Early access to medicines in Spain: a retrospective hospital study with a proposal for a regulatory change

Abstract

Background

Medicines awaiting price and reimbursement (P&R) decisions represent a delicate situation in the Spanish NHS. The interval between marketing authorisation (MA) and PR averages 616 days, leaving patients without standard access to EU-approved treatments for almost 2 years, despite their eventual availability. While early access before MA is regulated, no framework exists for the period between MA and PR. Responsibility for funding falls to hospitals or autonomous regions, with no harmonised criteria, leading to inequities in access.

Objectives

To describe our access model to medicines in special situations (MSS), and to evaluate the clinical and regulatory implications of this procedure.

Methods

We conducted a 2-year retrospective review of MSS access requests. Our tertiary hospital uses a transparent pathway: clinicians submit a consultation to Clinical Pharmacology and Hospital Pharmacy; departments issue independent reports to Hospital Administration, which decides, considering individual justification, clinical need, early access, and treatment benefit.

Results

A total of 295 request were received, and 78 involved products pending PR decision. Of them, 73 (93.6%) were granted local access. Subsequently, PR status changed to regular access in 64.1% of these requests (50/78), ensuring early access for 62.8% of patients (49/78). Onco-haematological patients granted early access showed an overall disease control rate of 63%, with a median survival of 16 months.

Conclusions

Until legislation regulates access to medicines awaiting reimbursement, or decision times are reduced, implementing robust local procedures is essential to guarantee fair and equitable MSS access, particularly for patients requiring treatment before national PR resolution.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12094-026-04349-x.

Introduction

Access to the latest innovative medicines while safeguarding public budgets and sustainability poses a significant challenge for modern healthcare systems [1, 2], which have to manage to provide early access to these therapies for patients with advanced diseases, rare conditions, or other conditions without effective therapeutic alternatives [3].

Regular access within the Spanish National Health System (NHS) is determined by three decision-making levels:

1. Marketing authorisation (MA) by the European Medicines Agency (EMA) or Agencia Española de Medicamentos y Productos Sanitarios (AEMPS).

2. Nationwide pricing and reimbursement decision by the Ministry of Health (MH).

3. Recommendations and policies governing medicines acquisition and use at regional and local levels.

Under European regulations, MA only considers quality, efficacy, and safety aspects. Afterward, each EU Member State determines conditions for incorporating medicines into its public health system based on national economic, organizational, social, political and health/epidemiological factors. In Spain, the evaluation and decision on the inclusion of a medicine in the NHS are conducted by the MH, which also sets the official purchase price for the public system. There may be medicines authorised by the regulatory agencies that are fully reimbursed by the NHS, not reimbursed at all, or reimbursed under restricted conditions with respect to MA [4].

Although pricing and reimbursement (P&R) of new medicines fall under the national government in Spain, the 17 autonomous regions hold decentralized responsibility for healthcare management, resource allocation, budgetary decisions, and the procurement of medicines [5]. While regional decisions and recommendations can influence medicine use, they cannot impose formal restrictions on access to a medicine already approved for use across the entire NHS [6].

Physicians are expected to prescribe medicines authorised and reimbursed for the patient’s clinical condition, using the standard paths for prescription, in what we will name as “regular access”. However, sometimes the most appropriate treatment for an individual patient is outside the scope of the authorised-reimbursed medicines. In such cases, the prescription falls under the framework of Medicines in Special Situations (MSS). In Spain, this is partially regulated at the national level⁶ and considered an “exceptional access”—not for being infrequent, but because it deviates from the standard regulatory pathway.

Special situations can be categorized depending on the combination of both regulatory agencies’ decision and the P&R status. The most relevant scenarios are:

1. The prescription of medicinal products that do not have a MA, regulated under Royal Decree 1015/2009 [7], which establishes the procedures and criteria for early access to investigational medicines (compassionate use). Compassionate use (CU) refers in Spain to both individual requests on a named patient basis or to a cohort of patients with a specific condition, to address the needs of patients with serious, life-threatening, or debilitating conditions when no other approved treatment is available, and they are ineligible for clinical trials. Compassionate use requires a treating physician's report justifying the need for the treatment after individual benefit/risk assessment, the patient’s written informed consent, and the approvals by the hospital administration, the pharmaceutical company providing the investigational medicine, and the AEMPS [8].

