Supremacy of modern morphometry in typing renal oncocytoma and malignant look-alikes.

Franziska Erlmeier, Annette Feuchtinger, Daniela Martina Borgmann, Martina Rudelius, Michael Autenrieth, Axel Walch, Gregor Weirich

Publikation: Beitrag in FachzeitschriftArtikelBegutachtung

12 Zitate (Scopus)


In the era of tumour type-specific therapies, the correct typing of renal tumours is of prime importance. As immunotyping and genotyping approaches are laborious and fall short of standardization, we used whole-scale computer-assisted morphometry instead. Three different types of renal tumours with different prognoses and therapies, notoriously prone to mistyping, were analysed. The sample of 335 tumours included clear cell renal cell carcinoma, chromophobe renal cell carcinoma and renal oncocytoma. The sample was analysed using H&E stains of tissue microarrrays in combination with an image-scanning software. Nuclear and cytoplasmic features were registered with the aid of computer-assisted morphometry. Features included shape, texture, colour and colour intensity for different cell compartments, e.g. nuclei and cytoplasm. The software passed several training steps for final validation. Using morphometry, we were able to classify the three renal tumour types correctly, with a 100 % specificity compared to the WHO typing. Nuclear features dominated the typing of chromophobe renal cell carcinoma, whereas cytoplasmic features were the leading classificators for renal oncocytoma. The grading of clear cell renal cell carcinoma attained a specificity of 80 %. In conclusion, modern morphometry may serve as a tool for typing renal epithelial tumours and additionally draws the attention to future nuclear research in chromophobe renal cell carcinoma.

Seiten (von - bis)147-156
FachzeitschriftHistochemistry and Cell Biology
PublikationsstatusVeröffentlicht - 4 Aug. 2015


Untersuchen Sie die Forschungsthemen von „Supremacy of modern morphometry in typing renal oncocytoma and malignant look-alikes.“. Zusammen bilden sie einen einzigartigen Fingerprint.