Does Methodological Guidance Produce Consistency? A Review of Methodological Consistency in Breast Cancer Utility Value Measurement in NICE Single Technology Appraisals

Does Methodological Guidance Produce Consistency? A Review of Methodological Consistency in... PharmacoEconomics Open (2018) 2:97–107 https://doi.org/10.1007/s41669-017-0040-5 REVIEW AR TICLE Does Methodological Guidance Produce Consistency? A Review of Methodological Consistency in Breast Cancer Utility Value Measurement in NICE Single Technology Appraisals 1 2 2 • • Micah Rose Stephen Rice Dawn Craig Published online: 5 July 2017 The Author(s) 2017. This article is an open access publication Abstract Since 2004, National Institute for Health and Care following NICE methodological guidance, historically STA Excellence (NICE) methodological guidance for technology guidance in breast cancer has generally not used NICE’s appraisals has emphasised a strong preference for using the preferred methods. Future STAs in breast cancer and reviews validated EuroQol 5-Dimensions (EQ-5D) quality-of-life of older guidance should ensure that utility measurement instrument, measuring patient health status from patients or methods are consistent with the NICE reference case to help carers, and using the general public’s preference-based valu- produce consistent, equitable decision making. ation of different health states when assessing health benefits in economic evaluations. The aim of this study was to review all NICE single technology appraisals (STAs) for breast Key Points for Decision Makers cancer treatments to explore consistency in the use of utility scores in light of NICE methodological guidance. A review of A comprehensive review of all completed National all published breast cancer STAs was undertaken using all Institute for Health and Care Excellence (NICE) publicly available STA documents for each included assess- single technology appraisals (STAs) and their ment. Utility scores were assessed for consistency with NICE- supporting documents identified 12 STAs with utility preferred methods and original data sources. Furthermore, scores derived from 14 published utility sources and academic assessment group work undertaken during the STA 3 redacted unpublished sources. process was examined to evaluate the emphasis of NICE- STA guidance published prior to 2016 lacked preferred quality-of-life measurement methods. Twelve compliance or exhibited poor compliance with NICE breast cancer STAs were identified, and many STAs used preferred methods for measuring quality-of-life (utility evidence that did not follow NICE’s preferred utility score scores), which reduces the consistency of decision measurement methods. Recent STA submissions show com- panies using EQ-5D and mapping. Academic assessment making both within and between disease areas. groups rarely emphasized NICE-preferred methods, and There appears to be a pattern of improved queries about preferred methods were rare. While there compliance, beginning in 2016. When NICE appears to be a trend in recent STA submissions towards periodically updates guidance to reflect new evidence and treatments, it is important that quality- of-life data are concurrently updated to conform to & Micah Rose m.j.rose@southampton.ac.uk NICE’s preferred methods. Southampton Health Technology Assessments Centre, University of Southampton, The University of Southampton 1 Background Science Park, Alpha House, Enterprise Road, Southampton SO16 7NS, UK The role of the National Institute for Health and Care Institute of Health and Society, Newcastle University, Excellence (NICE) is to improve health outcomes using the The Baddiley-Clark Building, Richardson Road, English National Health Service (NHS) and other public Newcastle upon Tyne NE2 4AX, UK 98 M. Rose et al. health and social care services. This is achieved through lead to inconsistent decision making, with potential adverse several research avenues, one of which will be explored in consequences for maximising population health. this study: producing evidence-based guidance and advice for health, public health and social care practitioners through the single technology appraisal (STA) process. 2 Objective STAs assess the clinical and cost effectiveness of health technologies to ensure that all NHS patients have This study assesses the consistency in utility measurement equitable access to the most clinically effective and cost- methods used in NICE STAs. Breast cancer was chosen as effective treatments available. STAs are designed to a case study for this review. Because the incidence of appraise a single health technology for a single indication breast cancer is consistently among the highest of any [1]. cancer in England, with 15.4% of new cancer registrations The STA process consists of several stages designed to in 2015 from breast cancer [6], we deemed that there would create rigorous, transparent decision making with ample be a higher likelihood of attaining a good sample size of stakeholder input into decisions. The primary evidence published STAs. sources are company evidence submissions and work conducted by an independent Evidence Review Group (ERG) that assesses the validity and quality of the company 3 Methods submissions (CSs). The ERG requests and incorporates clarifications from the company, and conducts additional 3.1 Evidence Selection analyses to address uncertainty and areas where the CS is lacking. All published guidance produced by the NICE STA and Cost-utility analyses (CUAs) are the primary type of multiple technology appraisal (MTA) programmes is avail- health economic analysis in NICE technology appraisals, a able on the NICE website [7]. Title searches were conducted type of cost-effectiveness analysis where the primary among these guidance to identify all published breast cancer measure of benefit is health-related quality-of-life STA guidance published between January 2006 and April (HRQoL), measured as quality-adjusted life-years 2017. The primary evidence documents for an STA are the (QALYs). In CUAs, HRQoL is generally measured CS and the ERG report. These documents, as well as the through preference-based utility scores and decrements following additional documents, were obtained and checked (utility values). for utility data: (1) company responses to requests for clar- In 2004, NICE established a reference case defining the ification from the ERG and NICE; (2) errata and addenda; preferred methods for economic submissions; guidance on and (3) documents produced by the NICE committee for measuring utility scores were as follows: utility scores public consultation and guidance production. Only STAs should be measured from patients or carers with valuation with published final funding decisions were included. Where of health states conducted by the general public using a any of the above documents referenced external sources, standardised and validated generic choice-based instrument these publications were obtained and reviewed for utility (time trade-off or standard gamble), with EuroQol 5-Di- data and additional sources. mensions (EQ-5D) as the preferred method of utility measurement [2]. Alternative methods required justifica- 3.2 Data Extraction tion. In 2008, NICE methods guidance recommended considering mapping to EQ-5D from other instruments Data were extracted using a form developed in Microsoft when EQ-5D was unavailable; this transforms other Excel (Microsoft Corporation, Redmond, WA, USA). The methods of measuring quality-of-life into EQ-5D data and extraction form was piloted on two STAs and further removed standard gamble from valuation methods [3]. The refined to ensure the capture of all relevant data. Data were latest guidance (2013) provides a checklist explicitly ask- extracted by one researcher (MJR) and checked by a sec- ing whether EQ-5D was used [4]. ond researcher (SJCR). The logic behind recommending one validated instru- A streamlined process was used to extract utility data. ment is that different methods of measurement produce First, data were extracted from the CS, and, second, the different results [2–4]. A systematic review in breast cancer ERG report was examined. Utility scores and utility utilities by Peasgood et al. found that measurement method decrements were identified in the ERG report, but only substantially influenced utility score, as did whose prefer- extracted where they differed from the CS. NICE appraisal ences were measured and whose values were ascribed to documents other than CSs and ERG reports were checked those preferences [5]. This measurement inconsistency can in the same way but none contained new utility data. A review of methodological consistency in breast cancer utility measurement in NICE STAs 99 Lastly, studies identified by STAs as utility value sources methodological guidance [2, 4]. Across the four EBC were extracted. STAs, there were 39 utility values (utility scores and utility Data on study and population characteristics, such as decrements). Three unpublished sources based on company year of publication, patient age, menopausal status, breast trials and ten published sources were used for utility values cancer stage, receptor status, treatments compared, and [15–24]. Among utility values, 21% (n = 8) were redacted previous treatments, were extracted. In addition, utility or not reported in the submission document, including all data were extracted for each health state and each adverse health-state utility scores in TA 107 (only utility decre- event. Health states were commonly represented by utility ments from Hillner and Smith were reported [16]) and scores, while adverse events and complications were health-state utility scores from TA 109 for remission commonly represented by utility decrements. Collectively, (disease-free survival [DFS]) and first locoregional recur- we define these as utility values. Additional data were rence. Among the redacted or not reported scores, there extracted on measurement methods, additional utility data were six utility scores that did not have the methods references, and mapping techniques. reported and two that were derived by mapping to EQ-5D from European Organisation for Research and Treatment 3.3 Synthesis and Analysis of Cancer (EORTC) Quality-of-Life Questionnaire-Core 30 (QLQ-C30) data in the company primary effectiveness The STAs were first divided into early breast cancer (EBC) trial (the Breast Cancer International Research Group 001 and metastatic breast cancer (MBC). The sources and Trial [BCIRG001]), both redacted [11, 25]. In summary, methods for the utility values were then compared with the there were 33 utility scores that had methods reported in relevant NICE reference case, which was defined by the EBC. Table 1 shows the consistency of unredacted utility year of STA submission. Next, utility scores identified in value sources, with the relevant NICE reference case for STAs were analysed for value consistency with their each utility source used in EBC STAs. The ERG on TA source. Additionally, an analysis of clarification requests 424 listed Essers et al. as the source of alternative utility by the ERG and NICE was undertaken. Feasibility of values in their analyses, however they did not identify the mapping was assessed by checking the availability of four original source papers these utility values were mapping algorithms using the Health Economics Research derived from in the study by Essers et al., i.e. Tengs and Centre (HERC) database of mapping studies [8]. Analyses Wallace, Carter et al., Hayman et al. and Van Hanswijck de included assessment of alternative utility values identified Jonge et al. [20, 21, 23, 24, 26]. Utility scores did not by ERGs and used in their exploratory analyses, where appear to be directly derived from the source papers as these differed from the CS. these papers were not cited by the ERG. We were unable to confirm utilities by Van Hanswijck de Jonge et al. as the poster presentation was not available to us; therefore, these 4 Results values are not included in Table 1 or in the following totals. Values derived from the study by Hayman et al. Twelve breast cancer STAs were identified, four evaluating were identical to those derived from the case report by EBC and eight evaluating MBC. The following sections Carter et al., and were therefore also not included [20, 21]. analyse STA consistency with preferred methods in the All percentages in this section are based on the 33 utility NICE reference case, and then consistency with source data scores with clear methods reporting. Among the transpar- and methods. ently reported utility scores, 36% (n = 12) of utility values were derived from patients and 39% (n = 13) were valued 4.1 Early Breast Cancer Technology Appraisals by the general public. The most consistent application of the reference case in EBC was the use of standard gamble There were four STAs relating to EBC: TA 107, TA 108, or time trade-off methods, with 45% (n = 15) of utility TA 109, and TA 424 [9–12]. EBC STAs and their utility scores using appropriate methods. Only 21% (n = 7) of score sources had broadly similar populations, with some utility scores used EQ-5D values. EBC utilities were often differences in epidermal growth factor receptor, oestrogen inconsistent with NICE reference case preferences. receptor and axillary lymph node involvement status, Two ERG groups evaluated four CSs of EBC treat- which may affect prognosis but was considered unlikely to ments. Of the four EBC STAs, only two included questions affect quality-of-life measurement [13]. No studies in the submitted to the company for clarification as the NICE meta-analysis by Peasgood et al. assessed quality-of-life STA process was still in its infancy in 2006. The clarifi- using the above factors [14]. cation questions submitted to the company in TA 107 did Three EBC STAs were assessed against 2004 method- not contain any questions about the derivation of utility ological guidance, and one was assessed against 2013 scores, and TA 424 primarily used methodologically 100 M. Rose et al. Table 1 Consistency of EBC utility sources with the NICE reference case Utility source, year STAs that use NICE-preferred assumptions (all should be yes) the utility source Utilities measured Valuation by the EQ-5D Choice-based method by patients general public used (not rating scale), TTO after 2008 Hillner and Smith 1991 [16] 107, 109 No No No No Sorensen et al. 2004 [33] 108 No No No Yes Ossa et al. 2004 [18] 109 No Yes No Yes Brown and Hutton 1998 [15] 109 No No No Yes Launois et al. 1996 [17] 109 No No No No b c Lloyd et al. 2006 [19] 424 No Yes No