gms | German Medical Science

70. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie (DGNC)
Joint Meeting mit der Skandinavischen Gesellschaft für Neurochirurgie

Deutsche Gesellschaft für Neurochirurgie (DGNC) e. V.

12.05. - 15.05.2019, Würzburg

Rapid and label-free recognition of brain tumours by Raman spectroscopy of intraoperative biopsies – a feasibility study of 210 patients

Schnelle und markierungsfreie Erkennung von Hirntumoren mittels Raman-Spektroskopie intraoperativer Biopsien – eine Machbarkeitsstudie an 210 Patienten

Meeting Abstract

  • presenting/speaker Roberta Galli - Technische Universität Dresden, Medizinische Fakultät Carl Gustav Carus – Klinisches Sensoring and Monitoring, Dresden, Deutschland
  • Ortrud Uckermann - Technische Universität Dresden, Universitätsklinikum Carl Gustav Carus – Klinik für Neurochirurgie, Dresden, Deutschland
  • Edmund Koch - Technische Universität Dresden, Medizinische Fakultät Carl Gustav Carus – Klinisches Sensoring and Monitoring, Dresden, Deutschland
  • Gabriele Schackert - Nationales Centrum für Tumorerkrankungen (NCT), Partnerstandort Dresden, Dresden, Deutschland; Technische Universität Dresden, Universitätsklinikum Carl Gustav Carus – Klinik für Neurochirurgie, Dresden, Deutschland
  • Gerald Steiner - Technische Universität Dresden, Medizinische Fakultät Carl Gustav Carus – Klinisches Sensoring and Monitoring, Dresden, Deutschland
  • Matthias Kirsch - Nationales Centrum für Tumorerkrankungen (NCT), Partnerstandort Dresden, Dresden, Deutschland; Technische Universität Dresden, Universitätsklinikum Carl Gustav Carus – Klinik für Neurochirurgie, Dresden, Deutschland

Deutsche Gesellschaft für Neurochirurgie. 70. Jahrestagung der Deutschen Gesellschaft für Neurochirurgie (DGNC), Joint Meeting mit der Skandinavischen Gesellschaft für Neurochirurgie. Würzburg, 12.-15.05.2019. Düsseldorf: German Medical Science GMS Publishing House; 2019. DocV081

doi: 10.3205/19dgnc096, urn:nbn:de:0183-19dgnc0961

Veröffentlicht: 8. Mai 2019

© 2019 Galli et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe http://creativecommons.org/licenses/by/4.0/.


Gliederung

Text

Objective: In brain surgery, novel technologies are continuously developed to achieve better tumour delineation and optimize the extent of resection. Raman spectroscopy is a pure optical method that enables to retrieve a signature of tissue biochemical composition in order to discern between tumour and normal tissue. While its potential was formerly demonstrated in the frame of biophysical research, we show here its translation in surgical practice.

Methods: Fresh unprocessed biopsies of patients undergoing brain tumour surgery were analysed over one year including all patients that gave consent (glioblastoma n=75, oligodendroglioma n=10, astrocytoma n=14, meningioma n=53, schwannoma n=8, other neoplastic entities n=17, brain metastases n=26). Non-neoplastic tissue was obtained from surgical treatment of patients affected by drug-resistant epilepsy (n=7). For Raman spectroscopy, tissue samples as received from the operating room were placed under a microscope (objective magnification 10x) and irradiated with a near-infrared laser to generate the Raman signal, which was registered by a Raman spectrometer. Measurements were performed by nurses trained for the purpose, without the supervision of a scientist. Five to seven spectra were acquired for each biopsy with total analysis duration of about 5 minutes. The spectra were subjected to dimensionality reduction by principal component analysis and a supervised classification approach was used to discriminate non-tumour tissue against tumour.

Results: All biopsies except one brain metastases were successfully measured and all 1070 measured spectra had a quality suited for classification, showing that the measurement can be performed routinely by medical personnel. The classification of non-neoplastic tissue against all tumour biopsies led to the correct recognition of 100% of non-neoplastic samples and of 97% of tumours. All recurrent glioblastoma, meningioma, schwannoma and metastases were correctly recognized. Primary glioblastoma was recognized with a correct rate of 94%. Oligodendroglioma and astrocytoma were recognized with correct rates of 90% and 86% respectively.

Conclusion: We demonstrated the feasibility of label-free recognition of brain tumours from non-neoplastic brain by Raman spectroscopy. Analysis of fresh biopsies is rapid and may be included in the surgical work flow. Furthermore, modern portable Raman devices exist that may enable in situ brain analysis for resection optimization.