information unavailable^a)	NA (information unavailable^a)	NA
^{aInformation on assessments in progress was not available for this agency}.
^{AEMPS Agencia Española de Medicamentos y Productos Sanitarios (Spanish Agency of Medicines and Medical Devices), AIFA Agenzia Italiana del Farmaco (Italian Medicines Agency), BIFIMED Buscador de la Información sobre la situación de financiación de los medicamentos (Search Engine for Information on the Funding Status of Medicines), CADTH Canadian Agency for Drugs and Technologies in Health, CEA cost−effectiveness analysis, EMA European Medicines Agency, FDA Food and Drug Administration, G−BA Gemeinsamer Bundesausschuss (Federal Joint Committee), HAS Haute Autorité de Santé (French National Authority for Health), NA not applicable, NICE National Institute for Health and Care Excellence, NHS National Health Service, PBAC Pharmaceutical Benefits Advisory Committee, SMC Scottish Medicines Consortium, SSN Servizio Sanitario Nazionale (National Health Service), TLV Tandvårds−och läkemedelsförmånsverket (Dental and Pharmaceutical Benefits Agency), TPR therapeutic positioning report}

[Please insert Table 1 here]

As each market and HTA system is unique, several market-specific assumptions were made to allow a reasonable comparison (Table 2).

Table 2

Market-specific assumptions used to compare decisions and outcomes
Agency	Assumptions
CEA markets
PBAC (Australia)	PBAC often conducts several assessments for a given drug / indication. The first HTA outcome will not always reflect the ultimate level of patient access decision granted to the drug/indication; the final outcome reflects multiple reassessments but the critical time and outcome are related to granting access. In addition, the new parallel assessment process (where regulatory and reimbursement assessments are conducted at the same time) means that there is the possibility of an HTA outcome before regulatory approval, but this does not reflect patient access.
CADTH (Canada)	No additional considerations.
NICE (England)	NICE may assess a drug/indication several times, most commonly before and after inclusion in the CDF. Therefore, the first and the most recent HTA/reimbursement outcomes may differ. Patient access was determined based on the first positive outcome (either full access or restricted access).
SMC (Scotland)	The SMC may assess a drug/indication several times, but only the most recent decision is available. Therefore, the first and the most recent HTA/reimbursement outcomes may differ. Patient access was determined based on the first positive outcome (either full access or restricted access).
TLV (Sweden)	No additional considerations.
Non-CEA markets
HAS (France)	The additional process after HTA for pricing negotiations was not considered in this analysis.
G-BA (Germany)	The date of the G-BA report was used to identify the first and last HTA/reimbursement outcomes; however, the patient access decision is assumed to have been made on the same date as the EMA marketing authorization. Germany was excluded from some analyses as a result. Additionally, if a molecule was first approved before 1 January 2011 (e.g., marketed as a different product), there is no G-BA assessment, but the product still has full access as per the EMA license and it is counted in the total sample.
AIFA (Italy)	Individual assessments are performed for each available formulation and/or package and the same indication can be reassessed multiple times when a new formulation is launched (e.g., from intravenous to subcutaneous) or a new package/dosage becomes available, without any change in the previous access decision. For consistency with other markets, any additional assessments for different packages/formulations in the same indication were not included in the assessment set. Of note, Italy classifies drugs into four classes. Class C(nn) comprises products that have been identified by AIFA but for which a price has not yet been negotiated; as such, these products are not reimbursed; this can include cases where an initial assessment that classifies a product in Class C(nn) later leads to a reimbursement decision.¹² The patient registry put in place for the drug (after AIFA approval) was also assessed to determine the access outcome.
AEMPS (Spain)	AEMPS reviews drugs and develops therapeutic positioning reports, which provide the basis for reimbursement decisions by the MoH. However, the MoHs also provides reimbursement decisions through BIFIMED, even prior to the completion of a formal HTA, to cover unmet need. In these cases, data were extracted from both sources to better reflect patient access decisions in the market.
FDA (US)	Assumes all drugs are available to all patients within FDA label indications.
^{AEMPS Agencia Española de Medicamentos y Productos Sanitarios (Spanish Agency of Medicines and Medical Devices), AIFA Agenzia Italiana del Farmaco (Italian Medicines Agency), BIFIMED Buscador de la Información sobre la situación de financiación de los medicamentos (Search Engine for Information on the Funding Status of Medicines), CADTH Canadian Agency for Drugs and Technologies in Health, CDF Cancer Drugs Fund, CEA cost−effectiveness analysis, EMA European Medicines Agency, FDA Food and Drug Administration, G−BA Gemeinsamer Bundesausschuss (Federal Joint Committee), HAS Haute Autorité de Santé (French National Authority for Health), HTA health technology assessment, MoH Ministry of Health, NICE National Institute for Health and Care Excellence, PBAC Pharmaceutical Benefits Advisory Committee, SMC Scottish Medicines Consortium, TLV Tandvårds−och läkemedelsförmånsverket (Dental and Pharmaceutical Benefits Agency)}

