NICE Technology Appraisals in the NHS in England (Innovation Scorecard) to March 2019

NICE resource impact assessments were used as the starting point to develop the estimates presented.

Resource impact templates are provided by NICE to support the implementation of most technology appraisal guidance when the resource impact is expected to be significant (greater than £5 million for England). The templates are an aid for financial planning purposes and enable users to estimate the local resource impact of implementing guidance at the time of publication. However, the template can be used to further develop and establish assumptions around an eligible population and an estimated treatment population.

The assumptions used to generate the resource impact templates are based on peer reviewed literature, data sources, expert opinion and other information. Data were sought from several different sources to refine the population numbers for the indications and circumstances detailed in the NICE technology appraisal guidance. The data reviewed included: epidemiological data such as the prevalence of a disease, proportions of patients at a stage of a disease, and their likely treatment history. For some medicines additional information was required, for example, the proportion of patients likely to discontinue treatment or choose an alternative.

NICE do not produce a resource impact template where the cost impact of a technology is not considered to be significant, or when estimating the cost impact is not possible. In the absence of a resource impact template an estimate was constructed using a stepwise process similar to that used to develop a resource impact template. The process involved:

1. A review of the available literature on the epidemiology of the indication(s) for the medicine.
2. Where appropriate, the use of primary data sources such as Hospital Episode Statistics (HES) and The Health Improvement Network (THIN) database (containing pseudonymous patient information extracted from a sample of GP practice clinical systems).
3. Consultation with clinical experts and consideration of expert opinion used in other templates/sources of evidence for the same therapeutic area.

For medicines with multiple indications, only those recommended by NICE were identified. Eligible populations were developed on an indication by indication basis. The separate indication estimates were then combined to produce an overall estimate. Indications not appraised or recommended by NICE were excluded from the estimate of the eligible population. Where a medicine has an indicated use not appraised or recommended by NICE this could give the impression of over usage, even if this is not the case.

Treatment population estimates were used to calculate the total expected volume of medicine at a national level. Where available, Defined Daily Doses (DDDs) as defined by the World Health Organisation (WHO) were used. For those medicines where DDDs were not available, or were developed for a different indication, an alternative measure such as daily dose or actual daily dose (ADD) was used.

The daily dose is the daily maintenance dose calculated from the recommended dosage as set out in the summary of product characteristics. The ADD is a prescribing measure developed for the Innovation Scorecard. ADDs assign a unique value for each presentation of a medicine based on units (such as tablets, capsules, patches) and the recommended frequency of daily use. For more information on ADD please refer to the Medicine Groupings Specification. 

The annual estimated usage volume was proportionally allocated to each quarter based on the number of days in a quarter.

Ranges for the estimate of the treatment population and the estimate of usage have been calculated for each estimate. The range provides an indication of the level of uncertainty in the estimate. They are calculated from confidence intervals and other specified measures at each step where available.

Manufacturer input was requested for new estimates or those where a material change had been made to the estimated treatment population calculation. The draft estimates were sent to the relevant manufacturers with a request for comments and feedback on the estimate calculations and the supporting data. Company feedback was then critically appraised and where appropriate, the draft estimate was updated. Where data used in the estimate is taken from a source that is routinely updated, such as the Quality and Outcomes Framework or population statistics, updating the estimate to reflect the latest publication is not considered to be a material change.

Data for observed use of the medicines under consideration were obtained by NHS Digital through its routine access to Prescribing Analysis and Cost Tabulation (PACT) data on primary care prescribing (supplied by the NHS BSA) and Hospital Pharmacy Audit Index (HPAI) data on secondary care prescribing (supplied by IQVIA, formerly IMS Health and Quintiles). Usage data were converted where appropriate, for example to DDDs or daily doses, to allow comparison with the estimates.

Although the secondary care database captures some data for drugs supplied through homecare services, it is incomplete. Therefore, the actual volume of medicine used may be higher than the volume reported in the observed use. This applies to all medicines in this report, which are prescribed as follows:

• Sacubitril valsartan: primary and secondary care
• Vedolizumab: secondary care only
• Afatinib, erlotinib, gefitinib and osimertinib: predominantly used in secondary care, but a very small volume of medicine is prescribed in primary care
• Abiraterone, cabazitaxel and enzalutamide: predominantly used in secondary care, but a small volume of medicine is prescribed in primary care
• Trifluridine–tipiracil: secondary care only
• Rifaximin (550mg): primary and secondary care
• Alirocumab and evolocumab: primary and secondary care
• Dimethyl fumarate and teriflunomide: predominantly used in secondary care, but a small volume of medicine is prescribed in primary care
• Etelcalcetide: secondary care

Observed use is compared with the estimated use to indicate whether use is higher or lower than expected. The estimated treatment population is calculated with an upper and lower range, to address the uncertainty in the underlying analyses. The ratio of observed to expected volume, and the upper and lower range, is calculated. Where the observed use is within the upper and lower estimate range, the use is as expected.

Several assumptions are made in order to develop the expected volume of medicine to be prescribed. This approach is due to limitations, which include:

• Lack of prevalence and incidence data at national level
• Some medicines have multiple indications of which one or more indications have not been recommended by NICE
• Medicines recommended as one of several options for treatment

Usage data are limited in coverage and quality. Problems include:

• Multiple indications for a single medicine (usage data give no information on the condition being treated)
• Failure of some hospitals to contribute data to the IQVIA data collection or are unable to provide full datasets
• Lack of available data from some mental health trusts
• Reporting of medicines supplied via the homecare route or by outsourced dispensing is not recorded in pharmacy systems
• Medicines that need to be diluted or manipulated to the individual patient’s requirements in specialist units; the way these are recorded in the pharmacy systems often does not allow calculation of the actual amount of drug used and sometimes these medicines are not recorded at all by the pharmacy

These limitations mean that caution must be exercised in interpreting the figures in the report as providing evidence of under or overuse of the medicines reviewed.

1. Population who meet SmPC

The SmPC states that sacubitril valsartan is indicated for the treatment of symptomatic chronic heart failure with reduced ejection fraction.

2. Prevalence of heart failure

The quality and outcomes framework (QOF) reported the number of people on the heart failure register (NHS Digital 2018). It is estimated that 832 people per 100,000 population have heart failure. The 95% confidence interval around the reported prevalence (95% CI 829 to 834) has been calculated using the exact binomial method (Pezzullo 2009).

The prevalence and confidence interval have been applied to the adult population in England:

3. Proportion of people with heart failure and reduced left ventricular systolic dysfunction

The national heart failure audit (National Institute for Cardiovascular Outcomes Research 2016) reported that 70.2% of patients who were admitted for heart failure had left ventricular systolic dysfunction. This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 2.

4. Proportion of people with New York Heart Association (NYHA) class II to IV symptoms

The manufacturer’s submission for TA388 (Novartis Pharmaceuticals UK Ltd 2015) reported that 89% of people with heart failure had a New York Heart Association (NYHA) heart failure classification symptom between II to IV. This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 3.

5. Proportion of people with a left ventricular ejection fraction of 35% or less

The resource impact assessment template for sacubitril valsartan estimates that 59.5% people with heart failure have a left ventricular ejection of 35% or less (NICE 2016). This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 4.

6. Proportion of people taking a stable dose of angiotensin‑converting enzyme (ACE) inhibitors or angiotensin II receptor‑blockers (ARBs)

The quality and outcomes framework heart failure register report the number of people taking a stable dose of angiotensin‑converting enzyme (ACE) inhibitor or angiotensin II receptor‑blocker (ARB). It was reported that 83.04% of people on the register are taking a stable dose of either (NHS Digital 2018). This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 5.

