Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE): an Updated Review

PharmacoEconomics - Open https://doi.org/10.1007/s41669-021-00295-2

REVIEW ARTICLE

Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE): An Updated Review

Ana‑Catarina Pinho‑Gomes1,2 · John Cairns3

Accepted: 2 August 2021 © The Author(s) 2021

Abstract This review discusses the methodological challenges to the evaluation of advanced therapy medicinal products (ATMPs) by the UK National Institute for Health and Care Excellence (NICE). We analysed the health technology appraisals (both published and in development) of ATMPs conducted by NICE in England until July 2021. A total of 14 health technology appraisals of ATMPs were included, of which two were in development and 12 had been published. There were ten gene therapy products (talimogene laherparepvec [TA410], strimvelis [HST7], tisagenlecleucel [TA554 and TA567], axicabtagene ciloleucel [TA559], voretigene neparvovec [HST11], autologous anti-CD19-transduced CD3+ cells [TA677], betibeglogene autotemcel [ID968], onasemnogene abeparvovec [HST15] and OTL-200 [ID1666]), one tissue engineered product (holoclar [TA467]) and three somatic cell therapy products (darvadstrocel [TA556] and autologous chondrocyte implantation [ACI] [TA477 and TA508]). Only three of these technologies were not recommended by NICE, although four were only recommended within the Cancer Drugs Fund. There was large uncertainty in the assessment of clinical efectiveness, as evidence relied mostly on small, single-arm, open-label studies. There were also several concerns in the cost-efectiveness evaluations, such as limited information on health-related quality of life, immature survival data due to short follow-up, and unclear curative potential. Substantial risk may be incurred with ATMPs, which have high upfront and possibly irrecoverable costs but uncertain long-term benefts. In conclusion, the challenges raised by the economic appraisal of ATMPs, albeit not unique, may be exacerbated by the uncertainty related to the often scant evidence base. Adaptations of the conventional decision- making process rather than completely new methods may improve appraisals of ATMPs.

* Ana‑Catarina Pinho‑Gomes [email protected]

1 School of Population Health and Environmental Sciences, Faculty of Life Sciences and Medicine, King’s College London, London SE1 1UL, UK 2 The George Institute for Global Health, London, UK 3 London School of Hygiene and Tropical Medicine, London, UK

Vol.:(0123456789) A.-C. Pinho‑Gomes, J. Cairns

by NICE and discuss the associated challenges and meth- Key Points for Decision Makers odological issues.

Advanced therapy medicinal products have transformative potential, but the limited evidence available presents 2 Methods signifcant challenges to economic evaluation. Evaluation of advanced therapy medicinal products is We searched for appraisals of ATMPs, both published and in pervaded with uncertainty because of the lack of robust development, on the NICE database of technology appraisal evidence in many key aspects. However, conventional guidance [5] and highly specialised technologies guidance methods of health technology appraisal employed by the in July 2021 [5]. We cross referenced this with the list of UK National Institute for Health and Care Excellence ATMPs published by the EMA [6]. For each technology, we may still be applied with some adjustments and perhaps reviewed the fnal appraisal or evaluation determination doc- greater fexibility than for other technologies. ument and associated documents. We extracted data about several domains pertaining to clinical efectiveness and cost Considering the methodological uncertainty in cost- efectiveness, broadly following the structure recommended efectiveness appraisals and the potential for irrecover- by the guide to the process of technology appraisal published able costs for health systems, decision makers may wish by NICE (clinical efectiveness, cost efectiveness, uncer- to consider new contractual arrangements that enable tainty, discounting, subgroups, additional benefts, gener- access to novel technologies whilst sharing the risks alisability to the national health service [NHS], innovation, between health systems and pharmaceutical companies. service reformulation, inequalities, patient access schemes, end of life) [7]. ATMPs for which marketing authorisation was withdrawn were not considered eligible for this study.

3 Results 1 Introduction To date, NICE has published guidance on 14 ATMPs Advanced therapy medicinal products (ATMPs) are medi- (Table 1): ten gene therapy products (talimogene laher- cines that replace or regenerate human cells, tissues or parepvec [TA410] [8], strimvelis [HST7] [9], tisagenle- organs to restore or establish normal function. The Euro- cleucel [TA554 and TA567] [10, 11], axicabtagene cilo- pean Medicines Agency (EMA) identifes three groups of leucel [TA559] [12], voretigene neparvovec [HST11] [13], ATMPs: gene therapy products, tissue engineered products autologous anti-CD19-transduced CD3+ cells [TA677] [14], and somatic cell therapy products [1]. In addition, some betibeglogene autotemcel [ID968] [15], onasemnogene abe- ATMPs may contain one or more medical devices as an inte- parvovec [HST15] [16] and OTL-200 [ID1666] [17]), one gral part of the medicine, which are referred to as combined tissue engineered product (holoclar [TA467] [18]) and three ATMPs (e.g., cells embedded in a biodegradable matrix or somatic cell therapy products (darvadstrocel [TA556] [19] scafold). and autologous chondrocyte implantation (ACI) [TA477 Although potential breakthroughs in this area of clinical and TA508] [20, 21]). The appraisal of sipuleucel-T was research are eagerly anticipated, unregulated use of novel excluded since the guidance and associated documents were therapies with unproven benefts can be not only inefective removed when the EMA withdrew marketing authorisation but also potentially harmful [2]. Furthermore, technology [22]. evaluations of ATMPs may present additional challenges Seven of these contained positive recommendations compared with those of conventional pharmaceutical treat- for use in specifc indications (talimogene laherparepvec, ments, and straightforward application of standard appraisal strimvelis, voretigene neparvovec, holoclar, onasemnogene methods recommended by the UK National Institute for abeparvovec and the two ACI), four recommended that the Health and Care Excellence (NICE) might be difcult or ATMP should be included in the Cancer Drugs Fund (CDF) even inappropriate [3]. Acknowledging the need for fexible (tisagenlecleucel, axicabtagene ciloleucel and autologous and adaptable methods to ensure that innovative technolo- anti-CD19-transduced CD3+ cells) and three did not rec- gies, such as ATMPs, are “fairly, efciently and robustly” ommend the ATMP for use in the NHS (darvadstrocel, beti- evaluated, NICE is currently considering substantial changes beglogene autotemcel and OTL-200). to its methods of health technology evaluation [4]. The aims The incremental cost-efectiveness ratios (ICERs) for of this study are to review evaluations of ATMPs conducted ATMPs recommended under the technology appraisal guidance [5] were within the £20,000–30,000 per Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) within the CDF as an option for treating relapsed or relapsed treating within the for CDF as an option in adults B-cell lymphoma refractory difuse large if the therapies, only systemic or more after two access agreement conditions in the are managed followed survival data and the lack of trialsurvival data com - and the data lack directly chemotherapy paring with salvage tisagenlecleucel means the size of this establish. beneft is difcult to AEs, is associated with frequent Tisagenlecleucel associated withand the managing and treating costs these in the should be refected cost-efectiveness modelling £49,963 per QALY gained with a 3-year cure point, cure with a 3-year gained £49,963 per QALY point cure with a 4-year gained £55,403 per QALY a small observational study. Lack of reliable data of reliable Lack a small observational study. of dataon long-term points. Lack survival and cure for immunoglobulins intravenous about need for cell transplant B-cell aplasia and stem but this defne population is difcult to tion NHS ble Tisagenlecleucel [TA 567] [TA Tisagenlecleucel Difuse large B-cell lymphoma Difuse large Tisagenlecleucel therapy is recommended for use for is recommended therapy Tisagenlecleucel CAR T-cell therapy T-cell CAR Single dose £282,000 Single Tisagenlecleucel is clinically efective, but immature but immature efective, is clinically Tisagenlecleucel £42,991 per QALY gained with a 2-year cure point, cure with a 2-year gained £42,991 per QALY Data from single-arm study with short follow-up and Data single-arm from with study short follow-up 3.5% for both costs and benefts both costs 3.5% for Considered those not eligible for stem cell transplant, cell transplant, stem thoseeligible for Considered not No benefts beyond those captured by QALY calcula - QALY those by captured benefts beyond No Findings are generalisable to clinical practice in the clinical practice to generalisable are Findings Innovative treatment, but reference case still applica - case still but reference treatment, Innovative - tinal large B-cell lymphoma tinal large - treat use within the for CDF as an option for B-cell or refractorying relapsed difuse large or primary B-cell lymphoma large mediastinal systemic or more in adults after two lymphoma if the conditions in thetherapies, managed only access agreement followed are - assessing compara data of comparator makes lack challenging efectiveness tive data for comparators are of low quality. Limited quality. of low are comparators data for data and no direct comparing axicabta - follow-up Uncer chemotherapy. ciloleucel with salvage gene of long-termtainty and survival in extrapolation points. Uncertaintycure and how about whether immunoglobulin will need intravenous pts many immature, so theimmature, of health duration benefts cannot cure show robustly tion NHS cable Axicabtagene ciloleucel [TA 559] ciloleucel [TA Axicabtagene Difuse large B-cell lymphoma and primary B-cell lymphoma medias - Difuse large Axicabtagene ciloleucel therapy is recommended is recommended ciloleucel therapy Axicabtagene CAR T-cell therapy T-cell CAR Single dose but priceSingle known not Axicabtagene ciloleucel is clinically efective, but the efective, ciloleucel is clinically Axicabtagene Over £50,000 peer QALY for salvage chemotherapy salvage for £50,000 peer QALY Over Data a small single-arm from (phase I/II) and study 3.5% for both costs and benefts because evidence is and benefts because evidence both costs 3.5% for No mention No No benefts beyond those captured by QALY calcula - QALY those by captured benefts beyond No Generalisable to patients and clinical practice in the patients and clinical practice to Generalisable Innovative treatment, but reference case still appli - case still but reference treatment, Innovative within the CDF as an option for treating relapsed relapsed treating within the for CDF as an option leukaemia or refractory B-cell acute lymphoblastic if the conditions in only ≤ 25 years, in people aged access agreement followed the are managed beneft vs. blinatumomab or salvage chemotherapy chemotherapy blinatumomab or salvage beneft vs. comparisons indirect as direct is based on naïve datacomparator lacking are sample sizes and diferent populations that sizes and diferent sample are Uncertainty compare. difcult to about whether stem need an allogeneic would pts many and how of data Lack aftercell transplant tisagenlecleucel. determineto side efects. Lack of treating the costs 30 data and reliable on survival beyond of valid efect months and curative some positive and negative separately and negative some positive tion NHS cable Tisagenlecleucel [TA 554] [TA Tisagenlecleucel B-cell acute lymphoblastic leukaemia B-cell acute lymphoblastic Tisagenlecleucel therapy is recommended for use for is recommended therapy Tisagenlecleucel CAR T-cell therapy T-cell CAR Single dose £282,000 Single Tisagenlecleucel is clinically efective, but the efective, is clinically Tisagenlecleucel Over £30,000 per QALY for blinatumomab for £30,000 per QALY Over chemotherapy salvage for £45,000 per QALY Over Only three single-arm (phase II) withOnly studies small 3.5% for costs and benefts costs 3.5% for - Philadelphia chromo analyse small to too Numbers No benefts beyond those captured by QALY calcula - QALY those by captured benefts beyond No Uncertain generalisability to routine practice in the practice routine to Uncertain generalisability Innovative treatment, but reference case still appli - case still but reference treatment, Innovative Summary of characteristics of the advanced therapy medicinal products included in thisSummary medicinal products therapy review of the of characteristics advanced 1 Table name and ATMP Number Conditions Recommendation Type of AMT Type Administration and cost Administration Clinical efectiveness Cost efectiveness (ICER) efectiveness Cost Uncertainty Discounting Subgroups Additional benefts Additional Generalisability to NHS to Generalisability Innovation A.