[Please insert Table 2 here]

For the US, it was assumed that full patient access was possible for all drug/indications from the point of regulatory approval given the decentralized system in which multiple payers are responsible for downstream access decisions. For Germany, full patient access was also assumed to be available for all drug/indications since, from the point of regulatory approval, the medicine can be made available to all indicated patients, and the role of HTA within the German system is to confirm or negotiate the price once the product is in use. For other markets, national patient access was determined based on the first HTA outcome recommending some degree of positive patient access (either full access or restricted access) or a final decision of ‘no access’, based on a negative outcome (regardless of the timing or number of previous assessments). In Italy, due to its unique role and widespread use, the patient registry put in place for the drug (after AIFA approval) was also assessed to determine the access decision in comparison to population eligible as per the EMA indication. It is recognized that a positive HTA recommendation does not always lead to full patient access and in some markets there may be subsequent steps, such as regional or local formulary decisions and pricing negotiations, that ultimately determine an individual patient’s ultimate access to a drug in practice.

Sample based on regulatory approvals

The sample included 129 unique drug/indications across 70 individual drugs (Figure S1). The number of regulatory and patient access decisions per market is presented in Table 3.

Table 3

Number of regulatory and patient access decisions per agency
Agency	Total number of regulatory approvals^a		Total number of patient access decisions^b
	Total	To end 2020	Total
CEA markets
PBAC (Australia)	80	73	61
CADTH (Canada)	88	80	64
NICE (England)	116	105	80
SMC (Scotland)	116	105	80
TLV (Sweden)	116	105	67
Non-CEA markets
HAS (France)	116	105	102
G-BA (Germany)	116	105	116
AIFA (Italy)	116	105	97
AEMPS (Spain)	116	105	85
FDA (US)	110	110	110^c
^{aDrug/indications, bdrug/indications with regulatory approval, cin the US, regulatory approval provides access to the market}
^{AEMPS Agencia Española de Medicamentos y Productos Sanitarios (Spanish Agency of Medicines and Medical Devices), AIFA Agenzia Italiana del Farmaco (Italian Medicines Agency), CADTH Canadian Agency for Drugs and Technologies in Health, CEA cost−effectiveness analysis, FDA Food and Drug Administration, G−BA Gemeinsamer Bundesausschuss (Federal Joint Committee), HAS Haute Autorité de Santé (French National Authority for Health), NICE National Institute for Health and Care Excellence, PBAC Pharmaceutical Benefits Advisory Committee, SMC Scottish Medicines Consortium, TLV Tandvårds−och läkemedelsförmånsverket (Dental and Pharmaceutical Benefits Agency)}

Of the 129 drug/indication regulatory approvals, 68 (53%) had been approved by all four regulatory agencies; the FDA was often the first agency to grant approval (76% of first approvals), followed by the EMA (22%). A total of 28 of the identified drugs had more than one indication approved by at least one regulatory agency, resulting in two to 10 indications for those drugs.

Number of patient access decisions

A total of 862 patient access decisions were included in the analysis (Fig. 1). The number of decisions in the non-CEA markets was markedly higher than in the CEA markets (n = 510 versus n = 352, respectively, a relative difference of 44.9%; Fig. 1); Canada and Australia had the fewest regulatory approvals (Table 3).

