Applying the Estimand Framework to Non­Inferiority Trials.

Most published applications of the estimand framework have focused on superiority trials. However, non-inferiority trials present specific challenges compared to superiority trials. The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use notes in their addendum on estimands and sensitivity analysis in clinical trials that there may be special considerations to the implementation of estimands in clinical trials with a non-inferiority objective yet provides little guidance. This paper discusses considerations that trial teams should make when defining estimands for a clinical trial with a non-inferiority objective. We discuss how the pre-addendum way of establishing non-inferiority can be embraced by the estimand framework including a discussion of the role of the Per Protocol analysis set. We examine what clinical questions of interest can be formulated in the context of non-inferiority trials and outline why we do not think it is sensible to describe an estimand as 'conservative'. The impact of the estimand framework on key considerations in non-inferiority trials such as whether trials should have more than one primary estimand, the choice of non-inferiority margin, assay sensitivity, switching from non-inferiority to superiority and estimation are discussed. We conclude by providing a list of recommendations, and important considerations for defining estimands for trials with a non-inferiority objective.

Why estimands are needed to define treatment effects in clinical trials.

BACKGROUND: The estimand for a clinical trial is a precise definition of the treatment effect to be estimated. Traditionally, estimates of treatment effects are based on either an ITT analysis or a per-protocol analysis. However, there are important clinical questions which are not addressed by either of these analyses. For example, consider a trial where patients take a rescue medication. The ITT analysis includes data after use of rescue, while the per-protocol analysis excludes these patients altogether. Neither of these analyses addresses the important question of what the treatment effect would have been if patients did not take rescue medication. MAIN TEXT: Trial estimands provide a broader perspective compared to the limitations of ITT and per-protocol analysis. Trial treatment effects depend on how events occurring after treatment initiation such as use of alternative medication or discontinuation of the intervention are included in the definition. These events can be accounted for in different ways, depending on the clinical question of interest. CONCLUSION: The estimand framework is an important step forward in improving the clarity and transparency of clinical trials. The centrality of estimands to clinical trials is currently not reflected in methods recommended by the Cochrane group or the CONSORT statement, the current standard for reporting clinical trials in medical journals. We encourage revisions to these guidelines.

Principles and recommendations for incorporating estimands into clinical study protocol templates.

Clinical study protocols are the foundation of good clinical studies. Prospective and multidisciplinary collaboration that pays attention to the design of all components of the study protocol can ensure that a clinical study will answer the research questions posed in a reliable manner that is meaningful for decision-makers and patients. The ICH E9(R1) addendum on estimands and sensitivity analysis in clinical trials provides a framework for clinical study planning to ensure alignment between study objectives, design, conduct, and analysis. The estimand or clinical question posed can be regarded as the backbone of the study and the clinical study protocol should reflect estimands accordingly. In practice, stakeholders are still learning how to embrace the estimand framework and how it impacts studies and study documents. In this paper, we anticipate that a protocol structure centred around estimands, or objectives rather than endpoints alone will prevail for all types of studies. To assist sponsors during this paradigm shift, this paper provides discussion and guidance for implementing the estimand framework in protocol templates.

The SimpleMix study with biphasic insulin aspart 30: a randomized controlled trial investigating patient-driven titration versus investigator-driven titration.

OBJECTIVE: The study aimed to confirm the efficacy, through non-inferiority, of patient-driven versus investigator-driven titration of biphasic insulin aspart 30 (BIAsp 30) in terms of glycemic control assessed by HbA1c change. METHODS: SimpleMix was a 20 week, open-label, randomized, two-armed, parallel-group, multicenter study in five countries (Argentina, China, India, Poland, and the UK). Patients with type 2 diabetes were randomized into either patient-driven or investigator-driven BIAsp 30 titration groups. RESULTS: Non-inferiority of patient-driven vs. investigator-driven titration based on change in HbA1c from baseline to week 20 could not be demonstrated. Mean (SE) estimated change from baseline to week 20 was -0.72 (0.08)% in the patient-driven group and -0.97 (0.08)% in the investigator-driven group; estimated difference 0.25% (95% CI: 0.04; 0.46). Estimated mean change (SE) in fasting plasma glucose from baseline to week 20 was similar between groups: -0.94 (0.21) mmol/L for patient-driven and -1.07 (0.22) mmol/L for investigator-driven (difference non-significant). Both treatment arms were well tolerated, and hypoglycemic episode rates were similar between groups, with a rate ratio of 0.77 (95% CI: 0.54; 1.09; p = 0.143) for all hypoglycemic episodes and 0.78 (95% CI: 0.42; 1.43; p = 0.417) for nocturnal hypoglycemic episodes. CONCLUSIONS: Non-inferiority of patient-driven versus investigator-driven titration with regard to change from baseline to end-of-treatment HbA1c could not be confirmed. It is possible that a clinic visit 12 weeks after intensification of treatment with BIAsp 30 in patients with type 2 diabetes inadequately treated with basal insulin may benefit patient-driven titration of BIAsp 30. A limitation of the study was the relatively small number of patients recruited in each country, which does not allow country-specific analyses to be performed. Overall, treatment with BIAsp 30 was well tolerated in both treatment groups.

