Article by Dr Jon Sisson
30 years ago, when I was a newly qualified doctor, there were 4 pharmacological options in the UK for type 2 diabetes mellitus (T2DM): metformin, sulphonylureas, acarbose, and insulin.
Acarbose had limited efficacy and was later withdrawn. The other three options remain important today. However, new options were needed that had equal or improved efficacy, were better tolerated, and could be used in a wider range of patients. This became more important as tighter treatment targets were introduced. For example, metformin commonly causes gastrointestinal adverse effects, whilst sulphonylureas and insulin carry the risks of hypoglycaemia and weight gain – the latter worsening insulin resistance further.
This review will focus on the new classes launched in the United Kingdom (UK) over those 30 years, touching on the implications for companies with developmental products aiming to file over the next 5-10 years. Those with minor roles (nateglinide, repaglinide, orlistat) or withdrawn for commercial reasons (albiglutide, dulagliptin) are not included.
The thiazolidinediones: A series of unfortunate events
The first of these was troglitazone, licenced in 1997 under the European Union (EU) Mutual Recognition procedure in a subset of member states, including the UK. It was however withdrawn within 6 weeks of its launch, due to adverse events of severe liver damage, resulting in death in some cases. The lessons from troglitazone (and others) are reflected in current non-clinical and clinical regulatory guidance on the evaluation of drug induced liver injury (DILI) (1,2).
Such idiosyncratic adverse events remain a challenge. Monitoring may not be fully effective in preventing serious outcomes, and not all patients with particular risk factors can be identified a priori, despite advances in the understanding of DILI. Even with a large population exposed in Phase III, rare events may not be seen – as we know from the “rule of 3” – and even this rule supposes that all events are both properly documented and correctly ascribed. Finally, if there is a rapid take-up after launch, cases can occur before any remedial action can be taken to update monitoring and safety warnings.
Despite the events with troglitazone, the class was effective in glycaemic control, and in 2000, rosiglitazone and pioglitazone were licenced across the EU, under the new centralised route. Both had efficacy comparable to metformin or sulphonylureas, without intrinsic risk of hypoglycaemia. There was no signal of liver toxicity from pre-clinical and clinical development, and more data was available on possible mechanisms for troglitazone-induced hepatotoxicity. There was some regulatory concern about fluid retention and heart failure, with some suggesting that further safety studies be performed prior to Marketing Authorisation. However, the majority at CHMP felt there was an unmet need in a subgroup of patients, and that with appropriate labelling and risk management, further studies could be undertaken as a post-marketing commitment.
After a decade of safety warnings being progressively added, rosiglitazone was suspended in 2010, and later the company chose not to renew the Marketing Authorisation. The suspension was primarily due to cardiovascular safety, although there were other issues under review.
Current CHMP guidance for diabetes products highlights the need to exclude drug-induced excess cardiovascular risk before marketing (3). The CHMP reflection paper on assessment of cardiovascular safety (4) advises a composite cardiovascular safety endpoint and cites an upper bound of 1.8 for the hazard ratio confidence interval. The choice of composite endpoint and upper bound may differ depending on the circumstances, and critically this approach requires an appropriate study population and length of follow-up to give an adequate number of observed cardiovascular events.
In 2007, a pioglitazone macrovascular outcome study failed on its primary composite endpoint, and showed increased incidence of oedema, weight gain and heart failure (5). Pioglitazone continues to have a role in some patients, but in a restricted group, in a second- or third-line role.
A new hope: the incretin-based therapies
The GLP-1 (Glucagon-like peptide 1) agonist exenatide was licenced in 2006. This was followed by liraglutide, lixisenatide, dulaglutide and semaglutide. These can be used in a range of combinations, including with insulin. Exenatide had the drawback of twice daily subcutaneous injection, liraglutide and lixisenatide required a single daily injection. When the prolonged release version of exenatide was approved, this permitted once weekly dosing, which is matched by dulaglutide and semaglutide.