7. Estimated treatment population

The resource impact assessment template for sacubitril valsartan projects the uptake of this medicine over time (NICE 2016). The projected uptake in 2017-18 is expected to be 24%; this figure has been adjusted +/- 10% to create a range (21.6% - 26.4%). These proportions have been applied as a scaling factor to the point estimate and the upper and lower ranges calculated in step 6.

8. Calculate estimated usage volume

The number of people likely to receive treatment with sacubitril valsartan (calculated in step 7) has been multiplied by the number of tablets taken per day, and then by the number of days in a year (365 days).

References

1. Office for National Statistics (2018) Annual Mid-year Population Estimates, 2017 ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed: 06/11/2018].
2. Pezzullo. (2009) Exact Binomial and Poisson Confidence Intervals calculator. Available from https://statpages.info/confint.html [Accessed 17/06/2016].
3. NHS Digital (2017). Quality and Outcomes Framework (QOF) for April 2017 - March 2018, England. Available from: https://digital.nhs.uk/data-and-information/publications/statistical/quality-and-outcomes-framework-achievement-prevalence-and-exceptions-data [Accessed 26/10/2018].
4. National Institute of Cardiovascular Outcomes Research (2016). National Heart Failure Audit, April 2014- March 2015. Available from: http://www.bsh.org.uk/resources/national-heart-failure-audit/ [Accessed: 17/11/2016].
5. Novartis Pharmaceuticals UK Ltd (2015). The manufacturer’s submission for NICE technology appraisal 388. Available from: https://www.nice.org.uk/guidance/ta388/evidence [Accessed: 17/08/2016].
6. NICE (2016) Resource impact assessment template for TA388. Manchester: NICE. Available from: https://www.nice.org.uk/guidance/ta388/resources [Accessed: 17/08/2016]. Additional expert clinical opinion was used to inform steps 5 and 7.

1. Population who meet SmPC

The SmPC states that the medicine is indicated for an adult population only. Therefore, the starting population is adults aged 18 years and over in England.

2. Prevalence of ulcerative colitis (UC) and Crohn’s disease (CD)

Rubin et al (2000) have reported the prevalence of ulcerative colitis and Crohn’s disease in the United Kingdom (UK). This study was carried out in 15 GP practices in North Tees and represented a population of 135,723 people. The prevalence figures were adjusted for age and sex and were reported as:

Prevalence of UC per 100,000 UK pop in 1995: 243.4 (95% CI 217.4 to 269.4)

Prevalence of CD per 100,000 UK pop in 1995: 144.8 (95% CI 124.8 to 168.8)

The prevalence figures and associated confidence intervals have been applied to the adult population in England:

3. Proportion of people with moderately to severely active disease

The manufacturer’s submission (Takeda 2014a) has reported that 52% of people with ulcerative colitis in the UK have moderately to severely active disease. Dretzke et al (2015) has reported that 40% of people with Crohn’s disease have moderately to severely active disease. These proportions have been applied as a scaling factor to the relevant point estimates and the upper and lower ranges calculated in step 2.

4. Proportion of people whose disease failed to respond to conventional therapy and are treated with a biologic

The proportion of people whose disease failed to respond to conventional therapy and are treated with a biologic has been estimated using data from the national clinical audit of biological therapies (Royal College of Physicians 2015).

To develop this estimate, the data reported in the flow diagram of people included in the audit from 2011 to 2015 was used to estimate that 10% of people with ulcerative colitis are treated with a biologic. This has been referred to as the ‘estimated number of people who are eligible for treatment’ because vedolizumab is a treatment option for people with ulcerative colitis which has failed to respond to conventional therapy (see SmPC and technology appraisal recommendations). The same data (Royal College of Physicians 2015) has been used to estimate that 56% of people have Crohn’s disease which has failed to respond to conventional therapy and have been treated with a biologic. The 95% confidence intervals around the audit data have been calculated, using the exact binomial method (Pezzullo 2009). Therefore 10% (95% CI 9.5 to 10.9%) of people with ulcerative colitis and 56% (95% CI 54.5 to 56.8%) of people with Crohn’s disease are estimated to be treated with a biologic. These figures have been applied to the point estimates and the upper and lower ranges calculated in step 3.

5. Proportion of people with Crohn’s disease who stop treatment with or whose disease has failed to respond to a tumour necrosis factor-alpha (TNFα) antagonist

The national clinical audit of biological therapies (Royal College of Physicians 2015) has reported the proportion of people with Crohn’s disease which has failed to respond to a TNFα antagonist. The audit reported follow-up treatment outcomes at 3, 6 and 12 months. Three month data has been used because the SmPC for biologics such as infliximab and adalimumab suggest that continued treatment beyond 4 to 12 weeks is not recommended unless people have a clinical response. Therefore, people are most likely to stop treatment in the first 3 months of starting therapy and this data would provide the most accurate figure of those eligible for treatment with vedolizumab. It is estimated that at 3 months, approximately 7% (95% CI 5.4 to 8.1%) of people with Crohn’s disease stopped treatment with adalimumab or infliximab either because of loss of response, poor response or side effects and 32% (95% CI 29.5 to 34.6%) of people did not experience clinical remission. To develop this estimate, the 95% confidence intervals around the audit data have been calculated, using the exact binomial method (Pezzullo 2009).

6. Number of people eligible to receive treatment with vedolizumab

The number of people estimated to be eligible for treatment with vedolizumab are based on steps 4 and 5 above. For ulcerative colitis, the figures from step 4 have been used because vedolizumab is a treatment option following conventional therapy.

For Crohn’s disease, the sum of the point estimates and the upper and lower ranges calculated in step 5 have been used.

7. Proportion of people likely to receive treatment with vedolizumab

The manufacturer’s submission for ulcerative colitis (Takeda 2014a) and the NICE costing report for Crohn’s disease (NICE 2015) have reported the projected uptake of vedolizumab. The proportions reported in the manufacturer’s submission are based on current uptake of medicines and market share assumptions. The projected uptake of vedolizumab in 2018 is estimated to be 30% for ulcerative colitis and 80% for Crohn’s disease. The uptake for people with Crohn’s disease is assumed to be higher because there has been previous intolerance or failure with biologics such as infliximab and adalimumab. These proportions have been applied as a scaling factor to the point estimates and the upper and lower ranges calculated in step 6.

8. Calculate estimated usage volume

The number of people expected to be treated with vedolizumab calculated in step 7 has been multiplied by the number of days in a year (365 days) to produce an expected volume of medicine per year in DDDs.

The DDD for vedolizumab is reported as 5.4mg/day (WHO 2015) and may underestimate the number of people receiving treatment.