-C. Pinho‑Gomes, J. Cairns Health, but the level of the in discount is commercial Health, but the level confdence. ICER higher than an what is considered acceptable so recommended use of NHS resources, data collect more CDF to on PFS, OS and immu - for usage noglobulin life-extending treatment at the treatment end of life life-extending Tisagenlecleucel [TA 567] [TA Tisagenlecleucel 556] [TA Darvadstrocel Not discussed Not Complex perianal in CD fstulas Complex No issues identifed No Company agreed a PAS with the Department of agreed a PAS Company Tisagenlecleucel meets both criteria meets Tisagenlecleucel a be considered to due to new infrastructure and training required to to infrastructure required and training new due to side efects and manage treatment administer Health, but the level of the discount is commercial Health, but the level - plausible cost-efective in confdence. The most ciloleucel vs. axicabtagene for ness estimates what is considered above are chemotherapy salvage an acceptable- so it is recom use of NHS resources, use within the collect further CDF to mended for data usage on PFS, OS and immunoglobulin considered a life-extending treatment at the treatment end of a life-extending considered life Axicabtagene ciloleucel [TA 559] ciloleucel [TA Axicabtagene need a phased implementation in theWill NHS 467] Holoclar [TA LSCD after eye burns LSCD after eye No issues identifed No Company agreed a PAS with the Department of agreed a PAS Company Axicabtagene ciloleucel meets both criteria ciloleucel meets Axicabtagene be to Health, but the level of the discount is commercial Health, but the level - plausible cost-efective in confdence. The most what above are tisagenlecleucel for ness estimates an acceptableis considered use of NHS resources, use within the CDF to for so it is recommended collect further- cell trans stem data on subsequent usage and immunoglobulin plant rates cannot be considered a life-extending treatment at treatment a life-extending be considered cannot of people with expectancy as the life the end of life leukae - or refractoryrelapsed acute lymphoblastic mia is uncertain Not discussed Not Tisagenlecleucel [TA 554] [TA Tisagenlecleucel 410] [TA laherparepvec Talimogene Unresectable metastatic melanoma Unresectable No issues identifed No Company agreed a PAS with the Department of agreed a PAS Company Tisagenlecleucel extends OS by > 3 months, but it OS by extends Tisagenlecleucel (continued) change in practice change 1 Table Number and ATMP name and ATMP Number name and ATMP Number Service reformulation/ Conditions Inequalities PAS End of life Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) - ing authorisation, for previously treated complex complex treated previously ing authorisation, for or mildly perianal in adults with fstulas non-active luminal CD active only modest beneft over and above placebo. Clinical and above beneft over modest only A 14% addi - in the pts beneft to NHS is unknown. a highly is disappointing for rate tional remission that one-of treatment is associated with complex, that The EMA considered further costs. high upfront and the mar necessary, on efcacy was information authorisationketing the is subject to submission of the number of an ongoing trial results with a larger registry and a global of pts Darvadstrocel [TA 556] [TA Darvadstrocel - within recommended, its market is not Darvadstrocel Allogeneic stem cells stem Allogeneic Single dose (up to three fstula tracts) costs £54,000 costs tracts) three dose (up to Single fstula Data on clinical efectiveness of darvadstrocel showed showed of darvadstrocel Data on clinical efectiveness corneal epithelial cells containing cells) is stem with pts to moderate for as an option recommended burns,if: only LSCD after eye severe or autograft, and or it is contraindicated in the PAS. in at of superfcial corneal neovascularisation corneal with corneal central quadrants, two least visual acuity. impaired and severely involvement, in people withHoloclar is recommended moderate both treating burns for LSCD after eye severe to only: eyes limbal autograft. of of theefectiveness clinical and cost evidence without in pts enough eyes two Holoclar in treating limbal autograft a conjunctival tissue for lial cells containing cells stem VAT remains about the comparator success rates and the about the success rates remains comparator of Holoclar efectiveness comparative Holoclar [TA 467] Holoclar [TA Holoclar (ex vivo expanded autologous human autologous expanded vivo Holoclar (ex and one eye treat used to -It is only limbal autograft had a conjunctival already -Pts have limbal a conjunctival enough tissue for is not -There it with the discount agreed provides company -The the presence LSCD is defned by severe to Moderate and of research -In the context a conjunctival enough tissue for there is not -When robust generate should be designed to research Such Ex vivo expanded autologous human corneal autologous epithe expanded - Ex vivo Single treatment for one eye costs £80,000 excluding £80,000 excluding costs one eye for treatment Single Holoclar is an efective treatment, but uncertainty treatment, Holoclar is an efective as an option for treating unresectable, regionally unresectable, treating regionally for as an option IIIB, IIIC or IVM1a) metastatic (stage or distantly lung bone, brain, to melanoma thatspread has not if: only or other internal organs, suitabletherapies is not and with the discount agreed in the PAS that kills cancer cells dose of 1,000,000 PFU/mL, followed by doses by dose of 1,000,000 PFU/mL, followed and then of 100,000,000 PFU/mL at 3 weeks of talimogene cost The acquisition 2 weeks. every is £1670 per 1 mL vial of eitherlaherparepvec 1,000,000 or 100,000,000 PFU/mL (excluding VAT) - clinical efective conclusions about the relative ipilimumab vs. laherparepvec ness of talimogene consid - because the trial widely used a comparator used in the and not NHS. Improved inefective ered ipilimumab vs. profle toxicity Talimogene laherparepvec [TA 410] [TA laherparepvec Talimogene Talimogene laherparepvec is recommended in adults is recommended laherparepvec Talimogene immuno - administered with systemically -Treatment laherparepvec talimogene provides company -The Modifed form of the herpes virusModifed form simplex type-1 Administered by intralesional injection at an initial intralesional by Administered Evidence presented was insufcient to draw frm draw insufcient to was Evidence presented (continued) 1 Table name and ATMP Number Recommendation Type of AMT Type Administration and cost Administration Clinical efectiveness A.-C. Pinho‑Gomes, J. Cairns QALY duration of beneft is uncertain. Lack of valid data of beneft is uncertain. duration of valid Lack CD with complex no data on HRQoL for on PROs; of data and time to rates on relapse Lack fstula. afterrelapse on understand to impact remission historynatural of disease uncertainty about long-term on efects and impact HRQoL those similar to pts in the RCT for porary in the NHS, as outcomes of management in the NHS than expected placebo armbetter were ble Darvadstrocel [TA 556] [TA Darvadstrocel £23,176–143,131 (estimates are highly uncertain) highly are £23,176–143,131 (estimates per One single RCT with only 1-year follow-up, so the follow-up, 1-year with only RCT One single 3.5% for both costs and benefts because there is both costs 3.5% for Not considered but, in any case, it was only considered considered only case, it was but, in any considered Not No comment No Uncertain generalisability to pts who receive contem - who receive pts to Uncertain generalisability Innovative treatment, but reference case still applica - case still but reference treatment, Innovative No comment No tival limbal allograft donor: a living related from tival per eyes) or £63,047 (two £42,139 (one eye) allograft: Keratolimbal £30,415 (one eye) QALY. BSC: £6948 per QALY. eyes) and £69,455(two If per QALY. eyes) and £12,669 (two (one eye) account, the into taken the efect on the donor was and fall decrease likely Holoclar would ICERs for of £20,000–30,000 per QALY within the range of nature need and the pt innovative Given gained. the the treatment, committee agreed that it would of cost this accept as a demonstration pragmatically efectiveness fve studies with follow-up of 12 months 14.5 up to with studies follow-up fve Uncertainty in the about the assumptions years. long-term and the of comparators use success rates Uncertainty about long-term success of eyedrops. successful at 1 was so if transplant of Holoclar, data for assumed. No success was lifelong year, eyes two LSCD is not a fatal or near-fatal condition or near-fatal a fatal LSCD is not eye, but recommendations apply to unilateral and unilateral to apply but recommendations eye, disease bilateral conjunctival limbal allograft transplantation)conjunctival and limbal autograft (in case of conjunctival donor eye captured not transplantation) were comparators are uncertain are comparators priate Holoclar [TA 467] Holoclar [TA Conjunctival limbal autograft dominated. Conjunc - Conjunctival Three studies with follow-up of ≤12 months, and with studies follow-up Three 3.5% rate for costs and benefts to be applied as and benefts to costs for 3.5% rate Holoclar was only cost efective for treating one treating for efective cost only Holoclar was Potentially negative efects on the donor (in case of negative Potentially Results for Holoclar are plausible, but results for for plausible, but results Holoclar are for Results - appro case still but reference treatment, Innovative No comment No £24,100 per QALY gained against BSC in pts BSC in pts against gained £24,100 per QALY with treatment suitable not for whose disease was immunotherapies administered systemically as based on post-hoc analysis of a subgroup in a analysis as based on post-hoc limited blinding Risk of bias due to RCT. single was Comparator loss of follow-up. and diferential - avail and not care of standard representative not data for and valid of reliable able in the NHS. Lack comparison with ipilimumab (theadequate most in the disease stage equivalent comparison for against dataof efectiveness