2. The prescription of medicines that have MA but have not yet received a P&R decision within the NHS. Exceptional or early access within the NHS to non-reimbursed medicines is not formally regulated. In these cases, the decision to provide access will not be based on an individual benefit–risk, but rather on the justification for early funding to address unmet medical needs that cannot wait, either for an individual or for a restricted group of patients. In the absence of a decision by the Ministry of Health, regional or local authorities—through hospitals or autonomous regions—will determine whether early access is granted.

3. The prescription of a medicine already on the market to treat a condition considered as not reimbursed by the NHS, i.e., due to an NHS P&R explicitly restrictive decision, a newly approved indication that is pending NHS decision, or for the off-label use of expensive medicinal products. All of them will need local/regional authorities (payers) approval.

Although all types of non-regular access to medicines are a matter of concern and debate within the NHS, the early access to medicines pending a P&R decision is the most sensitive one for several reasons. The significant delay in P&R decisions—an average of 616 days [9]—and that also involve new indications of an already approved medicine, results in a period of nearly 2 years during which patients cannot access an already EU-authorised treatment.

This situation generates uncertainty and frustration for both physicians and patients, aware of the existence of a medicine that could potentially alter the patient's prognosis [10] yet remains inaccessible within the NHS—despite the likelihood that it will become available, often too late for the patients who initially needed it.

While individual access can be requested under MSS procedure, there is no standardized procedure available across all hospitals or regions within the Spanish NHS. This leads to increased bureaucratic burden, limited transparency, and significant variability—and consequently inequity—across autonomous regions [11], between hospitals, and even among patients treated within the same hospital, as different physicians could be either unaware of the local procedure or unwilling to embark on a burdensome process with an uncertain outcome.

To address these limitations, a formal and transparent procedure was implemented at Puerta de Hierro University Hospital (HUPHM). Physician´s request for an MSS is made through the electronic medical record system and it generates automatically a parallel consultation to both the Clinical Pharmacology and Pharmacy Departments (PD). The Clinical Pharmacology Department (CPD) performs an assessment and issues a recommendation to the Hospital Administration (HA), considering the severity of the disease, the expected effectiveness and safety, plausibility of the therapeutic effect, justification for an individual exception, the urgent medical need for treatment and the lack of other approved treatments available, as well as the ineligibility for suitable clinical trials. The procedure allows a fast-track report when needed.

Based on the two independent reports, which include the estimated cost of the treatment by the PD, the HA will draw a final decision to grant/not grant the early access, which will be duly reported to the physician.

This study aims to describe the operation of a local procedure that has been found helpful by our physicians, to evaluate patient benefits in terms of clinical effectiveness as well as the costs, and provide useful information for the upcoming implementation of a pre-reimbursed medicines access procedure at the national level.

Methods

This study includes all the requests for MSS made between January 2023 and December 2024 at HUPHM. HUPHM is a tertiary, high complexity hospital [12], that accounts for 27,365 admissions per year with an average length of stay of 6.6 days and 646,887 outpatient consultations [13]. Some patients have been eligible for more than one MSS request during the study period.

Eligible requests were identified through the CPD inter-consultation record and HA files. Each patient's Medical Record was revised, and the National Mortality Index was cross‐referenced to determine their mortality status (up to December 2024). A relational database was created for the study. We included all consecutive eligible patients during 2023–2024 to minimize selection bias. Data were extracted by three reviewers and discrepancies were resolved by consensus.

Descriptive variables included: age, sex, requesting department, date of request, treatment context (SNOMED Term and ID), regulatory status, access program, compassionate use, P&R status, hospital formulary inclusion, costs of MSS and alternative (estimated and real annual cost), emergency situation, change of regulatory/P&R status during the study period, CPD and HA decisions and their dates.

