The United Kingdom Prospective Diabetes Study [1] has shown that beta-cell function declines over time, and insulin will be needed to achieve good glycaemic control in type 2 diabetes mellitus (T2DM). Recent studies show that initiating insulin early, especially when HbA1c (Glycated haemoglobin) is > 9% improves glycaemic control and patient outcome [2]. Meta-analysis by Kramer, et al. [3] showed that short term intensive insulin therapy for 2-3 weeks had favourable outcomes on recovery and maintenance of beta-cell function, glycaemic remission besides increment in beta-cell function and decrease insulin resistance.
The decision of health care professional (HCP) to start insulin therapy is precipitated by [4]:
• Worsening symptoms of hyperglycaemia.
• A persistently elevated HbA1c despite maximum tolerated doses of oral hypoglycaemic therapy.
• Inter-current illness or patient started on steroid therapy.
It is imperative to review the patient's diet and compliance with medicine before making a decision to start insulin therapy. National Institute for Health and Care Excellence (NICE) [5] recommends using insulin based therapy when HbA1c remains ≥ 7.5% or 58 mmol/mol after using suitable oral hypoglycemic agents or GLP-1 analogs if appropriate. Combination of most of the oral hypoglycaemic agents (OHAs) with insulin is very effective and safe, provided you consider mechanism of action of different types. Consider the on-going need for the OHAs while patient is on insulin. The Medicines and Healthcare products Regulatory Agency [6] has advised to be careful while using insulin with pioglitazone as it can increase risk of heart failure, especially in patients at risk. Continue metformin if there is no contraindication or intolerance [5]. In patients where metformin can't be used due to renal impairment, dipeptidyl peptidase-4 (DPP-4) inhibitors in dosage based on eGFR (estimated glomerular filtration rate) can be used along with insulin, as they are weight neutral and increases sensitivity of islet cells to glucose, without risk of hypoglycaemia. It would be prudent to continue agents like Sodium glucose co-transporter (SGLT2) inhibitors or Glucagon Like Peptide (GLP-1) RA if required. SGLT2 inhibitors act through insulin independent pathways and reduce the total insulin dose required due to the glycosuria. GLP-1 receptor analogs has been approved by NICE to be used along with basal insulin only. Sulphonylureas or insulin secretagogues should be avoided in combination of insulin in view of increased risk of hypoglycaemia, unless used in reduced doses. The various insulin types available are:
1) Rapid acting: Begins to work in 15 minutes, peak around 1 hour and continues to work for about 2-4 hours.
2) Ultra fast acting : Apidra (Fiasp) Which begins to act 4-7 minutes before regular apidra and lasts for around 3 hours.
3) Short acting: Reaches systemic circulation in 30 min, peaks after about 2-3 hours and remains active for around 3-6 hours.
4) Intermediate acting: Onset of action after 2-4 hours, peaks 4-12 hours later and remains active for about 12-18 hours.
5) Long acting: Gradually absorbed from injection site due to unique preparation. Its activity can range from 18-24 hours. The ultra-long acting insulin degludec, has plasma concentrations measurable beyond 24 hours allowing for flexible dosing.
The table, summarises various insulin types and preparations available in market (Table 1).
Table 1: Various insulin types and preparations available in market. View Table 1
There are many reasons that patients do not achieve glycaemic targets. These can be patient factors, disease related factors and also clinical inertia, for example, delay in intensification of treatment [7]. (Table 2)
Table 2: The causes of secondary drug failure to Oral hypoglycemic agents [31]. View Table 2
When initiating insulin in T2DM, NPH insulin is the preferred option either once or twice daily, according to the need, unless
• Individual's lifestyle is affected by symptomatic hypoglycaemia, or
• Twice daily insulin injection is deemed inappropriate due to other reasons.
In such situations a long-acting insulin analogue, detemir or glargine, should be considered [5]. The newer insulins like degludec, toujeo (high strength insulin) and biosimilars (example: Abasaglar) can also be used.
It can be initiated as 10 U/day or 0.1-0.2 U/kg/day. To reach the target fasting plasma glucose (FPG) level, titration of insulin is done by 10-15% or 2-4 units (U) once or twice weekly [8].
The duration of action of lantus and NPH is around 24 hours and 14 hours or more respectively. The effectiveness of basal insulin is estimated by FPG levels and titrated accordingly. The best time to dose basal insulin is either with evening meal or before bedtime [8] (Table 3).
Table 3: Evidence of analogue insulin used in the treatment of T2DM. View Table 3
The benefit of long-acting analogue insulins over NPH insulin, appears to be in the lower incidence of nocturnal hypoglycaemia, which must be a consideration when determining insulin choice. Due to its lesser day-to-day variability, degludec has the potential to further reduce the risk of hypoglycaemia than other long-acting basal insulins [9]. The DEVOTE study showed that degludec was non-inferior to glargine with respect to incidence of cardiovascular events with significant reductions in severe hypoglycaemia, especially nocturnal [10].