Patient access decisions in CEA versus non-CEA markets

The non-CEA markets had a higher number and proportion of all decisions that resulted in positive access for patients (either full access or restricted access) compared with CEA markets (n = 448 [87.8%] vs. n = 274 [77.8%]; Fig. 1). The number and proportion of full patient access decisions were higher in the non-CEA versus CEA markets (n = 335 [65.7%] vs. n = 124 [35.2%]), whereas restricted patient access decisions were more common in CEA markets (n = 150 [42.6%] vs. n = 113 [22.2%]; Fig. 1).

Patient access decisions across markets

Among the agencies in CEA markets, PBAC (Australia), CADTH (Canada), and NICE (England) granted more restricted patient access decisions than full access decisions, and a restricted access decision was more common than a ‘no access’ decision (Fig. 2). The opposite was true of the TLV (Sweden), where ‘no access’ (n = 31) was more common than restricted access (n = 10). The total number of overall decisions that resulted in positive patient access (either full access or restricted access) also differed across HTA agencies, with the TLV (Sweden) having the fewest (n = 36), followed by PBAC (Australia; n = 40), for which most (38/40) were restricted access (Fig. 2). Although the total number of positive patient access decisions (either full access or restricted access) from the SMC (Scotland) was slightly lower than for NICE (69 vs. 75), SMC had a higher proportion of full access decisions (47/69 [68.1%] vs. 31/75 [41.3%]). This was also the highest across the CEA markets.

Germany (n = 116), closely followed by the US (n = 110), had the most full patient access decisions, with no restricted access, based on the assumption that full patient access was available from the point of regulatory approval for the US and Germany.

Although the same EMA marketing authorization applies across the European Union (EU), this analysis identified a wide variation in the number of completed HTAs and patient access decisions. HAS (France) completed more HTAs than other European markets (102 for HAS, followed by 80 for each of NICE [England] and the SMC [Scotland]; Table 3). France also had the most full patient access decisions (n = 61) in Europe after Germany (Fig. 2), whereas Sweden (TLV) had the fewest full patient access decisions (n = 67) and the most ‘no access’ decisions (n = 31).

Across the European markets, there were fewer drug/indications with confirmed patient access decisions in the CEA than non-CEA markets (range: 67–80 in CEA markets vs. 85–116 in non-CEA markets; Table 3). The CEA markets also had more cases where regulatory approval had been granted prior to 2021 but a patient access decision had not yet been determined (range: 26–39 in CEA markets vs. 8–20 in non-CEA markets; Fig. 2).

Types of patient access restrictions

Of the 862 patient access decisions for individual drug indications, 263 (30.5%) were restricted access decisions with a total of 385 discrete restriction criteria applied (Fig. 3a), indicating that multiple criteria were applied to a single drug indication in many cases. The average number of discrete restriction criteria per decision ranged up to 2.1 for Canada (CADTH) in CEA markets and up to 1.8 for Spain (AEMPS) in non-CEA markets (Fig. 3b).

Disease characteristics (e.g., symptoms, features, disease stage or duration) were the most common type of restriction criteria in both CEA and non-CEA markets, accounting for a higher proportion of overall restriction criteria in the non-CEA markets (88/140 restrictions [62.9%]) than in the CEA markets (82/245 restrictions [33.5%]) (Fig. 3a). There was greater variability in the type of disease-related restrictions in CEA than non-CEA markets (Fig. 3b). CADTH (Canada) was the only agency that applied restriction criteria based on disease progression (16/77 restrictions), although restrictions based on disease characteristics (29/77 restrictions) were still the most common (Fig. 3b). Patient population restrictions were also more common in CEA than non-CEA markets. For example, inclusion in the Cancer Drugs Fund (CDF) was the most common type of restriction applied by NICE (England; 20/59 restrictions), although it still accounted for fewer than half of restrictions.