References

1. Office for National Statistics (2018) Annual Mid-year Population Estimates, 2017 ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed: 06/11/2018].
2. Rubin GP, Hungin, APS, Kelly PJ et al. (2000) Inflammatory bowel disease: Epidemiology and management in an English general practice population. Alimentary Pharmacology and Therapeutics 14: 1553-1559
3. Dretzke J, Edlin R, Round J et al. (2011) Use of tumour necrosis factor alpha (TNF α) inhibitors adalimumab and infliximab for Crohn’s disease. Systematic review and economic evaluation. Health Technology Assessment No 15.6. Available from: https://www.journalslibrary.nihr.ac.uk/hta/hta15060
4. Takeda (2014a) Manufacturer’s submission for NICE technology appraisal guidance 342. Available from: https://www.nice.org.uk/guidance/ta342/history [Accessed: 28/10/15]
5. Royal College of Physicians (2015). National clinical audit of biological therapies: adult report. UK inflammatory bowel disease (IBD) audit. London: Royal College of Physicians. Available from: https://www.rcplondon.ac.uk/projects/outputs/national-clinical-audit-report-biological-therapies-adult-report-2015 [Accessed: 01/10/2015]
6. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 16/02/17]
7. Datapharm Communications Ltd (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 16/02/2017].
8. NICE (2015). Costing report for vedolizumab for treating moderately to severely active Crohn's disease after prior therapy. Manchester: NICE. Available from: https://www.nice.org.uk/guidance/ta352/history [Accessed: 28/10/2015]
9. Pezzullo (2009) Exact Binomial and Poisson Confidence Intervals calculator. Available from: https://statpages.info/confint.html [Accessed 14/01/2016]

The following aims to estimate the expected use of the NICE positively appraised medicines afatinib, erlotinib, gefitinib and osimertinib in people with epidermal growth factor receptor (EGFR) positive advanced non-small-cell lung cancer (NSCLC).

Where these medicines have a licenced indication not positively appraised by NICE, usage has not been estimated. While this usage is expected to be small it may result in a lower expected usage when compared to observed usage.

1. Incidence of lung cancer

The SmPC states that these medicines are indicated in an adult population only.

Estimated incidence of lung cancer in adults aged 20 years or older was reported by the Office for National Statistics (2018), 2016 Cancer Registration Statistics for England.

2. Estimated number of people with non-small-cell lung cancer

The proportion of new cases of lung cancer that are non-small-cell lung cancer (NSCLC) in England is reported in the national lung cancer audit (2018) information sheet to be 88.4%. This proportion has been used to inform the upper and lower range and applied to the point estimate calculated in step 1.

3. Proportion of people presenting with stage IIIb (advanced) and stage IV (metastatic) NSCLC

The proportion of new cases of NSCLC that are diagnosed stage IIIb (advanced) or stage IV (metastatic) is reported in the national lung cancer audit for England information sheet to be 57.8%. This figure has been used to inform the upper and lower range. This proportion has been applied to the point estimate calculated in step 2.

4. Estimated proportion with stage IIIb/IV and PS 0–1 who receive systemic anticancer treatment

The proportion of people with good performance status (PS 0-1) and advanced or metastatic NSCLC who receive systemic anticancer treatment (SACT) in England is reported in the national lung cancer audit (2018) to be 62.5%. This proportion has been used to inform the upper and lower range and applied to the point estimate calculated in step 3.

5. Proportion of people with sufficient and insufficient sample for genotyping

The final appraisal determination produced for TA374 (NICE 2015) states that, to successfully test for EGFR mutation status, an adequate tissue sample and sufficient quality DNA is required. Zer (2014) estimated that between 15% and 35% of patients may not have sufficient samples for genotyping.

To calculate the proportion of people with sufficient samples for genotyping the inverse of these proportions has been applied. The midpoint (75%) has been applied to the point estimate and the proportions (65% and 85%) applied to the upper and lower range calculated in step 4.

6. Proportion of people who test positive for EGFR mutation

The proportion who test EGFR mutation positive was reported in Li et al (2013). Li reported that between 10% and 20% of people defined by the authors as ‘white’ had tumours that were EGFR mutation positive.

The reported proportions have been applied to the upper and lower range and the midpoint (15%) applied to the point estimate for people with sufficient samples for genotyping calculated in step 5.

7. Proportion of people who receive first line afatinib, erlotinib or gefitinib

It is expected that all the population in step 6 who test EGFR mutation positive will receive first line treatment with the tyrosine kinase inhibitors (TKIs) afatinib, erlotinib or gefitinib.

8. Proportion of people who progress following first line treatment with afatinib, erlotinib or gefitinib (previously treated EGFR-positive)

The proportion of people who have disease progression after first-line treatment with afatinib, erlotinib or gefitinib is stated in the manufacturer’s submission for osimertinib TA416 (AstraZeneca 2016) to be 65% of the incident population. This proportion has been applied as a scaling factor to the point estimate and the lower and upper range summed in step 7.

9. Proportion of people who harbour T790M mutation at progression (previously treated EGFR-population)

Based on genomic analysis, progression on a first-line EGFR TKI is characterised by an acquired secondary EGFR kinase domain mutation labelled T790M, causing resistance to the first-line EGFR TKI. The proportion of people who harbour T790M mutation at progression is reported in Mazza (2017) to be between 50% and 60%. The midpoint (55%) has been applied to the point estimate and the proportions (50% and 60%) applied to the upper and lower range calculated in step 8.

10. Proportion of people who receive osimertinib (previously treated EGFR-population)

It is assumed that all the people who harbour T790M mutation at progression (step 9) will receive osimertinib.

11. Number of people with insufficient sample for genotyping (unidentified EGFR status)

The final appraisal determination produced for TA374 (NICE 2015) states that, to successfully test for EGFR mutation status, an adequate tissue sample and sufficient quality DNA is required. Zer (2014) estimated that between 15% and 35% of patients may not have sufficient samples for genotyping.

To calculate the proportion with inadequate tissue sample, the midpoint (25%) has been applied to the point estimate and the proportions (15% and 35%) applied to the upper and lower range calculated in step 4.

12. Proportion of people who will receive second line therapy (unidentified EGFR status)

The proportion of the population expected to receive any second line therapy was stated in Zer (2014) to be between 30% and 50% of the eligible population. These proportions have been used to inform the upper and lower range and a mid-point (40%) has been applied to the point estimate calculated in step 11.

The final appraisal determination produced for TA374 (NICE 2015) states that an element of clinical judgement is needed when treating these patients. Patients in whom the disease has progressed after non-targeted chemotherapy may have erlotinib if their patient characteristics suggest that their tumour may be mutation-positive (for example, people who have never smoked or are light smokers, women, people of Asian family origin and people with adenocarcinoma histology). The proportion that will be identified is uncertain and is not calculated here. In clinical practice this may reduce the expected treatment population.

13. Treatment population for afatinib, erlotinib, gefitinib and osimertinib

The total treatment population for first line afatinib, erlotinib or gefitinib has been taken from step 7. The treatment population for second line osimertinib has been taken from step 10. The treatment population for second line erlotinib has been taken from step 12.

14. Calculate estimated usage volume

Defined Daily Doses (DDDs) were not available for afatinib, erlotinib, gefitinib and osimertinib so daily doses calculated from the SmPC have been used.

The SmPC recommended daily dose for afatinib is 40mg, for erlotinib is 150mg, for gefitinib is 250mg and for osimertinib is 80mg. The estimated treatment population (calculated in step 13) has been multiplied by estimated median progression-free survival per year to produce an expected volume of medicine per year in daily doses.

14a. Calculate estimated usage volume – first line afatinib, erlotinib and gefitinib

The total estimated treatment population for first line afatinib, erlotinib and gefitinib has been taken from step 13. Scluier et al (2015) reported that across 6 RCTs assessing the TKIs afatinib, erlotinib and gefitinib, compared with chemotherapy in people with NSCLC with active EGFR mutations, median progression free survival ranged between 9.2 and 13.1 months. To develop this estimate, the midpoint between this range was calculated (11 months) and applied to estimate the median progression free survival per year (11 months x 365/12 = 335 days).

