Data-driven surgical decision-making – the prediction of resectability in patients with glioblastoma using machine learning

Brain and Spine, vol. 1, pp. 100502

Abstract

Background: The surgical goal in LGG treatment is maximal safe resection.It is still not clear the impact of the extent of resection (EOR), with special regards to the residual tumour volume (RTV), correlated to overall survival (OS).The impact of tumour molecular subgrouping has to be defined in light of the former parameters.Methods: The patients included in this study presented with a lower-grade glioma (LGG) at the Neurological Institute "C.Besta" between 1994Besta" between and 2018 (n¼1918) (n¼1918).For this initial analysis, we considered those patients who underwent surgery between 2012 and 2018 (n¼327); all clinical, imaging, histopathological and molecular data were available for 299 patients.All specimens were analysed according to the 2016 WHO tumours classification; tumour volumes were determined using Medtronic software based on T1 and FLAIR MRI sequences acquired before the operation and one month after surgery.The EOR was defined as gross total resection (GTR), subtotal resection (STR) or partial resection (PR).Residual tumour volume (RTV) was measured.Results: GTR was confirmed to be a positive prognostic factor, improving OS in the IDH wild-type group (n¼96; P¼0.054); GTR in the 1p/19q co-deleted group (n¼101; P¼0.320) was associated with reduced benefit in OS.OS in the IDH mutant group (n¼102) seems to be independent of the percentage of resection (P¼0.530), while it declines as residual volume increases (IC 95%, 0.238-0.752;P¼0.000).Conclusions: EOR seems to have different survival impact according to different histological and molecular stratification: the benefit is diminished in the 1p/19q co-deleted group and maximal in the IDH wild-type group.RTV is reversely associated with OS in the IDH mutant group.

Authors 9

1. Julius M. Kernbach Aachen Faculty of Medicine Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA) Department of Neurosurgery

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Department of Neurosurgery, Faculty of Medicine, Aachen, Germany

   RWTH Aachen University, Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA), Aachen, Germany

2. Georg Neuloh Aachen Faculty of Medicine Department of Neurosurgery

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Department of Neurosurgery, Faculty of Medicine, Aachen, Germany

3. Jonas Ort Aachen Faculty of Medicine Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA) Department of Neurosurgery

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Department of Neurosurgery, Faculty of Medicine, Aachen, Germany

   RWTH Aachen University, Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA), Aachen, Germany

4. Karlijn Hakvoort Aachen Faculty of Medicine Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA) Department of Neurosurgery

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Department of Neurosurgery, Faculty of Medicine, Aachen, Germany

   RWTH Aachen University, Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA), Aachen, Germany

5. Lasse Dührsen

   University Medical Center Hamburg-Eppendorf

   Affiliation as printed

   University Hospital Hamburg-Eppendorf, Department of Neurosurgery, Hamburg, Germany

6. Franz Lennard Ricklefs

   University Medical Center Hamburg-Eppendorf

   Affiliation as printed

   University Hospital Hamburg-Eppendorf, Department of Neurosurgery, Hamburg, Germany

7. Stefan Josef Grau

   University Hospital Cologne

   Affiliation as printed

   University Hospital Köln, Department of Neurosurgery, Köln, Germany

8. Hans Clusmann Aachen Faculty of Medicine Department of Neurosurgery

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Department of Neurosurgery, Faculty of Medicine, Aachen, Germany

9. Daniel Delev Aachen Faculty of Medicine Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA) Department of Neurosurgery

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Department of Neurosurgery, Faculty of Medicine, Aachen, Germany

   RWTH Aachen University, Neurosurgical Artificial Intelligence Laboratory Aachen (NAILA), Aachen, Germany