For the time-to-event analysis of requests, time zero was defined as the date of the formal request made by the clinician. The event was defined as the date of the national positive P&R decision or change in regulatory status, and requests without an event were censored at the end of follow-up (the study end date).

For onco-haematology treatments, additional collected variables were: treatment intention (curative or palliative), best achieved response (full response, partial, stability, progression), adjuvant treatment, subsequent treatment, survival, and ECOG performance status.

For the cost-effectiveness analysis, hospital perspective was adopted, considering only expected vs actual medicine expenditure. These costs were calculated based on the hospital pharmacy price, expressed in euros. When the medicine was obtained through an expanded access program hold by the Pharmaceutical Company, the cost was 0€. Due to confidentiality agreements, disaggregated pricing data are not disclosed. The interventions compared were access to MSS vs standard treatment if any. It was assumed that the benefit of the alternative therapy was minimal or of limited clinical relevance. Given the 1-year time horizon, no discount rate was applied, as the impact of discounting on costs and outcomes over such a short period is minimal. Effectiveness was assessed only in oncology patients, based on the response rate achieved defined as complete response, partial response, or disease stabilisation. The rest of clinical conditions were too heterogeneous to have a common efficacy variable suitable for this study assessment.

Estimated cost was calculated by PD to be used for local decision-making according to the usual procedure. For this study the actual cost of treatment administration was also calculated taking into account discontinuations, delays in cycle administration and dose reductions.

The study was approved by the Ethics Committee for Investigation with Medicinal Products (CEIm) of Hospital Universitario Puerta de Hierro prior to study initiation. (Code PI 258–24).

Statistical analysis

A descriptive analysis of demographic, clinical and treatment data was performed. Median and interquartile ranges were used for continuous variables and frequencies and percentages for categorical variables.

Time-to-event data and Kaplan–Meier survival curves were used (R version 4.4.0) to estimate time before regular access. Hazard ratio was estimated using Cox proportional hazards regression and survival differences were assessed with log-rank test.

A p value < 0.05 was considered statistically significant.

Results

From January 2023 to December 2024, 295 requests were identified, involving MSS covering various MA and P&R scenarios (Fig. 1).

Fig. 1.

Flow diagram of requests and their local outcomes. Total number of requests, defined as clinicians’ applications for a MSS; approvals, indicating that authorization was granted by the Hospital Administration (HA) to access the medicine; and administrations, referring to cases in which the medicine was ultimately supplied to the patient

The requests originated from different clinical departments, with onco-haematology area accounting for over 50% of the total. In terms of regulatory status, 28 medicines were investigational and requested under compassionate use, 99 were off-label uses of marketed medicinal products, and 168 were approved medicines/indications with differing P&R statuses (Table 1).

Table 1.

Requests for medicines in special situations, Requests n = 295

	n (%)
Sex
Women	161 (54,6%)
Men	134 (45,4%)
Medical services involved
Medical Oncology	113 (38.3%)
Haematology	35 (11.9%)
Dermatology	34 (11.5%)
Cardiology	22 (7.5%)
Neurology	15 (5.1%)
Others*	76 (25.8%)
Therapeutic area
Onco-haematology	157 (53.2%)
Autoimmune diseases	34 (11.5%)
Cardiovascular and metabolic	22 (7.5%)
Infectious diseases	20 (6.8%)
Inflammatory diseases	17 (5.8%)
Urgent request**
Yes	56 (19.0%)
No	85 (28.8%)
No record	154 (52.2%)
EU regulatory status
Compassionate	28 (9.5%)
Approved	168 (57.0%)
Off-label use	99 (33.5%)
P&R status in the NHS
Non-reimbursed product	160 (54.2%)
Pending PR decision	78 (26.4%)
Non-reimbursed condition	30 (10.2%)
Reimbursed	27(9.1%)
Expanded access program***
Yes	50 (16.9%)
No	245 (83.1%)
Treatment administration status
Yes	233 (79.0%)
No	62 (21.0%)

