In a recent paper in the Journal of Clinical Pathology, Currie et al used immunohistochemistry on tissue microarrays to assess the expression of the hypoxia inducible factor 1α (HIF-1α) protein in breast cancer.¹ As we point out in a letter in this issue of the journal,² tissue microarrays are not very suitable for HIF-1α immunohistochemistry because focal perinecrotic HIF-1α staining is easily missed in the small tissue samples forming a tissue microarray. However, this perinecrotic type of HIF-1α expression is relevant in breast cancer.³

Furthermore, the authors assessed HIF-1α expression considering both the intensity and extent of nuclear and cytoplasmic reactivity. To our knowledge, there are no data supporting a functional role for cytoplasmic HIF-1α expression in breast cancer. Therefore, previous studies have only considered nuclear staining. Lastly, the authors do not describe how they arrive at a score in an individual case after considering “both the intensity and extent of nuclear and cytoplasmic reactivity”, and also the threshold used in statistical analysis is not provided. Further enlightenment on these matters would help us to appreciate the value of their findings. Perhaps, the lack of prognostic value of vascular endothelial growth factor D, which induces lymphangiogenesis, can be explained by the absence of lymphangiogenesis in invasive breast cancer.⁴