This study found that markets where CEA is not the primary focus of value assessments had more national-level patient access decisions with fewer restrictions on patient access beyond the regulatory label than markets that rely on CEA. Previous studies have examined different HTA processes and levels of patient access across individual markets,^1–4 and the current study agrees with previous results showing these differences and further elucidates the impact of the use of a CEA versus non-CEA approach to value assessment on patient access decisions for medicines. A previous analysis by Wang and colleagues¹ reported the first recommendations for drugs based on the initial assessments; we added to this, by capturing both initial and subsequent indications and including HTA agency source data, with data structured to allow for differences in the number of assessments conducted in each market.

Number and level of national patient access decisions

As well as the greater number of patient access decisions in the non-CEA versus CEA markets, the analyses indicated a difference in the number of decisions made by the different agencies. Potential factors influencing the smaller number of patient access decisions in CEA markets were the lower numbers of regulatory approvals by Health Canada (n = 88) and the Australian TGA (n = 80; CEA markets) in the study timeframe, which lagged considerably behind EMA (n = 116) and FDA (n = 110) approvals, and the reimbursement processes in the non-CEA markets of Germany, which allows national-level access immediately after regulatory approval, and the US. However, the same pattern of non-CEA markets having a greater number of patient access decisions than CEA-markets was observed when the focus was only on the European markets (which all shared the same regulatory approval in our study sample).

The analyses also identified variability between markets in the types of patient access decisions (full vs. restricted access), as even within CEA and non-CEA markets there was considerable variability in individual drug indication decisions. NICE (England) and SMC (Scotland) had the most overall positive patient access decisions (either full access or restricted access) in the CEA markets, consistent with previous findings by Wang and colleagues.¹ Despite the current study’s extended timelines for decision-making to the end of 2021, a portion of the ‘no access’ decisions by PBAC (Australia) may reflect an ongoing process of reassessment, as there are often multiple rounds of re-assessments and the agency’s initial decisions are often negative.

Even though the European markets all shared the same EMA license, there were considerable differences in patient access decisions across markets, further highlighting the potential impact of differences in market-specific processes. France had the highest proportion of full access decisions after HTA, while England had the lowest. Since January 2021, the UK grants a separate regulatory license through the MHRA, but in our sample only five drug/indications were approved by the EU after this date; a separate MHRA approval date was not identified, and the EMA approval was taken as the source.

Some of the variability in patient access decisions between markets is likely to be process-driven. For example, ‘restricted access’ decisions are not used in Germany, and decisions or recommendations about patient access beyond what is contained within the regulatory label are not made at a national level in the US. Although we considered national-level patient access to be automatic after regulatory approval in Germany and the US, ultimate patient access in these markets is affected by subsequent procedural steps and decisions. In the US, where there is a segmented healthcare system of numerous payers determining drug coverage, there are negotiations and process requirements after FDA approval that determine whether a member’s patients can have formulary access to the approved drugs of their health plan. Additionally, health plans often apply significant utilization management and patient cost-exposure requirements, which can affect individual patients’ ultimate access to and uptake of covered therapies⁵. In Germany, patients can have access to a drug immediately after EMA approval and drugs are freely priced for an initial period (12 months during our study period, changing to 6 months from 2023). However, subsequent pricing negotiations can sometimes lead to withdrawal of a drug from the market if no agreement is reached;⁶ in this sample, we identified only three such cases.

In other markets, patient access is ultimately also affected by process steps subsequent to national HTA. For example, pricing and access negotiations by Canadian provinces and Australian states, and formulary negotiations with local health authorities in England, also present additional downstream hurdles after the national HTA decision. To compare patient access decisions across markets in this study, we looked at the national decision rather than subsequent local steps, although we acknowledge that local decisions may also drive differences in in-market patient access.

Types of patient access restriction criteria

This study showed there were fewer patient access decisions in CEA than non-CEA markets and more than half of decisions in CEA markets resulted in no access or restricted access. The CEA markets also applied more restrictions (n = 245) than the non-CEA markets (n = 140) and a wider range of restriction criteria. The most frequent restriction types in both CEA and non-CEA markets were those based on disease characteristics and line of therapy; restrictions based on disease characteristics were more common in non-CEA markets and restrictions based on line of therapy were more common in CEA markets. Other restriction criteria, such as response to therapy or managed access agreements, were more common in the CEA than non-CEA markets. Although the types of restriction criteria varied among markets, differences in their descriptions limited our ability to fully categorize and compare restrictions and assess their impact.

