The Validation of Candidate Surrogates for a Time to Event Endpoint in Childhood Leukemia

The selection of the primary “endpoint” is a very important step in the design of clinical trials. Typically, the goal of a clinical trial is to assess the effect of treatment on this endpoint. It often happens, that the most sensitive and relevant clinical endpoint, which will be called the “true” endpoint, might be difficult to assess and to overcome this problem, a solution is to replace the true endpoint with another one, which is measured earlier, more conveniently or more frequently. Such “replacement” endpoint is termed “surrogate” and has the purpose of evaluating the effect of a specific treatment for a specific disease. Once a candidate surrogate is identified according to some specific properties, several formal methods are available for the validation, depending on the number of the trials performed. The first formal statistical approach dates back to 1989, when Prentice proposed a definition of surrogate endpoint and four criteria to validate it. The most important criterion among these is called “The Prentice’s Criterion”, which implies that “the full effect of treatment upon the true endpoint is captured by the surrogate”. More recently, a meta-analytic approach to the validation of a surrogate endpoint was proposed for a multi-trial context (Burzykowki, 2005). It consists in estimating associations at two different levels: the association between the surrogate and the clinical endpoint, called the “individual-level association”, and the association between the effect of treatment on the surrogate and on the clinical endpoint, called “trial-level association”. A good surrogate is one that has biological/clinical plausibility and is shown, statistically, to have strong individual-level and trial-level associations with the true endpoint. Many extensions of this meta-analytic approach exist to evaluate surrogacy when the candidate surrogate and the true clinical endpoint are not continuous or have different nature, e.g. both failure time endpoints or binary/ordinal surrogate and failure time endpoints. The motivating clinical question was the evaluation of Minimal Residual Disease (MRD) as a candidate surrogate endpoint in childhood acute lymphoblastic leukaemia (ALL). MRD, that quantifies the small numbers of leukemic cells circulating in the patient, has not yet been formally validated as a surrogate endpoint, whilst it is a well-established prognostic biomarker in ALL. The challenge has now evolved to the qualification of early MRD as an efficacy-response biomarker in the assessment of new drugs for the treatment of ALL. The main goal of the present work was to assess by the Prentice criteria and by the meta-analytic approach whether MRD, evaluated at the end of the induction treatment, can be considered a surrogate for Event Free Survival (EFS) in childhood B-lineage ALL patients treated with Dexamethasone or Prednisone in large collaborative randomized trials conducted in Europe and USA. From our analyses, MRD resulted a poor surrogate for EFS at the unit level, i.e. it does not permit reliable prediction of treatment effects on EFS but there is a considerable prognostic association at the individual level, after adjusting for treatment, i.e. the (global) odds of surviving event free is higher with negative/lower MRD. A secondary aim was to evaluate if EFS is a surrogate for Overall Survival (OS) and we found that EFS cannot be accepted as a surrogate for OS, although it is a good predictor of OS. Finally, as methods on the validation of a continuous surrogate for a failure time endpoint are missing, a proposal was made here. In line with the meta-analytic framework, a copula based approach was implemented and translated in a SAS macro.