Oral therapies based on the incretin pathway followed almost in parallel. The first of these DPP-IV (dipeptidase 4) inhibitors were sitagliptin and vildagliptin in 2007, followed by saxagliptin, linagliptin and alogliptin. Monotherapy with these is not associated with weight gain or hypoglycaemia, and indications include 2nd line monotherapy as well as dual and triple combination therapy.
In 2019, liraglutide gained an indication to treat children and adolescents aged 10 and over. In 2020, a novel formulation of semaglutide containing an absorption enhancer was approved as a once daily tablet.
With a more crowded market, benefits beyond blood glucose reduction and safety became more commercially important. The ideal therapy is often a once daily dose, without the need for titration or dose adjustment in special populations. The importance of controlling glucose variability throughout the day was more recognised, instead of a sole focus on HbA1c. The need for new therapies to have positive additional effects also intensified, especially weight loss. A separate liraglutide licence was approved, with adjunctive indications in weight management - for diabetics and non-diabetics. A similar licence for semaglutide is currently pending EC decision. Importantly, cardiovascular outcome studies have shown clear benefits for some of the GLP-1 agonists. Innovation in incretin-based therapies is continuing; for example, the GLP-1/GIP agonist tirzepatide has recently been submitted to regulators (6).
Transporter 2: the gliflozins
The first of the SGLT-2 (sodium glucose cotransporter 2) inhibitors was dapagliflozin in 2012, followed by canagliflozin, empagliflozin, and ertugliflozin. Indications vary, they can be used in monotherapy or combination. Recent regulatory updates have focused on longer-term outcome studies and expansion of therapeutic indications. Dapagliflozin now has additional indications in heart failure and renal disease. All in the class have reported cardiovascular outcome data.
The Next Generation
In 2007, metformin (Glucophage SR) received a UK licence for adjunctive use in pre-diabetes. There is an opportunity for new and established agents to expand into pre-diabetes and influence the level of consensus in this area. In established disease, there is increasing focus on lifestyle strategies; the DiRECT trial already showing that remissions sustained to 24 months are possible in more than a third of patients (7). Where this is not possible, research into halting or reversing beta-cell dysfunction is continuing, including research into the identification of patients most at risk for beta cell decline.
Alongside this and the development of new drugs, there are many opportunities to better understand and develop existing therapies. After all, we are still learning about metformin and sulphonylureas. Experience has told us that the launch posology may not be the optimal one, and that apparent “me-too” products can drive incremental improvements. Alternative delivery systems have been studied for the GLP-1 agonists, further formulation work for these and other therapies may allow greater adherence to long-term therapy.
There is also an opportunity for agents currently used in T2DM to be considered as adjunctive treatments in type 1 diabetes, with recent research including liraglutide and dapagliflozin. The dual SGLT-1 and 2 inhibitor sotagliflozin has recently been licenced in this context.
For existing and new therapies, companies can benefit from further exploring the clinical and metabolic outcomes of long-term treatment. Not only can this provide differential advantages over competitors and influence treatment guidelines, it can potentially enable a widening of the therapeutic indications and extend the lifecycle. Patient populations can also be extended, particularly in children and adolescents with T2DM, given increasing prevalence in this group.
Other than the pharmaceutical options covered in this article, there have been significant strides in the general management of T2DM, and this will continue to be important. Digitally enabled monitoring technology will have an increasing role in initial development, in generating real-world data and to optimise individual treatment.
Whatever the sequels, tranScrip is ready to be the best supporting actor.
The views and opinions expressed are those of the author.
Note: Current product information can be found on the EMA (8) and MHRA (9) websites. For centrally authorised products the EMA website includes initial assessment reports and details of all post-licencing activity.
- Reflection paper on non-clinical evaluation of drug induced liver injury (DILI). EMEA/CHMP/SWP/150115/2006
- Drug-Induced Liver Injury: Premarketing Clinical Evaluation. FDA 2009
- Guideline on clinical investigation of medicinal products in the treatment or prevention of diabetes mellitus. CPMP/EWP/1080/00 Rev. 2
- Reflection paper on assessment of cardiovascular safety profile of medicinal products. EMA/CHMP/50549/2015
- Lancet 2005;366:1279-1289
- Lancet Diabetes Endocrinol. 2019;7(5):344-355