Generalizability of study findings

This study focused on oncology and orphan drugs, as these are therapeutic and innovative technology areas of increasing policy focus, and account for an increasing share of HTAs. These types of drugs are increasingly being approved by regulators with earlier evidence packages and/or smaller patient numbers due to high unmet medical need, clinical equipoise, and promising clinical data.^7,8 HTA agencies approach the limited availability of information after a product’s initial regulatory approval differently, which may influence the patient access decision. For example, non-CEA markets focus more on the available clinical effectiveness data in access decision-making, whereas CEA markets, which rely on cost-effectiveness modeling over a life-time horizon, typically require data extrapolation (e.g., for overall survival) and assumptions beyond the clinical data that may generate further uncertainty for decision-making and cause delays and/or restricted patient access decisions. For instance, the time horizon over which the costs and benefits of treatment are evaluated is one of the most commonly reported methodological issues that payers face, together with costing assumptions and model structure.⁹ The non-CEA markets may therefore be more favorably positioned to take early patient access decisions for innovative drugs.

Our study focused on current national-level approaches to patient access decisions in markets with established systems, and the extent to which the findings represent other markets is unknown. New and future programs and policy changes may also have a differential impact on patient access to medicines across markets. It is currently unclear what impact the expected EU Joint Clinical Assessment process from 2025 may have on patient access, as many in-market patient access decisions are made incorporating nuances that exist at a local or regional level.¹⁰ Future comparative analyses may also be affected by separate MHRA regulatory approvals for the UK - in our sample there were no additional MHRA decisions for any drug/indication, but divergence from the EMA indication may be possible in future.

Finally, some regulatory and HTA decisions in our study may have been affected by the coronavirus disease-19 (COVID-19) pandemic due to shifting priorities and resources, although this should have affected most markets at the same time.

Limitations

Although this study provides a recent, in-depth analysis, some limitations apply. First, our approach of using initial EMA approvals as a means of identifying drugs/indication may have limited the sample size compared with anchoring on the FDA approval; however, despite the fact that regulatory approvals are often granted earlier in the US, using the FDA as the basis for selection would have identified a significant number of drug/indications without regulatory approval in other markets (and hence no HTA), which would not advance data for analyses.

Second, given variability across markets, several market-specific assumptions had to be made to compare decisions across markets and may not be a true reflection of ultimate patient access. For example, in Italy, drugs with AIFA approval are often required to have a patient registry to capture data on their use in clinical practice. Registries define specific criteria for patient access; for consistency with the other markets in our analysis, any criteria included in the registry (beyond the stated regulatory indication) were considered to be restrictions, although many criteria are simply aligning the registry with the clinical trial population (Fig. 2) and clinical practice. The extent of the impact of these registries on patient access decisions could be further explored. Also, in the UK, inclusion in the CDF was considered to be a limitation on patient access; CDF inclusion is a condition that can be applied to approval by NICE, creating access where there may not otherwise have been but where full national-level access has not been granted; the impact of this restriction on the number of patients who can access the drug/indication was not quantified in this study.

Third, our analyses focused on the overall use of a CEA versus non-CEA approach to determining national-level patient access decisions for medicines and did not further examine the reasons behind individual HTA agency decisions on any particular medicine. We are unable to confirm any causal effect of a CEA versus non-CEA approach on individual decisions as cost-effectiveness thresholds, other acceptance criteria, and HTA systems and pathways also vary between markets. However, our findings show an overall trend for CEA markets issuing fewer decisions than non-CEA markets and further work is now needed to identify the reasons behind this and other factors affecting patient access at the national and regional level.

Finally, the type of restriction affects the number of patients who can have access to a drug; our analysis did not investigate or account for these differences.

Suggestions for future analyses

This study identified that CEA markets issued fewer decisions than non-CEA markets, even in the EU markets where the drug regulatory approval numbers are the same. A future study could investigate these differences that may be related to process efficiency, complexity of decision-making, or the wider use of more non-disease characteristic restrictions and could examine if regulatory decisions made by the UK MHRA have any impact on differences between markets.