*Other medical departments involved: Allergology, Gastroenterology, Endocrinology, Infectious Diseases, Internal Medicine, Intensive Care, Nephrology, Respiratory Medicine, Ophthalmology, Oncology (Radiotherapy), Paediatrics, Rheumatology, and Urology

**Request for treatment initiation within 7 days

***Access programme with free of charge provision by the manufacturer. The majority were compassionate use medicines (28/50; 56%), followed by other situations

Timelines are shown in Supplementary Fig. 1. The median interval from clinician request to treatment initiation was 14 days (IQR 6–27.75).

Considering reimbursement status of all requests (Fig. 2), a total of 240 out of 295 requests (81.4%) were for approved but non-reimbursed medicines/indications: 33 approved medicines with a formal non-reimbursement NHS decision, 78 awaiting P&R decision, 30 approved with additional P&R restrictions that were not complied in the requested case and 99 off-label uses subject to approval due to novelty or cost. In 27 cases, the request was for approved and reimbursed medicines not yet included in the hospital formulary, with need for acquisition approval due to their high cost.

Fig. 2.

Flow diagram of status evolution during study follow-up

A total of 243 out of 295 requests (82.4%) received HA approval, and 233/295 (78.9%) were ultimately received by the patient. The approved medicine was not administered in 10 patients at the data cutoff: 5 due to clinical deterioration, 3 because approval was obtained at the end of the study period and administration had not yet occurred, and in 2 cases the reason could not be determined from the available records. No statistical age–sex differences in treatment approval or administration were observed.

Regarding the type of indication and medicine (Table 2 and SM Table 1), there is predominance of onco-haematology, followed by conditions related to lipid metabolism, infectious diseases and dermatological disorders. The regulatory or reimbursement category that triggers the request is distributed differently by indications. Approved medicines awaiting P&R are frequent in the oncological setting, while the indication for 'disorder of lipoprotein and/or lipid metabolism' was associated with non-reimbursed conditions of non-statin lipid-lowering medications that are nevertheless recommended by current treatment guidelines.

Table 2.

Top ten indications

Indications (SNOMED term)	Most frequent reason*** for request % (n)	Total (N)
Malignant neoplasm of breast	Pending P&R, 70% (21)	30
Disorder of lipoprotein AND/OR lipid metabolism	Non-reimbursed (Restriction) 94.7% (18)	19
Malignant neoplasm of lung	Pending P&R, 30.8% (4) Non-reimbursed, 30.8% (4)	13
Infectious disease	Not included in HF, 41.7% (5)	12
Pruritic rash*	Off label**50% (6)	12
Malignant neoplasm of prostate	Pending P&R 72.7% (8)	11
Malignant neoplasm of kidney	Off label 54.6% (6)	11
Amyloidosis	Off label 50% (4)	8
Malignant neoplasm of thyroid gland	Off label 37.5% (3)	8
Malignant lymphoma	Off label 62.5% (5)	8

*Ancestor that includes chronic prurigo, prurigo nodularis and subacute prurigo

**Only new or expensive medicines need approval for off-label uses, otherwise off-label use is prescribed by the medical doctor with no previous approval

***Reason at the time the request was submitted

After the initial request, subsequent national regulatory and P&R changes have been analysed by the end of the 2-year study's observation period. Of 78 requests for medicines pending P&R, 93.6% (73/78) received local access and during the study, 50 out of 78 changed their P&R status to regular access. At our hospital level this meant early access for 62.8% (49/78) of the cases, with a median of 129 days (IQR 7–273) before the national positive P&R decision. Overall, 37.6% (111/295) of cases underwent a regulatory status change during the study period, with 23.7% (70/295) ultimately transitioning to regular access, defined as national approved and reimbursed medicines. (Fig. 2 and SM Table 2).