No Lidgren et al. 2007 [22] 424 Yes Yes Yes Yes b,d Tengs and Wallace 2000 [23] 424 No Variable No No b,d Carter et al. 1998 [20] 424 No No No No EBC early breast cancer, EQ-5D EuroQoL-5 dimensions, NICE National Institute for Health and Care Excellence, STAs single technology appraisals, TTO time trade-off Assessed against 2004 NICE reference case Assessed against 2013 NICE reference case No mapping methods used to derive EQ-5D Referenced as Essers et al. [26], and used only in Evidence Review Group analyses, Essers et al. used multiple sources [33] appropriate scores, resulting in no questions about NICE- citedbythe ERG[20, 23, 24]. Tengs and Wallace preferred methods [27, 28]. While each of the ERG groups gathered together a large sample of utility scores derived commented on the methods of deriving utility scores in from a variety of studies using a variety of methods; their reports, no commentary indicated an NICE preference however, none that were cited in the study by Essers for EQ-5D data [29–32]. In fact, as Table 1 shows, one et al. complied with the 2013 NICE guidance [4, 23, 26]. ERG group that examined company utility values derived Carter et al. reported on a case study of a 74-year-old from EQ-5D explicitly chose to test alternative utility woman who used standard gamble elicitation derived values that consistently did not use NICE-preferred meth- from healthcare workers to assess her quality-of-life ods or comply with the 2013 NICE reference case [4, 32]. [20]. The utility value reported by Essers et al. does not Utility scores used in EBC had utility values that were perfectly match the value reported by Carter et al. generally consistent with their cited sources, but sources [20, 26]. Thethird studywas aposter thatwecould not rarely used NICE-preferred methods. Among studies acquire [24]. Mixing these different instruments could assessed against the 2004 NICE reference case, Hillner and provide inconsistent results as it is well-known that Smith used a rating scale [16], Sorensen et al. used a alternative methods of utility measurement produce dif- chained standard gamble method [33, 34], Brown and ferent results [2]. Hutton used the standard gamble method [15], Ossa et al. used a bespoke time trade-off method based on the EQ-5D 4.2 Metastatic Breast Cancer Technology [18], and redacted scores in submissions used undisclosed Appraisals scales or EQ-5D. One STA was assessed against the 2013 NICE reference case, TA 424. Between the CS and the There were eight technology appraisals of MBC: TA 116, ERG report, there were five sources for utility scores TA 214, TA 239, TA 250, TA 263, TA 295, TA 371, and reported, i.e. Lidgren et al., Lloyd et al., Tengs and Wal- TA 423 [12, 35–41]. Some appraisals were for locally lace, Carter et al., and Van Hanswijck de Jonge et al. advanced breast cancer (LABC) or MBC. Similar to how [19, 20, 22–24]. The company used five utility scores from human epidermal growth factor receptor 2 (HER2) and Lidgren et al. and one score from Lloyd et al. All five oestrogen receptor status in EBC were handled, the same scores from Lidgren et al. were derived by mapping utility scores were used for LABC as those used for MBC. EORTC QLQ-C30 to EQ-5D. Lloyd et al. conducted a The populations for the STAs were broadly similar, with standard gamble with regression to derive scores. most technology appraisals using utility scores derived The ERG used alternative utility scores from Essers from one study, i.e. Lloyd et al. [19]. No utility values were et al. [26], which were derived from three studies not redacted in MBC STAs. A review of methodological consistency in breast cancer utility measurement in NICE STAs 101 All STAs were assessed against the NICE reference case were utilised and how the utility scores were generated [5]. relevant to the year of their production. Only TA 116 was As such, no STA conducted on breast cancer has produced assessed against the 2004 reference case [40], and six STAs cost-effectiveness results that can be compared like-for- were assessed against the 2008 reference case: TA 214, like with other disease areas. TA 239, TA 250, TA 263, TA 295 and TA 371 The eight MBC STAs included three ERG groups and [35–39, 46]. While TA 371 was published after the 2013 five companies [35–41, 46]. Only one ERG group explic- NICE reference case had been published, the STA began itly expressed NICE’s preference for EQ-5D data and before the 2013 methods guidance had been published, queried whether mapping algorithms to EQ-5D were con- therefore it has also been assessed against the 2008 refer- sidered [50]. The company reply indicated that no search ence case. One MBC STA was assessed against the 2013 had been conducted but that a current trial was collecting NICE reference case, i.e. TA 423 [41]. In total, 135 utility both EQ-5D and FACT-B data with a goal of producing values were reported. such an algorithm, with data expected in the next Six sources were used among the eight MBC STAs. 18–24 months (that submission was in 2010 [54]). The Evaluating the six sources for consistency with the NICE company has since had three NICE STA guidance pub- reference case is more parsimonious than evaluating each lished using the study by Lloyd et al. for utilities, with the of the 135 utility values. Table 2 presents the six MBC latest submission produced in 2013 [19, 36, 46, 55]. utility sources and assesses them against the NICE refer- In 2008, NICE officially advocated mapping methods to ence case. Five of six sources used regression methods to convert other quality-of-life scales to EQ-5D [3]. Only the derive utility scores [5, 19, 41–43]. Lloyd et al. built a last of the seven breast cancer STAs produced after the regression based on chained standard gamble results from 2008 NICE methods guidance used mapping to produce the general population [19], while Cooper et al. [42] used a EQ-5D values [35–39, 46]. Each of the six STAs that did meta-regression of utility scores for MBC, with most val- not use mapping algorithms used trial evidence with ues derived from three sources, all of which used the effectiveness measures that could be mapped to EQ-5D at standard gamble approach, with oncology nurses and the time of submission to NICE. oncologists providing responses and valuations Four clinical effectiveness studies used in STAs used the [15, 17, 44]. Peasgood et al. conducted a systematic review FACT-B disease-specific quality-of-life instrument. The with separate meta-regressions of utility scores in EBC and E2100 trial was used by TA 214 (submitted May 2010) and MBC, derived using a variety of measurement methods and TA 263 (submitted December 2011) for effectiveness a variety of different groups of individuals valuing health evidence [36, 54, 56], while the CONFIRM trial was used states; the meta-regressions included many covariates, for effectiveness evidence in TA 239 (submitted November including valuation methods [5]. TA 423 used the mapping 2011) [38, 57]. The study referred to as Study 201 was used algorithm developed by Crott and Briggs to map EORTC- in the CS for TA 250 (submitted March 2011) [37, 58], and QLQC30 trial data to EQ-5D utility scores for both health the EMILIA trial was used for effectiveness evidence in states and for utility decrements due to adverse events TA 371 (submitted December 2013) [46, 59, 60]. FACT-B [41, 43, 45]. contains the FACT-G cancer questionnaire plus additional As Table 2 shows, most utility scores used in STAs of breast cancer-specific questions [61]. Cheung et al. and MBC did not use the preferred methods of utility mea- Teckle et al. mapped FACT-G to EQ-5D in a breast cancer surement specified in the relevant NICE methods guidance. population [62, 63]. During TA 214, the company in From 2006 onwards, seven of eight CSs for MBC STAs TA 263 indicated that an in-process trial using FACT-B used Lloyd et al., a source that did not use NICE-preferred and EQ-5D data would produce a mapping algorithm [54], methods [35–41, 46]. All eight ERG reports for MBC and use of individual patient data in this algorithm could STAs used utility values derived from Lloyd et al. allow sophisticated control for adverse events and disease [32, 47–53]. As a consequence, many of the utility values states. Examining the latest version of the HERC database are out of line with NICE-preferred methods used in other of mapping studies identified no published algorithm based STAs in other disease areas, making the utility values in on that trial [8]. breast cancer difficult to compare with those in other The Bolero-2 trial was used for effectiveness evidence areas—the currency of comparison is not universal and in TA 295 (submitted February 2013); Bolero-2 used the there is no means of conversion [19]. EORTC QLQ-C30 and the breast cancer-specific ques- Peasgood et al. found that measurement methodology tionnaire (QLQ-BR23) [2, 35, 64]. EORTC QLQ-C30 was (who measured, who valued, what instrument) was often mapped to EQ-5D by Kind [25], Crott and Briggs (breast highly influential on utility values [5]. While the work of cancer population) [45], and Kim et al. (MBC population) Peasgood et al. was used for sensitivity analysis in TA 214, [65]. Additionally, Kim et al. included the QLQ-BR23 in no explanation was given as to which regression variables mapping [65]. In 2013, mapping was possible at the time of 102 M. Rose et al. Table 2 Comparison of MBC utility methods with NICE reference case Utility source, year STAs utilising NICE preferred assumptions (all should be yes) source Utilities Valuation by EQ-5D Choice-based Mapping measured the used method used? by patients general public (not rating scale); if after 2008, TTO Lloyd et al. 2006 [19] 116 No Yes No Yes Not relevant a b d Lloyd et al. 2006 214 , 239, 250, 263 , 295, 371 , 423 No Yes No No No g,h [19] (ERG) Cooper et al. 2003 214 No No No No No [42] Hutton et al., 1996 250 No No No No No [44] Peasgood et al. 2010 214 Partially Partially Partially Partially No [5] h e TA 423 CS [41] 423 Yes Yes Yes Yes Yes Hudgens et al. 2014 423 Yes Yes Yes Yes Yes [43] MBC metastatic breast cancer, EQ-5D EuroQoL-5 dimensions, TTO time trade-off, CS company submission, ERG Evidence Review Group report, NICE National Institute for Health and Care Excellence, STAs single technology appraisals Used only in sensitivity analysis Cites Fleeman et al. [51], used the Lloyd et al. [19], regression but fixed age at 47 years and weighted by treatment response Cited as Winstanley et al. [87] in the company submission but all utility values are derived from Cooper et al. [42] TA371 began in 2012, before the 2013 NICE Methods Guidance was released [4] Crott and Briggs [45] Assessed against 2004 NICE reference case Assessed against 2008 NICE reference case Assessed against 2013 NICE reference case submission for TA 295. No justification was provided as to that did not use a mapping algorithm generated from that why mapping methods were not used by the company. trial [12, 36, 46, 55, 66]. In addition to nonreference Table 3 provides a checklist for assessing consistency methods being prevalent across breast cancer STAs, clin- with the NICE reference case with regard to utility mea- ical guideline (CG) 81 and an HTA of lapatinib plus tras- surement. The checklist is designed to force technology tuzumab with an aromatase inhibitor also used utility appraisal participants (whether the company or appraisal values that were not valued using EQ-5D or mapped to EQ- group) to fully examine consistency with the NICE refer- 5D [47, 67]. These guidance were produced with direct ence case, help formulate clarification requests and NICE oversight or by an NICE-contracted technology requests for additional analyses to be sent to the company, appraisal group. and encourage ERGs to ensure that they are knowledgeable The findings of this review relate to breast cancer STAs about available mapping algorithms for the disease area conducted by NICE for the English NHS. It is unclear being studied in a technology appraisal. whether the results are generalisable to other disease areas. We have not made, nor can we make, any conclusions about methodological consistency in utility score mea- 5 Discussion surement in NICE technology appraisals in general. The review did not formally assess other forms of NICE The results of this review show that NICE methodological guidance, such as MTAs, diagnostic guidance (DG), or CG. guidance has not led to consistent application of NICE- For completeness, we have conducted rapid informal preferred methods of utility measurement in breast cancer reviews of the utility scores used in other breast cancer STAs. While mapping algorithms were often available, guidance produced by and for NICE. We identified two they were rarely assessed, recommended, or used. In one NICE CGs related to breast cancer: CG 80 and CG 81 instance, a company stated that their trial data would allow [67, 68]. CG 80 did not use QALYs or utility scores [68], mapping [54], and subsequently submitted five submissions whereas CG81 used utility scores that were derived from A review of methodological consistency in breast cancer utility measurement in NICE STAs 103 Table 3 Reference case assessment checklist for technology appraisal submissions Was EQ-5D used to value health? Yes No Were utility scores derived from the relevant patient population? Were utility scores derived from the relevant patient population? Were values for these utility scores derived from the general Were values for these utility scores derived from the general population? population? Was EQ-5D measured directly, or mapped from another Was the preferred time trade-off utility score measurement method used? instrument? Were any mapping algorithms available and not used? Was sufficient justification provided for the choice of utility score measurement? EQ-5D EuroQoL-5 dimensions Cooper and colleagues and were thus inconsistent with In the study by Lloyd et al., EQ-5D was collected at NICE-preferred methods, as stated earlier [42, 67]. Four baseline in a nonclinical valuation population but not in a additional technology appraisals were identified through clinical study for the disease population [19]. One ERG the NICE website (TA 34) and through CG 81, which stated that Lloyd et al. might have been the best available replaced TA 30, TA 54, and TA 62, all of which were not utility data; however, this ERG did not consider whether a available from the NICE website, and, although TA 34 was mapping algorithm could have been used to produce EQ- available, it did not use QALYs or utility scores [69]. 