Further research could also investigate the extent to which patient access restrictions in CEA and non-CEA markets affect the number of patients who can access a drug within a market. Our analyses focused on the effect of CEA on national-level patient access decisions as, without this initial approval, no access would be granted regardless of downstream decisions. Other barriers to access, such as e.g., regional decisions and out-of-pocket costs for US patients, may also have important impacts on real-world access in different markets.

National-level positive patient access decisions for innovative oncology and orphan drugs with first approval by the EMA between 2016 and 2019 were more limited in CEA than in non-CEA markets. Patient access restrictions beyond the regulatory label were also applied more often in CEA than non-CEA markets, highlighting that the chosen approach to the assessment of value has a significant impact on patient access decisions, thus creating the potential for variability in patient access across markets. This demonstrates the importance of ensuring that any approach to value assessment is tailored to local values, supports timely decision-making, and is fit for purpose.

AEMPS Agencia Española de Medicamentos y Productos Sanitarios (Spanish Agency of
Medicines and Medical Devices)

AIFA Agenzia Italiana del Farmaco (Italian Medicines Agency)

BIFIMED Buscador de la Información sobre la situación de financiación de los
medicamentos (Search Engine for Information on the Funding Status of Medicines)

CADTH Canadian Agency for Drugs and Technologies in Health

CDF Cancer Drugs Fund

CEA Cost-effectiveness analysis

COVID-19 Coronavirus disease-19

EMA European Medicines Agency

EU European Union

FDA Food and Drug Administration

G-BA Gemeinsamer Bundesausschuss (Federal Joint Committee)

HAS Haute Autorité de Santé (French National Authority for Health)

HTA Health Technology Assessment

MoH Ministry of Health

MHRA Medicines and Healthcare Products Regulatory Agency

NA Not applicable

NICE National Institute for Health and Care Excellence

PBAC Pharmaceutical Benefits Advisory Committee

SMC Scottish Medicines Consortium

SSN Servizio Sanitario Nazionale (National Health Service)

TGA Therapeutic Goods Administration

TPR Therapeutic positioning report

TLV Tandvårds-och läkemedelsförmånsverket (Dental and Pharmaceutical Benefits
Agency)

UK United Kingdom

US United States

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and materials

All data for this analysis are based on publicly available data from the individual agency web sites.

Competing interests

GK-M, CA and JMcK are, or were, employees of Avalere Health at the time of these analyses.

AA, SH and DR are employees of Janssen Pharmaceuticals and AG is an employee of Johnson & Johnson. All Janssen Pharmaceuticals and Johnson & Johnson employees may have shares in Johnson & Johnson and other pharmaceutical companies.

Funding

This study was funded by Janssen Global Services, LLC.

Authors' contributions

GK-M, JMcK and SH contributed to the study concept and design; acquisition, analysis and interpretation of data; and drafting and critical revision of the manuscript for important intellectual content. JMcK also supervised the study and manuscript development and SH was involved in obtaining funding.

CA contributed to the analysis and interpretation of data and drafting and critical revision of the manuscript for important intellectual content.

AA contributed to the study concept and design; acquisition, analysis and interpretation of data; and drafting and critical revision of the manuscript for important intellectual content.

AG contributed to the study concept and design; analysis and interpretation of data; and drafting and critical revision of the manuscript for important intellectual content.

DR contributed to the study concept and design; analysis and interpretation of data; drafting and critical revision of the manuscript for important intellectual content; and obtaining funding.

All authors reviewed and approved the final manuscript.