14b. Calculate estimated usage volume – subsequent osimertinib

The total estimated treatment population for subsequent osimertinib use has been taken from step 13. The committee for TA416 (NICE 2016) noted the pooled AURA dataset (AURA extension and AURA2) for people having osimertinib as second and later lines of treatment. It noted that the estimate of median progression-free survival from this analysis was 11.0 months (335 days).

14c. Calculate estimated usage volume – second line erlotinib

The total treatment population for second line erlotinib has been taken from step 13. Cappuzzo (2010) reported a multicentre, randomised, placebo-controlled phase 3 study on second line erlotinib. Progression free survival was reported as 12.3 weeks (86 days).

14d. Total estimated usage volume – afatinib, erlotinib, gefitinib and osimertinib

The total volume in daily doses for afatinib, erlotinib, gefitinib and osimertinib has been summed from steps 14a, 14b and 14c.

References

1. Office for National Statistics (2018) Cancer Registration Statistics, England: 2016. ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/datasets/cancerregistrationstatisticscancerregistrationstatisticsengland [Accessed: 12/03/2018]
2. National Cancer Registration and Analysis Service (2018). Royal College of Physicians. National lung cancer audit 2018 (for the audit period 2016). London: RCP, 2017. Available from https://www.rcplondon.ac.uk/projects/outputs/nlca-annual-report-2017 [Accessed 03/04/2018]
3. TA374 (2015) Erlotinib and gefitinib for treating non-small-cell lung cancer that has progressed after prior chemotherapy. Final appraisal determination, section 4.3.6. Available from https://www.nice.org.uk/guidance/ta374 [accessed 29/01/2018]
4. Zer A, Natasha B. Leigh l. Second-Line Therapy in Non–Small-Cell Lung Cancer: The DELTA Between Different Genotypes Widens. Journal of Clinical Oncology 32, no. 18 (June 2014) 1874-1881.
5. Li T, Kung H, Mack P et al. (2013). Genotyping and Genomic Profiling of Non–Small-Cell Lung Cancer: Implications for Current and Future Therapies. Journal of clinical oncology. Vol 31, number 8. March 2013.
6. TA416 (NICE 2016). Manufacturers submission (AstraZeneca 2016) Available from https://www.nice.org.uk/guidance/ta416/chapter/3-Evidence [Accessed 29/01/2018]
7. Mazza V, Cappuzzo F. Treating EGFR mutation resistance in non-small cell lung cancer – role of osimertinib. The Application of Clinical Genetics. 2017; 10:49-56. doi:10.2147/TACG.S103471.
8. Scluier J, Berghmans T, Meert A (2015) Advances in target therapy in lung cancer. European respiratory review 24: 23-29. Journal of Clinical Oncology 32, no. 18 (June 2014) 1874-1881.
9. Cappuzzo F, Ciuleanu T, Stelmakh L (2010). Erlotinib as maintenance treatment in advanced non-small-cell lung cancer: a multicentre, randomised, placebo-controlled phase 3 study. The Lancet Oncology. Volume 11, Issue 6, June 2010, Pages 521-529
10. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 17/01/2018]
11. Datapharm Communications Limited (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 17/01/2018]

1. Population who meet SmPC

The SmPC states that abiraterone, cabazitaxel and enzalutamide are indicated for an adult male population only. Therefore, the starting population is adult male new registrations for prostate cancer diagnosis in England.

The ONS publishes new adult cancer registrations for England. This estimate uses the number of new registrations for diagnosis code ICD-10 C61, malignant neoplasm of prostate, in 2016 (ONS, 2018).

2. Prevalence of metastatic hormone-relapsed prostate cancer

It is estimated that between 11.2% (Morgan 2010) and 19.5% (Stevenson 2011) of people with prostate cancer have metastatic hormone-relapsed cancer. These estimates have been used to develop the upper and lower range. The midpoint of the two figures, 15.35%, has been applied to the point estimate.

The prevalence figure and reported range has been applied to the adult population with new prostate cancer in England:

3. Proportion of people who have metastatic hormone relapsed prostate cancer that are expected to receive treatment with abiraterone or enzalutamide before treatment with docetaxel chemotherapy is indicated

Abiraterone and enzalutamide are indicated before chemotherapy in patients with no or mild symptoms. Autio (2013) estimated that 75% of patients with metastatic hormone relapsed prostate cancer have no or mild symptoms and are therefore eligible for treatment with abiraterone or enzalutamide. Expert opinion suggests that 80% of these patients will receive treatment with abiraterone or enzalutamide. The reported proportion and estimated number to receive treatment have been applied to the point estimate and upper and lower range calculated in step 2.

4. Proportion of people that require further treatment after abiraterone or enzalutamide

Expert opinion suggests that treatment with abiraterone or enzalutamide provides clinical benefit in 65% of patients. For the remaining 35% of patients, expert opinion suggests that 55% will require further treatment and go on to receive docetaxel chemotherapy. Of those that receive chemotherapy at that stage, it is estimated that 30% will subsequently receive treatment with cabazitaxel. The estimated number to receive treatment has been applied to the point estimate and upper and lower range calculated in step 3.

5. Proportion of people who have metastatic hormone relapsed prostate cancer that are expected to receive treatment with abiraterone, cabazitaxel or enzalutamide after treatment with first-line docetaxel chemotherapy

Abiraterone, cabazitaxel and enzalutamide are also indicated for use after docetaxel chemotherapy in patients whose disease has progressed during or after docetaxel chemotherapy. Expert opinion suggests that for 55% of patients, first-line treatment with docetaxel chemotherapy will not be successful and that 75% of those patients will then go on to have further treatment with abiraterone, cabazitaxel or enzalutamide. The estimated number to receive treatment has been applied to the point estimate and upper and lower range calculated in step 4.

6. Calculate estimated usage volume

The daily dose for abiraterone has been reported as 1,000mg (WHO 2016) and the daily dose for enzalutamide as 160mg (WHO 2016). The recommended dose of cabazitaxel is 25 mg/m2 every 3 weeks (SmPC 2016). The average body surface area of adult male cancer patients in the UK is reported to be 1.91 m2 (Sacco 2010). Therefore, the daily dose for cabazitaxel has been calculated as 2.27mg. These medicines are used in primary and secondary care; therefore the observed mgs have been extracted for primary and secondary care.

The number of people who have metastatic hormone relapsed prostate cancer likely to receive treatment with abiraterone or enzalutamide before treatment with docetaxel chemotherapy (calculated in step 3) has been multiplied by 365 (days in a year) then multiplied by the recommended daily dose to produce an expected volume of medicine per year in daily doses.

TA391 (NICE 2016) states that treatment with cabazitaxel is stopped when the disease progresses or after a maximum of 10 cycles (whichever happens first). 10 cycles have been calculated as 210 days (SmPC). The number of people who have metastatic hormone relapsed prostate cancer likely to receive treatment with cabazitaxel after treatment with abiraterone or enzalutamide and subsequent docetaxel chemotherapy (calculated in step 4) has been multiplied by 210 (suggested treatment duration in days), then multiplied by the recommended daily dose to produce an expected volume of medicine per year in daily doses.

It has not been possible to identify the proportional split of people who will receive treatment with abiraterone, cabazitaxel or enzalutamide after treatment with first-line docetaxel chemotherapy (calculated in step 5), therefore this total patient group has been assumed to receive treatment for 365 days in a year. This may overestimate the total number of daily doses as the shorter treatment duration of cabazitaxel is not reflected in this calculation.