5D utilities. Using utility scores derived by mapping as the CG 81 did not use methodologically appropriate HRQoL base case, and the values used in the study by Lloyd et al. data [67]. Two DGs were identified: DG 8 and DG 10. utilities as a sensitivity analysis would have been in line DG 8 used health-state utility scores that were inconsistent with NICE guidance, would allow transparent discussion of with preferred methods, but did include utility decrements any limitations of mapping algorithms, and would allow that were consistent with NICE methodology; the majority decision makers to decide whether they prefer consistency of utility values used in DG 8 were inconsistent with NICE within breast cancer STAs or consistency with method- methods [70]. The utility values in DG 10 used NICE- ological guidance [3, 19]. preferred methods [71]. A limited amount of data were There are many mapping studies but support for map- available for TAs produced prior to 2002, but it is clear ping methods is not universal. Doble and Lorgelly assessed from other published guidance in breast cancer that the external validity of mapping algorithms from EORTC inconsistency with methodological guidance has histori- QLQ-C30 to EQ-5D and found that existing algorithms cally been a problem in these evidence streams, in a similar generally perform poorly, with seven of ten algorithms fashion to STAs. having statistically significantly different observed and The findings of this study are similar to previous work predicted QALYs [73]. The best performing algorithm was evaluating utility value measurement in STAs. Tosh et al. the most complicated and computationally intensive algo- [72] found that only 56% of values used in technology rithm, which limits its application [73, 74]. Pennington and appraisals were compliant with the 2004 NICE reference Davis found that choice of mapping algorithm could sig- case [2], and only 49% used EQ-5D [72]. Technology nificantly affect cost-effectiveness results [75]. Introducing appraisal participants have shown a lack of awareness of mapping algorithms may add complexity to the technology the full specification of NICE’s methodological prefer- appraisal process without producing better estimates of ences for utility measurement, and have not emphasised EQ-5D valuations. The impending transition from EQ-5D- mapping to EQ-5D. This can be remedied in future tech- 3L to EQ-5D-5L now that a UK scoring algorithm for EQ- nology appraisals by working from the full reference case 5D-5L is available adds further complications to the debate guidance and using the framework laid out in Table 3 for over the appropriateness of mapping as the greater sensi- assessing utilities. tivity of EQ-5D-5L may make it more amenable to map- The study by Lloyd et al. was used most across the 12 ping [76, 77]. published STAs, and was used in some form by all eight A search for STAs currently in development revealed MBC STAs and one EBC STA [19]. Their study was eight STAs [78]; four were suspended, withdrawn, or dis- conducted by United BioSource Corporation, a consultancy continued before any evidence submissions were received company, and Eli Lilly, a pharmaceutical company and one from the company. In addition, one STA was suspended of the companies that made a submission to NICE [19, 40]. after evidence submission by the company and analysis by 104 M. Rose et al. the ERG; this STA used utility scores from the study by and analyses from the company clarification response into Lloyd et al., with no objection from the ERG and no ERG analysis, check an executable model, and produce a request for additional analyses using NICE-preferred report usually spanning more than 100 pages [86]. It is methods [79]. NICE currently has four guidance actively in important for ERGs to focus on the factors most likely to development for MBC [78], two of which do not have affect cost-effectiveness conclusions, and survival data publicly available submissions from the manufacturers frequently have the largest impact on cancer model [80, 81]. One STA began in 2013, and, in this STA, the CS outcomes. and ERG report used the study by Lloyd et al. for utility However, ERGs have frequently failed to comment on data [19, 46, 82]. The most recent STA (CS provided in the consistency of utility scores in breast cancer STAs with September 2016) to have CSs, clarification questions, and NICE-preferred methods. In fact, in TA 423, the ERG an ERG report conforms well to NICE-preferred methods advised using noncompliant data from the study by Lloyd [83, 84]. et al. [19], and, in TA 424, the ERG used utility scores Why EQ-5D values were so infrequently used in CSs is from sources that did not comply with the NICE reference an unanswered question. An in-progress systematic review, case for their alternative utility value analyses [32, 48]. We currently published only as a conference abstract, identified believe that failure to use NICE preferences may be pre- 17 studies that provided EQ-5D utility scores for breast vented through the use of Table 3. cancer [85]. Only three published breast cancer STAs used In the current methodological guidance (2013), the EQ-5D values. Two EBC STAs used EQ-5D—one used section on valuing health effects for the reference cases has mapping (TA 109, scores redacted) and one used directly 12 subsections, with the use of EQ-5D being reiterated in measured EQ-5D scores (TA 424). The clinical evidence nine of these [4]. The emphasis in the guidance is clearly sources used in the companies’ submissions for MBC all on consistency across different disease areas and between had data that could be mapped to EQ-5D at the time of NICE guidance. Unfortunately, the written emphasis has submission. Only one MBC STA used mapping from not, until very recently, been followed by consistent EORTC QLQ-C30 to derive appropriate EQ-5D utility actions. scores and received criticism for using these appropriate In breast cancer STAs, NICE and ERG groups have methods [41, 48]. We believe there are three potential often not requested analyses, and have frequently not explanations for why alternative utility values have not conducted supplemental analyses, using EQ-5D utilities. been used: (1) alternative values have been identified and The quantity of EQ-5D mapping algorithms for breast are unfavourable to companies; (2) there is a lack of belief cancer populations has increased, as evidenced in the in mapping methodology; (3) and/or the values from the HERC database of mapping studies [8]. This database study by Lloyd et al. are viewed as accepted and further facilitates the use of mapping methods to transform clinical work viewed as unnecessary or inappropriate in spite of the effectiveness data into EQ-5D utility scores in current and dataset’s inconsistency with NICE-preferred methods. future breast cancer STAs. NICE needs to clarify whether None of these potential explanations helps foster informed the use of mapping methodologies is their preferred second consistent decision making across disease areas. option, and, if it is, they should cease to accept utility The last two published MBC guidance (TA 423 and values derived from noncompliant measurement methods TA 424) both used EQ-5D utility scores, with one STA in breast cancer STAs (and more broadly), and ensure that using mapping to derive these values [12, 41]. The most their preference for EQ-5D utility scores is adhered to, recent breast cancer STA guidance in development also otherwise NICE’s repeated emphasis of their preference for used EQ-5D [83]. It appears probable that NICE, ERGs and EQ-5D has little meaning and they increase the risk of companies are improving compliance to NICE’s methods; inconsistent decision making. however, STA guidance issued prior to this improvement in compliance are still binding and still have the potential to cause inequitable decisions. The NICE should ensure 6 Conclusion that when these guidance are updated, the utility methods in these guidance are also made conformant with NICE- Historically, breast cancer STAs have shown a broad lack preferred methods. of compliance with preferred methods for measuring utility While some ERGs have criticised utility score usage in scores in the NICE reference case. However, the last two STAs, the primary focus in breast cancer STAs has been on guidance published in December 2016 adhere to the NICE survival rather than utilities, which is not inappropriate. reference case, and the latest STA guidance in develop- Time is limited in STAs. Only 8 weeks are available to ment also conforms to NICE-preferred methods evaluate a multi-hundred page document, formulate clari- [12, 41, 83]. It will be the continued responsibility of the fication questions to the company, incorporate new data ERG groups and NICE to ensure that future guidance and A review of methodological consistency in breast cancer utility measurement in NICE STAs 105 standardised incidence rates for all cancer sites by age and sex: reviews of guidance maintain this forming trend. It would Office for National Statistics; 2017. be inappropriate to let noncompliant utilities used in past 7. National Institute for Health and Care Excellence (NICE). NICE guidance remain the preferred utilities in future guidance. published guidance and advice list. 2017. Available at: https:// To fail to update to methodologically appropriate utility www.nice.org.uk/guidance/published?type=ta. Accessed 3 Apr scores when updating guidance would be to grant these 8. Dakin H. Review of studies mapping from quality of life or guidance using methods that are inconsistent with the clinical measures to EQ-5D: an online database. Health Qual Life NICE reference case a durable precedent, which is a Outcomes. 2013;11:151. doi:10.1186/1477-7525-11-151. decision that should not stand. 9. Bristol-Myers Squibb Pharmaceuticals Ltd. Taxol (paclitaxel) for the adjuvant treatment of early breast cancer: submission to the National Institute for Health and Clinical Excellence (CS TA108); 2006. Author Contributions MR was responsible for revision of the 10. Roche. Herceptin (trastuzumab): NICE submission. Achieving research proposal, drafting the protocol, collaboratively designing the clinical excellence in the adjuvant treatment of her-2 positive data extraction form, piloting the extraction form, performing data breast cancer (CS TA107). 2006. extraction and checking, synthesising findings, and writing all drafts 11. Sanofi-Aventis. Docetaxel for the adjuvant treatment of early of this article. SR contributed to the production of the protocol, col- node-positive breast cancer. Single Technology Appraisal. Sub- laboratively designing the data extraction form, piloting the extraction mission to the National Institute for Health and Clinical Excel- form, performing data extraction and checking, providing feedback, lence (CS TA109). 2006. and editing all drafts of this article. DC identified the topic, wrote the 12. Roche. Neoadjuvant Parjeta (pertuzumab) for the treatment of initial research proposal, contributed to the production of the protocol, HER2-positive early breast cancer (CS TA424). 2016. had oversight of the review, and provided feedback and editing of all 13. Cheang MCU, Chia SK, Voduc D, Gao D, Leung S, Snider J, drafts of this article. et al. Ki67 Index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst. Compliance with Ethical Standards 2009;101(10):736–50. 14. Peasgood T, Ward S, Brazier J. A review and meta analysis of This research involved no human or animal participation. health state utility values in breast cancer. Discussion paper. Sheffield: The University of Sheffield; 2010. Funding This study applied for and received no funding. 15. Brown RE, Hutton J. Cost-utility model comparing docetaxel and paclitaxel in advanced breast cancer patients. Anticancer Drugs. Conflict of interest Micah Rose, Stephen Rice and Dawn Craig 1998;9(10):899–907. declare no conflicts of interest. 16. Hillner BE, Smith TJ. Efficacy and cost effectiveness of adjuvant chemotherapy in women with node-negative breast cancer. a Data Availability Statement The data identified and analysed dur- decision-analysis model. N Engl J Med. 1991;324(3):160–8. ing the current study are available from the corresponding author 17. Launois R, Reboul-Marty J, Henry B, Bonneterre J. A cost-utility upon reasonable request. analysis of second-line chemotherapy in metastatic breast cancer. Docetaxel versus paclitaxel versus vinorelbine. PharmacoEco- Open Access This article is distributed under the terms of the nomics. 1996;10(5):504–21. Creative Commons Attribution-NonCommercial 4.0 International 18. Ossa DF, Briggs A, Cowell W, Littlewood T, Sculpher M. Utility License (http://creativecommons.org/licenses/by-nc/4.0/), which per- associated with severity of cancer-related anaemia (CRA): a mits any noncommercial use, distribution, and reproduction in any societal valuation. Value Health. 2004;7(6):680. medium, provided you give appropriate credit to the original 19. Lloyd A, Nafees B, Narewska J, Dewilde S, Watkins J. Health author(s) and the source, provide a link to the Creative Commons state utilities for metastatic breast cancer. Br J Cancer. license, and indicate if changes were made. 2006;95(6):683–90. 20. Carter K, Ritchey N, Castro F, Caccamo L, Kessler E, Erickson B, et al. Treatment of early-stage breast cancer in the elderly: a health-outcome-based approach. Med Decis Mak. References 1998;18:213–9. 21. Hayman J, Hillner B, Harris J, Weeks J. Cost-effectiveness of 1. National Institute for Health and Care Excellence (NICE). Guide routine radiation therapy following conservative surgery for to the single technology appraisal process 2009. London: NICE; early-stage breast cancer. J Clin Oncol. 1998;16(3):1022–9. 22. Lidgren M, Wilking N, Jonsson B, Rehnberg C. Health related 2. National Institute for Health and Care Excellence (NICE). Guide quality of life in different states of breast cancer. Qual Life Res. to the methods of technology appraisal 2004. London: NICE; 2007;16(6):1073–81. 23. Tengs T, Wallace A. One thousand health-related quality-of-life 3. National Institute for Health and Care Excellence (NICE). Guide estimates. Med Care. 2000;38(6):583–637. to the methods of technology appraisal 2008. London: NICE; 24. Van Hanswijck de Jonge P, Doyle S, Farina C, Walker M. Poster 2008. presentation 190 P: elicitation of UK Health utilities in primary, 4. National Institute for Health and Care Excellence (NICE). Guide to recurrent and metastatic breast cancer. In: 31st European Society the methods of technology appraisal 2013. London: NICE; 2013. Medical Oncology (ESMO), 29 Sep–3 Oct 2006, Istanbul. 