Acknowledgements

This work was conducted by Avalere Health. Medical writing support, under the direction of the authors, was provided by Claire Lavin, PhD, on behalf of Avalere Health, funded by Janssen Global Services, LLC

Wang T, Sola B, McAuslane N. Centre for Innovation in Regulatory Science R&D Briefing 83: Review of HTA outcomes and timelines in Australia, Canada and Europe 2016–2020. Centre for Innovation in Regulatory Science. 2022. London, UK.
European Federation of Pharmaceutical Industries and Associations (EFPIA). The root cause of unavailability and delay to innovative medicines: Reducing the time before patients have access to innovative medicines. June 2020.https://www.efpia.eu/media/554527/root-causes-unvailability-delay-cra-final-300620.pdf.Accessed May 27, 2024.
EUnetHTA. An analysis of HTA and reimbursement procedures in EUnetHTA partner countries: final report.https://www.eunethta.eu/wp-content/uploads/2018/02/WP7-Activity-1-Report.pdf. Accessed May 27, 2024.
IQVIA. EFPIA Patients W.A.I.T. Indicator 2021 Survey. Updated July 2022. https://www.efpia.eu/media/676539/efpia-patient-wait-indicator_update-july-2022_final.pdf. Accessed May 27, 2024.
Fusco N, Sils B, Graff JS, Kistler K, Ruiz K. Cost-sharing and adherence, clinical outcomes, health care utilization, and costs: A systematic literature review. J Manag Care Spec Pharm. 2023;29(1):4-16.
Staab TR, Walter M, Nesurini SM, Dintsios C-M, von der Schulenburg, J-MG, Amelung VE, et al. “Market withdrawals” of medicines in Germany after AMNOG: a comparison of HTA ratings and clinical guideline recommendations. Health Econ Rev. 2018;8(1):23.
European Medicines Agency. Conditional marketing authorisation. https://www.ema.europa.eu/en/human-regulatory/marketing-authorisation/conditional-marketing-authorisation. Accessed May 27, 2024.
Food and Drug Administration. Accelerated approval program. https://www.fda.gov/drugs/nda-and-bla-approvals/accelerated-approval-program. Accessed May 27, 2024.
Masucci L, Beca J, Sabharwal M, Hoch JS. Methodological issues in economic evaluations submitted to the Pan-Canadian Oncology Drug Review (pCODR). PharmacoEconomics Open 2017: 1: 255-263.
European Medicines Agency (2022) – EUnetHTA 21 Bilateral. Meeting minutes June 2022. https://www.eunethta.eu/wp-content/uploads/2022/10/EUnetHTA21-EMA-Bilateral-Meeeting-june-2022_en.pdf. Accessed May 27, 2024.
Fattore C, Canali B, Candelora L, Fiorentino F, La Malfa P, Mazza G, et al. Analysis of intra-regional variability of drugs’ time to access in Italy. Poster presented as ISPOR Europe 2022. https://www.ispor.org/docs/default-source/euro2022/isporeu22urbinati-pdf.pdf?sfvrsn=6477b48c_0. Accessed May 27, 2024.
Negrini C, Camera A, Dobreva M, Ciampa B, Molchanova E, Pilli S, et al. Faster Drug Market Access In Italy: is Class C-Nn The Answer? Value Health. 2015. https://doi.org/10.1016/j.jval.2015.09.1584

Competing interest reported. GK-M, CA and JMcK are, or were, employees of Avalere Health at the time of these analyses. AA, SH and DR are employees of Janssen Pharmaceuticals and AG is an employee of Johnson & Johnson. All Janssen Pharmaceuticals and Johnson & Johnson employees may have shares in Johnson & Johnson and other pharmaceutical companies.

SupplementalFigure122Oct24d1.pptx

Download PDF

Version 1

posted

You are reading this latest preprint version

Impact of the use of cost-effectiveness analysis (CEA) on patient access to innovative medicines for cancer and rare diseases: A comparison of CEA versus non-CEA markets

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Methods

Included markets

Sample selection

Identification of patient access decisions

Results

Sample based on regulatory approvals

Number of patient access decisions

Patient access decisions in CEA versus non-CEA markets

Patient access decisions across markets

Types of patient access restrictions

Discussion

Number and level of national patient access decisions

Types of patient access restriction criteria

Generalizability of study findings

Limitations

Suggestions for future analyses

Conclusions

Abbreviations

Declarations

Ethics approval and consent to participate

Consent for publication

Availability of data and materials

Competing interests

Funding

Authors' contributions

Acknowledgements

References

Additional Declarations

Supplementary Files

Status:

Version 1