clinical on Lack NHS). means that similar disease stage for comparator be assessed. could not overall efectiveness cost survival in the overestimated largely Company economic model whom systemic immunotherapy is unsuitable immunotherapy whom systemic tions - localised treat whom alternative for pts suitable for in the NHS available widely not ments are cable Talimogene laherparepvec [TA 410] [TA laherparepvec Talimogene £23,900 per QALY gained against dacarbazine and against gained £23,900 per QALY Evidence of low quality for clinical efectiveness clinical efectiveness for quality Evidence of low 3.5% for both costs and benefts both costs 3.5% for Clinical and cost-efective option only for pts for for pts for only option Clinical and cost-efective No benefts beyond those captured by QALY calcula - QALY those by captured benefts beyond No Not directly addressed but comment that it would be but comment that addressed it would directly Not Innovative treatment, but reference case still appli - case still but reference treatment, Innovative Not discussed Not (continued) change in practice change 1 Table Number and ATMP name and ATMP Number Cost efectiveness (ICER) efectiveness Cost Uncertainty Discounting Subgroups Additional benefts Additional Generalisability to NHS to Generalisability Innovation Service reformulation/ Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) would apply if the technology were recommended were if the technology apply would CD3+ cells is recommended for use within the for CD3+ cells is recommended or refractory relapsed treating for CDF as an option previously in adults who have mantle cell lymphoma if the recommended It is only had a BTK inhibitor. access agreementconditions in the for managed - CD3+ cells treat anti-CD19-transduced autologous followed ment are £46,898 and between range to arrangement) likely but committee favoured gained, £72,920 per QALY gained the of £58,223 per QALY estimate longer and have more time before their disease time before more and have longer relapses Darvadstrocel [TA 556] [TA Darvadstrocel CD3+ cells anti-CD19-transduced Autologous [TA677] No issues No or refractoryRelapsed mantle cell lymphoma Company has a commercial arrangement (PAS) that arrangement has a commercial (PAS) Company No comment No Treatment with anti-CD19-transduced autologous Treatment ICER (with discount agreed in the commercial CAR T-cell therapy T-cell CAR Price was submitted as commercial in confdence submitted as commercial Price was Preliminary evidence suggests that pts may live for for live that may pts suggests Preliminary evidence . 2 cm up to 4 up to 2 , with a greater 2 cm cm 2 2 cm cm diference observed in defects > 2 in defects observed diference Health, but the level of the discount is commercial Health, but the level in confdence option for treating symptomatic articular symptomatic treating cartilage for option if: of the knee in adults only defects cartilage defects in investigating clinicians experienced assessed by measure using a validated knee cartilage damage knee osteoarthritis)for and ture at 2 years for defects of 1–4 defects for at 2 years ture (chondrospheres) transportation. Costs may vary in diferent settings settings vary in diferent transportation. may Costs discounts procurement because of negotiated Chondrosphere improves outcomes at 4 years and outcomes at 4 years improves Chondrosphere microfracture vs. defects larger for utility values after 5 years, the £4360 ICERs are after 5 years, utility values vs. only chondrosphere for gained per QALY gained and £5294 per QALY only microfracture vs. chondrosphere by followed chondrosphere for vs. chondrosphere ICER for only. microfracture gained, be < £20,000 per QALY to BSC is likely > 2 defects for Holoclar [TA 467] Holoclar [TA 508] [TA using chondrosphere ACI No issues No of the articular knee Symptomatic cartilage defects Company agreed a PAS with the Department of agreed a PAS Company Not applicable Not ACI using chondrosphere is recommended as an is recommended using chondrosphere ACI articular repair to surgery had previous -Pt has not osteoarthritic the to knee is minimal (as damage is > 2 -Defect - as microfrac as efective is at least Chondrosphere Autologous chondrocytes combined in a scafold chondrocytes Autologous £10,000 per culture per pt, including cell costs and including cell costs per pt, £10,000 per culture When microfracture is assumed to return is assumed to baseline to When microfracture 2 cm 2 cm Health, but the level of the discount is commercial Health, but the level in confdence if: of thematic articular knee only cartilage defects in investigating clinicians experienced assessed by measure using a validated knee cartilage damage knee osteoarthritis)for implantation tiated procurement discounts. Recommendations discounts. Recommendations tiated procurement of £16,000 based on a maximum cell cost are - microfrac vs. 2–5 years over symptoms improves works (thesuitableture ACI most comparator). knee repair when there has been no previous better and there is no osteoarthritic- and microf damage, >2 defects suitable is not for racture £20,000 per QALY gained but likely to be under to but likely gained £20,000 per QALY subgroups for gained £20,000 per QALY Talimogene laherparepvec [TA 410] [TA laherparepvec Talimogene 477] [TA ACI No issues identifed No of the articular knee Symptomatic cartilage defects Company agreed a PAS with the Department of agreed a PAS Company - sympto treating for as an option recommended ACI surgery knee repair had previous -Pt has not -Osteoarthritic the to knee is minimal (as damage is > 2 -Defect The case for end-of-life considerations was not made not was considerations end-of-life The case for -Procedure is done at a tertiary-Procedure centre referral Traditional ACI or matrix-associated chondrocyte or matrix-associated chondrocyte ACI Traditional Costs may vary in diferent settings because of nego - settings vary in diferent may Costs Although uncertain, that ACI there is some evidence ICER for the whole eligible population may exceed exceed the whole eligible population may ICER for (continued) 1 Table name and ATMP Number name and ATMP Number Inequalities Conditions PAS Recommendation End of life Type of AMT Type Administration and cost Administration Clinical efectiveness Cost efectiveness (ICER) efectiveness Cost A.-C. Pinho‑Gomes, J. Cairns - of pts in ongoing, phase III, multicentre, open- in ongoing, phase III, multicentre, of pts compar of evidence Lack label, single-arm study. CD3+ cells anti-CD19-transduced ing autologous with treatment. thecommon alternative most directly survival dataImmature mean that long-term survival uncertain.- remain cure and potential efective Cost starts, when treatment age to and sensitive ness very older than are NHS pts those included in the trial. Uncertainty HRQoL among long- about whether population term general to is comparable survivors and sex of same age selection) ble Autologous anti-CD19-transduced CD3+ cells anti-CD19-transduced Autologous [TA677] Uncertainty due to short follow-up and small number Uncertainty short due to follow-up 3.5% for both costs and benefts both costs 3.5% for No comment No comment No Evidence is generalisable to pts in the pts to NHS (after pt Evidence is generalisable Innovative treatment, but reference case still applica - case still but reference treatment, Innovative No comment No 2 cm 2 data on the long-term beneft of chondrosphere. Uncertainty that about assumption benefts of likely which cease at 5 years, microfracture ICERs. Despite uncertaintyunderestimated about - and microfrac chondrosphere comparison between in available widely not are of ACI other forms ture, as comparators the be considered NHS so cannot pts (as per TA477): pts knee widely available in the NHS, and there no good are available widely people with defects for options surgical alternative > 2 cm are already captured within captured already are the economic model - ling sphere can be delivered by other centres besides other centres by can be delivered sphere TA477 tertiary as for centres referral ACI using chondrosphere [TA 508] [TA using chondrosphere ACI Evidence from two RCT (phases II and III). Limited RCT two Evidence from 3.5% for both costs and benefts as per reference case and benefts as per reference both costs 3.5% for ACI likely to be cost efective only in subgroups of only efective be cost to likely ACI knee repair had previous not -Pts who have minimal osteoarthritic the to damage -Pts who have -Pts with of > 2 articular cartilage defects There is an unmet need because currently ACI is not is not need because currently ACI is an unmet There Findings from trials likely generalisable to the to NHS generalisable trials from Findings likely Chondrosphere is innovative, but the health benefts is innovative, Chondrosphere - that chondro comment, but acknowledgement No - 2 cm study. Lack of robust evidence to compare long- compare to evidence of robust Lack study. with microfracture, of ACI term efectiveness in thepartially because of diferences popula - on of evidence trial.tions included in each Lack types of ACI, benefts of diferent the relative of data Lack vary. preferences although clinicians’ of requiring knee probability estimate accurately to Uncer of ACI. after failure surgery replacement tain cost at which ACI treatments would be made would treatments ACI tain at which cost the to confdential. NHS as discounts are available Uncertainty for about the modelled utility values is successful or microfracture in whom ACI pts about means estimates and unsuccessful, which utility well Uncertainty on how ICER vary widely. in ACI the refect for population considered values the NHS pts: knee refected thoserefected of NHS pts However, innovation in thisbring case did not innovation However, within captured the what was beyond pts beneft for modelling as the licensed laboratory the that only makes is at the time of TA477 available technology ACI afliated with a tertiary NHS orthopaedic referral hospital ACI [TA 477] [TA ACI Evidence from fve RCTs and one observational and one observational RCTs fve Evidence from 3.5% for both costs and benefts as per reference case and benefts as per reference both costs 3.5% for ACI likely to be cost efective only in subgroups of only efective be cost to likely ACI knee repair had previous not -Pts who have minimal osteoarthritic the to damage -Pts who have -Pts with of > 2 articular cartilage defects Not considered Not reported trialsUncertainty utility values by whether ACI, although not new, is technically innovative. innovative. is technically althoughnew, not ACI, ACI is only recommended in tertiary recommended is only centres referral ACI (continued) change in practice change 1 Table Number and ATMP name and ATMP Number Uncertainty Discounting Subgroups Additional benefts Additional NHS to Generalisability Innovation Service reformulation/ Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) tive, and collecting furthertive, data on PFS, OS and age startswhen treatment uncertainty will reduce in the - anti-CD19-trans autologous Therefore, evidence. use as an for recommended duced CD3+ cells are has a commercial Company withinoption the CDF. and a managed arrangement discount PAS (simple access agreement access including a commercial anti-CD19- autologous agreement). This makes the to NHS with a CD3+ cells available transduced in confdence discount, but size is commercial - treat the criteria a life-extending be considered to it are receiving because pts ment at the end of life < 24 months, and it could extend for live to likely 3 months at least by their life Autologous anti-CD19-transduced CD3+ cells anti-CD19-transduced Autologous [TA677] 15] [HST abeparvovec Onasemnogene SMA No comment No Early estimates suggest treatment could be cost efec - could be cost treatment suggest estimates Early Autologous anti-CD19-transduced CD3+ cells meet CD3+ cells meet anti-CD19-transduced Autologous mutations severe arthritis is a formal severe of the knee (which authorisation), in thecontraindication marketing clinicians allowing by mitigated but those were of based on severity ACI assess suitabilityto for osteoarthritis ACI using chondrosphere [TA 508] [TA using chondrosphere ACI 11] [HST neparvovec Voretigene gene RPE65 gene caused by Inherited dystrophies retinal Inequalities could arise from excluding pts with pts could ariseInequalities excluding from No comment No Not applicable Not severe arthritis is a formal severe of the knee (which authorisation), in thecontraindication marketing clinicians allowing by mitigated but those were of based on severity ACI assess suitabilityto for osteoarthritis may be available to the to NHS be available may ACI [TA 477] [TA ACI 7] [HST Strimvelis Inequalities could arise from excluding pts with pts could ariseInequalities excluding from ADA–SCID No comment, but mention that discounted pricesNo Not applicable Not (continued) 1 Table name and ATMP Number name and ATMP Number Inequalities Conditions PAS End of life A.