I.D.S. STRATEGY FOR ACTIVITIES RELATED TO
ISLET AUTOANTIBODIES 1998-2001
This document outlines the proposed strategy for the range of IDS activities relating to islet autoantibody measurement over the next three years. It has been drawn up by a working party appointed by the IDS council, and includes issues relating to standardisation, workshops and proficiency programmes.
The overall objective of the strategy is to ensure that:
- Antibodies are measured in such a way as to achieve the highest sensitivity and specificity.
- All laboratories obtain comparable results when measuring islet autoantibodies in the same set of samples.
The most recent evaluation, the JDF sponsored 1st IDS Combined Antibody Workshop, showed:
- There was one or more assay methods for each marker which achieved high sensitivity and specificity in some laboratories, but there remained wide variation in sensitivity between laboratories as a whole, even using the same reference control cohort, similar methods, and thresholds giving the same specificity.
- A number of laboratories produced very comparable ranking of individual samples, although reporting in widely differing units.
- Even these laboratories, however, differed in sensitivity and specificity achieved when in-house thresholds were used to define 'positive' and 'negative'.
This implies that:
- Methods that achieve high sensitivity and specificity are available.
- These methods are not always implemented in a way which achieves high sensitivity and specificity.
- Introduction of common units may improve comparability of results.
- The definition of positive and negative affects comparability and ability to achieve high sensitivity and specificity.
The major issues that need to be addressed in future IDS activities therefore include assay methods, common units and the way in which antibody 'positive' and 'negative' is defined.The immediate needs are:
- mechanisms to improve general implementation of the best assay methods
- a means to assess new methods and new markers
- evaluation of the effect of use of reference sera and the introduction of common units in improving assay performance.
These three should improve the ability of assays to measure antibodies reproducibly down to a low level which should enable us to address the remaining issue of comparing thresholds between laboratories.
It is proposed that the structure of IDS autoantibody activities are revised, with roles of both workshop and proficiency programmes redefined. Specific issues will be addressed by workshops involving a limited number of laboratories, and the current proficiency programme will be expanded to take over part of the role of previous workshops. The revised proficiency programme, which will involve all laboratories, will serve the dual roles of a) evaluating the effectiveness of the recommendations of the workshops in laboratories as a whole, and b) providing laboratories with an ongoing external assessment of their own performance in order to improve and maintain assay quality.
1. Evaluation and improvement of general implementation of assay methods
a) Evaluation. Proficiency programmes involving small numbers of samples are likely to identify major assay problems but are insufficient to help the majority of laboratories to improve performance. It is therefore proposed that proficiency rounds involve larger numbers of sera. The programme will aim to assess whether laboratories can measure antibodies reproducibly over a range which includes low levels. The proficiency will therefore take on the form and analysis of the most recent GAD and combined antibody workshops, circulating a set of sera which includes up to 50 disease and 50 control samples. Analysis will be based on disease sensitivity and specificity, and reproducibility. It will be undertaken at intervals of approximately 18 months to coincide with IDS meetings. A core set of samples will be included which can be used as a vehicle for continually evaluating performance within the proficiency programme.
b) Improvement of individual assay performance. The IDS will try to establish a network of regional reference centres or regional wet workshops. Laboratories performing poorly in the proficiency programme would be offered the opportunity to take part in such activities. In addition, a scheme may be set up whereby laboratories encountering problems could be linked with another which appeared to be performing well so that exchange visits could be arranged.
2. Assessment of new methods and markers
A mechanism is needed by which the sensitivity and specificity of innovative assay methods can be compared with best available methods identified through the workshop programme. Potential new antibody markers require similar rapid and effective evaluation. To achieve this, a set of sera equivalent to those produced in the 2nd GAD antibody workshop or the combined antibody workshop will be collected and initially assayed for antibodies to ICA and GAD, IA-2/ICA512, and insulin in 10 laboratories selected on the basis of assay performance in the combined antibody workshop. The 10 laboratories would be selected for performance in each assay and therefore might not be the same for all markers. This will provide a median and range of assay sensitivity and specificity as a reference point for future comparisons. These sera, appropriately blinded, will then be available specifically for use in rapid evaluation of new methods, and the results could be included in the original publication and description of ne w methods, with due acknowledgement to the IDS. The same set of sera could also be used in evaluating potential new antibody markers. This would avoid the need for repeated workshops when each new method is proposed, or at least provide a screening process to determine which methods and markers warrant further standardisation.
The aim is to collect a set of sera containing 50 samples from patients with newly diagnosed type I diabetes and 50 control subjects. At least 50 ml of each sample will be required, which will be divided into 0.2 ml aliquots.
3. Evaluation of reference sera and introduction of common units
A protocol has been drawn up to address the issue of evaluation of the WHO reference serum and to determine the best way to use the reference serum to improve inter-laboratory concordance and intra-laboratory reproducibility for measurement of ICA, GAD antibodies and IA-2 antibodies. There is currently no reference standard for IAA, and attempts will be made to rectify this as soon as possible.
4. Relating thresholds between laboratories
This is likely to be the most difficult goal to achieve, and as stated above, is dependent on the ability of assays to measure antibodies reproducibly down to a low level. In the recent IDS antibody workshops, the use of an adjusted sensitivity after 'standardising' specificity with the common set of control samples overcame some of the difficulties in comparing assay performance. A large common set of control samples might therefore facilitate the comparison of thresholds used in different assays and circumstances, without putting any restrictions on how thresholds are defined. Ways to the obtain and make available such a set of samples will be addressed.