The steepness from the protection curve must sometimes be bounded to achieve convergence and methods for doing so are outlined. curve functions improved model evaluation criteria for every dataset. Standard errors based on the noticed information were found to be unreliable; bootstrap estimates of precision were to be preferred. In most instances, case-cohort designs resulted in little loss of precision. Some results achieved suggested measures intended for utility. == Conclusions == The original scaled logit model can be improved upon. Evaluation criteria permit well-fitting models and useful results to be recognized. The proposed methods provide a comprehensive set of tools for quantifying the relationship between immunological assays and protection from disease. == Electronic supplementary material == The online edition of this article (doi: 10. 1186/s12874-015-0096-9) contains supplementary material, which is available to certified users. Keywords: Vaccine, Correlate of protection, Correlate of immunity, Surrogate endpoint, Immunological assay == Background == Immunological assays measure characteristics of the immune system, such as antibody concentrations or the ability of serum to neutralize pathogens in vitro, which are induced by an immune stimulus such as disease or vaccination, and which are associated with protection from disease. The relationship between an immunological assay and protection from disease is of considerable interest in vaccines research. In early phase clinical trials of a new vaccine, an immune response noticed by an immunological assay suggests the possibility the vaccine might be protective. In later on phase trials, values of immunological assays are used for dose selection and dose ranging, and to assess the effect of co-administration with other vaccines. In vaccine efficacy trials, data on post-vaccination assay values and subsequent disease occurrence may be used to predict protection in other settings. Correlates of protection threshold values of specific immunological assays believed to be associated with protection from disease have Ceftriaxone Sodium Trihydrate been established for many vaccine-preventable diseases [1], and are used as surrogates for protection in the development of combination vaccines. A number of elements need to be demonstrated for an assay to reliably substitute for observation of clinical disease. First it should be shown that increasing assay values correlate with reduction in the rate of disease; standard statistical methods are available for this purpose, such as logistic regression [2], and the assay is then described as a correlate of risk [3, 4]. When vaccination both increases assay values and reduces the rate of disease the term correlate of protection has been suggested [5]; elsewhere the term correlate of vaccine-induced protection has been used when this condition is met [6, 7]. It is desirable to show that an assay meets criteria for a surrogate endpoint, such as the Prentice criteria [8] or that it explains a high proportion of treatment effect [9]. Alternatively or additionally it can be useful Rabbit polyclonal to ACSF3 to demonstrate that the property measured by the assay is causally or mechanistically related to protection from disease (rather than both merely reflecting some common, unobserved characteristic such as robustness of the immune system). Immunologists understanding of the mechanisms of action from the immune system have shown such mechanistic associations [1012], and statistical methods have been developed Ceftriaxone Sodium Trihydrate seeking to demonstrate causal associations [1316]. Ceftriaxone Sodium Trihydrate Finally, it is necessary to quantify the relationship between assay value and protection the level of protection at each assay value which is the subject of this research. Established correlates of protection have been characterized as threshold values of immunological assays; interpretation of such thresholds can however be problematic whether the threshold is one at which protection can be viewed as complete, or whether it represents a population average measure differentiating susceptible from protected individuals and statistical methods with different interpretations have been developed [1721]. In fact the relationship is likely continuous. Natural variability between individuals means that at any given assay value some individuals will be protected and some not, and if protection does in fact increase with increasing assay value then the proportion of individuals protected will increase in a smooth continuous manner, which may be represented by a protection curve. Most data on the relationship between assay value and subsequent protection from disease comes from settings in which the exposure of subjects to the pathogen of interest is not guaranteed, to ensure that although common statistical methods such as logistic regression may demonstrate a relationship between assay prices and disease they cannot explicitly quantify safeguard. Absence of disease may reveal protection, or merely insufficient exposure. A scaled logit model modelling disease happening as a function of assay value simply by an visibility parameter and a parametric protection contour.