References

1. Office for National Statistics (2018) Cancer Registration Statistics, England: 2016. ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/datasets/cancerregistrationstatisticscancerregistrationstatisticsengland [Accessed: 06/11/2018]
2. Kirby M, Hirst C and Crawford E (2011) Characterising the castration-resistant prostate cancer population: a systematic review
3. Morgan C, McEwan P, Chamberlain G et al. (2010) Castration-resistant prostate cancer (CRPC): a UK epidemiology study
4. Stevenson M, Jones M, Kearns B et al. (2011) Cabazitaxel for the second line treatment of hormone refractory, metastatic prostate cancer: A Single Technology Appraisal.  ScHARR, the University of Sheffield
5. Autio K, Bennett A, Jia X et al. (2013) Prevalence of pain and analgesic use in men with metastatic prostate cancer using a patient-reported outcome measure
6. NICE Adoption and Impact Programme Reference Panel (2016), clinical experts with an interest in metastatic castration-resistant prostate cancer.
7. Sacco J, Botten J, Macbeth F et al (2010) The average body surface area of adult cancer patients in the UK: a multicentre retrospective study
8. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 16/02/2017]
9. Datapharm Communications Limited (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 16/02/2017].

1. Incidence of colorectal cancer

The SmPC states that trifluridine–tipiracil for previously treated metastatic colorectal cancer is indicated in an adult population only. Therefore, the starting population is adults aged 18 years or older in England with colorectal cancer.

Estimated incidence of colorectal cancer for adult males and females was reported by the Office for National Statistics (2018), 2016 Cancer Registration Statistics for England.

2. Proportion of new cases diagnosed as metastatic colorectal cancer (stage 4)

The proportion of new cases of colorectal cancer that are diagnosed as metastatic (stage 4) in England has been published by Public Health England (22%). This is cancer diagnosis by stage (where known). To develop this estimate, the 95% confidence intervals around the data have been calculated using the exact binomial method (Pezzullo 2009). These figures have been used to inform the upper (22.34% males, 22.39% females) and lower (21.66% males, 21.62% females) range. These proportions have been applied to the point estimate calculated in step 1.

3. Proportion of new cases that are diagnosed as colorectal cancer (stage 2 or 3)

The proportion of new cases of colorectal cancer that are diagnosed at stage 2 or 3 in England has been published by Public Health England (47%). This is cancer diagnosis by stage (where known). To develop this estimate, the 95% confidence intervals around the data have been calculated using the exact binomial method (Pezzullo 2009). These figures have been used to inform the upper (47.41% males, 47.47% females) and lower (46.59% males, 46.53% females) range. These proportions have been applied to the point estimate calculated in step 1.

4. Proportion of people diagnosed with stage 2 or 3 colorectal cancer that progresses to metastatic (stage 4) disease following diagnosis

The proportion of people diagnosed with stage 2 or 3 colorectal cancer that progresses to metastatic (stage 4) disease following diagnosis is between 50% and 60% (Kish 2016). These figures have been used to inform the upper and lower range. A midpoint (55%) has been used to calculate the point estimate. These proportions have been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 3.

5. Number of people with metastatic colorectal cancer (stage 4)

The number of males and females with metastatic disease has been calculated by summing those diagnosed at stage 4 with those diagnosed at stage 2 or 3 whose disease then progresses to stage 4 (steps 2 and 4).

6. Proportion with metastatic colorectal cancer who have first-line and second-line treatments

The proportion of people with metastatic colorectal cancer who receive first-line and second-line treatments (42.5%) has been taken from the manufacturer’s submission for TA405 (Servier Laboratories Ltd 2016). This proportion has been applied as a scaling factor to the point estimate and the lower and upper range calculated in step 5.

7. Proportion of people with metastatic colorectal cancer who are eligible for treatment after other available therapies

The proportion of people with metastatic colorectal cancer who are eligible for treatment after other available therapies (33%) has been taken from table 83 of the manufacturer’s submission for TA405 (Servier Laboratories Ltd 2016). This proportion has been applied as a scaling factor to the point estimate and the lower and upper range calculated in step 6.

8. Proportion of the eligible population who will receive treatment with trifluridine–tipiracil

The proportion of the eligible population who will receive treatment with trifluridine–tipiracil has been estimated as 40% in year 3, based on the manufacturer’s submission for TA405 (Servier Laboratories Ltd 2016). This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 7.

9. Calculate estimated usage volume

The recommended starting dose of trifluridine–tipiracil in adults is 35 mg/m2/dose administered orally twice daily on days 1 to 5 and days 8 to 12 of each 28-day cycle (SmPC 2017). The average body surface area for people with colorectal cancer is 1.96m2 in males and 1.71m2 in females (Sacco 2010). A body surface area of 1.96m2 equates to a total daily dose of 130mg and a body surface area of 1.71m2 equates to a total daily dose of 120mg (SmPC 2017). In a 28 day period the average dose is calculated as 1,300mg (10 x 130mg) for males and 1,200mg (10 x 120mg) for females.

A phase 3 trial (Mayer 2015) to assess the efficacy and safety of trifluridine–tipiracil in a global population of 800 patients reported median overall survival of 7.1 months (217 days).

To calculate the annual use, the 28 day use has been multiplied by 7.75 (217/28). This equates to usage of 10,075mg for males and 9,300mg for females. This has been applied to the eligible population calculated in step 8, applied as a scaling factor to the point estimate and the upper and lower range.

References

1. Office for National Statistics (2018) Cancer Registration Statistics, England: 2016. ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/datasets/cancerregistrationstatisticscancerregistrationstatisticsengland [Accessed: 06/11/2018]
2. Public health England. National Cancer Registration and Analysis Service (2016). Cancer survival by stage at diagnosis for England (experimental statistics): Adults diagnosed 2012, 2013 and 2014 and followed up to 2015. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/bulletins/cancersurvivalbystageatdiagnosisforenglandexperimentalstatistics/adultsdiagnosed20122013and2014andfollowedupto2015 [Accessed 10/07/2017]
3. Pezzullo (2009) Exact Binomial and Poisson Confidence Intervals calculator. Available from: https://statpages.info/confint.html [Accessed 14/01/2017]
4. Kish T, Uppal P. (2016). Trifluridine/tipiracil (lonsurf) for the treatment of metastatic colorectal cancer. Pharmacy and Therapeutics, 41(5), p.314.
5. Servier Laboratories Ltd (2016) The manufacturer’s submission for NICE technology appraisal 405. Available from: https://www.nice.org.uk/guidance/ta405 [Accessed: 24/01/2017].
6. Sacco JJ, Botten J, Macbeth F, Bagust A, Clark P. The Average Body Surface Area of Adult Cancer Patients in the UK: A Multicentre Retrospective Study. Shea BJ, ed. PLoS ONE. 2010;5(1): e8933. doi:10.1371/journal.pone.0008933.
7. Mayer RJ, Van Cutsem E, Falcone A, et al. (2015) Randomized trial of TAS-102 for refractory metastatic colorectal cancer. N Engl J Med. 2015;372(20):1909–1919. 
8. Datapharm Communications Ltd (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 07/07/2017].

1. Population who meet SmPC

The SmPC states that the medicine is indicated for an adult population only. Therefore, the starting population is adults aged 18 years and over in England.

2. Prevalence of cirrhosis

Fleming et al (2008) has reported that the prevalence of cirrhosis for the whole population in the United Kingdom (UK) is 76.3 per 100,000 population.