The length of the follow-up depended on the time of inclusion into this study, with requests submitted in the earlier period of the study having greater opportunities to undergo a regulatory change (SM Table 2). As expected, the proportion of medicine requests that eventually gained regular access to the system raised up to 40.0% (32/80) when considering only those medicines followed by at least 1.5 years after the request.

The therapeutic area of onco-haematology has been analysed in greater detail (Tables 3 and 4). Up to 157 requests were submitted in this subgroup, with 125 different medicine schemes requested for 150 different Snomed CT terms. Most patients had a good performance status, and 18.5% were treated with curative intent. Patients were followed up during a median of 280 days (IQR 149–531). The majority of patients received the requested treatment. The median treatment duration was 166 days, with a standard deviation of 195 days. Median overall survival of 131 patients who received the treatment was 16.04 months (95% CI, 13.38–17.03), compared to 8.05 months (95% CI, 5.52–16.73) for those 26 who did not receive the treatment (HR 0.43, 95% CI 0.25–0.74, p = 0.002). When considering only the 45 patients who received treatment during the first semester of 2023 (median follow-up of 19.26 months), the median overall survival increases to 20.74 months (95% CI 19.79–21.86). The survival comparison between treated and non-treated patients should be interpreted as exploratory given the non-randomized design of this study.

Table 3.

Onco-haematology patient and request characteristics, Requests n = 157

	€ (IQR)	n (%)
Sex
Women		89 (56,7%)
Men		68 (43,3%)
ECOG Performance Status
0		44 (28.0%)
1		53 (33.8%)
2		10 (6.4%)
3		1 (0.01%)
4		0 (0%)
5		0 (0%)
NR		49 (31.2%)
Treatment intention
Curative		24 (15.3%)
Palliative		111 (70.7%)
NR		22 (14.0%)
Best achieved response		n = 127a
Complete		16 (12.6%)
Partial		41 (32.3%)
Stable disease		23 (18.1%)
Progression		22 (17.3%)
Death		25 (19.7%)
Overall disease control rateb		80 (63%)
Objective response ratec		57 (44.9%)
Finally received treatment
Yes		131 (83.4%)
No		26 (16.6%)
Treatment duration		n = 131
< 6 months		69 (51.7%)
≥ 6 months– < 1 year		29 (22.1%)
≥ 1 year– < 1,5 years		22 (16.8%)
≥ 1,5 years		11 (8.4%)
Cost–median cost (IQR)—€
Estimated costd	34,185.42 € (19,034.72€–50,538.99€)
Actual coste	8,977.00 € (492.34€–23,419€)
Total actual cost (per year)	2,350,534.89€
Cost-effectiveness–median cost (IQR)
Per overall disease control rate	14,250.99 € (781.49€–37,173.01€)
Per objective response rate	20,001.39 € (1,096.53€–52,158.12€)

^aAmong patients who received access to treatment, response could be assessed in 127 out of 131 individuals

^bComplete + Partial response + Stable disease

^cComplete + Partial response

^dMedian costs calculated per Pharmacy, used by hospital administration to make decisions

^eAccounting for treatment discontinuations, dose reductions, and delays due to toxicity

Table 4.

Baseline characteristics in the onco-haematology subgroup (treated vs not treated)

	Treated patients	Not treated patients
	n = 131	n = 26
Sex
Women	75 (57,3%)	14 (53,8%)
Men	56 (42,7%)	12 (46,2%)
Age
Mean	59,1	65,1
SD	15,4	13,2
ECOG performance status
0	43 (32,8%)	1 (3.8%)
1	51(38,9%)	2 (7.7%)
2	8 (6,0%)	2 (7.7%)
3	1 (0,7%)	0(0.01%)
4	0 (0%)	0 (0%)
5	0 (0%)	0 (0%)
NR	28 (21,3%)	21 (80.8%)
Treatment intention
Curative	24 (18.3%)	0 (0,0%)
Palliative	105 (80.2%)	6 (23,1%)
NR	2 (1.5%)	20 (76,9%)

Overall, 44.9% of patients achieved complete or partial responses, while 18.1% experienced disease stabilisation, with 37% showing progression or death.