-C. Pinho‑Gomes, J. Cairns option for treating 5q SMA with treating a biallelic mutation for option and a clinical diagnosis of type 1 in the SMN1 gene if: SMA in babies, only the national multidisciplinaryagreed by team. It is these groups if: for recommended only needed is not tomy arrangement. auditable criteriaciplinary team should develop to be allocated to abeparvovec enable onasemnogene them ≥70% will give babies in whom treatment to Onasem - sit independently. of being able to chance as an option is recommended abeparvovec nogene 5q SMA with presymptomatic a biallelic treating for mutation three and up to copies of in the SMN1 gene if only in babies. It is recommended the SMN2 gene the access agreement conditions in the are managed followed health benefts for babies aged ≤6 months with babies aged type health benefts for in babies who have is little evidence There 1 SMA. >6 months. Onasem - aged are when they treatment out - in better result to is likely abeparvovec nogene before particularly early, is given comes if treatment abeparvovec Onasemnogene of symptoms. the onset long-term health benefts, but the have to is likely of beneft is uncertainextent Onasemnogene abeparvovec [HST 15] [HST abeparvovec Onasemnogene Onasemnogene abeparvovec is recommended as an is recommended abeparvovec Onasemnogene ≤6 months, or aged are They 7–12 months, and their is aged treatment are They - or a tracheos > 16 h/day for Permanent ventilation the to it according commercial provides The company 7–12 months, the babies aged national multidis - For Adeno-associated virus vector-based gene therapy gene virusAdeno-associated vector-based Single dose. £1,795,000 (excluding VAT) dose. £1,795,000 (excluding Single Onasemnogene abeparvovec provides important provides abeparvovec Onasemnogene marketing authorisation, as an option for treating treating authorisation, for as an option marketing in RPE65 -mediated inherited dystrophies retinal inherited withpts retinal vision loss caused by confrmed from biallelic RPE65 muta- dystrophy cells. sufcient viable retinal tions and who have provides if the company only It is recommended the to according commercial neparvovec voretigene arrangement terms of improving vision and preventing vision vision and preventing terms of improving deterioration is and disease progression. There be efect to treatment for a biological rationale maintained Voretigene neparvovec [HST 11] [HST neparvovec Voretigene Voretigene neparvovec is recommended, within is recommended, its neparvovec Voretigene Adeno-associated virus vector-based gene therapy gene virusAdeno-associated vector-based Single dose. £613,410 per patient (excluding VAT) dose. £613,410 per patient (excluding Single Voretigene neparvovec has considerable beneft in has considerable neparvovec Voretigene authorisation, as an option for treating ADA–SCID ADA–SCID treating authorisation, for as an option cell stem related when no suitable HLA-matched donor is available is €594,000 (excluding VAT; at an exchange rate of rate at an exchange VAT; is €594,000 (excluding £505,000) to £0.85, this€1 to equates tion rate is likely to be similar for Strimvelis and Strimvelis be similar for to is likely tion rate the improve nor HSCTs Strimvelis Neither HSCT. and a non-immunological aspects of ADA–SCID, successful substantial proportion who have of pts impairments. Strim - lifelong will have treatment AEs cause fewer to is expected treatment velis than during other options because of the treatment conditioning needed; there is also busulfan lower no risk disease. The risk of graft host of cancer, vs. and be excluded small, cannot although probably surveillance requires Strimvelis [HST 7] [HST Strimvelis Strimvelis is recommended, within its marketing within is recommended, its marketing Strimvelis Ex vivo gene therapy gene Ex vivo Single-dose treatment and lifelong efect. The price and lifelong treatment Single-dose - infec Severe HSCT. OS vs. improves Strimvelis (continued) 1 Table Number and ATMP name and ATMP Number Recommendation Type of AMT Type Administration and cost Administration Clinical efectiveness Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) open-label single-arm with studies, small sample of 18–24 size (13–33 babies). Short follow-up months means efect on long-term survival is mile - long motor and it is unclear whether unknown, maintained. are Supportive that achieved are stones but thesuitable considered care most comparator, limitations, and many historynatural have studies data Immature be comparable. not outcomes may uncertaintyintroduced in economic models, which rather based on assumptions than actual out - were of long-term efects). No comes (e.g., extrapolation data drug on disease-specifc HRQoL from trials, so health-state utilities add uncertainty the to economic needed of care model. Uncertainties the around level in the long term, with the possibility that it may horizon. Uncertainties on the a lifetime over increase HRQoL efect on carers’ subsequent plausible, but a weighting of 1.86, which is lower is lower of 1.86, which plausible, but a weighting applied was gain, QALY for than could be allowed substantial due to uncertainty the criteria despite using a 1.5% discount rate for - uncertainty about long-term benefts and achieve ment of normal health or near-normal groups according to genetic mutations and type of groups genetic to according SMA Onasemnogene abeparvovec [HST 15] [HST abeparvovec Onasemnogene ICER not provided ICER not Evidence from two completed and three ongoing completed two Evidence from QALY gain was 18.62 in the scenario most considered was gain QALY Onasemnogene abeparvovec was considered to meet meet to considered was abeparvovec Onasemnogene Treatment recommendations vary for diferent sub - diferent vary for recommendations Treatment - count rate. £60,908–86,118 per QALY gained at gained £60,908–86,118 per QALY count rate. 1.5% discount rate small RCT (phase III) with 3–4 years’ follow-up. follow-up. (phase III) with 3–4 years’ small RCT decades but for last to efect is expected Treatment - of treat so duration available, limited follow-up ment efect is uncertain; of 40-year assumption data on No reasonable. efect considered treatment disease-specifc HRQoL. Long-term efects and uncertainties in economic key were utility values depends on largely on budget model. Impact the by estimated was which number of eligible pts, in England as 86 pts company applied of the uncertainty about whether voretigene nepar of the uncertainty voretigene about whether the criteria met using a discount fully for vovec uncertain highly pts whether of 1.5%. It was rate be consid - would neparvovec voretigene receiving “normal health” and or near-normal have to ered thewhether long-term would benefts of treatment be achieved gene mutations RPE65 gene diferent caused by trophies is uncertain plausible biologically but not Voretigene neparvovec [HST 11] [HST neparvovec Voretigene £114,956–155,750 per QALY gained at 3.5% dis - gained £114,956–155,750 per QALY Small study (phase I) with 7.5 years’ follow-up and follow-up (phase I) with Small study 7.5 years’ QALY gain was 12.1–17.7, so weighting of 1.2 was of 1.2 was 12.1–17.7, so weighting was gain QALY A 3.5% discount rate was deemed preferable because deemed preferable was A 3.5% discount rate - inherited across varies dys treatment Whether retinal an HLA-matched unrelated donor and a haploi - unrelated an HLA-matched at 3.5%. £54,072 respectively, dentical donor, an an HSCT from vs. and £49,429 per QALY donor and a haploidentical unrelated HLA-matched at 1.5% respectively, donor, studies. Several RCTs; thestarted frst 15 years RCTs; Several studies. - ran of disease-specifc HRQoL from ago. Lack domised trials. Long-term survival after an HSCT uncertain is highly but is one of the or Strimvelis afecting model results. infuential factors most bridging treatment Uncertain of need for duration adenosine glycol-modifed with polyethylene on impact Actual deaminase and associated costs. is which depends on number of pts, NHS budget can Treatment per year. two be one to to expected so costs in Italy, in one centre be delivered only hospitalisation,will be incurred accommoda - for This adds substantial uncertainty as tion and travel. will be made in Euros payments it is unclear how rate exchange because of the fuctuating foreign incurred during a pt additional costs by and how will (e.g., if hospitalisation is extended) their stay be covered from an HLA-matched donor and a haploidenti - an HLA-matched from of 1.40 and 1.96 was weighting QALY cal donor. applied, respectively sidered because of uncertaintysidered pts about whether have to be considered would withtreated Strimvelis “normal health” the and whether or near-normal be achieved long-term would benefts of treatment because of the limited evidence Strimvelis [HST 7] [HST Strimvelis £91,910 and £84,172 per QALY vs. an HSCT from an HSCT from vs. £91,910 and £84,172 per QALY Rare disease, so small number of pts included in disease, so small number of pts Rare QALY gain of 14.0 and 19.6 for Strimvelis vs. HSCT vs. Strimvelis of 14.0 and 19.6 for gain QALY Both discount rates (1.5 and 3.5%) should be con - Both discount rates No comment No (continued) 1 Table name and ATMP Number Cost efectiveness (ICER) efectiveness Cost Uncertainty QALY weighting QALY Discounting Subgroups A.