High-strength insulin products such as insulin glargine 300 units/ml (Toujeo) have been developed for people with type 1 or type 2 diabetes who have large daily insulin requirements to reduce the number and volume of injections. In 3 randomised controlled trials (RCTs) in 2496 adults with type 2 diabetes, Toujeo had similar efficacy to insulin glargine 100 units/ml (Lantus) in terms of HbA1c reduction [11].
Toujeo is not bioequivalent to insulin glargine 100 units/ml (Lantus) and dose adjustment is needed when patients are switched from Lantus or other basal insulins to Toujeo or vice versa.
Premixed insulin therapy is suitable for patients who cannot count carbohydrates or those who have consistent eating patterns and routine lifestyle. Initiating insulin therapy with premixed insulin once daily, opens the possibility of a stepwise approach to intensify therapy to twice or thrice daily to achieve target HbA1c levels [12]. NICE [5] recommends to start premixed insulin if HbA1c > 9.0%.
A twice daily dose has been recommended for initiation of premix insulin but a once daily dose has been found to be effective for patients who are reluctant to start insulin and require < 20 units/day [13]. Kilo, et al. [14] evaluated the effectiveness of initiating patients on a simple once daily regimen with insulin aspart 30/70, NPH insulin and biphasic human insulin 30/70, in combination with metformin. The results showed a 1.1%-1.3% reduction of Hba1c for all groups with FPG reductions of 31% for aspart 30/70, 37% for NPH and 28% for biphasic human insulin, concluding that each one could safely be added to metformin.
Premixed insulin can be initiated according to the following common standards [15]:
• 10 U/day once daily either in the morning or evening or 0.3-0.5 U/kg/day, depending on glucose level.
• 0.2 U/kg/day for Hba1c < 9% and 0.4 U/kg/day for Hba1c > 9%. Patients with Hba1c > 9% are at increased risk of complications, hence optimal glycaemic control is required.
• Twice a day premixed insulin, if total daily dose of insulin (TDD) is more than 20 U, dosed along with breakfast and evening meal based upon carbohydrate content in meal and daily activity.
• The 50:50 ratio can be considered for premix analogues, if a single dose exceeds 30 U.
Tanaka and Ishii [16] found lispro mix 50/50 insulin, which is high proportion insulin containing 50% rapid and 50% intermediate acting insulin, controlled the post-prandial glucose (PPG) and stabilised the diurnal blood glucose variations in patients who had poor control on insulin 70/30 or 75/25. Lispro 50/50 had more rapid onset of action with lower risk of nocturnal hypoglycaemia.
The initial dose can be titrated once or twice weekly for 12 weeks and every two weeks thereafter, according to the FPG targets as per algorithm below [15] (Table 4).
Table 4: Premixed insulin dose titration. View Table 4
The addition of a single prandial insulin injection to the already existing basal regimen before the main meal or the meal consistent with highest PPG, is referred to as a "basal-plus" strategy [17,18]. This has been an effective method when intensifying insulin therapy, before a full basal-bolus regimen is to be implemented [19]. Careful patient evaluation and timing are necessary due to the complex nature of this regimen and the need for multiple daily injections and multiple daily glucose monitoring. Therefore younger, highly motivated, active individuals with variable eating habits with both type 1 and T2DM will be best suited for such a regimen (Table 5).
Table 5: Evidence for the use of the basal-plus strategy. View Table 5
From the results obtained in table, not all differences observed were statistically significant when comparing them in the use of this strategy to basal-bolus and premixed regimens.
Basal-bolus regimen is an intensified insulin therapy when target glycaemic control is not achieved with basal insulin. This regimen mimics the physiological insulin secretion from the pancreas. The intermediate or long acting insulin is given as basal insulin. The rapid or short-acting insulin is given as bolus insulin before meal. This regimen needs frequent and active self-monitoring, knowledge on insulin-carbohydrate ratio and correction factors (CF), and titration of insulin dose to achieve target glycaemic control [20].
The basal-bolus insulin regimen can be initiated by intensifying the basal or basal plus regimen. To initiate, the TDD (0.6-1.0 units/kg) is first calculated. Then 50% of TTD is given as basal insulin which is administered at bedtime. The remaining 50% is divided between bolus doses for breakfast, lunch and dinner [21].
There are different methods of dose titration, adjusting 2-3 units in appropriate dose every 3-7 days to achieve the target. Generally, titration of basal insulin is based on FPG, while bolus insulin is based on pre prandial or PPG levels at mentioned in table 6 [21].
