The United Kingdom now has its first biosimilar insulin, following the 2015 launch of Abasaglar, a version of U100 insulin glargine made by Eli Lilly. The same insulin, but branded Basaglar, has also been approved by the Food and Drugs Administration (FDA) in the United States, this following on from their first biosimilar approval – Zarxio (filgrastim) – in March 2015. The term ‘biosimilar’ is preferred to ‘generic’ because the complexity of insulin production techniques mean that small differences from the already-approved reference product cannot be excluded. The biosimilar must show it has no clinically meaningful differences in terms of safety and effectiveness from the reference agent and only minor differences in clinically inactive components are allowable in biosimilar products.

These events are in contrast to the decision made in November 2015 by the Committee for Medicinal Products for Human Use (CHMP) to recommend refusal for marketing authorisation of Solumarv. Solumarv is another biosimilar insulin, this time a short-acting one, made by Marvel Lifesciences Limited and the reference product is Humulin S.

The different responses probably indicate that the regulatory authorities take into account more issues other than just the clinical studies. For example, the CHMP commented that Marvel ‘did not define the manufacturing process for Solumarv in sufficient detail. As such, it was not possible to show that Solumarv used in clinical studies was representative of batches intended for the market and that its quality was comparable to Humulin S’s’. There are also concerns about companies’ ability to maintain supplies of their biosimilar product; since the biosimilars are inevitably cheaper than reference insulins, widespread switching is possible and any supply issues could have major consequences. In this respect Eli Lilly’s heritage in diabetes is important – they were the first company to commercially produce insulin in 1921 and have done so ever since.

Perhaps these considerations should also apply to generic molecules, where pharmacy-led switching is now commonplace? A case in point is that of pioglitazone (Actos, produced by Takeda) which went off-patent in 2012. Immediately, generic versions became available at a large price discount and NICE in the first draft of its type 2 diabetes guideline update moved pioglitazone into the preferred second-line option following metformin. Then in December 2015, United Kingdom Medicines Information (UKMi) issued a memo on ‘shortage of supply’ for all strengths of pioglitazone. Limitation of prescribing to 28 days was suggested as an option to reduce the risk of stocks running out (odd, given that almost all patients renew their prescribed medications on a monthly basis) with medication review recommended for those patients who cannot obtain supplies. What was not mentioned was that the shortage was down to supply-chain issues in the generic companies, with Takeda being able to make up the short-fall (which would probably not have happened if patients had remained on or been initiated with the branded reference product). I suspect the costs of this episode will never be calculated and there will be little account taken of the concerns and inconvenience suffered by patients.

More insulin biosimilars will emerge over the coming years and more anti-diabetic medicines will lose patent protection. The potential for future mishap seems very real…

Professor Steve Bain

The project to move the management of patients with type 2 diabetes (T2DM) from secondary care specialists to primary care began in the 1970’s but gained major momentum over the last ten years. It is a central pillar of health policy in the UK but on what evidence is this based?

Most NHS initiatives are based on cost reduction but is this a cheaper option? As a secondary care specialist, my costs for outpatient review are modest; an office, computer, sphygmomanometer, monofilament and support staff. I presume that these are no different to primary care but that the attributed costs are higher due to a costing formula? Furthermore, if my practice is physically moved into primary care, then the numbers of patients reviewed will decrease since the SpR training grades are not be in attendance. Finally, my availability to supervise on-call and in-patient general medicine patients is reduced, needing more resource expenditure in those areas.

Has the increase in primary care management of T2DM led to improvements in outcomes? It seems not, in terms of the surrogate marker of HbA1c, which remains stubbornly unchanged. Furthermore, ‘treatment inertia’ indicating slow escalation of pharmacotherapy including oral medication and insulin is reported to be worse in the UK than in most developed economies. This is not a criticism of general practitioner colleagues. There are currently eight different classes of anti-diabetic medications available in the UK with over thirty commonly used insulin preparations, three of which have been licenced for use in the last twelve months. As a specialist who focusses on diabetes, I struggle to keep up-to-date; why would anyone expect a generalist practitioner and supporting practice nurse to have this specialist knowledge?

The major push for a primary care takeover of T2DM is the massive numbers of people with the condition. Surely this has to be the focus? We need measures to reverse the epidemic and clever means (involving information technology) to allow specialist management of people with this complex disease.