Cost of treatment

Regarding the total cost of the treatments accessed via the MSS procedure, the overall median was 33,762.77€ (IQR 16,000.68 €–52,716.00 €). For accepted requests, the median cost was 31,195.98€ (IQR 15,592.68€–47,472.00€), whereas for rejected requests, the median cost rose to 64,820.00€ (IQR 34,021.00 €–116,200.00 €).

In the onco-haematology subgroup, estimated treatment costs were higher than actual costs once treatment modifications and discontinuations were considered. Using the real expenditure, and assuming negligible costs for alternatives, cost-effectiveness was expressed as cost per overall disease control or objective response (Table 3).

Discussion

The demand for early access to medicines, before regular access, has become a common challenge for national health systems. Different countries, regions and hospitals are providing diverse solutions, with very limited evidence on their outcomes. This study aimed to inform of a procedure that ensures transparency, agility, common criteria and consistency in access decisions, with accountability regarding its impact on health and associated costs.

This study analysed nearly 300 special-access requests over 2 years at a terciary hospital in Madrid. A key strength is the substantial number of requests (n = 268) for access to medicines with regulatory approval, thus with an already established positive risk–benefit ratio, but in different stages of the decision-making process for regular access within the NHS. Our cohort also has a significant number of requests for access to off-label uses (n = 99), which in Spain are considered to fall within the scope of responsibility of the prescribing physician and does not require prior authorisation by the regulatory agency (AEMPS). Nevertheless, authorisation at local or regional level is usually required for off-label uses of recently marketed or high-cost medicines. These off-label uses are outside the NHS coverage. Therefore, beyond the medical justification, local and/or regional payers authorisation for the NHS to assume the cost of treatments is required.

Overall, requests for non-financed medicinal products, either off-label, pending P&R or medicinal products with a negative P&R decision, account for 90.5% of our cohort, with only 9.5% compassionate uses of investigational medicinal products. This is underlined, because it addresses the most conflicting situation at present in our system, due to the fact that it often operates with criteria and procedures that differ substantially between hospitals and regions, and lack of transparency and accountability, what determines inequity within the NHS.

The Spanish NHS provides a wide access to marketed medicines [6, 14] under free-of-charge conditions for all patients treated at the hospital level. However, the assessment and the decision-making system to grant the NHS coverage operates with substantial delays. According to our model’s data, 64% of medicinal products pending P&R in our cohort ended up receiving regular access at a later stage; for many patients—particularly those with cancer—waiting for that regular access is not an option as it may be too late. Furthermore, even for medicinal products granted a P&R positive decision, several drawbacks still exist to get access due to the involvement of multiple decision-making bodies at regional level—including hospital-level strategies, regional authorities, and evaluation committees—which increase complexity and introduce further delays to regular access.

This complex and conflicting situation increases the need for a well-established procedure for granting early access within the NHS to medicines with regulatory approval but not yet reimbursed or accessible in Spanish hospitals to address unmet medical needs when awaiting for the regular access within the NHS is not an option for the patient. An important point to highlight is that this “non-reimbursement” status is not a static stage in the P&R process: as price negotiations advance, some of these non-reimbursed medicines will sooner or later transition to a reimbursed status and gain regular access within the NHS.

Reductions in access time for innovative therapies in Spain have been reported [14, 20], but it remains far from optimal. Indeed, figures are based on the first individual patient gaining access through local MSS procedures, but do not reflect the reality of the population in clinical practice. MSS pathway is nominal (patient-based), heavily dependent on the physician’s initiative and local instructions. However, even for those situations, where one application is started, it remains unknown how many patients are granted early access. Thus, these data do not represent actual availability [15]. There is an ongoing reform of the system [16] that will probably shorten the time between regulatory approval and reimbursement decision with subsequent regular access. This could reduce the current level of conflict, but no matter how fast the procedure may be, there will always be patients who need it sooner, underlying the need for a standardised procedure for early access to innovative medicinal products.