To develop this estimate, the 95% confidence interval around this prevalence (95% CI 73.6 to 79.1) has been calculated using the exact binomial method (Pezzullo 2009).

The prevalence figure and calculated confidence interval has been applied to the adult population in England:

3. Proportion of people with cirrhosis where overt hepatic encephalopathy is present

Bajaj (2010) has reported that between 30% and 45% of people with cirrhosis have overt hepatic encephalopathy. The mid-point figure (37.5%) and reported range (30% to 45%) has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 2.

4. Proportion of people with overt hepatic encephalopathy who will receive treatment

The manufacturer’s submission for TA337 (Norgine 2013) has reported that 91% of people with overt hepatic encephalopathy will receive treatment for the condition. This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 3.

5. Proportion of people likely to receive treatment with rifaximin

The manufacturer’s submission for TA337 (Norgine 2013) has reported that 40% of people are likely to receive treatment with rifaximin at year 5 in 2018 (rifaximin was first marketed in Jan 2013). This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 4 for all time periods.

6. Calculate estimated usage volume

The SmPC reports that the recommended daily dose for rifaximin is 1,100mg (2 x 550mg tablets). The number of people likely to receive treatment with rifaximin (calculated in step 5) has been multiplied by 365 (days in a year) to produce an expected number of daily doses.

References

1. Office for National Statistics (2018) Annual Mid-year Population Estimates, 2017 ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed: 06/11/2018].
2. Fleming KM, Aithal GP, Solaymani-Dodaran M et al. (2008) Incidence and prevalence of cirrhosis in the United Kingdom, 1992-2001: a general population -based study. Journal of Hepatology 49: 732-8.
3. Pezzullo. (2009) Exact Binomial and Poisson Confidence Intervals calculator. Available from https://statpages.info/confint.html [Accessed 14/01/2016]
4. Bajaj J (2010) Review article: the modern management of hepatic. encephalopathy. Alimentary Pharmacology & Therapeutics 31: 537-547.
5. Amodio P, Del Piccolo F, Petteno E (2001) Prevalence and prognostic value of quantified electroencephalogram (EEG) alterations in cirrhotic patients. Journal of Hepatology 35: 37-45.
6. Romero-Gomez M, Boza F, Garcia-Valdecasas MS (2001) Subclinical hepatic encephalopathy predicts the development of overt hepatic encephalopathy. American Journal of Gastroenterology 96: 2718–23.
7. Rifaximin (TARGAXAN® 550), Norgine Pharmaceuticals Ltd. Single technology appraisal (STA) Specification for manufacturer/sponsor submission of evidence (2013). Available from: https://www.nice.org.uk/guidance/ta337/resources [Accessed: 14/11/2016]
8. Datapharm Communications Limited (2017) The electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 16/02/17]
9. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 16/02/2017]

1. Adult population England

The following estimates the population of adults with primary hypercholesterolaemia (familial and non-familial), or mixed dyslipidaemia who are eligible for treatment with a PCSK9 inhibitor (alirocumab or evolocumab).

The SmPC states that alirocumab and evolocumab are indicated for an adult population. Evolocumab is also indicated in adolescents aged 12 years and over with homozygous familial hypercholesterolaemia in combination with other lipid-lowering therapies. Due to data availability the eligible adolescent population has not been calculated. The starting population for this estimate is adults aged 20 years or older in England; this was because of the prevalence data available. Excluding those aged 18 to 19 may underestimate the eligible population.

2. People with LDL-C of 4 mmol/litre or over and at high risk of CVD

The number of people with primary non-familial hypercholesterolaemia or mixed dyslipidaemia has been estimated through CVD risk and LDL-C levels. The number of people at high CVD risk and with LDL-C of 4 mmol/litre or over was estimated in the manufacturer submission to NICE for alirocumab (Sanofi/ Regeneron 2016) to be 0.19% of the adult population. This proportion has been applied to the population identified in step 1.

3. People with LDL-C of 3.5 mmol/litre or over and a very high risk of CVD

The number of people with primary non-familial hypercholesterolaemia or mixed dyslipidaemia has been estimated through CVD risk and LDL-C levels. The number of people at very high risk of CVD with LDL-C of 3.5 mmol/litre or over was estimated in the manufacturer submission to NICE for alirocumab (Sanofi/ Regeneron 2016) to be 0.17% of the adult population. This proportion has been applied to the adult population stated in step 1.

4. Total at high risk or very high risk of CVD with specified LDL-C levels

The estimated number of people at high risk or very high risk of CVD with specified LDL-C levels from steps 2 and 3 is summed below.

5. Maximal tolerated lipid‑lowering therapy

It is expected that people would reach third line therapy following maximal tolerated lipid-lowering therapy being reached.

A retrospective cohort study using Clinical Practice Research Datalink records linked with Hospital Episode Statistics data (Danese 2017), reported on the management of lipid-lowering therapy in approximately 24,000 patients with cardiovascular events in the UK. It suggests that 20% of patients are receiving high intensity statins 12 months after their cardiovascular event. In addition to this another 2% of patients are similarly treated with a combination of statins plus ezetimibe.

This proportion (22%) has been applied to the point estimate in step 4.

6. Estimated treatment population for alirocumab or evolocumab

The proportion of this population at high risk or very high risk of CVD with specified LDL-C levels to be treated (uptake) with alirocumab or evolocumab is estimated in the NICE resource impact template for alirocumab (NICE 2016) to be 31% in year 2 (2016). This figure has been adjusted +/- 10% to create a range (27.9% - 34.1%). These proportions have been applied as a scaling factor to the point estimate in step 5.

7. Primary heterozygous familial hypercholesterolaemia with LDL-C of 5 mmol/litre or over and no history of CVD

The number of people with primary heterozygous familial hypercholesterolaemia and no history of CVD with LDL-C of 5 mmol/litre or over was estimated in the manufacturer submission to NICE for alirocumab (Sanofi/ Regeneron 2016) to be 0.01% of the adult population. This proportion has been applied to the adult population stated in step 1.

8. Primary familial heterozygous hypercholesterolaemia with LDL-C of 3.5 mmol/litre or over and a history of CVD

The number of people with primary heterozygous familial hypercholesterolaemia and a history of CVD with LDL-C of 3.5 mmol/litre or over was estimated in the manufacturer submission to NICE for alirocumab (Sanofi/ Regeneron 2016) to be 0.02% of the adult population. This proportion has been applied to the adult population stated in step 1.

9. Total primary heterozygous familial hypercholesterolaemia

The estimated number of people with primary heterozygous familial hypercholesterolaemia and at high risk or very high risk of CVD with specified LDL-C levels has been summed from steps 7 and 8.

10. Maximal tolerated lipid‑lowering therapy primary heterozygous familial hypercholesterolaemia

It is expected that people would reach third line therapy following maximal tolerated lipid-lowering therapy being reached.

A retrospective cohort study using Clinical Practice Research Datalink records linked with Hospital Episode Statistics data (Danese 2017), reported on the management of lipid-lowering therapy in approximately 24,000 patients with cardiovascular events in the UK. It suggests that 20% of patients are receiving high intensity statins 12 months after their cardiovascular event. In addition to this another 2% of patients are similarly treated with a combination of statins plus ezetimibe.

This proportion (22%) has been applied to the point estimates and the upper and lower ranges calculated in step 9.