-C. Pinho‑Gomes, J. Cairns with SMA and their carers by achieving motor motor achieving with SMA and their by carers enabling as sitting and walking, such milestones small gains Even participation and school. in society side to from roll as ability to function, such in motor side, hold the head or lift the arms into can translate and carers pts for of life quality improved greatly nosis is becoming more common nosis is becoming more reference case should be adjusted reference in a small number of highly specialised centres specialised centres in a small number of highly because there expertise. concentrate This is a need to long distances. travel need to may means families be required may staf training Additional munities led the committee to allow treatment to be to treatment munities led the allow committee to 6 months in certain beyond circumstances ofered Health, but the level of the in discount is commercial Health, but the level confdence Onasemnogene abeparvovec [HST 15] [HST abeparvovec Onasemnogene Signifcant improvement in the quality of life of pts of pts of life in the quality Signifcant improvement Evidence is generalisable to the diag - to NHS as earlier Evidence is generalisable Innovative technology, but no comment about whether but no comment about whether technology, Innovative Onasemnogene abeparvovec would only be delivered be delivered only would abeparvovec Onasemnogene Possible delay in diagnosis in disadvantaged com - in diagnosis disadvantaged delay Possible Company agreed a PAS with the Department of agreed a PAS Company Not applicable Not attend mainstream school, retain their retain independ - school, attend mainstream their partence, take in social activities and achieve reduce would neparvovec Voretigene full potential. not would as parents on families fnancial impact for and pay care provide to up work give need to - neparvo home adaptations. Voretigene expensive incurred the decrease expenditure by would vec departments thatnon-NHS government provide vision loss afected by families support for difcult to predict in pts with less severe diagnoses with in pts predict difcult to less severe because of individual variability whether reference case should be adjusted reference whether staf training to reduce risk of AEs. It is feasible to to risk reduce staf to training of AEs. It is feasible in NHS clinical practice thisimplement technology Health, but the level of the discount is commercial Health, but the level in confdence Voretigene neparvovec [HST 11] [HST neparvovec Voretigene [ID968] autotemcel Betibeglogene With sustained vision, children would be able to be able to would sustained vision, children With TDT Relevance of study results to clinical practice is clinical practice to results of study Relevance Innovative technology, but no comment about technology, Innovative Not specifcally addressed but mention the need for but mention the addressed need for specifcally Not No issues No Company agreed a PAS with the Department of agreed a PAS Company Not applicable Not would be cost savings and benefts with Strimvelis and benefts with Strimvelis savings be cost would incurred outside the NHS and personal social - in stafng and infra changes services. major No be needed structure would velis in clinical practice may be lower than that in be lower may in clinical practice velis greater lead to clinical the clinical trial, may which beneft whether reference case should be adjusted reference whether with Strimvelis, so no additional infrastructure or with Strimvelis, centres be needed at specialist staf would training in England transplant related to lack of donors for certain of donors for lack to related transplant common is more in whom ADA–SCID ethnicities origins) and Somalian family (Irish traveller Strimvelis [HST 7] [HST Strimvelis OTL-200 [ID1666] Improvements to carer-related quality of life. There There of life. quality carer-related to Improvements Metachromatic leukodystrophy Metachromatic The age of the population who would receive Strim- receive of the population who would The age Innovative technology, but no comment about technology, Innovative People would need to travel to Italy for treatment treatment for Italy to travel need to would People Potential to reduce inequalities in waiting time for time for in waiting inequalities reduce to Potential No comment No Not applicable Not (continued) change in practice change 1 Table Number and ATMP name and ATMP Number name and ATMP Number Additional benefts Additional Conditions Generalisability to NHS to Generalisability Innovation Service reformulation/ Inequalities PAS End of life Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE)

/beta 6 0 beta within its marketing authorisation, for treating TDT TDT treating authorisation,within for its marketing a have who do not ≥12 years aged in pts than £30,000 per QALY gained than £30,000 per QALY tually no longer need blood transfusions or need need blood transfusions no longer tually them less often cells/mL dispersion for a one-time infusion is cells/mL dispersion for VAT). price, (list £1,450,000 per pt excluding arrangement, has a commercial which Company had been applied if the technology have would recommended β-thalassaemia, using a lentiviral gene delivery delivery β-thalassaemia, gene using a lentiviral system genotype, when HSCT is appropriate, but an HLA- genotype, available HSC donor is not related matched Betibeglogene autotemcel [ID968] autotemcel Betibeglogene Betibeglogene autotemcel is not recommended, recommended, is not autotemcel Betibeglogene The most plausible ICER was considerably higher considerably plausible ICER was The most - that either some pts even suggests evidence Modest Cost of betibeglogene autotemcel 1.2 to 20×10 1.2 to autotemcel of betibeglogene Cost β-globin gene therapy for sickle cell anaemia and sickle for therapy gene β-globin - leukod metachromatic treating authorisation, for biallelic mutations in the characterised by ystrophy enzyme activity in that ARSA reduce gene ARSA who have: children signs or symptoms independently walk and who can still symptoms, decline. no cognitive and have withtreatment OTL-200started that in the was published. Children this guidance was NHS before outside this treatment may recommendation having the to funding arrange - continue without change this guidance was them before ments in place for and their NHS clinician con - published, until they sider it appropriate This decision should be stop. to theor young clinician and the child by made jointly person or their or carers parents - be a cost-efec considers to NICE normally range (£100,000 per QALY) use of NHS resources tive of consensual response criteria.of consensual response Less than 50% of a full response have to patients expected VAT; company submission). Company has a com - submission). Company company VAT; if the apply would arrangement,mercial which recommended were technology gene (ex vivo genetically modi - genetically vivo (ex gene human ARSA and stem CD34+ haematopoietic fed autologous cells) progenitor OTL-200 [ID1666] OTL-200 is not recommended, within its marketing OTL-200 within recommended, its marketing is not types, with no clinical juvenile or early Late infantile clinical signs or type, with early juvenile The early afect intended to is not This recommendation ICER is commercial in confdence but outside theICER is commercial Clinical efectiveness is highly unclear due to lack lack unclear due to is highly Clinical efectiveness Single dose. List price is £2,875,000 (excluding price dose. List Single is £2,875,000 (excluding Gene therapy medicinal product that expresses the that medicinal product Gene therapy expresses (continued) 1 Table name and ATMP Number Recommendation Cost efectiveness (ICER) efectiveness Cost Clinical efectiveness Administration and cost Administration Type of AMT Type A.-C. Pinho‑Gomes, J. Cairns genotype 0 /beta 0 beta due to recent improvements in treatment. Clinical in treatment. improvements recent due to efcacy based on three single- small, multicentre, arm in total), with (24 pts limited follow-up studies is ongoing. study Long-term follow-up (5 years). population as HRQoL based on UK Chart Review trial insufcient data were about long-term efect and life expectancy associ - expectancy about long-term efect and life ated with TDT per QALY gained was considered acceptable considered was gained per QALY Betibeglogene autotemcel [ID968] autotemcel Betibeglogene Not applicable Not Uncertainty about life expectancy of pts with of pts TDT expectancy Uncertainty about life Only indicated for pts without pts indicated for Only 3.5% for both costs and benefts due to uncertainty and benefts due to both costs 3.5% for No comment No No comment No No comment No Innovative treatment so an ICER closer to £30,000 so an ICER closer to treatment Innovative - weight between 1 and 3, but the exact weighting weighting 1 and 3, but the exact between weight uncertainwas and depended on the distribution of subgroups if pooled centre trialscentre with in total). size (25 pts small sample and cognitive motor to Clinical outcomes related of life. quality refect adequately not function may lifetime a pt’s for OTL-200 could be efective of the infused cells maintains because the progeny successful engraft - correction.the gene However, ment and migration the of cells into nervous central so the disease 2 years, up to could take system there efect. is a treatment could progress before Uncertainty included in analy pts whether as to the to population of eligible generalisable ses are Uncertaintypts. and cryopreserved about fresh of OTL-200, be formulations as the latter would fresh from was but evidence used commercially Uncertainty of progressionformulation. about rate HRQoL No and stabilisationresponse. of treatment data collected as part of the trials. ICER Pooled subgroups of diferent prevalence by weighted adds uncertainty because some subgroups have outcomes, the distribution of subgroups is worse subgroup in each pts there few very are unknown, distribution to of and the sensitive ICER is very subgroups of diferent subgroupsof diferent about long-term of normal efect and restoration or health near-normal - qualita but considered quantifed completely not and their children committee. Benefts for by tively families and full benefts not captured by economic model by captured and full benefts not OTL-200 [ID1666] QALY gain commercial in confdence. QALY in confdence. QALY commercial gain QALY Non-randomised, open-label, prospective, single- open-label, prospective, Non-randomised, Single ICER was presented weighted by prevalence prevalence by weighted presented ICER was Single 3.5% for both costs and benefts due to uncertainty and benefts due to both costs 3.5% for Full efect of benefts beyond direct health direct benefts efect of benefts beyond Full No comment No No comment No Innovative treatment that can be life transforming, transforming, that treatment can be life Innovative (continued) change in practice change 1 Table Number and ATMP name and ATMP Number QALY weighting QALY Uncertainty Subgroups Discounting Additional benefts Additional Generalisability to NHS to Generalisability Service reformulation/ Innovation Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) - - ran Cancer DrugsCancer CDF Crohn’s disease, Crohn’s CD value-added tax value-added chimeric antigen receptor, receptor, chimeric antigen CAR Bruton’s tyrosine kinase, tyrosine Bruton’s BTK transfusion-dependent β-thalassaemia, transfusion-dependent VAT criteria efective, and collecting furtherefective, data on PFS, OS starts when treatment and age the will reduce autologous uncertainty Therefore, in the evidence. - recom CD3+ cells are anti-CD19-transduced within use as an option the CDF. mended for arrangement has a commercial (simple Company access agreement and a managed discount PAS access agreement).including a commercial This CD3+ anti-CD19-transduced autologous makes the to NHS with a discount, but cells available in confdence. Betibeglogene size is commercial range a wider age to be available would autotemcel be accessible only it would However, than HSCT. a specifc genotype to tion groups is outside the remit of a technology tion groups is outside the of a technology remit seen in ethnic In theappraisal. is mostly UK, TDT groups being in minority populations, the largest - fam Indian or Bangladeshi people with Pakistani, backgrounds ily Betibeglogene autotemcel [ID968] autotemcel Betibeglogene Betibeglogene autotemcel does not meet end-of-life end-of-life meet does not autotemcel Betibeglogene Early estimates suggest treatment could be cost could be cost treatment suggest estimates Early Incidence of a condition in diferent popula - Incidence of a condition in diferent best supportive care, care, supportive best BSC technology appraisal, TDT appraisal, technology Health, but the level of the discount is commercial Health, but the level in confdence treatment “before the onset of cognitive decline,” decline,” of cognitive the onset “before treatment so it is critical that ensure to those with pre-existing in learning disadvantaged not disabilities are diagnosis in Delayed accessing the technology. - also cre groups may disadvantaged from children ate inequalities OTL-200 [ID1666] Not applicable Not Company agreed a PAS with the Department of agreed a PAS Company Marketing authorisation states that pts should have authorisation states Marketing that should have pts advanced therapy medicinal product, product, medicinal therapy advanced ATMP RCT life-year, quality-adjusted patient-reported units, PROs patient(s), QALY outcomes, pt(s) PFU plaque-forming survival, PFS progression-free patient access scheme, PAS (continued) highly HST highly cell transplant, HSCT homologous stem cell, HSC homologous stem of life, HLA HRQoL health-related quality antigen, human leukocyte Medicines Agency, EMA European aryl ARSA medicinal therapy, AMT advanced event, AE adverse combined immunodefciency, adenosine deaminase defciency–severe implantation, ADA–SCID chondrocyte autologous Fund, Fund, , OS overall and Excellence Care NHS national health for service, Institute NICE National LSCD ratio, cell defciency, limbal stem ICER incremental cost-efectiveness specialised technologies, survival, sulphatase A, spinal muscular atrophy, TA trial,domised controlled SMA spinal muscular atrophy, 1 Table name and ATMP Number ACI End of life PAS Inequalities A.