Our data show that a well-established, transparent procedure, with assessment of requests by physicians specialized in Clinical Pharmacology using pre-defined criteria to justify the approval and a timely response, provides clear benefits for patients at an affordable cost for the system. In our 2-year local experience, up to 40% of the approved requests for oncological cases are cases, where regular NHS access was achieved later on, at a time that would have been too late for our study patients. With a longer follow-up, the percentage of medicinal products finally granted P&R is likely to increase.

In our cohort, oncological patients who were granted special access to non-reimbursed medicines showed an overall disease control rate of 63%, with a median survival of 16 months. Of note, in all cases, the responsible physician and the CPD had agreed on the lack of adequate alternatives for the patient. The actual median annual cost of the requested medicine—accounting for treatment discontinuations, dose reductions, and delays due to toxicity—was 8,977.00 € (IQR 492.34€–23,419€). We consider that the success of the procedure requires an adequate and restrictive selection of candidate patients for early access. Physicians must be aware of the regulatory framework and the hospital criteria for approval, which should ideally have been agreed upon by consensus. Predefining a close monitoring of treatment response and early stopping rules, will contribute to an optimal use of resources. Some indicators of good alignment in our case are the 82.4% approval rate for oncologic requests, where the basal performance status was rather good (61.8% with ECOG 0–1) and up to 15.3% of cases were a curative attempt. These principles do not apply only to oncologic MSS requests, but rather are deemed universal and thus, applicable to every MSS request regardless of the medical condition.

Contrary to CU decisions, which should be based on the patient's individual benefit–risk, access to approved but non-reimbursed medicines rely on the balance between the benefit to the patient and the provider's decision to assume payment in a condition, where there is still no NHS coverage decision. Making this decision in committees or local bodies, with disparate and poorly understood criteria, results in unacceptable inequity in the NHS. Therefore, despite the relative success of our programme, we consider it more appropriate that early access to non-reimbursed medicines follows a common procedure within the NHS.

Although obviously the application and the decision have to be made at local or regional level, under the responsibility of the healthcare provider and payer, this should ideally follow a common process at national level through a standardized procedure with known and agreed criteria that ensures equitable early access within the NHS, irrespectively of the region or institution, where the patient resides [17]. The ongoing review of Spain's Medicines Law offers a strategic opportunity to include provisions for such interim access mechanisms [16].

Early access to innovative therapies not only benefits patients, particularly those with life-threatening or rare conditions, but also creates opportunities to generate reliable real-world evidence on medicine effectiveness and safety during periods of uncertainty [18]. On the other hand, for medicines with a demonstrated potential, it is advisable to incorporate them into NHS as swiftly as possible, since delays in access impose significant opportunity costs on patients awaiting treatment [10].

To be effective, any national procedure should incorporate the following key elements: predictability, with available and pre-established decision-making criteria; due justification based on national-level standards; transparency by all stakeholders involved; timeliness; responsibility and accountability, with due guarantees for the healthcare system and patients (SM Table 3).

A key aspect is the initial market price for early access, which could be negotiated centrally within a very short timeframe. Ideally, this period should be short enough to avoid any delay in the medicine’s NHS access. Alternatively, to gain time—albeit at the cost of an initial investment—the initial price proposed by the company could be temporarily accepted, with subsequent clawbacks or adjustments applied based on the final agreed price for the NHS. This enables early access while addressing uncertainty and ensuring accountability.

There may be situations in which, following early access a decision is ultimately made not to fund the medicine within the NHS. In such cases, it is considered appropriate that no new treatments be initiated but treatments that have already started should be allowed to continue, although criteria for treatment continuation may be established.

Several EU countries (e.g., France and Italy) and the United Kingdom, have implemented early access programs based on population-level frameworks [19]. Spain, by contrast, continues operating under a nominative approach. Nonetheless, the limited availability of published data on early access is a common issue across countries and does not appear to be exclusive to the Spanish context [8].