11. Estimated treatment population for alirocumab or evolocumab primary heterozygous familial hypercholesterolaemia

The proportion of this population to be treated (uptake) with alirocumab or evolocumab is estimated in the NICE resource impact template for alirocumab (NICE 2016) to be 41% in year 2 (2016). This figure has been adjusted +/- 10% to create a range (36.9% - 45.1%). These proportions have been applied as a scaling factor to the point estimate and the upper and lower ranges calculated in step 10.

12. Calculate estimated usage volume (ADD)

The SmPC recommended dose for alirocumab is 75mg administered once every 2 weeks.  For those requiring a larger LDL-C reduction, a dose of 150mg once every 2 weeks or 300mg once every 4 weeks is recommended. The ADD for the 75mg/1ml solution for injection has been calculated as 5.4mg (75 x 26 / 365). The ADD for the 150mg/1ml solution for injection has been calculated as 10.7mg (150 x 26 / 365).

The SmPC recommended dose for evolocumab is 140mg every 2 weeks or 420mg once monthly. Evolocumab is recommended by NICE as an option for treating primary hypercholesterolaemia or mixed dyslipidaemia, only if the dosage is 140 mg every 2 weeks. The ADD has been calculated as 10mg (140 x 26 / 365). This is the same as the WHO DDD for evolocumab.

The number of people who are estimated to receive treatment with alirocumab or evolocumab (summed from step 6 and 11) has been multiplied by 365 (days in a year) to produce an expected volume of medicine per year in ADDs.

References

1. Office for National Statistics (2018) Annual Mid-year Population Estimates 2017. ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed: 06/11/2018].
2. Alirocumab for treating primary hypercholesterolaemia and mixed dyslipidaemia, TA393, manufacturer’s submission (Sanofi/Regeneron), 2016. Available from: https://www.nice.org.uk/guidance/ta393/evidence [Accessed: 27/06/2017].
3. Danese, M.D., Gleeson, M., Kutikova, L., Griffiths, R.I., Khunti, K., Seshasai, S.R.K. and Ray, K.K., 2017. Management of lipid-lowering therapy in patients with cardiovascular events in the UK: a retrospective cohort study. BMJ open, 7(5), p.e013851.
4. NICE (2016) Costing template: TA394. Manchester: NICE. Available from: https://www.nice.org.uk/guidance/ta394/resources [Accessed: 01/09/2017].
5. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 01/09/2017]
6. Datapharm Communications Limited (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 01/09/2017].

1. Population who meet SmPC

The SmPC states that both medicines are indicated for an adult population only. Therefore, the starting population is males and females aged 18 years and over in England.

2. Prevalence of multiple sclerosis

Mackenzie et al (2013) has reported the prevalence (and 95% confidence intervals) of multiple sclerosis for the whole population in the United Kingdom (UK):

Female prevalence per 100,000 UK pop in 2010: 285.8 (95% CI 277.7 to 293.1)

Male prevalence per 100,000 UK pop in 2010: 113.1 (95% CI 108.6 to 117.7)

The relevant prevalence figure and associated 95% confidence interval has been applied to the adult population in England:

3. Proportion of people with relapsing-remitting multiple sclerosis

Kobelt et al (2006) has reported that 35.5% of patients with multiple sclerosis in the UK have relapsing-remitting disease. This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 2.

4. Proportion of people who have previously received, or are currently receiving disease modifying therapy

The NICE costing template for TA320 (NICE 2014) has estimated that around 38.75% of people have previously received or are currently receiving disease modifying therapy. This is based on a longitudinal study of multiple sclerosis in south west England (Zejicek 2010). The associated range (33.75% to 43.75%) has been reported as part of the sensitivity analysis in the NICE costing template (NICE 2014). These proportions have been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 3.

5. Proportion of people likely to receive treatment with dimethyl fumarate or teriflunomide

The NICE costing template for TA320 (NICE 2014) has estimated that 55% of people will receive treatment with either dimethyl fumarate or teriflunomide. This proportion has been applied as a scaling factor to the point estimate and the upper and lower range calculated in step 4.

6. Calculate estimated usage volume

The number of people who are likely to receive treatment with dimethyl fumarate or teriflunomide (calculated in step 5) has been multiplied by 365 (days in a year) to produce an expected volume of medicine per year in daily doses. The daily dose is taken from the WHO DDD for dimethyl fumarate (480mg) and teriflunomide (14mg).

References

1. Office for National Statistics (2018) Annual Mid-year Population Estimates, 2017 ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed: 06/11/2018].
2. Mackenzie IS, Morant SV, Bloomfield GA et al. (2013). Incidence and prevalence of multiple sclerosis in the UK 1990–2010: a descriptive study in the General Practice Research Database. Journal of Neurol Neurosurg Psychiatry 10: 1136.
3. Kobelt G, Berg J, Lindren P (2006) Costs and quality of life of multiple sclerosis in the United Kingdom. European Journal of Health Economics 7: 96-104.
4. Zajicek J, Ingram W, Vickery J et al. (2010) Patient-orientated longitudinal study of multiple sclerosis in south west England (The south west impact of multiple sclerosis project, SWIMS)1: protocol and baseline characteristics of cohort, BioMedCentral Neurology 10: 88.
5. NICE (2014) Costing template: TA320. Manchester: NICE. Available from: https://www.nice.org.uk/guidance/ta320/resources [Accessed: 10/08/2015].
6. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 16/02/2017]
7. Datapharm Communications Limited (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 16/02/2017].

1. Population in England

Etelcalcetide is recommended for use in adults only.

2. Proportion of adults having dialysis

In their 2017 annual report, the UK Renal Registry reported that, in 2016, 24,663 people in England were having dialysis. Based on the mid-year population estimate for England in that year (43,482,790), this was 56.7 per 100,000 of the adult population. This proportion has been applied to the population identified in step 1 and 95% confidence intervals were calculated (56.0 to 57.4 per 100,000) using the exact binomial method (Pezzullo 2009).

3. Proportion of adults having haemodialysis

The UK Renal Registry reported that 21,560 (87.4%) of those dialysis patients were on haemodialysis. This proportion has been applied to the dialysis population calculated in step 2 and 95% confidence intervals were calculated (87.0 to 87.8%) using the exact binomial method.

4. Less the proportion of adults having home haemodialysis

Unlike most other treatments for secondary hyperparathyroidism, etelcalcetide is administered either during dialysis into the venous line at the end of treatment during rinse-back, or intravenously immediately after. This method of administration will be less suitable for those on home haemodialysis as they would need to visit the dialysis centre for administration of etelcalcetide 3 times a week, defeating the benefit of home haemodialysis.

The UK Renal Registry reported that, in 2016, 4.4% of all dialysis patients were on home haemodialysis. This proportion has been applied to the dialysis population reported in step 2 and then deducted from the haemodialysis population calculated in step 3.

5. Proportion of adults with secondary hyperparathyroidism and uncontrolled parathyroid hormone levels

Treatment with a calcimimetic is commenced after patients have tried standard therapy (dietary control, phosphate binders and vitamin D analogues) and failed to attain either a satisfactory reduction in parathyroid hormone levels or to reach target levels, which current guidance recommends should be in the range of 2 to 9 times the upper limit of normal for patients with CKD G3a to G5.

At this stage, parathyroidectomy is an option – however, it is expected that most patients and their clinicians would prefer to opt for a minimally invasive pharmacological-therapy, before resorting to an invasive surgical procedure with its associated risks. Therefore, the number of patients having parathyroidectomy at this stage is likely to be negligible and not considered in this estimate.