-C. Pinho‑Gomes, J. Cairns quality-adjusted life-year (QALY) threshold usually con- because ATMPs are typically prescribed as end-of-line sidered by NICE, whereas the ATMPs with ICERs of treatment to people with rare diseases, which limits sample £30,000–50,000 were recommended for the CDF. For the size. However, this does not necessarily explain why studies three ATMPs approved under the highly specialised tech- were non-blinded and single-armed, which raises concerns nologies guidance [5], which applies a threshold of £100,000 about the pharmaceutical industry sponsoring those studies. per QALY, the ICERs for strimvelis were under the thresh- The limited evidence available raised important issues for old irrespective of discount rate, but those for voretigene assessing clinical efectiveness. First, short follow-up meant neparvovec were over the threshold when a 3.5% discount that survival data were immature, so the survival estimates rate was applied. In the case of onasemnogene abeparvo- and overall duration of efects were highly uncertain. In this vec, using a discount rate of 1.5%, the ICER was under the context, scenario analyses exploring the efects of diferent weighted threshold of £186,000 per QALY. assumptions about long-term benefts might be advanta- Technology appraisals also considered whether there geous, as shown by their use for the evaluation of onasem- were externalities or additional benefts beyond those cap- nogene abeparvovec [4]. Second, single-arm studies did not tured by QALYs. These additional benefts for individuals provide direct evidence of efcacy against comparators, and included avoidance of negative impact of alternative treat- fnding suitable comparator(s) with valid and reliable data ments (holoclar) and improved independence, social par- available was difcult. For instance, data from previous stud- ticipation and ability to achieve full potential (strimvelis, ies often did not refect contemporary practice because of the voretigene neparvovec and onasemnogene abeparvovec). fast evolution of therapeutics, or they included populations Positive externalities were taken into account qualitatively with diferent disease stages and prognoses. Comparison in the recommendations made for highly specialised tech- with historical cohorts that studied the natural progression of nologies, such as reduced fnancial impact on families and disease was also naïve (e.g., OTL-200). This rendered quan- society overall due to cost savings for the NHS and non-NHS tifcation of benefts and potential harms associated with dif- government departments (e.g., social care). Chimeric anti- ferent treatment modalities challenging, which translated gen receptor (CAR) T-cell therapies were deemed to have no into uncertainty in the evaluation of clinical efectiveness. additional benefts, whereas the appraisals of darvadstrocel, The criteria for life-extending end-of-life treatments, betibeglogene autotemcel and ACIs did not comment on which include (1) life-extending treatments for people with additional efects and externalities. a short life expectancy (normally < 24 months) and (2) pro- viding a gain in overall survival of > 3 months [5], were introduced to allow the QALYs gained by treatments that 4 Discussion improve patient survival in terminal stages of disease to be valued more highly. Although these criteria have allowed This review suggests that appraisals of ATMPs present spe- approval of drugs that would have otherwise been denied cifc, and arguably greater, challenges to NICE committees to NHS patients, all bar one have been cancer drugs [23]. than do other technologies. Uncertainty pervades clinical In keeping with this, of the ATMPs, only three of the CAR and cost efectiveness in all appraisals because of the sparse T-cell therapies (TA559, TA567 and TA677) met the criteria or even absent evidence on utility values, long-term efects for life-extending end-of-life treatment. Perhaps acknowl- and costs of treatment versus comparators. These issues are edging the limited scope and questionable fairness of the partly due to the nature of the conditions, which are typically end-of-life criteria, NICE is considering replacing these rare and severe, and partly due to the technologies, which criteria by a modifer for severity of disease [4]. The actual are novel and yet to be fully understood. We explore the key number of technologies that will beneft from a severity issues related to the evaluation of clinical and cost efective- modifer depends on how it is defned and applied, but they ness of ATMPs, and we discuss the potential implications of can be expected to form a broader range of conditions and adoption of ATMPs for the NHS and society overall. to refect more accurately societal values than the previous end-of-life criteria. 4.1 Clinical efectiveness NICE has statutory and ethical duties to support innovative technologies, so greater risks may be accepted to enable Overall, evidence on the clinical efectiveness of ATMPs access to highly innovative technologies, which have valu- was limited and weak, but this varied substantially across able benefts for patients and society [24]. Although NICE technologies. The main difculty in determining the clini- defnes which technologies qualify as innovative [5], there cal efectiveness of ATMPs was related to the limitations is much room for interpretation, and ambiguity has resulted of the studies available, which were mostly single-armed, in signifcant variability in implementation [25]. This raises non-blinded studies with small sample sizes and short fol- concerns as to whether innovation, which is not an inde- low-up. The poor quality of the studies available may be pendent social value, may be jeopardising the core values Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) of health and equity in the NHS [26]. The possibility of and preferred the 3.5% discount rate in most cases. How- considering innovation as a modifer also does not garner ever, for strimvelis and voretigene neparvovec, committees NICE support at present, as there is no evidence that society stated that both discount rates would be taken into consid- values health benefts from innovative technology more than eration, acknowledging the uncertainty about whether those equivalent benefts from less innovative technology [27]. In therapies would restore normal or near-normal health. For keeping with this, no adjustment to the reference case was onasemnogene abeparvovec, a 1.5% discount was applied, warranted for any of the ATMPs, despite all being consid- despite, arguably, similar uncertainty about whether efects ered highly innovative. would be sustained in the long term and patients would achieve normal or near-normal health. This suggests that a 4.2 Cost Efectiveness lower threshold for considering the 1.5% discount rate may be applied in highly specialised technology appraisals than Overall, there was substantial uncertainty in estimates of in standard technology appraisals. The common disagree- cost efectiveness of ATMPs because of the lack of valid ment between companies and committees on discounting and reliable data to inform the economic models. This was demonstrates the subjectivity involved in the interpretation reflected in the disagreements about model parameters of the fulflment of the eligibility criteria. Furthermore, con- between committees, evidence review groups and compa- cerns have been raised that the non-reference case discount nies. The poor quality of the clinical studies was further rate is rarely applied as it sets a high bar for the technologies compounded by a lack of data on patient-reported outcomes, it otherwise appears to support by requiring no signifcant particularly health-related quality of life. This meant that irrecoverable costs and implying that a high degree of cer- utility values specifc to the condition of interest were often tainty is needed [29]. For all these reasons, NICE is consid- missing, and extrapolation from the literature introduced ering a proposal to use a reference case discount rate of 1.5% further uncertainty to the models (e.g., holoclar, strimvelis, per year for both costs and health efects [4], which is the voretigene neparvovec, betibeglogene autotemcel, onasem- discount rate for health values recommended by the govern- nogene abeparvovec, OTL-200). In some cases, economic ment [30]. Lowering the reference case discount rate may models may have overestimated utility gains of treatment increase the cost efectiveness of ATMPs, particularly those (e.g., talimogene laherparepvec, betibeglogene autotemcel). that have high upfront costs and long-term health benefts. For CAR T-cell therapies, overestimation of long-term sur- ATMPs are, in general, very expensive, and patient access vival and cure points may have exerted a paramount efect schemes have been used to lower the ICER and thus facili- on cost efectiveness, as those were key determinants of the tate NICE approval [31]. These are agreements between the economic models. In addition, QALY valuation when efects Department of Health and pharmaceutical companies that are predicted to be lifelong remains complex because of con- enable companies to ofer discounts or rebates that reduce cerns about whether diferent social values apply to evalua- the cost of a drug for the NHS. Simple discount schemes tions of potential cures and substantial uncertainty in long- have been preferred over complex schemes, such as provi- term outcomes [28]. Nonetheless, the lifelong nature of the sion of free stock, dose caps or payments by results (i.e., efects of strimvelis, voretigene neparvovec, onasemnogene performance or outcome-based schemes) [32]. It is argu- abeparvovec and OTL-200 was assumed based on biologi- able, though, that the latter could be appropriate to ATMPs, cal plausibility, and QALYs were weighted (i.e., granted an when there is substantial uncertainty on long-term efects increased value) to refect the added value of large QALY and high