5. Peasgood T, Ward SE, Brazier J. Health-state utility values in 25. Kind P. Measuring the value of quality of life in cancer: an index breast cancer. Expert Rev Pharmacoecon Outcomes Res. based on EORTC QLQC-30. J Clin Oncol. 2005;23(16):531S-S. 2010;10(5):553–66. 26. Essers B, Seferina S, Tjan-Heijnen V, Severens J, Nova ´kA, 6. Office for National Statistics. Cancer registration statistics, Pompen M, et al. Transferability of model-based economic England: first release, 2015 Cancer diagnoses and age- evaluations: the case of trastuzumab for the adjuvant treatment of 106 M. Rose et al. HER2-positive early breast cancer in the Netherlands. Value 44. Hutton J, Brown R, Borowitz M, Abrams K, Rothman M, Health. 2010;13(4):375–80. Shakespeare A. A new decision model for cost-utility compar- 27. National Institute for Health and Care Excellence (NICE). Tras- isons of chemotherapy in recurrent metastatic breast cancer. tuzumab for early breast cancer clarification questions (PFC1, PharmacoEconomics. 1996;9(Suppl 2):8–22. TA107). 2006. 45. Crott R, Briggs A. Mapping the QLQ-C30 quality of life cancer 28. National Institute for Health and Care Excellence (NICE). Tras- questionnaire to EQ-5D patient preferences. Eur J Health Econ. tuzumab for early breast cancer clarification questions (PFC2, 2010;11(4):427–34. TA107). 2006. 46. Roche. Trastuzumab emtansine for treating HER2-positive, 29. Chilcott J, Lloyd Jones M, Wilkinson A. Docetaxel for the locally advanced or metastatic breast cancer after treatment with adjuvant treatment of early node-positive breast cancer: a single trastuzumab and a taxane (CS TA371): Roche; 2013. technology appraisal (ERG TA109). Sheffield: The School of 47. Fleeman N, Bagust A, Boland A, Dickson R, Dundar Y, Moonan Health and Related Research (ScHARR); 2006. M, et al. Lapatinib and trastuzumab in combination with an 30. Ward S, Pilgrim H, Hind D. Trastuzumab for the treatment of aromatase inhibitor for the first-line treatment of metastatic hor- primary breast cancer in HER2 positive women: a single tech- mone receptor-positive breast cancer which over-expresses nology appraisal (ERG TA107). Sheffield: University of Shef- human epidermal growth factor 2 (HER2): a systematic review field, School of Health and Related Research (ScHARR); 2006. and economic analysis. Health Technol Assess. 31. Centre for Health Economics (CHE), Regional Drug and Thera- 2011;15(42):1–93, iii–iv. peutics Centre. The use of paclitaxel in the management of early 48. Fleeman N, Bagust A, Richardson M, Houten R, Krishan A, stage breast cancer (ERG TA108). 2006. Beale S, et al. Eribulin for treating locally advanced or metastatic 32. Squires H, Pandor A, Thokala P, Stevens J, Kaltenthaler E, breast cancer after chemotherapy [ID964] (ERG Report TA 423): Clowes M. Pertuzumab for the neoadjuvant treatment of HER2- Liverpool Reviews and Implementation Group; 2016. positive breast cancer (ERG TA424). 2016. 49. Fleeman N, Beale S, Davis H, Dickson R, Loncaster J, O’Reilly 33. Sorensen SV, Brown R, Benedict A, Flood E, Revicki D. Patient- S, et al. Bevacizumab in combination with capecitabine for the rated utilities in postmenopausal early breast cancer (EBC): a first-line treatment of metastatic breast cancer: a Single Tech- cross-country comparison. Value Health. 2004;7(6):641–2. nology Appraisal (ERG TA263). Liverpool: Liverpool Reviews 34. Locker GY, Mansel R, Cella D, Dobrez D, Sorensen S, Gandhi & Implementation Group (LRiG), University of Liverpool; 2011. SK, et al. Cost-effectiveness analysis of anastrozole versus 50. Rodgers M, Soares M, Epstein D, Yang H, Fox D, Eastwood A. tamoxifen as primary adjuvant therapy for postmenopausal Bevacizumab in combination with a taxane for the first-line women with early breast cancer: a US healthcare system per- treatment of HER2-negative metastatic breast cancer (ERG spective. The 5-year completed treatment analysis of the ATAC TA214). York: Center for Reviews and Dissemination (CRD), (‘Arimidex’, Tamoxifen Alone or in Combination) trial. Breast CHE Technology Assessment Group; 2010. Cancer Res Treat. 2007;106(2):229–38. 51. Fleeman N, Bagust A, Beale S, Blundell M, Dwan K, Pilkington 35. Novartis. Single technology appraisal (STA): Everolimus (Afin- G, et al. Everolimus in combination with an aromatase inhibitor itor) in combination with exemestane for the treatment of for the treatment of breast cancer after prior endocrine therapy: a advanced or metastatic HER2 negative, hormone receptor posi- Single Technology Appraisal. LRiG, The University of Liverpool tive breast cancer after prior endocrine therapy (CS TA295): (ERG TA295). Liverpool: Liverpool Reviews and Implementa- Novartis; 2012. tion Group (LRiG), The University of Liverpool; 2013. 36. Roche. Single technology appraisal (STA): Bevacizumab in 52. Bagust A, Blundell M, Boland A, Dickson R, Dundar Y, Fleeman combination with capecitabine for the treatment of metastatic N, et al. Fulvestrant for the treatment of locally advanced or breast cancer (CS TA263): Roche; 2011. metastatic breast cancer (ERG TA239). Liverpool: Liverpool 37. Eisai Ltd. Single technology appraisal (STA): Eribulin for the Reviews and Implementation Group (LRiG); 2011. treatment of locally advanced breast cancer or metastatic breast 53. Jones J, Takeda A, Tan SC, Cooper K, Loveman ECA, Murray N. cancer (CS TA250). 2011. Evidence Review Group Report commissioned by the NHS R&D 38. AstraZeneca UK Ltd. Single Technology Appraisal (STA): Ful- HTA Programme on behalf of NICE: gemcitabine for metastatic vestrant 500 mg for the treatment of postmenopausal women with breast cancer (ERG TA116). Southampton: Southampton Health oestrogen receptor positive, locally advanced or metastatic breast Technology Assessments Centre; 2006. cancer (CS TA239). 2010. 54. Roche. Bevacizumab in combination with a taxane for the first- 39. Roche. Bevacizumab in combination with taxanes for the treat- line treatment of metastatic breast cancer: manufacturer response ment of HER2-negative 1st line metastatic breast cancer. Roche: to points for clarification (PFC TA214). London. 2010. submission to the National Institute for Health and Clinical 55. Roche. Single technology appraisal—bevacizumab in combina- Excellence (CS TA214). 2010. tion with capecitabine for the first-line treatment of metastatic 40. Eli Lilly and Company Ltd. Gemcitabine for the treatment of breast cancer: points for clarification (PFC TA263). Manchester: metastatic breast cancer. Single technology appraisal (STA). Roche; 2012. Submission to the National Institute for Health and Clinical 56. Eastern Cooperative Oncology Group. Paclitaxel with or without Excellence (CS TA116). 2006. bevacizumab in treating patients with locally recurrent or meta- 41. Eisai Ltd. Eribulin for treating locally advanced or metastatic static breast cancer. NCT00028990. ClinicalTrials.gov. Bethesda brest cancer after chemotherapy (CS TA 423). 2016. (MD): National Library of Medicine (US); 2002. TM 42. Cooper NJ, Abrams KR, Sutton AJ, Turner D, Lambert PC. 57. AstraZeneca. Comparison of fulvestrant (FASLODEX A Bayesian approach to Markov modelling in cost-effectiveness 250 mg and 500 mg in postmenopausal women with oestrogen analyses: application to taxane use in advanced breast cancer. J R receptor positive advanced breast cancer progressing or relapsing Stat Soc Ser A Stat Soc. 2003;166:389–405. after previous endocrine therapy (CONFIRM). NCT00099437. 43. Hudgens S, Briggs A, Velikova G, Forsythe A, McCutcheon S, ClinicalTrials.gov. Bethesda (MD): National Library of Medicine Kind P. Impact of treatment with eribulin or capecitabine for (US); 2004. metastatic breast cancer (MBC) on EQ-5D utility derived from 58. Eisai, Inc. A study of E7389 in advanced/metastatic breast cancer EORTC QLQ-C30. In: European Society of Medical Oncology; patients. NCT00097721. ClinicalTrials.gov. Bethesda (MD): 2014. National Library of Medicine (US); 2004. A review of methodological consistency in breast cancer utility measurement in NICE STAs 107 59. Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, 74. Longworth L, Yang Y, Young T, Mulhern B, Hernandez Alava et al. Trastuzumab emtansine for HER2-positive advanced breast M, Mukuria C, et al. Use of generic and condition-specific cancer. N Engl J Med. 2012;367(19):1783–91. measures of health-related quality of life in NICE decision- 60. Welslau M, Dieras V, Sohn JH, Hurvitz SA, Lalla D, Fang L, making: a systematic review, statistical modelling and survey. et al. Patient-reported outcomes from EMILIA, a randomized Health Technol Assess. 2014;18(9):1–224. phase 3 study of trastuzumab emtansine (T-DM1) versus cape- 75. Pennington B, Davis S. Mapping from the Health Assessment citabine and lapatinib in human epidermal growth factor receptor Questionnaire to the EQ-5D: the impact of different algorithms 2-positive locally advanced or metastatic breast cancer. Cancer. on cost-effectiveness results. Value Health. 2014;17(8):762–71. 2014;120(5):642–51. 76. Devlin N, Shah K, Feng Y, Mulhern B, van Hout B. Valuing 61. FACIT.org. FACT questionnaires. Available at: http://www.facit. health-related quality of life: an EQ-5D-5L value set for England org/FACITOrg/Questionnaires. Accessed 8 May 2014. (OHE research paper 16/01). Office of Health Economics; 2016. 62. Cheung Y-B, Thumboo J, Gao F, Ng G-Y, Pang G, Koo W-H, 77. Feng Y, Devlin N, Shah K, Mulhern B, van Hout B. New et al. Mapping the English and Chinese versions of the Functional methods for modelling EQ-5D-5L value sets: an application to Assessment of Cancer Therapy-General to the EQ-5D Utility English data (OHE research paper 16/02). Office of Health Index. Value Health. 2009;12(2):371–6. Economics; 2016. 63. Teckle P, McTaggart-Cowan H, Van der Hoek K, Chia S, 78. National Institute for Health and Care Excellence (NICE). NICE Melosky B, Gelmon K, et al. Mapping the FACT-G cancer- In Development Technology Appraisal Guidance List. 2017. specific quality of life instrument to the EQ-5D and SF-6D. Available at: http://www.nice.org.uk/guidance/ Health Qual Life Outcomes. 2013;11:203. indevelopment?type=TA. Accessed 25 Apr 2017. 64. Novartis Pharmaceuticals. Everolimus in combination with 79. Jones J, Takeda A, Picot J, von Keyserlingk C, Clegg A. Evi- exemestane in the treatment of postmenopausal women with dence Review Group Report commissioned by the NHS R&D estrogen receptor positive locally advanced or metastatic breast HTA Programme on behalf of NICE: lapatinib for HER2 over- cancer who are refractory to letrozole or anastrozole expressing breast cancer. Southampton: Southampton Health (BOLERO2). NCT00863655. ClinicalTrials.gov. Bethesda (MD): Technology Assessments Centre; 2007. National Library of Medicine (US); 2009. 80. National Institute for Health and Care Excellence (NICE). Breast 65. Kim EJ, Ko SK, Kang HY. Mapping the cancer-specific EORTC cancer (brain metastases)—etirinotecan pegol [ID881]. 2017. QLQ-C30 and EORTC QLQ-BR23 to the generic EQ-5D in Available at: https://www.nice.org.uk/guidance/indevelopment/ metastatic breast cancer patients. Qual Life Res. gid-ta10066. Accessed 25 Apr 2017. 2012;21(7):1193–203. 81. National Institute for Health and Care Excellence (NICE). Ful- 66. Roche. Pertuzumab in combination with trastuzumab and doc- vestrant for untreated hormone-receptor positive metastatic breast etaxel for the treatment of HER2-positive metastatic breast can- cancer [ID951]. 2017 Available at: https://www.nice.org.uk/ cer. STA submission: Roche; 2013. guidance/indevelopment/gid-ta10106. Accessed 25 Apr 2017. 67. National Collaborating Centre for Cancer. Advanced breast 82. Fleeman N, Bagust A, Beale S, Dwan K, Dickson R, Proudlove cancer: diagnosis and treatment (CG 81). 2009. C, et al. Pertuzumab in combination with trastuzumab and doc- 68. National Collaborating Centre for Cancer. Early and locally etaxel for the treatment of HER2 positive metastatic or locally advanced breast cancer: diagnosis and treatment (CG 80): recurrent unresectable breast cancer: a Single Technology National Institute for Health and Care Excellence; 2009. Appraisal. Liverpool: Liverpool Reviews and Implementation 69. Lewis R, Bagnall A-M, Forbes C, Shirran E, Duffy S, Kleijnen J, Group (LRiG), The University of Liverpool; 2013. et al. A rapid and systematic review of the clinical effectiveness 83. Pfizer. Palbociclib in combination with an aromatase inhibitor for and cost-effectiveness of trastuzumab for breast cancer: National previously untreated metastatic, hormone receptor-positive, Institute for Health and Care Excellence; 2001. HER2-negative breast cancer [ID915 CS]. 2017. 70. Huxley N, Jones-Hughes T, Coelho H, Snowsill T, Cooper C, 84. Fleeman N, Stainthorpe A, Bagust A, Richardson M, Nolan S, Cooper K, et al. A systematic review and economic evaluation of Houten R, et al. Palbociclib in combination with an aromatase intraoperative tests [RD-100i one-step nucleic acid amplification inhibitor for previously untreated metastatic, hormone receptor- (OSNA) system and Metasin test] for detecting sentinel lymph positive, HER2-negative breast cancer [ID915 ERG Report]. node metastases in breast cancer. Health Technol Assess. 2017. 2012;19(2):1–215. 85. Hughes R, Mitchell CR, Bishop RS, Fotheringham I. Health-state 71. Ward S, Scope A, Rafia R, Pandor A, Harnan S, Evans P. Gene utility values in breast and prostate cancer measured using the expression profiling and expanded immunohistochemistry tests to EQ-5D: a systematic review of the literature. Value Health. guide the use of adjuvant chemotherapy in breast cancer man- 2014;17(7):A645–6. agement. Health Technol Assess. 2011;17(44):1–302. 86. National Institute for Health and Care Excellence (NICE). Guide 72. Tosh JC, Longworth LJ, George E. Utility values in National to the processes of technology appraisal 2014. London: NICE; Institute for Health and Clinical Excellence (NICE) Technology 2014. Appraisals. Value Health. 2011;14(1):102–9. 87. Winstanley J, Murray N. National Institute for Health and Care 73. Doble B, Lorgelly P. Mapping the EORTC QLQ-C30 onto the Excellence (NICE). CG81: Advanced breast cancer: diagnosis EQ-5D-3L: assessing the external validity of existing mapping and treatment. Cardiff: National Collaborating Centre for Cancer. algorithms. Qual Life Res. 2016;25(4):891–911. 2009. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png PharmacoEconomics - Open Springer Journals

Does Methodological Guidance Produce Consistency? A Review of Methodological Consistency in Breast Cancer Utility Value Measurement in NICE Single Technology Appraisals

Free
11 pages
Loading next page...
