Reliability of GIST diagnosis based on partial information

In several medical diagnostic procedures there is the absence for a “golden standard” to follow, especially when new advances in the understanding of the biology are done or new techniques are discovered. In such situations, different opinions appear in scientific literature and they are based on disparate case studies, each one with its peculiarity and heterogeneity of data. Due to these discrepancies, it is reasonable the physician tries to “replicate” such evaluations to have a better (scientifically based) idea of which diagnostic procedure to adopt. But there are situations where new sophisticated techniques are required and, either they are so expensive that only few selected trials can be performed, or they are at such initial stage that it is hard to fully trust on them. Moreover, the physician could have the opportunity to make only a “retrospective” study and usually reliability and accuracy of such data are not very high.