upfront costs [33]. National discounts are known to gains throughout life, thus enhancing their cost efective- NICE committees, and cost-efectiveness models are based ness. However, this is controversial as there is no evidence on discounted prices. However, “companies sometimes pro- that society places additional value on technologies that are vide confdential [local] discounts to the NHS, making the potentially curative, and NICE is not supportive of a specifc real cost of cells difcult to ascertain” (TA477), in which modifer for potentially curative treatments at present [4]. case committees accept models based on the approximate NICE guidance recommends using a 3.5% discount list price for the technology. In addition, the exact details rate for both costs and benefts, but a 1.5% discount rate is of discounts are commercial in confdence, which, albeit allowed when appraising treatments that “restore people who understandable, compromises transparency and may under- would otherwise die or have severely impaired life to full or mine patients’ trust in how decisions are made about ration- near full health, and when this is sustained over a very long ing of healthcare resources in the NHS [34]. On the other period (normally at least 30 years)” [7]. Although compa- hand, patient access schemes may be disproportionately nies claimed that their products were eligible for the 1.5% beneftting certain technologies. CAR T-cell therapies were discount rate (e.g., tisagenlecleucel, holoclar, betibeglo- recommended for inclusion in the CDF, which allows their gene autotemcel and OTL-200), committees noted that “it use in the NHS while further data are collected to support a is rarely considered appropriate to change the discount rate” robust appraisal of their cost efectiveness and a subsequent A.-C. Pinho‑Gomes, J. Cairns fnal recommendation. It is arguable whether similar fund- The frst could reduce inequalities in waiting time for ing arrangements should be potentially available to all tech- transplant that result from a lack of suitable donors for cer- nologies irrespective of the underlying disease (e.g., dar- tain ethnicities in whom the condition for which it is rec- vadstrocel could have been approved under such a scheme ommended (i.e., adenosine deaminase defciency–severe whilst further data were being collected by a large trial). combined immunodefciency) is more common, such as The 2019 general election featured a promise to replace the Irish traveller and Somalian family origins. This then CDF with an Innovative Medicines Fund, but this is yet to leads to increased waiting times and potentially worse out- be implemented. comes for those individuals. Conversely, ACI could result The four ATMPs evaluated as highly specialised technol- in inequalities as marketing authorisation excludes those ogies listed children, and specifcally very young children, with severe osteoarthritis. However, this was mitigated by as the primary benefciaries. In these cases, the higher valu- allowing clinicians to assess patients’ suitability for ACI ation of QALYs assigned to treatments that have potentially based on severity of osteoarthritis. In addition, restricting lifelong efects (implemented via a higher cost-efectiveness ACI in TA477 to tertiary referral centres may lead to geo- threshold) captured the increase in beneft that would result graphical inequalities as ACI may not be widely available from delivering a potentially transformative treatment to across the country. For onasemnogene abeparvovec, the children while remaining in line with NICE’s view that a potential for a delayed diagnosis in disadvantaged chil- modifer based on age is not appropriate [4]. As ATMPs dren led the committee to allow treatment to be ofered to are expected to target severe genetic diseases that mani- children aged > 6 months in certain circumstances, despite fest in infancy, the valuation and measurement of QALYs scant evidence on clinical efectiveness beyond 6 months in children presents substantial challenges. NICE recom- of age. NICE’s commitment to addressing inequalities mends using a generic measure with good psychometric is illustrated by the fact that it considered adding impact performance in the relevant age group and reporting who on inequalities as a modifer to future health technology has completed the questionnaire [5]. Nevertheless, the dif- appraisals [4]. This is based on equality legislation [37] culty of valuing and measuring utility in children introduces and evidence suggesting that the UK population prioritises additional uncertainty to cost-efectiveness evaluations, and seeking a fair distribution of health across society and is further research is warranted to refne methods for assessing willing to trade-of less health overall if the health is gen- health-related quality of life in children [35]. erated in disadvantaged groups, particularly for socioeco- Subgroup analysis may be especially important for nomic disadvantage [38]. However, further work is needed ATMPs, as ICERs are typically high and may vary signif- to explore which sources of inequality to include, whether cantly across subgroups. NICE allows committees to rec- direct and indirect efects would be considered, and how ommend treatment for a selected subgroup of patients irre- and in which circumstances this could be implemented in spective of whether the technology is found to be clinically health technology evaluations. and cost efective for the whole population, provided that Although none of the ATMPs included in this review the decision is ethically and methodologically sound [5]. required major service redesign and staf training, ATMPs Although ten of the 14 technologies analysed in this review are by nature likely to have signifcant service delivery considered whether treatment efects would be diferent in efects. NICE prefers that the full additional cost of introduc- certain subgroups of patients, only six made specifc sub- ing a technology is included in the evaluation. However, it group recommendations (talimogene laherparepvec, holo- concedes that some of the costs may be apportioned between clar, ACIs, onasemnogene abeparvovec and betibeglogene other technologies, particularly when “high-cost medical autotemcel). The remaining three evaluations considered devices have other uses beyond the evaluated indication that data were lacking to determine whether there were (either now or in the future) or for technologies that need meaningful diferences in treatment efects across subgroups substantial service redesign to introduce them (which may (tisagenlecleucel, voretigene neparvovec) or that it was clini- afect other technologies or people not having treatment with cally difcult to identify those subgroups (tisagenlecleucel). a technology)” [4]. This fexibility to allow non-reference This clearly illustrates the challenges in assessing the credi- case analysis may be especially suitable for ATMPs, as costs bility and relevance of diferences between subgroups, which may be shared by diferent stakeholders and technologies. have also been acknowledged by NICE [4]. In addition, ATMPs may be associated with additional costs The impact on inequalities may be particularly rele- that would not commonly be considered in other technol- vant for the evaluation of innovative technologies, such ogy evaluations, such as costs associated with travelling to as ATMPs, as there may be a greater risk of creating or specialist centres abroad. For strimvelis, these costs were exacerbating inequalities [36]. In this review, all ATMPs not incorporated in the economic model, and it was unclear except strimvelis, ACI and onasemnogene abeparvovec how those costs would be covered in the long term. How- were considered to have a neutral efect on inequalities. ever, whether this should be included in economic models Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE) to fully refect the cost for the NHS is arguable, especially Nonetheless, whereas voretigene neparvovec, strimvelis and if, as expected, these issues become increasingly common. onasemnogene abeparvovec were recommended despite the scant evidence on long-term benefts, betibeglogene auto- 4.3 Implications for the National Health Service temcel was not recommended, at least partially because of and Society Overall “the potential to commit the NHS to irrecoverable costs”. This illustrates the difcult compromise between allowing Although ATMPs offer potentially important benefits fexibility for non-reference case analysis (e.g., by accepting for patients, their families and society overall, their high greater risks) and allocating resources to maximise popula- upfront and often one-of costs can pose challenges related tion health [41]. to afordability and implementation in the NHS [39]. Many Third, ATMPs may have wider consequences for patients, of the mitigation measures are commercial in nature (e.g., their families and carers, the NHS and other governmental patient access schemes) and not directly in the remit of sectors and ultimately society at large. Those consequences NICE technology evaluations. However, NICE is account- are often benefcial and hence associated with positive utility able to the NHS, the government and the public it serves, (e.g., reduced burden for paid and unpaid carers). However, and thus the wider implications of health technology evalu- they can be associated with disutility, particularly when ations of ATMPs for the NHS and society overall cannot be ATMPs extend life expectancy but with signifcant disabil- overlooked. ity, leading to an increased need for caregiving over a life- The frst concern raised by the implementation of ATMPs time horizon. For instance, for onasemnogene abeparvovec, by the NHS relates to the generalisability of research con- caregiver disutility was not included in the model because ducted elsewhere. This question, albeit shared with other it was difcult to quantify the disutility for carers and it technologies, is paramount for ATMPs because of the scant would increase the ICER, which was considered “counter- evidence available. Committees sought advice from clinical intuitive”. In other ATMP evaluations, wider benefts con- experts and complemented this with data from observational tributed qualitatively to the appraisal of the evidence and studies in England, whenever possible (e.g., darvadstrocel). the fnal recommendation, as they were hard to quantify or Findings were considered generalisable to clinical practice intangible (e.g., OTL-200). This is partly due to the lack of in the NHS for axicabtagene ciloleucel, tisagenlecleucel in data on these benefts, as no studies had included them as TA567 and ACI in TA508. However, in other cases, gener- outcomes. However, even if data were available, the refer- alisability was questionable because outcomes for the com- ence case does not contemplate wider benefts that may be parator did not represent contemporary outcomes in the NHS accrued because of treatment, particularly when these are (e.g., relapse rates were much higher for the comparator of non-health benefts (e.g., reduced need for social care) or darvadstrocel in the single trial available than rates reported when they fall on someone other than the person receiving in the NHS), the characteristics of the patients in the stud- treatment (e.g., parents and carers in general). NICE guid- ies did not match those of typical NHS patients (tisagen- ance acknowledges that “care delivered by the NHS could lecleucel in TA554, darvadstrocel, holoclar, strimvelis) or have other benefts that are considered socially valuable but heterogeneity in treatment efects was possible depending are not directly related to health and are not easily