Parent and health professional perspectives in the management of adolescents with diabetes: development of assessment instruments for international studies.

OBJECTIVE: Assessment of quality of life (QOL) in adolescents with diabetes requires patient, parent and health professional input. Psychometrically robust instruments to assess parent and professional perspectives are required. RESEARCH DESIGN AND METHODS: Questionnaires concerning adolescent QOL were developed for completion by parents and health professionals. In an international study assessing QOL in 2,101 adolescents with diabetes (median age 14 years, range 10-18; from 17 countries including Europe, Japan and North America), parents and health professionals completed their respective questionnaires between March and August 1998. RESULTS: Feasibility and acceptability of the new questionnaires were indicated by high questionnaire completion rates (adolescents 92%; parents 89%; health professionals 94%). Internal consistency was confirmed (Cronbach's alpha coefficients 0.80 parent; 0.86 health professional). Correlations of Diabetes Quality of Life Questionnaire for Youths (DQOLY) scores with parent and health professional global QOL ratings were generally low (r ranging from 0.12 to 0.36). Parent-rated burden decreased incrementally across adolescence, particularly for girls. Professional-rated burden followed a similar profile but only after age 15 years. Until then, burden was rated as uniformly high. Clinically relevant discrepancies in parent and professional burden scores were noted for one-parent families and families where adolescents had been referred for psychological help. In both cases, health professionals but not one-parent families perceived these as high burden situations. The clinical significance of this relates to the significantly poorer metabolic control recorded for adolescents in both situations. CONCLUSIONS: Parent and health professional questionnaires were found to have adequate internal consistency, and convergent and discriminant validity in relation to key clinical and QOL outcomes. The questionnaires are brief, easy to administer and score. They may also enable comparisons across countries and languages to facilitate development of international health outcome parameters. The inclusion of the parent and health professional perspectives completes a comprehensive assessment of adolescent QOL relevant to diabetes.

Insulin injection regimens and metabolic control in an international survey of adolescents with type 1 diabetes over 3 years: results from the Hvidore study group.

UNLABELLED: The optimal insulin regimen for paediatric patients with type 1 diabetes remains controversial. Therefore this multicentre study was performed in adolescents over a 3-year period to assess metabolic control, severe hypoglycaemia, and weight gain in relation to insulin injection regimens. Out of 2873 children and adolescents in an international survey in 1995, 872 adolescents (433 boys, 439 girls, mean age in 1995 11.3+/-2.2 years) were restudied in 1998, relating insulin regimens to HbA(1c) measured in a central laboratory. In addition, the daily dose of insulin, changes in body mass index (BMI), and events of severe hypoglycaemia were evaluated. Over 3 years, the use of multiple injection regimens increased from 42% to 71%: 251 patients remained on twice daily insulin, 365 remained on multiple injections and 256 shifted from twice daily insulin to multiple injections. In all three subgroups an increase in insulin dose, a deterioration of metabolic control, and an increase in BMI were observed. Metabolic control deteriorated less than expected over 3 years during adolescence (HbA(1c) 1995: 8.7+/-1.6%; 1998 observed: 8.9+/-1.6%, HbA(1c) expected for 1998: 9.0%). BMI increased more than expected, the increase was greatest in patients switching from twice daily to multiple injections, and higher in females compared to males. CONCLUSION: in this international study, metabolic control was unsatisfactory in many adolescents with type 1 diabetes irrespective of the insulin regimen. No improvement in metabolic control was observed in this cross-sectional survey, over 3 years in any of the subgroups. Even the group switching from twice to multiple injections did not improve blood glucose control and the increase in body mass index was most pronounced in this group. Conclusive evidence, however, should be based on prospectively planned, randomised therapeutic trials in paediatric patients.