Professor Steve Bain

The long-awaited NICE guideline ‘Type 2 diabetes in adults: management’ (new guideline [NG] 28) was finally published on 2nd December 2015, almost eleven months after the first draft was released. The first draft caused uproar in the diabetes community and led to the highly unusual step of NICE issuing a second draft for consultation in June 2015. The whole process has led to a guideline which is much more in-keeping with international practice, as outlined in the position statement from the American Diabetes Association and European Association for the Study of Diabetes, updated in January 2015.

NG28 replaces the previous NICE clinical guideline (CG) 66, published in 2008 and the subsequent glycaemic update (CG87) which was published one year later. The technology assessments relating to liraglutide (TA203) and exenatide prolonged-release are also superseded by this document. This means that TAs for the three SGLT2 inhibitors remain extant.

The guideline contains recommendations for managing type 2 diabetes in adults, and focuses on patient education, dietary advice, managing cardiovascular risk, managing blood glucose levels, and identifying and managing long-term complications.

The advice on patient education is standard, with quality-assured structured education programmes mandated and regular audit of outcomes. I guess the only issue here (and, of course, it is a huge one) is whether local health economies have the funding to support this activity. Dietary advice and blood pressure management are similarly uncontentious with the vast majority of it being a re-stating of the recommendations in CG66 (from 2008). Lipid management is not dealt with and is left to CG181 published in July 2014; as an aside, I do regard this as contentious, given the focus on non-HDL cholesterol and targets based on pre-statin lipid levels. Both of these recommendations make routine audit of lipid management much more difficult.

Antiplatelet advice has changed to ‘Do not offer antiplatelet therapy (aspirin or clopidogrel) for adults with type 2 diabetes without cardiovascular disease’. Practitioners will need to decide how to manage those patients who are currently receiving aspirin, based on NICE’s previous advice (which recommended it for all T2DM patients over the age of 50 years, unless blood pressure was sub-optimally controlled). I would be inclined to leave things unchanged, given that the ASCEND Trial investigating the use of aspirin in ‘low risk’ patients with T2DM is due to complete in 2017.

Self-monitoring of blood glucose (SMBG) is generally discouraged in NG28 but by the time the various caveats (including DVLA requirements) are taken into account, there will be few patients for whom SMBG is completely excluded.

By far the most controversial areas of the drafts of this guideline involved blood glucose management and some of these remain (although they are greatly tempered). Targets for HbA1c are problematic. Despite widespread evidence of poor achievement of HbA1c targets in the UK (attributed to ‘clinical inertia’), NG28 recommends that levels for treatment escalation (58mmol/mol) are higher than the target for treatment (53mmol/mol). This ‘treat-to-failure’ approach is unlikely to promote better overall control.

The controversies over drug therapy for blood glucose control have largely receded following the down-playing of repaglinide and the formal inclusion of SGLT2 inhibitors into the treatment algorithm (which is likely to be the most widely read page of the entire document). DPP-4 inhibitors are now placed at the top of the lists of first and second-line intensification steps (and for monotherapy in metformin-intolerant patients) and modified-release metformin is included as an option.

The place of GLP-1 receptor agonists has diminished somewhat in NG28 in that they are now only an option when triple oral therapy including metformin has failed and they are not included in the metformin-intolerant algorithm at all. As before, there is a BMI threshold for GLP-1RA initiation (introduced on the basis of cost, rather than clinical evidence) and stopping rules apply which are made up of both HbA1c (11mmol/mol) and weight (3%) reductions. At least the combination of GLP-1RA and insulin is recognised (in a specialist environment) and the 1.2mg limit on liraglutide seems to have disappeared.

The advice on insulin is necessarily vague such as ‘offer NPH insulin injected once or twice daily according to need’ and wouldn’t encourage inexperienced professionals to embark on this therapy (probably a good thing). The premix advice doesn’t seem to have any evidence base behind it (e.g. a higher HbA1c of 75 mmol/mol promoting its use) and it is unclear why NG28 would imagine that any patient would prefer not to inject immediately before a meal when the option is doing so 30 minutes before.

Overall NG28 is to be welcomed and the challenge will be to get clinicians and patients to follow it in a way that prevailing high levels of HbA1c in the UK indicate was not the case for CG87 (which essentially advised the same targets). Another positive is the commitment of NICE to allow for more frequent review and update of the guideline since the pace of change in diabetes therapies is so rapid.

Professor Steve Bain