captured on individual variability (voretigene neparvovec). in a cost per QALY analysis” [7]. Nonetheless, incorporat- Second, ATMPs are associated with a larger fnancial risk ing techniques “that consider the trade-of between health for the NHS than many other technologies [39]. They often benefts and non-health benefts quantitatively” into decision have high upfront and, depending on contractual arrange- making was considered unsuitable in 2013, and it did not ments, potentially irrecoverable costs, yet the full benefts feature in the case for change of health technology evalua- may take many years to accrue or may not be permanent. tion in 2020 [4]. NICE recommends that non-reference case Furthermore, the high cost per patient means there is high analysis is used when there are substantial identifable health volatility, which may be challenging to accommodate in benefts not captured by QALYs and emphasises the need annual budgets and increase fnancial risks if there is an to develop methods to formally incorporate qualitative evi- unexpectedly high number of cases. The quintessential ques- dence into decision making [4]. tion is how much risk the NHS should incur and in which circumstances, as well as whether the risk should be shared by pharmaceutical companies (e.g., by using outcome- 5 Conclusion based contracts) [40]. NICE considers that greater risks could be accepted for “conditions for which it is recognised NICE evaluation of ATMPs revealed signifcant challenges, that evidence generation is complex and difcult, such as mainly related to large uncertainty about long-term and rare diseases, innovative technologies, technologies that potentially curative efects, high upfront costs, discount- provide large benefts”, all of which apply to ATMPs [4]. ing, innovation, benefts above and beyond QALYs and A.-C. Pinho‑Gomes, J. Cairns apportioning of costs. However, these challenges are not 4. National Institute for Health and Care Excellence. Reviewing our unique to ATMPs, so completely diferent methods may methods for health technology evaluation: consultation. https:// www.nice. org. uk/ about/ what- we- do/ our- progr ammes/ nice- guida not be required. Adaptations to the conventional decision- nce/chte-methods-consultation. Cited 12 Dec 2020. making process, such as the use of fexible non-reference 5. National Institute for Health and Care Excellence. Guidance and case analysis, integration of diferent sources of evidence, advice list. https://www. nice. org. uk/ guida nce/ publi shed? type= hst . and special funding arrangements, may improve appraisal Cited 12 Dec 2020 6. Summaries of scientifc recommendations on classifcation of of ATMPs. advanced therapy medicinal products. https://www.ema.europa. eu/en/ human- regul atory/ marke ting- autho risat ion/ advan ced- thera Declarations pies/advan ced- thera py- class ifca tion/ summa ries- scien tifc- recom mendations- class ifca tion- advan ced- thera py- medic inal- produ cts . Funding No sources of funding were used to conduct this study or Cited 20 Dec 2020 prepare this manuscript. 7. National Institute for Health and Care Excellence. Guide to the processes of technology appraisal. PMG19. England 2018. 8. National Institute for Health and Care Excellence. Talimogene Conflict of interest Ana-Catarina Pinho-Gomes and John Cairns have laherparepvec for treating unresectable metastatic melanoma. no conficts of interest that are directly relevant to the content of this https://www. nice. org. uk/ guida nce/ ta410/ histo ry . Cited 6 Jul 2021. article. 9. National Institute for Health and Care Excellence. Strimvelis for treating adenosine deaminase defciency–severe combined Availability of data and material All data are provided in the manu- immunodefciency. https://www.nice.org.uk/guidance/hst7/histo script. ry. Cited 6 Jul 2021 10. National Institute for Health and Care Excellence. Tisagenlecleu- Ethics approval Not applicable. cel for treating relapsed or refractory B-cell acute lymphoblastic leukaemia in people aged up to 25 years. https://www. nice.org. Consent Not applicable. uk/guidance/ta554/history. Cited 6 Jul 2021 11. national Institute for Health and Care Excellence. Tisagenlecleucel Author contributions ACPG and JC designed the study. ACPG for treating relapsed or refractory difuse large B-cell lymphoma searched for, extracted and analysed the data and drafted the manu- after 2 or more systemic therapies. https://www. nice. org. uk/ guida script. ACPG and JC contributed to interpretation of the results and to nce/ta567/history. Cited 6 Jul 2021 the drafting, review and approval of the fnal manuscript. ACPG and JC 12. National Institute for Health and Care Excellence. Axicabtagene are the guarantors and take responsibility for the contents of this article. ciloleucel for treating difuse large B-cell lymphoma and primary mediastinal large B-cell lymphoma after 2 or more systemic thera- Open Access pies. https://www. nice. org. uk/ guida nce/ ta559/ histo ry . Cited 6 July This article is licensed under a Creative Commons Attri- 2021 bution-NonCommercial 4.0 International License, which permits any 13. National Institute for Health and Care Excellence. Voretigene non-commercial use, sharing, adaptation, distribution and reproduction neparvovec for treating inherited retinal dystrophies caused by in any medium or format, as long as you give appropriate credit to the RPE65 gene mutations. https://www. nice. org. uk/ guida nce/ hst11/ original author(s) and the source, provide a link to the Creative Com- history. Cited 6 Jul 2021 mons licence, and indicate if changes were made. The images or other 14. National Institute for Health and Care Excellence. Autologous third party material in this article are included in the article's Creative anti-CD19-transduced CD3+ cells for treating relapsed or refrac- Commons licence, unless indicated otherwise in a credit line to the tory mantle cell lymphoma. https://www.nice.org.uk/guidance/ material. If material is not included in the article's Creative Commons ta677/history. Cited 6 Jul 2021. licence and your intended use is not permitted by statutory regula- 15. National Institute for Health and Care Excellence. Betibeglogene tion or exceeds the permitted use, you will need to obtain permission autotemcel for treating transfusion-dependent beta-thalassaemia. directly from the copyright holder. To view a copy of this licence, visit https://www.nice.org.uk/guidance/indevelopment/gid-ta10334. http://creativecommons.org/licenses/by-nc/4.0/. Cited 6 Jul 2021. 16. National Institute for Health and Care Excellence. Onasemnogene abeparvovec for treating spinal muscular atrophy. https://www. nice.org.uk/guidance/hst15/history. Cited 6 Jul 2021. References 17. National Institute for Health and Care Excellence. OTL-200 for treating metachromatic leukodystrophy. https://www.nice.org.uk/guida nce/indevelopment/gid-hst10028/documents. Cited 13 Jul 2021. 1. Regulation (EC) No 1394/2007 of the European Parliament and of 18. National Institute for Health and Care Excellence. Holoclar for the Council of 13 November 2007 on advanced therapy medicinal treating limbal stem cell defciency after eye burns. https://www. products and amending directive 2001/83/EC and regulation (EC) nice.org.uk/guidance/ta467/history. Cited 6 Jul 2021. No 726/2004. The Ofcial Journal of the European Union2007. 19. National Institute for Health and Care Excellence. Darvadstrocel 2. Committee for Advanced Therapies. EMA warns against using for treating complex perianal fstulas in Crohn’s disease. https:// unproven cell-based therapies EMA/CAT/94295/2020. Amster- www.nice.org.uk/guidance/ta556/history. Cited 6 Jul 2021. dam: European Medicines Agency; 2020. 20. National Institute for Health and Care Excellence. Autologous 3. Hettle R, Corbett M, Hinde S, Hodgson R, Jones-Diette J, chondrocyte implantation for treating symptomatic articular carti- Woolacott N, et al. The assessment and appraisal of regenerative lage defects of the knee. https://www. nice. org. uk/ guida nce/ ta477/ medicines and cell therapy products: an exploration of methods history. Cited 6 Jul 2021. for review, economic evaluation and appraisal. Health Technol 21. National Institute for Health and Care Excellence. Autologous Assess. 2017;21(7):1–204. chondrocyte implantation using chondrosphere for treating Evaluation of Advanced Therapy Medicinal Products by the National Institute for Health and Care Excellence (NICE)

symptomatic articular cartilage defects of the knee. https://www. 33. Towse A, Fenwick E. Uncertainty and cures: discontinuation, irre- nice.org.uk/guidance/ta508/history. Cited 6 Jul 2021. versibility, and outcomes-based payments: what is diferent about 22. Withdrawal of the marketing authorisation in the European Union a one-of treatment? Value Health. 2019;22(6):677–83. EMA/303072/2015. London: European Medicines Agency; 2015. 34. Drummond M, Sorenson C. Nasty or nice? A perspective on the p. EMA/303072/2015. use of health technology assessment in the United Kingdom. 23. Macaulay R, Patel P. Applying Nice end of life criteria in the ERA Value Health. 2009;12(Suppl 2):S8-13. of drug approvals based on single-ARM trial data. Value Health. 35. Ravens-Sieberer U, Erhart M, Wille N, Wetzel R, Nickel J, Bull- 2017;20(9):A457. inger M. Generic health-related quality-of-life assessment in 24. National Institute for Health and Care Excellence. Our principles. children and adolescents: methodological considerations. Phar- London. macoeconomics. 2006;24(12):1199–220. 25. Charlton V, Rid A. Innovation as a value in healthcare priority- 36. Weiss D, Rydland HT, Øversveen E, Jensen MR, Solhaug setting: the UK experience. Soc Just Res. 2019;32(2):208–38. S, Krokstad S. Innovative technologies and social inequali- 26. Bryan S, Lee H, Mitton C. ‘Innovation’ in health care coverage ties in health: a scoping review of the literature. PLoS ONE. decisions: all talk and no substance? J Health Serv Res Policy. 2018;13(4):e0195447. 2013;18(1):57–60. 37. Equality Act. London: Parliament of the UK. 2010. 27. Kennedy I. Appraising the value of innovation and other benefts: 38. McNamara S, Holmes J, Stevely AK, Tsuchiya A. How averse a short study for NICE. London: National Institute for Health and are the UK general public to inequalities in health between Care Excellence; 2009. socioeconomic groups? A systematic review. Eur J Health Econ. 28. Pearson SD, Ollendorf DA, Chapman RH. New cost-efectiveness 2020;21(2):275–85. methods to determine value-based prices for potential cures: what 39. Spoors J, Miners A, Cairns J, Palnoch D, Summerfeld A, McEn- are the options? Value Health. 2019;22(6):656–60. tee J, et al. Payer and implementation challenges with advanced 29. O’Mahony JF, Paulden M. NICE’s selective application of dif- therapy medicinal products (ATMPs). BioDrugs. 2021;35(1):1–5. ferential discounting: ambiguous, inconsistent, and unjustifed. 40. Jönsson B, Hampson G, Michaels J, Towse A, von der Schu- Value Health. 2014;17(5):493–6. lenburg JMG, Wong O. Advanced therapy medicinal products 30. The Green Book: central government guidance on appraisal and and health technology assessment principles and practices for evaluation London: HM Treasury. 2020. value-based and sustainable healthcare. Eur J Health Econ. 31. Seoane-Vazquez E, Shukla V, Rodriguez-Monguio R. Innovation 2019;20(3):427–38. and competition in advanced therapy medicinal products. EMBO 41. Gonçalves E. Advanced therapy medicinal products: value judge- Mol Med. 2019;11(3):e9992. ment and ethical evaluation in health technology assessment. Eur 32. Macaulay R, Jamali E. PHP218: nice patient access schemes— J Health Econ. 2020;21(3):311–20. who, what, why, when and how. Value Health. 2015;18:A335–766.