 
/lp/springer_journal/does-methodological-guidance-produce-consistency-a-review-of-kG1y420HQz
Publisher
Springer International Publishing
Copyright
Copyright © 2017 by The Author(s)
Subject
Medicine & Public Health; Pharmacoeconomics and Health Outcomes
ISSN
2509-4262
eISSN
2509-4254
D.O.I.
10.1007/s41669-017-0040-5
Publisher site
See Article on Publisher Site

Abstract

PharmacoEconomics Open (2018) 2:97–107 https://doi.org/10.1007/s41669-017-0040-5 REVIEW AR TICLE Does Methodological Guidance Produce Consistency? A Review of Methodological Consistency in Breast Cancer Utility Value Measurement in NICE Single Technology Appraisals 1 2 2 • • Micah Rose Stephen Rice Dawn Craig Published online: 5 July 2017 The Author(s) 2017. This article is an open access publication Abstract Since 2004, National Institute for Health and Care following NICE methodological guidance, historically STA Excellence (NICE) methodological guidance for technology guidance in breast cancer has generally not used NICE’s appraisals has emphasised a strong preference for using the preferred methods. Future STAs in breast cancer and reviews validated EuroQol 5-Dimensions (EQ-5D) quality-of-life of older guidance should ensure that utility measurement instrument, measuring patient health status from patients or methods are consistent with the NICE reference case to help carers, and using the general public’s preference-based valu- produce consistent, equitable decision making. ation of different health states when assessing health benefits in economic evaluations. The aim of this study was to review all NICE single technology appraisals (STAs) for breast Key Points for Decision Makers cancer treatments to explore consistency in the use of utility scores in light of NICE methodological guidance. A review of A comprehensive review of all completed National all published breast cancer STAs was undertaken using all Institute for Health and Care Excellence (NICE) publicly available STA documents for each included assess- single technology appraisals (STAs) and their ment. Utility scores were assessed for consistency with NICE- supporting documents identified 12 STAs with utility preferred methods and original data sources. Furthermore, scores derived from 14 published utility sources and academic assessment group work undertaken during the STA 3 redacted unpublished sources. process was examined to evaluate the emphasis of NICE- STA guidance published prior to 2016 lacked preferred quality-of-life measurement methods. Twelve compliance or exhibited poor compliance with NICE breast cancer STAs were identified, and many STAs used preferred methods for measuring quality-of-life (utility evidence that did not follow NICE’s preferred utility score scores), which reduces the consistency of decision measurement methods. Recent STA submissions show com- panies using EQ-5D and mapping. Academic assessment making both within and between disease areas. groups rarely emphasized NICE-preferred methods, and There appears to be a pattern of improved queries about preferred methods were rare. While there compliance, beginning in 2016. When NICE appears to be a trend in recent STA submissions towards periodically updates guidance to reflect new evidence and treatments, it is important that quality- of-life data are concurrently updated to conform to & Micah Rose m.j.rose@southampton.ac.uk NICE’s preferred methods. Southampton Health Technology Assessments Centre, University of Southampton, The University of Southampton 1 Background Science Park, Alpha House, Enterprise Road, Southampton SO16 7NS, UK The role of the National Institute for Health and Care Institute of Health and Society, Newcastle University, Excellence (NICE) is to improve health outcomes using the The Baddiley-Clark Building, Richardson Road, English National Health Service (NHS) and other public Newcastle upon Tyne NE2 4AX, UK 98 M. Rose et al. health and social care services. This is achieved through lead to inconsistent decision making, with potential adverse several research avenues, one of which will be explored in consequences for maximising population health. this study: producing evidence-based guidance and advice for health, public health and social care practitioners through the single technology appraisal (STA) process. 2 Objective STAs assess the clinical and cost effectiveness of health technologies to ensure that all NHS patients have This study assesses the consistency in utility measurement equitable access to the most clinically effective and cost- methods used in NICE STAs. Breast cancer was chosen as effective treatments available. STAs are designed to a case study for this review. Because the incidence of appraise a single health technology for a single indication breast cancer is consistently among the highest of any [1]. cancer in England, with 15.4% of new cancer registrations The STA process consists of several stages designed to in 2015 from breast cancer [6], we deemed that there would create rigorous, transparent decision making with ample be a higher likelihood of attaining a good sample size of stakeholder input into decisions. The primary evidence published STAs. sources are company evidence submissions and work conducted by an independent Evidence Review Group (ERG) that assesses the validity and quality of the company 3 Methods submissions (CSs). The ERG requests and incorporates clarifications from the company, and conducts additional 3.1 Evidence Selection analyses to address uncertainty and areas where the CS is lacking. All published guidance produced by the NICE STA and Cost-utility analyses (CUAs) are the primary type of multiple technology appraisal (MTA) programmes is avail- health economic analysis in NICE technology appraisals, a able on the NICE website [7]. Title searches were conducted type of cost-effectiveness analysis where the primary among these guidance to identify all published breast cancer measure of benefit is health-related quality-of-life STA guidance published between January 2006 and April (HRQoL), measured as quality-adjusted life-years 2017. The primary evidence documents for an STA are the (QALYs). In CUAs, HRQoL is generally measured CS and the ERG report. These documents, as well as the through preference-based utility scores and decrements following additional documents, were obtained and checked (utility values). for utility data: (1) company responses to requests for clar- In 2004, NICE established a reference case defining the ification from the ERG and NICE; (2) errata and addenda; preferred methods for economic submissions; guidance on and (3) documents produced by the NICE committee for measuring utility scores were as follows: utility scores public consultation and guidance production. Only STAs should be measured from patients or carers with valuation with published final funding decisions were included. Where of health states conducted by the general public using a any of the above documents referenced external sources, standardised and validated generic choice-based instrument these publications were obtained and reviewed for utility (time trade-off or standard gamble), with EuroQol 5-Di- data and additional sources. mensions (EQ-5D) as the preferred method of utility measurement [2]. Alternative methods required justifica- 3.2 Data Extraction tion. In 2008, NICE methods guidance recommended considering mapping to EQ-5D from other instruments Data were extracted using a form developed in Microsoft when EQ-5D was unavailable; this transforms other Excel (Microsoft Corporation, Redmond, WA, USA). The methods of measuring quality-of-life into EQ-5D data and extraction form was piloted on two STAs and further removed standard gamble from valuation methods [3]. The refined to ensure the capture of all relevant data. Data were latest guidance (2013) provides a checklist explicitly ask- extracted by one researcher (MJR) and checked by a sec- ing whether EQ-5D was used [4]. ond researcher (SJCR). The logic behind recommending one validated instru- A streamlined process was used to extract utility data. ment is that different methods of measurement produce First, data were extracted from the CS, and, second, the different results [2–4]. A systematic review in breast cancer ERG report was examined. Utility scores and utility utilities by Peasgood et al. found that measurement method decrements were identified in the ERG report, but only substantially influenced utility score, as did whose prefer- extracted where they differed from the CS. NICE appraisal ences were measured and whose values were ascribed to documents other than CSs and ERG reports were checked those preferences [5]. This measurement inconsistency can in the same way but none contained new utility data. A review of methodological consistency in breast cancer utility measurement in NICE STAs 99 Lastly, studies identified by STAs as utility value sources methodological guidance [2, 4]. Across the four EBC were extracted. STAs, there were 39 utility values (utility scores and utility Data on study and population characteristics, such as decrements). Three unpublished sources based on company year of publication, patient age, menopausal status, breast trials and ten published sources were used for utility values cancer stage, receptor status, treatments compared, and [15–24]. Among utility values, 21% (n = 8) were redacted previous treatments, were extracted. In addition, utility or not reported in the submission document, including all data were extracted for each health state and each adverse health-state utility scores in TA 107 (only utility decre- event. Health states were commonly represented by utility ments from Hillner and Smith were reported [16]) and scores, while adverse events and complications were health-state utility scores from TA 109 for remission commonly represented by utility decrements. Collectively, (disease-free survival [DFS]) and first locoregional recur- we define these as utility values. Additional data were rence. Among the redacted or not reported scores, there extracted on measurement methods, additional utility data were six utility scores that did not have the methods references, and mapping techniques. reported and two that were derived by mapping to EQ-5D from European Organisation for Research and Treatment 3.3 Synthesis and Analysis of Cancer (EORTC) Quality-of-Life Questionnaire-Core 30 (QLQ-C30) data in the company primary effectiveness The STAs were first divided into early breast cancer (EBC) trial (the Breast Cancer International Research Group 001 and metastatic breast cancer (MBC). The sources and Trial [BCIRG001]), both redacted [11, 25]. In summary, methods for the utility values were then compared with the there were 33 utility scores that had methods reported in relevant NICE reference case, which was defined by the EBC. Table 1 shows the consistency of unredacted utility year of STA submission. Next, utility scores identified in value sources, with the relevant NICE reference case for STAs were analysed for value consistency with their each utility source used in EBC STAs. The ERG on TA source. Additionally, an analysis of clarification requests 424 listed Essers et al. as the source of alternative utility by the ERG and NICE was undertaken. Feasibility of values in their analyses, however they did not identify the mapping was assessed by checking the availability of four original source papers these utility values were mapping algorithms using the Health Economics Research derived from in the study by Essers et al., i.e. Tengs and Centre (HERC) database of mapping studies [8]. Analyses Wallace, Carter et al., Hayman et al. and Van Hanswijck de included assessment of alternative utility values identified Jonge et al. [20, 21, 23, 24, 26]. Utility scores did not by ERGs and used in their exploratory analyses, where appear to be directly derived from the source papers as these differed from the CS. these papers were not cited by the ERG. We were unable to confirm utilities by Van Hanswijck de Jonge et al. as the poster presentation was not available to us; therefore, these 4 Results values are not included in Table 1 or in the following totals. Values derived from the study by Hayman et al. Twelve breast cancer STAs were identified, four evaluating were identical to those derived from the case report by EBC and eight evaluating MBC. The following sections Carter et al., and were therefore also not included [20, 21]. analyse STA consistency with preferred methods in the All percentages in this section are based on the 33 utility NICE reference case, and then consistency with source data scores with clear methods reporting. Among the transpar- and methods. ently reported utility scores, 36% (n = 12) of utility values were derived from patients and 39% (n = 13) were valued 4.1 Early Breast Cancer Technology Appraisals by the general public. The most consistent application of the reference case in EBC was the use of standard gamble There were four STAs relating to EBC: TA 107, TA 108, or time trade-off methods, with 45% (n = 15) of utility TA 109, and TA 424 [9–12]. EBC STAs and their utility scores using appropriate methods. Only 21% (n = 7) of score sources had broadly similar populations, with some utility scores used EQ-5D values. EBC utilities were often differences in epidermal growth factor receptor, oestrogen inconsistent with NICE reference case preferences. receptor and axillary lymph node involvement status, Two ERG groups evaluated four CSs of EBC treat- which may affect prognosis but was considered unlikely to ments. Of the four EBC STAs, only two included questions affect quality-of-life measurement [13]. No studies in the submitted to the company for clarification as the NICE meta-analysis by Peasgood et al. assessed quality-of-life STA process was still in its infancy in 2006. The clarifi- using the above factors [14]. cation questions submitted to the company in TA 107 did Three EBC STAs were assessed against 2004 method- not contain any questions about the derivation of utility ological guidance, and one was assessed against 2013 scores, and TA 424 primarily used methodologically 100 M. Rose et al. Table 1 Consistency of EBC utility sources with the NICE reference case Utility source, year STAs that use NICE-preferred assumptions (all should be yes) the utility source Utilities measured Valuation by the EQ-5D Choice-based method by patients general public used (not rating scale), TTO after 2008 Hillner and Smith 1991 [16] 107, 109 No No No No Sorensen et al. 2004 [33] 108 No No No Yes Ossa et al. 2004 [18] 109 No Yes No Yes Brown