Intraindividual variability of CD30 expression in mycosis fungoides -implications for diagnostic evaluation and therapy.

AIMS: To evaluate the frequency, subcellular localization, and variability of CD30 expression in mycosis fungoides.

METHODS

This retrospective multicenter study investigated 135 biopsy specimens of 95 patients with mycosis fungoides for CD30 expression and CD30 staining pattern in relation to histomorphologic criteria, eosinophils, and tumour stage. We used two different CD30 cutoffs: (i) any CD30 positivity (important for in-label treatment with an anti-CD30 antibody-based drug), and (ii) ≥10% (most commonly used cutoff in therapy studies). The histologic slides were digitally evaluated with a slide viewer. None of the patients were treated with an anti-CD30 antibody-based drug.

RESULTS

We found any CD30 expression in 90% of the samples (tumour cells and tumour microenvironment). More than 60% of the samples had CD30 expression ≥10%. CD30 staining was predominantly cytoplasmic and membranous, a Golgi pattern was only found in three samples. In patients with multiple biopsies (69 samples), a highly variable CD30 expression was found, especially in biopsies taken at different timepoints. CD30 and eosinophils were more common in advanced disease stage and cytoplasmic CD30 staining in intraepidermal lymphocytes was significantly associated with the presence of eosinophils.

CONCLUSIONS

90% of mycosis fungoides biopsies showed CD30 expression and are principally eligible for anti-CD30-antibody treatment. Because of the high intraindividual variability, investigation of multiple tissue samples for CD30 expression improves the assessment of this therapeutic target. Analysis of only a single skin biopsy might lead to misinterpretation of CD30 and bears the risk of inadequate and potentially unfavourable therapeutic decisions.