What is the best second line therapy for people aged 65 years or over with type 2 diabetes?: systematic review and network meta-analysis

Research question What is the most effective and safe second line therapy for people aged 65-years and over with type 2 diabetes(T2DM)?

Background One-in-six people aged 65-years or over has T2DM and over half of prevalent T2DM cases are in this age group.

Older adults have a higher risk of adverse cardiovascular and renal outcomes, and potentially more adverse-effects from drugs.

Evidence suggests that the relationships between therapy and outcome may differ between older and younger people so the common approach of trial participant selection (recruiting people with a wide range of ages) may not allow a full description of the responses in this age group.

Aims and Objectives To perform a systematic review and network meta-analysis (NMA) using either stratified summary data or individual patient data (IPD) of all relevant randomised controlled trials to determine the rankings of second line therapies for patients aged 65-years or over with T2DM. Objective 1.

Calculate the relative effectiveness of drugs and combinations of drugs in patients aged 65-years or over who are also receiving metformin. Objective 2.

Calculate the relative effectiveness of drugs and combinations of drugs in patients aged 65-years or over who are not receiving metformin Methods We will update the search in a recent NMA and perform risk of bias assessments on newly identified studies. We will obtain individual patient data (IPD) for all relevant studies using 5 different processes.

Including data repositories, contacting data owners, and authors of papers. The statistical analysis will be a two-stage process. First, IPD will be analysed independently for each trial using appropriate regression models for each outcome. Models will adjust for variables that were stratified by in randomisation.

We will also adjust for baseline values of HbA1c, sex, BMI, presence of cardio-vascular disease, presence of chronic kidney disease in all models.

Second, these summary treatment effect estimates and their standard errors will be used in a NMA to synthesise the evidence in a single mathematically coherent model.

This will allow us to rank treatments by efficacy and safety, and then estimate the relative effectiveness of all treatments with confidence intervals.

Timelines for delivery Pre-grant phase(12/19-05/20): Update literature search, apply to trial portals and request data from study teams, develop protocol. 05/20-01/21: Finalise protocol, continue data acquisition and develop computer code for evidence synthesis. 11/20-06/21: Clean data, quality assess newer studies and analyse data 07/21-10/21: Update database and perform NMA.

Write-up study and disseminate. Many processes overlap during these phases. PPI will be throughout the project.

Anticipated Impact and Dissemination We will identify the best second line treatments for patients with T2DM who are 65-years or older. Findings will be disseminated to patients in collaboration with (e.g.

Foundation for Diabetes Research in Older People); to clinicians via diabetes, older people and general practice conferences, networks and publications; policy makers through the NICE diabetes management guidelines committees.

We envisage the results will change practice by impacting on evidence-based prescribing for older people, reduce adverse effects, and potentially improved cardiovascular health, and quality of life.