and Hutton 1998 [15] 109 No No No Yes Launois et al. 1996 [17] 109 No No No No b c Lloyd et al. 2006 [19] 424 No Yes No No Lidgren et al. 2007 [22] 424 Yes Yes Yes Yes b,d Tengs and Wallace 2000 [23] 424 No Variable No No b,d Carter et al. 1998 [20] 424 No No No No EBC early breast cancer, EQ-5D EuroQoL-5 dimensions, NICE National Institute for Health and Care Excellence, STAs single technology appraisals, TTO time trade-off Assessed against 2004 NICE reference case Assessed against 2013 NICE reference case No mapping methods used to derive EQ-5D Referenced as Essers et al. [26], and used only in Evidence Review Group analyses, Essers et al. used multiple sources [33] appropriate scores, resulting in no questions about NICE- citedbythe ERG[20, 23, 24]. Tengs and Wallace preferred methods [27, 28]. While each of the ERG groups gathered together a large sample of utility scores derived commented on the methods of deriving utility scores in from a variety of studies using a variety of methods; their reports, no commentary indicated an NICE preference however, none that were cited in the study by Essers for EQ-5D data [29–32]. In fact, as Table 1 shows, one et al. complied with the 2013 NICE guidance [4, 23, 26]. ERG group that examined company utility values derived Carter et al. reported on a case study of a 74-year-old from EQ-5D explicitly chose to test alternative utility woman who used standard gamble elicitation derived values that consistently did not use NICE-preferred meth- from healthcare workers to assess her quality-of-life ods or comply with the 2013 NICE reference case [4, 32]. [20]. The utility value reported by Essers et al. does not Utility scores used in EBC had utility values that were perfectly match the value reported by Carter et al. generally consistent with their cited sources, but sources [20, 26]. Thethird studywas aposter thatwecould not rarely used NICE-preferred methods. Among studies acquire [24]. Mixing these different instruments could assessed against the 2004 NICE reference case, Hillner and provide inconsistent results as it is well-known that Smith used a rating scale [16], Sorensen et al. used a alternative methods of utility measurement produce dif- chained standard gamble method [33, 34], Brown and ferent results [2]. Hutton used the standard gamble method [15], Ossa et al. used a bespoke time trade-off method based on the EQ-5D 4.2 Metastatic Breast Cancer Technology [18], and redacted scores in submissions used undisclosed Appraisals scales or EQ-5D. One STA was assessed against the 2013 NICE reference case, TA 424. Between the CS and the There were eight technology appraisals of MBC: TA 116, ERG report, there were five sources for utility scores TA 214, TA 239, TA 250, TA 263, TA 295, TA 371, and reported, i.e. Lidgren et al., Lloyd et al., Tengs and Wal- TA 423 [12, 35–41]. Some appraisals were for locally lace, Carter et al., and Van Hanswijck de Jonge et al. advanced breast cancer (LABC) or MBC. Similar to how [19, 20, 22–24]. The company used five utility scores from human epidermal growth factor receptor 2 (HER2) and Lidgren et al. and one score from Lloyd et al. All five oestrogen receptor status in EBC were handled, the same scores from Lidgren et al. were derived by mapping utility scores were used for LABC as those used for MBC. EORTC QLQ-C30 to EQ-5D. Lloyd et al. conducted a The populations for the STAs were broadly similar, with standard gamble with regression to derive scores. most technology appraisals using utility scores derived The ERG used alternative utility scores from Essers from one study, i.e. Lloyd et al. [19]. No utility values were et al. [26], which were derived from three studies not redacted in MBC STAs. A review of methodological consistency in breast cancer utility measurement in NICE STAs 101 All STAs were assessed against the NICE reference case were utilised and how the utility scores were generated [5]. relevant to the year of their production. Only TA 116 was As such, no STA conducted on breast cancer has produced assessed against the 2004 reference case [40], and six STAs cost-effectiveness results that can be compared like-for- were assessed against the 2008 reference case: TA 214, like with other disease areas. TA 239, TA 250, TA 263, TA 295 and TA 371 The eight MBC STAs included three ERG groups and [35–39, 46]. While TA 371 was published after the 2013 five companies [35–41, 46]. Only one ERG group explic- NICE reference case had been published, the STA began itly expressed NICE’s preference for EQ-5D data and before the 2013 methods guidance had been published, queried whether mapping algorithms to EQ-5D were con- therefore it has also been assessed against the 2008 refer- sidered [50]. The company reply indicated that no search ence case. One MBC STA was assessed against the 2013 had been conducted but that a current trial was collecting NICE reference case, i.e. TA 423 [41]. In total, 135 utility both EQ-5D and FACT-B data with a goal of producing values were reported. such an algorithm, with data expected in the next Six sources were used among the eight MBC STAs. 18–24 months (that submission was in 2010 [54]). The Evaluating the six sources for consistency with the NICE company has since had three NICE STA guidance pub- reference case is more parsimonious than evaluating each lished using the study by Lloyd et al. for utilities, with the of the 135 utility values. Table 2 presents the six MBC latest submission produced in 2013 [19, 36, 46, 55]. utility sources and assesses them against the NICE refer- In 2008, NICE officially advocated mapping methods to ence case. Five of six sources used regression methods to convert other quality-of-life scales to EQ-5D [3]. Only the derive utility scores [5, 19, 41–43]. Lloyd et al. built a last of the seven breast cancer STAs produced after the regression based on chained standard gamble results from 2008 NICE methods guidance used mapping to produce the general population [19], while Cooper et al. [42] used a EQ-5D values [35–39, 46]. Each of the six STAs that did meta-regression of utility scores for MBC, with most val- not use mapping algorithms used trial evidence with ues derived from three sources, all of which used the effectiveness measures that could be mapped to EQ-5D at standard gamble approach, with oncology nurses and the time of submission to NICE. oncologists providing responses and valuations Four clinical effectiveness studies used in STAs used the [15, 17, 44]. Peasgood et al. conducted a systematic review FACT-B disease-specific quality-of-life instrument. The with separate meta-regressions of utility scores in EBC and E2100 trial was used by TA 214 (submitted May 2010) and MBC, derived using a variety of measurement methods and TA 263 (submitted December 2011) for effectiveness a variety of different groups of individuals valuing health evidence [36, 54, 56], while the CONFIRM trial was used states; the meta-regressions included many covariates, for effectiveness evidence in TA 239 (submitted November including valuation methods [5]. TA 423 used the mapping 2011) [38, 57]. The study referred to as Study 201 was used algorithm developed by Crott and Briggs to map EORTC- in the CS for TA 250 (submitted March 2011) [37, 58], and QLQC30 trial data to EQ-5D utility scores for both health the EMILIA trial was used for effectiveness evidence in states and for utility decrements due to adverse events TA 371 (submitted December 2013) [46, 59, 60]. FACT-B [41, 43, 45]. contains the FACT-G cancer questionnaire plus additional As Table 2 shows, most utility scores used in STAs of breast cancer-specific questions [61]. Cheung et al. and MBC did not use the preferred methods of utility mea- Teckle et al. mapped FACT-G to EQ-5D in a breast cancer surement specified in the relevant NICE methods guidance. population [62, 63]. During TA 214, the company in From 2006 onwards, seven of eight CSs for MBC STAs TA 263 indicated that an in-process trial using FACT-B used Lloyd et al., a source that did not use NICE-preferred and EQ-5D data would produce a mapping algorithm [54], methods [35–41, 46]. All eight ERG reports for MBC and use of individual patient data in this algorithm could STAs used utility values derived from Lloyd et al. allow sophisticated control for adverse events and disease [32, 47–53]. As a consequence, many of the utility values states. Examining the latest version of the HERC database are out of line with NICE-preferred methods used in other of mapping studies identified no published algorithm based STAs in other disease areas, making the utility values in on that trial [8]. breast cancer difficult to compare with those in other The Bolero-2 trial was used for effectiveness evidence areas—the currency of comparison is not universal and in TA 295 (submitted February 2013); Bolero-2 used the there is no means of conversion [19]. EORTC QLQ-C30 and the breast cancer-specific ques- Peasgood et al. found that measurement methodology tionnaire (QLQ-BR23) [2, 35, 64]. EORTC QLQ-C30 was (who measured, who valued, what instrument) was often mapped to EQ-5D by Kind [25], Crott and Briggs (breast highly influential on utility values [5]. While the work of cancer population) [45], and Kim et al. (MBC population) Peasgood et al. was used for sensitivity analysis in TA 214, [65]. Additionally, Kim et al. included the QLQ-BR23 in no explanation was given as to which regression variables mapping [65]. In 2013, mapping was possible at the time of 102 M. Rose et al. Table 2 Comparison of MBC utility methods with NICE reference case Utility source, year STAs utilising NICE preferred assumptions (all should be yes) source Utilities Valuation by EQ-5D Choice-based Mapping measured the used method used? by patients general public (not rating scale); if after 2008, TTO Lloyd et al. 2006 [19] 116 No Yes No Yes Not relevant a b d Lloyd et al. 2006 214 , 239, 250, 263 , 295, 371 , 423 No Yes No No No g,h [19] (ERG) Cooper et al. 2003 214 No No No No No [42] Hutton et al., 1996 250 No No No No No [44] Peasgood et al. 2010 214 Partially Partially Partially Partially No [5] h e TA 423 CS [41] 423 Yes Yes Yes Yes Yes Hudgens et al. 2014 423 Yes Yes Yes Yes Yes [43] MBC metastatic breast cancer, EQ-5D EuroQoL-5 dimensions, TTO time trade-off, CS company submission, ERG Evidence Review Group report, NICE National Institute for Health and Care Excellence, STAs single technology appraisals Used only in sensitivity analysis Cites Fleeman et al. [51], used the Lloyd et al. [19], regression but fixed age at 47 years and weighted by treatment response Cited as Winstanley et al. [87] in the company submission but all utility values are derived from Cooper et al. [42] TA371 began in 2012, before the 2013 NICE Methods Guidance was released [4] Crott and Briggs [45] Assessed against 2004 NICE reference case Assessed against 2008 NICE reference case Assessed against 2013 NICE reference case submission for TA 295. No justification was provided as to that did not use a mapping algorithm generated from that why mapping methods were not used by the company. trial [12, 36, 46, 55, 66]. In addition to nonreference Table 3 provides a checklist for assessing consistency methods being prevalent across breast cancer STAs, clin- with the NICE reference case with regard to utility mea- ical guideline (CG) 81 and an HTA of lapatinib plus tras- surement. The checklist is designed to force technology tuzumab with an aromatase inhibitor also used utility appraisal participants (whether the company or appraisal values that were not valued using EQ-5D or mapped to EQ- group) to fully examine consistency with the NICE refer- 5D [47, 67]. These guidance were produced with direct ence case, help formulate clarification requests and NICE oversight or by an NICE-contracted technology requests for additional analyses to be sent to the company, appraisal group. and encourage ERGs to ensure that they are knowledgeable The findings of this review relate to breast cancer STAs about available mapping algorithms for the disease area conducted by NICE for the English NHS. It is unclear being studied in a technology appraisal. whether the results are generalisable to other disease areas. We have not made, nor can we make, any conclusions about methodological consistency in utility score mea- 5 Discussion surement in NICE technology appraisals in general. The review did not formally assess other forms of NICE The results of this review show that NICE methodological guidance, such as MTAs, diagnostic guidance (DG), or CG. guidance has not led to consistent application of NICE- For completeness, we have conducted rapid informal preferred methods of utility measurement in breast cancer reviews of the utility scores used in other breast cancer STAs. While mapping algorithms were often available, guidance produced by and for NICE. We identified two they were rarely assessed, recommended, or used. In one NICE CGs related to breast cancer: CG 80 and CG 81 instance, a company stated that their trial data would allow [67, 68]. CG 80 did not use QALYs or utility scores [68], mapping [54], and subsequently submitted five submissions whereas CG81 used utility scores that were derived from A review of methodological consistency in breast cancer utility measurement in NICE STAs 103 Table 3 Reference case assessment checklist for technology appraisal submissions Was EQ-5D used to value health? Yes No Were utility scores derived from the relevant patient population? Were utility scores derived from the relevant patient population? Were values for these utility scores derived from the general Were values for these utility scores derived from the general population? population? Was EQ-5D measured directly, or mapped from another Was the preferred time trade-off utility score measurement method used? instrument? Were any mapping algorithms available and not used? Was sufficient justification provided for the choice of utility score measurement? EQ-5D EuroQoL-5 dimensions Cooper and colleagues and were thus inconsistent with In the study by Lloyd et al., EQ-5D was collected at NICE-preferred methods, as stated earlier [42, 67]. Four baseline in a nonclinical valuation population but not in a additional technology appraisals were identified through clinical study for the disease population [19]. One ERG the NICE website (TA 34) and through CG 81, which stated that Lloyd et al. might have been the best available replaced TA 30, TA 54, and TA 62, all of which were not utility data; however, this ERG did not consider whether a available from the NICE website, and, although TA 34 was mapping algorithm could have been used to produce EQ- available, it did not use QALYs or utility scores [69]. 