Nuclear immunomarkers, such as oestrogen and progesterone receptors, TTF-1, microphthalmia transcription factor, Myo-D1, myogenin, cyclins, Rb2/p130, CDX-2, p57, and Fli-1 are generally easier to interpret than are cytoplasmic or membrane markers. Transcription factors usually yield an "all or none" signal, often of in excess of 90% of nuclei. There is much less likelihood of confusion with biotin or other sources of false-positive cytoplasmic staining. Staining is usually largely independent of the degree of differentiation.
However difficulty can arise for the following reasons:
light nuclear immunoreactivity can be masked by nuclear hyperchromasia
light nuclear immunoreactivity can be masked by high non-specific cytoplasmic staining. This happens occasionally in my experience with TTF-1.
It is helpful to compare the nuclei with appropriate negative and positive internal control nuclei. A red chromogen such as AEC (3-amino-9-ethylcarbazole) may be helpful.
Some nuclear markers such as Fli-1, when positive, stain over 50% of nuclei and circumspection is required if only a few nuclei stain1. One paper set the threshold for positivity for CDX-2 at 25% of nuclei2. A much lower threshold can be safely applied for TTF-1 as a marker of lung tumours. It is important to know the threshold level of nuclear staining for the marker being used.
References
This page last revised 29.5.2003.