The technology appraisal for cinacalcet recommends that treatment should only be initiated in patients with ‘very uncontrolled’ parathyroid hormone levels, defined as greater than 800pg/ml. However, it is expected that some clinicians may consider treatment with a calcimimetic when levels exceed the target upper limit of 9 times the upper limit of normal for patients with CKD G3a to G5, despite standard treatment. This translates approximately to 585 to 677pg/ml. Therefore, a literature search was undertaken to identify the number of dialysis patients who are reported to have parathyroid levels above 800pg/ml to inform the lower range and the number of patients with levels above 9 times the upper limit of normal to inform the upper range of patients who have uncontrolled hyperparathyroidism and are likely to receive treatment with cinacalcet.

Fernandez-Martin et al, as part of the COSMOS study, which surveyed European haemodialysis centres (including the UK), reported that 7.4% of haemodialysis patients had a parathyroid hormone level greater than 800pg/ml. This proportion has been applied to the population calculated in step 4 and used to estimate the lower limit.

The UK Renal Registry states that 18.5% of haemodialysis patients have a parathyroid hormone level above 677pg/ml (9 times the upper limit of normal). This proportion has been applied to the population calculated in step 4 and used to estimate the upper limit.

The point estimate is taken as the mean of the lower and upper proportions (13.0%).

6. Proportion of adults with an adjusted calcium level at or above the lower limit of normal

The SmPC for both etelcalcetide and cinacalcet advise that corrected serum calcium should be at or above the lower limit of the normal range prior to administration of first dose, a dose increase, or reinitiation after a dose stop.

The proportion of patients with an adjusted serum calcium greater than 2.2mmol/L, as reported by the UK Renal Registry (88.3%), was used to calculate the proportion of those patients estimated in step 5, who would be appropriate for initiating treatment with a calcimimetic.

7. Proportion of adults for whom cinacalcet is not suitable

NICE recommends that etelcalcetide should only be used if a calcimimetic is indicated but cinacalcet is not suitable. It is expected that most patients will be considered for etelcalcetide if they cannot tolerate cinacalcet, their condition does not respond to it, or they are unlikely to adhere to the regimen. While other reasons for not being suitable for cinacalcet are possible (such as administration difficulties or absorption disorders), these are thought to be a small proportion and have not been considered in this estimate.

7a. Proportion of adults who do not tolerate cinacalcet

Withdrawals due to adverse events in published trial data were used to estimate the proportion of patients that may not tolerate cinacalcet and discontinue treatment. A literature search was undertaken to identify clinical trials which included a European or English population, which evaluated cinacalcet in patients on dialysis. Six studies were identified, and they reported a range of withdrawals due to adverse events from 4.0% to 18.1%. These figures have been used to estimate the range of proportions of people who may discontinue cinacalcet due to tolerability and a point estimate calculated from the mean of all values reported in all six papers (11.5%).

7b. Proportion of adults whose condition does not respond to cinacalcet

NICE recommends that treatment with cinacalcet should only be continued if a reduction in parathyroid hormones of 30% is achieved. This was the measure used to estimate the proportion of patients whose condition does not respond to cinacalcet in this estimate. Using the same criteria as above, two trials were identified which reported this measure as an outcome, and they reported that 63 and 64% of patients achieved a 30% reduction in parathyroid hormone levels. This is translated into 37 and 36% of patients whose condition does respond to cinacalcet and has been used to calculate an upper and lower range. The mean has been calculated as the point estimate (36.5%).

7c. Proportion of adults who do not adhere to cinacalcet

People with CKD are usually taking many oral medicines and non-adherence is acknowledged to be a concern in this population. In the EVOLVE trial, it was reported that 3.5% of people taking cinacalcet discontinued the trial due to ‘non-compliance’. Whilst adherence in a clinical trial is likely to be higher than in real-life, given the lack of data in the real-world setting, this proportion has been used to estimate the proportion of people who may not adhere to cinacalcet. This is likely to be an underestimation.

8. Estimated treatment population

Adding together the number of people who do not tolerate cinacalcet (step 7a), those whose condition does not respond (step 7b) and those who do not adhere to the regimen (step 7c) gives the number of people who are eligible for treatment with etelcalcetide. Parathyroidectomy is an option at this stage; however, the numbers of patients opting for parathyroidectomy rather than a less-invasive pharmacological therapy is thought to be negligible. It has therefore been assumed that, once accounting for those in whom cinacalcet is not suitable, most patients will be prescribed etelcalcetide. Therefore, all those eligible to receive etelcalcetide are included in the estimated treatment population.

9. Calculate estimated usage volume

The DDD for etelcalcetide is reported as 2.1mg (WHO 2017).

The number of people estimated to receive etelcalcetide (calculated in step 8) has been multiplied by the number of days in a year (365) to give the annual usage in DDDs.

References

1. Office for National Statistics (2018) Annual Mid-year Population Estimates 2017. ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed 06/11/2018]
2. Office for National Statistics (2017) Annual Mid-year Population Estimates 2016. ONS: Newport. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/datasets/populationestimatesforukenglandandwalesscotlandandnorthernireland [Accessed 06/11/2018]
3. Pezzullo. (2009) Exact Binomial and Poisson Confidence Intervals calculator. Available from https://statpages.info/confint.html [Accessed 17/06/2016].
4. Byrne C, Caskey F, Castledine C et al. (2018) UK Renal Registry 20^th Annual Report of the Renal Association. Nephron 2018;137: Suppl.1
5. Fernandez-Martin JL, Carrero JJ, Benedik M et al. (2013) COSMOS: the dialysis scenario of CKD-MBD in Europe. Nephrology Dialysis Transplantation 28: 1922- 1935
6. Block GA, Martin KJ, de Francisco ALM et al. (2004) Cinacalcet for secondary hyperparathyroidism in patients receiving hemodialysis. New England Journal of Medicine 350: 1516-1525
7. Urena-Torres P, Bridges I, Christiano C et al. (2013) Efficacy of cinacalcet with low-dose vitamin D in incident haemodialysis subjects with secondary hyperparathyroidism. Nephrology Dialysis Transplantation 28: 1241 - 1254
8. The EVOLVE Trial Investigators (2012) Effect of cinacalcet on cardiovascular disease in patients undergoing dialysis. New England Journal of Medicine 367: 2482 - 2492
9. Ketteler M, Martin KJ, Wolf M et al (2012) Paricalcitol versus cinacalcet plus low-dose vitamin D therapy for the treatment of secondary hyperparathyroidism in patients receiving haemodialysis: results of the IMPACT SHPT study. Nephrology Dialysis Transplantation 27: 3270 - 3278
10. Messa P, Macario F, Yaqoob M et al (2008) The OPTIMA Study: Assessing a new cinacalcet (Sensipar/Mimpara) treatment algorithm for secondary hyperparathyroidism. Clinical Journal of the American Society of Nephrology 3: 36 – 45
11. Raggi P, Chertow GM, Urena-Torres P et al (2011) The ADVANCE study: a randomized study to evaluate the effects of cinacalcet plus low-dose vitamin D on vascular calcification in patients on haemodialysis. Nephrology Dialysis Transplantation 26: 1327 – 1339
12. WHO (2017) ATC/DDD Index 2017. [WWW] WHO. Available from: https://www.whocc.no/atc_ddd_index/ [Accessed: 10 May 18]
13. Datapharm Communications Limited (2017) The Electronic Medicines Compendium [WWW]. Available from: https://www.medicines.org.uk:443/emc/ [Accessed: 10 May 18]