5D utilities. Using utility scores derived by mapping as the CG 81 did not use methodologically appropriate HRQoL base case, and the values used in the study by Lloyd et al. data [67]. Two DGs were identified: DG 8 and DG 10. utilities as a sensitivity analysis would have been in line DG 8 used health-state utility scores that were inconsistent with NICE guidance, would allow transparent discussion of with preferred methods, but did include utility decrements any limitations of mapping algorithms, and would allow that were consistent with NICE methodology; the majority decision makers to decide whether they prefer consistency of utility values used in DG 8 were inconsistent with NICE within breast cancer STAs or consistency with method- methods [70]. The utility values in DG 10 used NICE- ological guidance [3, 19]. preferred methods [71]. A limited amount of data were There are many mapping studies but support for map- available for TAs produced prior to 2002, but it is clear ping methods is not universal. Doble and Lorgelly assessed from other published guidance in breast cancer that the external validity of mapping algorithms from EORTC inconsistency with methodological guidance has histori- QLQ-C30 to EQ-5D and found that existing algorithms cally been a problem in these evidence streams, in a similar generally perform poorly, with seven of ten algorithms fashion to STAs. having statistically significantly different observed and The findings of this study are similar to previous work predicted QALYs [73]. The best performing algorithm was evaluating utility value measurement in STAs. Tosh et al. the most complicated and computationally intensive algo- [72] found that only 56% of values used in technology rithm, which limits its application [73, 74]. Pennington and appraisals were compliant with the 2004 NICE reference Davis found that choice of mapping algorithm could sig- case [2], and only 49% used EQ-5D [72]. Technology nificantly affect cost-effectiveness results [75]. Introducing appraisal participants have shown a lack of awareness of mapping algorithms may add complexity to the technology the full specification of NICE’s methodological prefer- appraisal process without producing better estimates of ences for utility measurement, and have not emphasised EQ-5D valuations. The impending transition from EQ-5D- mapping to EQ-5D. This can be remedied in future tech- 3L to EQ-5D-5L now that a UK scoring algorithm for EQ- nology appraisals by working from the full reference case 5D-5L is available adds further complications to the debate guidance and using the framework laid out in Table 3 for over the appropriateness of mapping as the greater sensi- assessing utilities. tivity of EQ-5D-5L may make it more amenable to map- The study by Lloyd et al. was used most across the 12 ping [76, 77]. published STAs, and was used in some form by all eight A search for STAs currently in development revealed MBC STAs and one EBC STA [19]. Their study was eight STAs [78]; four were suspended, withdrawn, or dis- conducted by United BioSource Corporation, a consultancy continued before any evidence submissions were received company, and Eli Lilly, a pharmaceutical company and one from the company. In addition, one STA was suspended of the companies that made a submission to NICE [19, 40]. after evidence submission by the company and analysis by 104 M. Rose et al. the ERG; this STA used utility scores from the study by and analyses from the company clarification response into Lloyd et al., with no objection from the ERG and no ERG analysis, check an executable model, and produce a request for additional analyses using NICE-preferred report usually spanning more than 100 pages [86]. It is methods [79]. NICE currently has four guidance actively in important for ERGs to focus on the factors most likely to development for MBC [78], two of which do not have affect cost-effectiveness conclusions, and survival data publicly available submissions from the manufacturers frequently have the largest impact on cancer model [80, 81]. One STA began in 2013, and, in this STA, the CS outcomes. and ERG report used the study by Lloyd et al. for utility However, ERGs have frequently failed to comment on data [19, 46, 82]. The most recent STA (CS provided in the consistency of utility scores in breast cancer STAs with September 2016) to have CSs, clarification questions, and NICE-preferred methods. In fact, in TA 423, the ERG an ERG report conforms well to NICE-preferred methods advised using noncompliant data from the study by Lloyd [83, 84]. et al. [19], and, in TA 424, the ERG used utility scores Why EQ-5D values were so infrequently used in CSs is from sources that did not comply with the NICE reference an unanswered question. An in-progress systematic review, case for their alternative utility value analyses [32, 48]. We currently published only as a conference abstract, identified believe that failure to use NICE preferences may be pre- 17 studies that provided EQ-5D utility scores for breast vented through the use of Table 3. cancer [85]. Only three published breast cancer STAs used In the current methodological guidance (2013), the EQ-5D values. Two EBC STAs used EQ-5D—one used section on valuing health effects for the reference cases has mapping (TA 109, scores redacted) and one used directly 12 subsections, with the use of EQ-5D being reiterated in measured EQ-5D scores (TA 424). The clinical evidence nine of these [4]. The emphasis in the guidance is clearly sources used in the companies’ submissions for MBC all on consistency across different disease areas and between had data that could be mapped to EQ-5D at the time of NICE guidance. Unfortunately, the written emphasis has submission. Only one MBC STA used mapping from not, until very recently, been followed by consistent EORTC QLQ-C30 to derive appropriate EQ-5D utility actions. scores and received criticism for using these appropriate In breast cancer STAs, NICE and ERG groups have methods [41, 48]. We believe there are three potential often not requested analyses, and have frequently not explanations for why alternative utility values have not conducted supplemental analyses, using EQ-5D utilities. been used: (1) alternative values have been identified and The quantity of EQ-5D mapping algorithms for breast are unfavourable to companies; (2) there is a lack of belief cancer populations has increased, as evidenced in the in mapping methodology; (3) and/or the values from the HERC database of mapping studies [8]. This database study by Lloyd et al. are viewed as accepted and further facilitates the use of mapping methods to transform clinical work viewed as unnecessary or inappropriate in spite of the effectiveness data into EQ-5D utility scores in current and dataset’s inconsistency with NICE-preferred methods. future breast cancer STAs. NICE needs to clarify whether None of these potential explanations helps foster informed the use of mapping methodologies is their preferred second consistent decision making across disease areas. option, and, if it is, they should cease to accept utility The last two published MBC guidance (TA 423 and values derived from noncompliant measurement methods TA 424) both used EQ-5D utility scores, with one STA in breast cancer STAs (and more broadly), and ensure that using mapping to derive these values [12, 41]. The most their preference for EQ-5D utility scores is adhered to, recent breast cancer STA guidance in development also otherwise NICE’s repeated emphasis of their preference for used EQ-5D [83]. It appears probable that NICE, ERGs and EQ-5D has little meaning and they increase the risk of companies are improving compliance to NICE’s methods; inconsistent decision making. however, STA guidance issued prior to this improvement in compliance are still binding and still have the potential to cause inequitable decisions. The NICE should ensure 6 Conclusion that when these guidance are updated, the utility methods in these guidance are also made conformant with NICE- Historically, breast cancer STAs have shown a broad lack preferred methods. of compliance with preferred methods for measuring utility While some ERGs have criticised utility score usage in scores in the NICE reference case. However, the last two STAs, the primary focus in breast cancer STAs has been on guidance published in December 2016 adhere to the NICE survival rather than utilities, which is not inappropriate. reference case, and the latest STA guidance in develop- Time is limited in STAs. Only 8 weeks are available to ment also conforms to NICE-preferred methods evaluate a multi-hundred page document, formulate clari- [12, 41, 83]. It will be the continued responsibility of the fication questions to the company, incorporate new data ERG groups and NICE to ensure that future guidance and A review of methodological consistency in breast cancer utility measurement in NICE STAs 105 standardised incidence rates for all cancer sites by age and sex: reviews of guidance maintain this forming trend. It would Office for National Statistics; 2017. be inappropriate to let noncompliant utilities used in past 7. National Institute for Health and Care Excellence (NICE). NICE guidance remain the preferred utilities in future guidance. published guidance and advice list. 2017. Available at: https:// To fail to update to methodologically appropriate utility www.nice.org.uk/guidance/published?type=ta. Accessed 3 Apr scores when updating guidance would be to grant these 8. Dakin H. Review of studies mapping from quality of life or guidance using methods that are inconsistent with the clinical measures to EQ-5D: an online database. Health Qual Life NICE reference case a durable precedent, which is a Outcomes. 2013;11:151. doi:10.1186/1477-7525-11-151. decision that should not stand. 9. Bristol-Myers Squibb Pharmaceuticals Ltd. Taxol (paclitaxel) for the adjuvant treatment of early breast cancer: submission to the National Institute for Health and Clinical Excellence (CS TA108); 2006. Author Contributions MR was responsible for revision of the 10. Roche. Herceptin (trastuzumab): NICE submission. Achieving research proposal, drafting the protocol, collaboratively designing the clinical excellence in the adjuvant treatment of her-2 positive data extraction form, piloting the extraction form, performing data breast cancer (CS TA107). 2006. extraction and checking, synthesising findings, and writing all drafts 11. Sanofi-Aventis. Docetaxel for the adjuvant treatment of early of this article. SR contributed to the production of the protocol, col- node-positive breast cancer. Single Technology Appraisal. Sub- laboratively designing the data extraction form, piloting the extraction mission to the National Institute for Health and Clinical Excel- form, performing data extraction and checking, providing feedback, lence (CS TA109). 2006. and editing all drafts of this article. DC identified the topic, wrote the 12. Roche. Neoadjuvant Parjeta (pertuzumab) for the treatment of initial research proposal, contributed to the production of the protocol, HER2-positive early breast cancer (CS TA424). 2016. had oversight of the review, and provided feedback and editing of all 13. Cheang MCU, Chia SK, Voduc D, Gao D, Leung S, Snider J, drafts of this article. et al. Ki67 Index, HER2 status, and prognosis of patients with luminal B breast cancer. J Natl Cancer Inst. Compliance with Ethical Standards 2009;101(10):736–50. 14. Peasgood T, Ward S, Brazier J. A review and meta analysis of This research involved no human or animal participation. health state utility values in breast cancer. Discussion paper. Sheffield: The University of Sheffield; 2010. Funding This study applied for and received no funding. 15. Brown RE, Hutton J. Cost-utility model comparing docetaxel and paclitaxel in advanced breast cancer patients. Anticancer Drugs. Conflict of interest Micah Rose, Stephen Rice and Dawn Craig 1998;9(10):899–907. declare no conflicts of interest. 16. Hillner BE, Smith TJ. Efficacy and cost effectiveness of adjuvant chemotherapy in women with node-negative breast cancer. a Data Availability Statement The data identified and analysed dur- decision-analysis model. N Engl J Med. 1991;324(3):160–8. ing the current study are available from the corresponding author 17. Launois R, Reboul-Marty J, Henry B, Bonneterre J. A cost-utility upon reasonable request. analysis of second-line chemotherapy in metastatic breast cancer. Docetaxel versus paclitaxel versus vinorelbine. PharmacoEco- Open Access This article is distributed under the terms of the nomics. 1996;10(5):504–21. Creative Commons Attribution-NonCommercial 4.0 International 18. Ossa DF, Briggs A, Cowell W, Littlewood T, Sculpher M. Utility License (http://creativecommons.org/licenses/by-nc/4.0/), which per- associated with severity of cancer-related anaemia (CRA): a mits any noncommercial use, distribution, and reproduction in any societal valuation. Value Health. 2004;7(6):680. medium, provided you give appropriate credit to the original 19. Lloyd A, Nafees B, Narewska J, Dewilde S, Watkins J. Health author(s) and the source, provide a link to the Creative Commons state utilities for metastatic breast cancer. Br J Cancer. license, and indicate if changes were made. 2006;95(6):683–90. 20. Carter K, Ritchey N, Castro F, Caccamo L, Kessler E, Erickson B, et al. Treatment of early-stage breast cancer in the elderly: a health-outcome-based approach. Med Decis Mak. References 1998;18:213–9. 21. Hayman J, Hillner B, Harris J, Weeks J. Cost-effectiveness of 1. National Institute for Health and Care Excellence (NICE). Guide routine radiation therapy following conservative surgery for to the single technology appraisal process 2009. London: NICE; early-stage breast cancer. J Clin Oncol. 1998;16(3):1022–9. 22. Lidgren M, Wilking N, Jonsson B, Rehnberg C. Health related 2. National Institute for Health and Care Excellence (NICE). Guide quality of life in different states of breast cancer. Qual Life Res. to the methods of technology appraisal 2004. London: NICE; 2007;16(6):1073–81. 23. Tengs T, Wallace A. One thousand health-related quality-of-life 3. National Institute for Health and Care Excellence (NICE). Guide estimates. Med Care. 2000;38(6):583–637. to the methods of technology appraisal 2008. London: NICE; 24. Van Hanswijck de Jonge P, Doyle S, Farina C, Walker M. Poster 2008. presentation 190 P: elicitation of UK Health utilities in primary, 4. National Institute for Health and Care Excellence (NICE). Guide to recurrent and metastatic breast cancer. In: 31st European Society the methods of technology appraisal 2013. London: NICE; 2013. Medical Oncology (ESMO), 29 Sep–3 Oct 2006, Istanbul. 