The real-world budget impact from our cohort was significantly lower than expected, as initial projections—used in the original application—were approximately 3.8 times higher. This discrepancy suggests that some access decisions may have been based on overestimated forecasts, potentially leading to rejecting certain requests due to perceived budgetary constraints. These findings highlight the need for further studies that integrate economic evaluations within early access programs to more accurately determine their true budgetary impact.

We acknowledge several limitations in this study. First, as an observational and retrospective analysis, it is subject to inherent biases and constraints. Second, the study is based on data from a single institution, with a particular assessment process, whose results may not be extrapolated to other hospitals. Third, it must be noted that the evidence supporting MSS requests can vary widely—from therapies with strong evidence of benefit that are used in anticipation of P&R approval, to other requests based on limited or weak evidence, including treatments that may never be reviewed or approved by regulatory authorities. Fourth, due to the wide heterogeneity of the included requests and diseases, the cost-effectiveness analysis was limited to the Oncology subgroup. This approach limits the generalizability of the findings to the rest of the sample studied, and even within the oncology subgroup, given the heterogeneity in treatment aims (curative vs palliative). Cost estimation was limited to drug acquisition costs and did not include other relevant cost components (e.g., administration, monitoring, hospitalizations, or adverse event management), which may underestimate the overall economic burden. Effectiveness was assessed using response rate in the oncology subgroup due to the short follow-up period, and the choice/definition of the comparator relied on assumptions that may not fully capture heterogeneity in usual care. Finally, for cases in which the MSS was finally not administered, we assumed a response value of 0. Although this could have overestimated the MSS benefit, we consider the impact to be low, because in most requests, no other treatment alternatives exist. In support of this assumption, it may be noted that the actual proportion of MSS-treated patients is very low relative to the total number of treatments administered at the hospital, and that the efforts involved in submitting a MSS request suggest that physicians would not have done so if they had believed a feasible alternative existed.

Conclusion

This study highlights the value of a structured early access procedure for medicines in a tertiary healthcare setting, providing greater clarity on the associated costs and timelines before regular access.

Until national legislative changes are introduced to regulate access to medicines awaiting P&R decisions—or until such decisions are made promptly—it is essential to implement a structured, national-level, early access procedure to ensure fair and equitable access, particularly for patients requiring treatment before an official NHS resolution. Early access to high-value therapies is critical both to avoid missed treatment opportunities for patients and prevent value loss for the Healthcare System.

A structured, transparent decision-making framework would support consistent and equitable prioritisation of treatments while remaining compliant with regulatory and financial requirements.

To promote sustainability and fairness, mechanisms, such as clawback systems, pay for performance or conditional approval could be introduced, ensuring value is preserved if treatments ultimately prove to be no cost-effective. At the EU level, harmonising early access programme (EAP) regulations for high-need medicines and establishing country-specific platforms to inform physicians of available therapies could improve coordination, encourage timely referrals, and reduce inequities across regions and countries following regulatory approval.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial or non-for-profit sectors. Article publishing charges was provided by Fundación para la Investigación Biomédica del Hospital Universitario Puerta de Hierro Majadahonda.

Data availability

Due to institutional confidentiality agreements, particularly regarding economic data, the datasets generated and/or analyzed during the current study are not publicly available. Partial data is available for scientific research upon a reasonable request to the authors and subject to institutional approval.

Declarations

Conflict of interest

The authors declare that article publishing charges was provided by Fundación para la Investigación Biomédica del Hospital Universitario Puerta de Hierro Majadahonda. The authors declares that there are no conflicts of interest regarding the publication of this article.

Ethical approval and informed consent

The study protocol and statistical analysis plan were approved by the Research Ethics Committee at the Hospital Universitario Puerta de Hierro (protocol code PI-258/24) with a waiver of informed consent, due to its retrospective nature. The study was conducted in accordance with the provisions of the Declaration of Helsinki (2024) and local procedures for confidentiality and personal data protection.