• Increase or decrease insulin by 10% depending on blood glucose reading shown in Table 6 [21].
Table 6: Insulin dose optimisation according to blood glucose readings. View Table 6
• To titrate the bolus dose, add or deduct the correction dose to or from the bolus dose.
Correction dose = Excess or deficit blood glucose × CF.
CF = 1500/TDD (will give the value in mg/dl, Used when patient is on rapidly acting insulin)
• To titrate the basal dose, some use Treat-to-Target Trial's titration schedule as per table [22] (Table 7).
Table 7: Treat-to-Target trial's titration schedule for basal insulin. View Table 7
Giugliano, et al. [23] concluded it to be the best regimen to attain the target glycaemic control. The patients treated with basal-bolus regimen had statistically significant achievement of HbA1c target of < 7% compared to biphasic insulin, 63.5% vs. 50.8% (odd ratio 0.57 [95% CI 0.36-0.90]) (p = 0.034). No difference was seen in terms of incidence of hypoglycaemia or weight gain between the two regimens.
In Treating to Target in Type 2 Diabetes (4-T) study, 67.7% of the patients in biphasic group, 73.6% in the prandial group and 81.6% in the basal group were later intensified to basal-bolus regimen, and two-third of them reached the glycaemic target [24].
Insulin analogues closely mimic physiological endogenous insulin secretion. Compared to human insulin formulations, they have a more predictable onset and duration of action, along with increased flexibility for dose administration and meal times [25].
A comparative analysis of various studies comparing premix human insulin vs. premix insulin analogues is given in the Table 8.
Table 8: Comparing premix human insulin vs. premix insulin analogues. View Table 8
In conclusion, premix analogues are similar to human insulin but some trials have shown better HbA1c reduction. The main advantage is better post-prandial glucose control and less delayed hypoglycaemia.
When initiating insulin, the HCP could consider premixed or basal insulin, depending on clinical situation (Table 9).
Table 9: Comparing pre-mixed and basal insulin initiation regimens in T2DM. View Table 9
These studies show that glycaemic control can be better with pre-mixed rather than basal insulin, but with higher risk of hypoglycaemic episodes and weight gain.
Fasting glucose levels contribute to HbA1c at all levels of glycaemia, but when HbA1c is less than 9%, prandial glucose levels become more important. However, further analysis of DURABLE trial showed that only premixed lispro insulin decreased prandial glucose. Where initial HbA1c was less than 9%, there is greater improvement in HbA1c with pre-mixed insulin rather than basal [26].
Thus for the following patients, pre-mixed insulin may provide better glycaemic control [27]:
• HbA1C > 8.5%.
• Poor patient compliance with high demands of basal bolus regimen.
• Low (< 150 mg/dl) fasting or pre-prandial glucose but high HbA1c.
Therefore, despite ADA [8] advising to start patients on basal insulin, there may be situations where premixed insulin maybe more suitable.
A biosimilar medicine is a biological medicine that is developed to be very closely similar to an existing biological medicine (the reference medicine). The active substance of both is essentially the same biological substance but, just like the reference medicine, the biosimilar has some element of natural variability. Biosimilars undergo a regulatory process which demands extensive comparability studies that demonstrate similarity to the reference medicine. Biosimilars have the potential to offer considerable cost savings [28].
Insulin glargine biosimilar (Abasaglar) is licensed for the treatment of diabetes mellitus both type 1 and type 2. Abasaglar is a basal insulin for once daily use and is bioequivalent to insulin glargine (Lantus). Abasaglar is available as cartridges of 100 units/ml for use in the reusable pen or as a pre filled Abasaglar pen 100 units/ml [28].
The efficacy and safety of Abasaglar is non-inferior to Lantus and this evidence is based on the 2 phase III studies of insulin glargine biosimilar (Abasaglar) in people with type 1 and type 2 diabetes, ELEMENT 1 [29] and ELEMENT 2 [30], respectively.
HCPs should individualise the treatment and discuss with patients, regarding various options and educate them on insulin therapy. Basal insulin is preferred as an augmentation therapy when adding on to the oral hypoglycaemic agents. Replacement therapy includes basal-bolus regimen with correction dose or consider premixed insulin if appropriate [21].
Glucose control, adverse effects, cost, adherence and quality of life should be considered while choosing therapy. Titration of insulin over time is critical for glycaemic control and to prevent diabetes related complications [21]. Though basal-bolus resembles physiological insulin secretion, premixed analogues are good option with less demanding glucose monitoring and injection schedule. The rule should be tailoring the insulin treatment to suit patient and not the other way around [27].
Amit dey, Gloria Nomusa Cele, Karen Blackwood, Meela Ali, Naved Akhtar, Oye Akindele and Swati Dholakia.