5. Peasgood T, Ward SE, Brazier J. Health-state utility values in 25. Kind P. Measuring the value of quality of life in cancer: an index breast cancer. Expert Rev Pharmacoecon Outcomes Res. based on EORTC QLQC-30. J Clin Oncol. 2005;23(16):531S-S. 2010;10(5):553–66. 26. Essers B, Seferina S, Tjan-Heijnen V, Severens J, Nova ´kA, 6. Office for National Statistics. Cancer registration statistics, Pompen M, et al. Transferability of model-based economic England: first release, 2015 Cancer diagnoses and age- evaluations: the case of trastuzumab for the adjuvant treatment of 106 M. Rose et al. HER2-positive early breast cancer in the Netherlands. Value 44. Hutton J, Brown R, Borowitz M, Abrams K, Rothman M, Health. 2010;13(4):375–80. Shakespeare A. A new decision model for cost-utility compar- 27. National Institute for Health and Care Excellence (NICE). Tras- isons of chemotherapy in recurrent metastatic breast cancer. tuzumab for early breast cancer clarification questions (PFC1, PharmacoEconomics. 1996;9(Suppl 2):8–22. TA107). 2006. 45. Crott R, Briggs A. Mapping the QLQ-C30 quality of life cancer 28. National Institute for Health and Care Excellence (NICE). Tras- questionnaire to EQ-5D patient preferences. Eur J Health Econ. tuzumab for early breast cancer clarification questions (PFC2, 2010;11(4):427–34. TA107). 2006. 46. Roche. Trastuzumab emtansine for treating HER2-positive, 29. Chilcott J, Lloyd Jones M, Wilkinson A. Docetaxel for the locally advanced or metastatic breast cancer after treatment with adjuvant treatment of early node-positive breast cancer: a single trastuzumab and a taxane (CS TA371): Roche; 2013. technology appraisal (ERG TA109). Sheffield: The School of 47. Fleeman N, Bagust A, Boland A, Dickson R, Dundar Y, Moonan Health and Related Research (ScHARR); 2006. M, et al. Lapatinib and trastuzumab in combination with an 30. Ward S, Pilgrim H, Hind D. Trastuzumab for the treatment of aromatase inhibitor for the first-line treatment of metastatic hor- primary breast cancer in HER2 positive women: a single tech- mone receptor-positive breast cancer which over-expresses nology appraisal (ERG TA107). Sheffield: University of Shef- human epidermal growth factor 2 (HER2): a systematic review field, School of Health and Related Research (ScHARR); 2006. and economic analysis. Health Technol Assess. 31. Centre for Health Economics (CHE), Regional Drug and Thera- 2011;15(42):1–93, iii–iv. peutics Centre. The use of paclitaxel in the management of early 48. Fleeman N, Bagust A, Richardson M, Houten R, Krishan A, stage breast cancer (ERG TA108). 2006. Beale S, et al. Eribulin for treating locally advanced or metastatic 32. Squires H, Pandor A, Thokala P, Stevens J, Kaltenthaler E, breast cancer after chemotherapy [ID964] (ERG Report TA 423): Clowes M. Pertuzumab for the neoadjuvant treatment of HER2- Liverpool Reviews and Implementation Group; 2016. positive breast cancer (ERG TA424). 2016. 49. Fleeman N, Beale S, Davis H, Dickson R, Loncaster J, O’Reilly 33. Sorensen SV, Brown R, Benedict A, Flood E, Revicki D. Patient- S, et al. Bevacizumab in combination with capecitabine for the rated utilities in postmenopausal early breast cancer (EBC): a first-line treatment of metastatic breast cancer: a Single Tech- cross-country comparison. Value Health. 2004;7(6):641–2. nology Appraisal (ERG TA263). Liverpool: Liverpool Reviews 34. Locker GY, Mansel R, Cella D, Dobrez D, Sorensen S, Gandhi & Implementation Group (LRiG), University of Liverpool; 2011. SK, et al. Cost-effectiveness analysis of anastrozole versus 50. Rodgers M, Soares M, Epstein D, Yang H, Fox D, Eastwood A. tamoxifen as primary adjuvant therapy for postmenopausal Bevacizumab in combination with a taxane for the first-line women with early breast cancer: a US healthcare system per- treatment of HER2-negative metastatic breast cancer (ERG spective. The 5-year completed treatment analysis of the ATAC TA214). York: Center for Reviews and Dissemination (CRD), (‘Arimidex’, Tamoxifen Alone or in Combination) trial. Breast CHE Technology Assessment Group; 2010. Cancer Res Treat. 2007;106(2):229–38. 51. Fleeman N, Bagust A, Beale S, Blundell M, Dwan K, Pilkington 35. Novartis. Single technology appraisal (STA): Everolimus (Afin- G, et al. Everolimus in combination with an aromatase inhibitor itor) in combination with exemestane for the treatment of for the treatment of breast cancer after prior endocrine therapy: a advanced or metastatic HER2 negative, hormone receptor posi- Single Technology Appraisal. LRiG, The University of Liverpool tive breast cancer after prior endocrine therapy (CS TA295): (ERG TA295). Liverpool: Liverpool Reviews and Implementa- Novartis; 2012. tion Group (LRiG), The University of Liverpool; 2013. 36. Roche. Single technology appraisal (STA): Bevacizumab in 52. Bagust A, Blundell M, Boland A, Dickson R, Dundar Y, Fleeman combination with capecitabine for the treatment of metastatic N, et al. Fulvestrant for the treatment of locally advanced or breast cancer (CS TA263): Roche; 2011. metastatic breast cancer (ERG TA239). Liverpool: Liverpool 37. Eisai Ltd. Single technology appraisal (STA): Eribulin for the Reviews and Implementation Group (LRiG); 2011. treatment of locally advanced breast cancer or metastatic breast 53. Jones J, Takeda A, Tan SC, Cooper K, Loveman ECA, Murray N. cancer (CS TA250). 2011. Evidence Review Group Report commissioned by the NHS R&D 38. AstraZeneca UK Ltd. Single Technology Appraisal (STA): Ful- HTA Programme on behalf of NICE: gemcitabine for metastatic vestrant 500 mg for the treatment of postmenopausal women with breast cancer (ERG TA116). Southampton: Southampton Health oestrogen receptor positive, locally advanced or metastatic breast Technology Assessments Centre; 2006. cancer (CS TA239). 2010. 54. Roche. Bevacizumab in combination with a taxane for the first- 39. Roche. Bevacizumab in combination with taxanes for the treat- line treatment of metastatic breast cancer: manufacturer response ment of HER2-negative 1st line metastatic breast cancer. Roche: to points for clarification (PFC TA214). London. 2010. submission to the National Institute for Health and Clinical 55. Roche. Single technology appraisal—bevacizumab in combina- Excellence (CS TA214). 2010. tion with capecitabine for the first-line treatment of metastatic 40. Eli Lilly and Company Ltd. Gemcitabine for the treatment of breast cancer: points for clarification (PFC TA263). Manchester: metastatic breast cancer. Single technology appraisal (STA). Roche; 2012. Submission to the National Institute for Health and Clinical 56. Eastern Cooperative Oncology Group. Paclitaxel with or without Excellence (CS TA116). 2006. bevacizumab in treating patients with locally recurrent or meta- 41. Eisai Ltd. Eribulin for treating locally advanced or metastatic static breast cancer. NCT00028990. ClinicalTrials.gov. Bethesda brest cancer after chemotherapy (CS TA 423). 2016. (MD): National Library of Medicine (US); 2002. TM 42. Cooper NJ, Abrams KR, Sutton AJ, Turner D, Lambert PC. 57. AstraZeneca. Comparison of fulvestrant (FASLODEX A Bayesian approach to Markov modelling in cost-effectiveness 250 mg and 500 mg in postmenopausal women with oestrogen analyses: application to taxane use in advanced breast cancer. J R receptor positive advanced breast cancer progressing or relapsing Stat Soc Ser A Stat Soc. 2003;166:389–405. after previous endocrine therapy (CONFIRM). NCT00099437. 43. Hudgens S, Briggs A, Velikova G, Forsythe A, McCutcheon S, ClinicalTrials.gov. Bethesda (MD): National Library of Medicine Kind P. Impact of treatment with eribulin or capecitabine for (US); 2004. metastatic breast cancer (MBC) on EQ-5D utility derived from 58. Eisai, Inc. A study of E7389 in advanced/metastatic breast cancer EORTC QLQ-C30. In: European Society of Medical Oncology; patients. NCT00097721. ClinicalTrials.gov. Bethesda (MD): 2014. National Library of Medicine (US); 2004. A review of methodological consistency in breast cancer utility measurement in NICE STAs 107 59. Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, 74. Longworth L, Yang Y, Young T, Mulhern B, Hernandez Alava et al. Trastuzumab emtansine for HER2-positive advanced breast M, Mukuria C, et al. Use of generic and condition-specific cancer. N Engl J Med. 2012;367(19):1783–91. measures of health-related quality of life in NICE decision- 60. Welslau M, Dieras V, Sohn JH, Hurvitz SA, Lalla D, Fang L, making: a systematic review, statistical modelling and survey. et al. Patient-reported outcomes from EMILIA, a randomized Health Technol Assess. 2014;18(9):1–224. phase 3 study of trastuzumab emtansine (T-DM1) versus cape- 75. Pennington B, Davis S. Mapping from the Health Assessment citabine and lapatinib in human epidermal growth factor receptor Questionnaire to the EQ-5D: the impact of different algorithms 2-positive locally advanced or metastatic breast cancer. Cancer. on cost-effectiveness results. Value Health. 2014;17(8):762–71. 2014;120(5):642–51. 76. Devlin N, Shah K, Feng Y, Mulhern B, van Hout B. Valuing 61. FACIT.org. FACT questionnaires. Available at: http://www.facit. health-related quality of life: an EQ-5D-5L value set for England org/FACITOrg/Questionnaires. Accessed 8 May 2014. (OHE research paper 16/01). Office of Health Economics; 2016. 62. Cheung Y-B, Thumboo J, Gao F, Ng G-Y, Pang G, Koo W-H, 77. Feng Y, Devlin N, Shah K, Mulhern B, van Hout B. New et al. Mapping the English and Chinese versions of the Functional methods for modelling EQ-5D-5L value sets: an application to Assessment of Cancer Therapy-General to the EQ-5D Utility English data (OHE research paper 16/02). Office of Health Index. Value Health. 2009;12(2):371–6. Economics; 2016. 63. Teckle P, McTaggart-Cowan H, Van der Hoek K, Chia S, 78. National Institute for Health and Care Excellence (NICE). NICE Melosky B, Gelmon K, et al. Mapping the FACT-G cancer- In Development Technology Appraisal Guidance List. 2017. specific quality of life instrument to the EQ-5D and SF-6D. Available at: http://www.nice.org.uk/guidance/ Health Qual Life Outcomes. 2013;11:203. indevelopment?type=TA. Accessed 25 Apr 2017. 64. Novartis Pharmaceuticals. Everolimus in combination with 79. Jones J, Takeda A, Picot J, von Keyserlingk C, Clegg A. Evi- exemestane in the treatment of postmenopausal women with dence Review Group Report commissioned by the NHS R&D estrogen receptor positive locally advanced or metastatic breast HTA Programme on behalf of NICE: lapatinib for HER2 over- cancer who are refractory to letrozole or anastrozole expressing breast cancer. Southampton: Southampton Health (BOLERO2). NCT00863655. ClinicalTrials.gov. Bethesda (MD): Technology Assessments Centre; 2007. National Library of Medicine (US); 2009. 80. National Institute for Health and Care Excellence (NICE). Breast 65. Kim EJ, Ko SK, Kang HY. Mapping the cancer-specific EORTC cancer (brain metastases)—etirinotecan pegol [ID881]. 2017. QLQ-C30 and EORTC QLQ-BR23 to the generic EQ-5D in Available at: https://www.nice.org.uk/guidance/indevelopment/ metastatic breast cancer patients. Qual Life Res. gid-ta10066. Accessed 25 Apr 2017. 2012;21(7):1193–203. 81. National Institute for Health and Care Excellence (NICE). Ful- 66. Roche. Pertuzumab in combination with trastuzumab and doc- vestrant for untreated hormone-receptor positive metastatic breast etaxel for the treatment of HER2-positive metastatic breast can- cancer [ID951]. 2017 Available at: https://www.nice.org.uk/ cer. STA submission: Roche; 2013. guidance/indevelopment/gid-ta10106. Accessed 25 Apr 2017. 67. National Collaborating Centre for Cancer. Advanced breast 82. Fleeman N, Bagust A, Beale S, Dwan K, Dickson R, Proudlove cancer: diagnosis and treatment (CG 81). 2009. C, et al. Pertuzumab in combination with trastuzumab and doc- 68. National Collaborating Centre for Cancer. Early and locally etaxel for the treatment of HER2 positive metastatic or locally advanced breast cancer: diagnosis and treatment (CG 80): recurrent unresectable breast cancer: a Single Technology National Institute for Health and Care Excellence; 2009. Appraisal. Liverpool: Liverpool Reviews and Implementation 69. Lewis R, Bagnall A-M, Forbes C, Shirran E, Duffy S, Kleijnen J, Group (LRiG), The University of Liverpool; 2013. et al. A rapid and systematic review of the clinical effectiveness 83. Pfizer. Palbociclib in combination with an aromatase inhibitor for and cost-effectiveness of trastuzumab for breast cancer: National previously untreated metastatic, hormone receptor-positive, Institute for Health and Care Excellence; 2001. HER2-negative breast cancer [ID915 CS]. 2017. 70. Huxley N, Jones-Hughes T, Coelho H, Snowsill T, Cooper C, 84. Fleeman N, Stainthorpe A, Bagust A, Richardson M, Nolan S, Cooper K, et al. A systematic review and economic evaluation of Houten R, et al. Palbociclib in combination with an aromatase intraoperative tests [RD-100i one-step nucleic acid amplification inhibitor for previously untreated metastatic, hormone receptor- (OSNA) system and Metasin test] for detecting sentinel lymph positive, HER2-negative breast cancer [ID915 ERG Report]. node metastases in breast cancer. Health Technol Assess. 2017. 2012;19(2):1–215. 85. Hughes R, Mitchell CR, Bishop RS, Fotheringham I. Health-state 71. Ward S, Scope A, Rafia R, Pandor A, Harnan S, Evans P. Gene utility values in breast and prostate cancer measured using the expression profiling and expanded immunohistochemistry tests to EQ-5D: a systematic review of the literature. Value Health. guide the use of adjuvant chemotherapy in breast cancer man- 2014;17(7):A645–6. agement. Health Technol Assess. 2011;17(44):1–302. 86. National Institute for Health and Care Excellence (NICE). Guide 72. Tosh JC, Longworth LJ, George E. Utility values in National to the processes of technology appraisal 2014. London: NICE; Institute for Health and Clinical Excellence (NICE) Technology 2014. Appraisals. Value Health. 2011;14(1):102–9. 87. Winstanley J, Murray N. National Institute for Health and Care 73. Doble B, Lorgelly P. Mapping the EORTC QLQ-C30 onto the Excellence (NICE). CG81: Advanced breast cancer: diagnosis EQ-5D-3L: assessing the external validity of existing mapping and treatment. Cardiff: National Collaborating Centre for Cancer. algorithms. Qual Life Res. 2016;25(4):891–911. 2009.

Journal

PharmacoEconomics - OpenSpringer Journals

Published: Jul 5, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off