Advertisement

Strahlentherapie und Onkologie

, Volume 194, Issue 6, pp 580–590 | Cite as

Semantic imaging features predict disease progression and survival in glioblastoma multiforme patients

  • Jan C. Peeken
  • Josefine Hesse
  • Bernhard Haller
  • Kerstin A. Kessel
  • Fridtjof Nüsslin
  • Stephanie E. Combs
Original Article

Abstract

Background

For glioblastoma (GBM), multiple prognostic factors have been identified. Semantic imaging features were shown to be predictive for survival prediction. No similar data have been generated for the prediction of progression. The aim of this study was to assess the predictive value of the semantic visually accessable REMBRANDT [repository for molecular brain neoplasia data] images (VASARI) imaging feature set for progression and survival, and the creation of joint prognostic models in combination with clinical and pathological information.

Methods

189 patients were retrospectively analyzed. Age, Karnofsky performance status, gender, and MGMT promoter methylation and IDH mutation status were assessed. VASARI features were determined on pre- and postoperative MRIs. Predictive potential was assessed with univariate analyses and Kaplan–Meier survival curves. Following variable selection and resampling, multivariate Cox regression models were created. Predictive performance was tested on patient test sets and compared between groups. The frequency of selection for single variables and variable pairs was determined.

Results

For progression free survival (PFS) and overall survival (OS), univariate significant associations were shown for 9 and 10 VASARI features, respectively. Multivariate models yielded concordance indices significantly different from random for the clinical, imaging, combined, and combined + MGMT models of 0.657, 0.636, 0.694, and 0.716 for OS, and 0.602, 0.604, 0.633, and 0.643 for PFS. “Multilocality,” “deep white-matter invasion,” “satellites,” and “ependymal invasion” were over proportionally selected for multivariate model generation, underlining their importance.

Conclusions

We demonstrated a predictive value of several qualitative imaging features for progression and survival. The performance of prognostic models was increased by combining clinical, pathological, and imaging features.

Keywords

Biomarker VASARI Prognostic model Radiomics Semantic features 

Semantische Bildeigenschaften prognostizieren Tumorprogression und Überleben von Patienten mit Glioblastoma multiforme

Zusammenfassung

Einleitung

Für das Glioblastoma multiforme wurden bereits multiple prognostische Faktoren identifiziert. Semantische Bildeigenschaften zeigten eine prognostische Aussagekraft für das Überleben. Die Wertigkeit zur Vorhersage der Krankheitsprogression bleibt unklar. Ziele der Studie waren die prognostische Wertigkeit der sog. VASARI-Bildeigenschaften für Krankheitsprogression und Überleben zu evaluieren und kombinierte Vorhersagemodelle unter Beachtung klinischer und pathologischer Informationen zu generieren.

Methoden

Von 189 Patienten wurden retrospektiv Alter, Karnofsky-Index, Geschlecht, MGMT-Promotor-Methylierung und IDH-Mutationsstatus erfasst. “Visually Accessable REMBRANDT [Repository for Molecular Brain Neoplasia Data] Images”(VASARI)-Eigenschaften wurden in prä- und postoperativen MRT-Studien bestimmt. Die Vorhersagekraft einzelner Eigenschaften wurde univariat und anhand von Kaplan-Meyer-Kurven analysiert. Nach Variablenselektion und Resampling wurden multivariate Cox-Regressionsmodelle generiert. Die prädiktive Vorhersagekraft der Modelle wurde mit Testpatienten ermittelt und die Modelle wurden miteinander verglichen. Die Selektionsfrequenz einzelner Variablen und Variablenpaare wurden analysiert.

Ergebnisse

Mit dem progressionsfreien Überleben (PFS) und dem Gesamtüberleben (OS) waren 9 bzw. 10 VASARI-Eigenschaften in der univariaten Analyse signifikant assoziiert. Die Vorhersagekraft der multivariaten Modelle (klinisch, Bildgebung, kombiniert, kombiniert + MGMT) waren mit Konkordanzindizes von je 0,657, 0,636, 0,694 und 0,716 für OS und 0,602, 0,604, 0,633 und 0,643 für PFS signifikant besser als der Zufall. Die Eigenschaften „Multilokalität“, „Invasion tiefer weißer Hirnstrukturen“, „Satellitenherde“ und „ependymale Invasion“ wurden überproportional für die Generierung der multivariaten Modelle selektiert und unterstreichen deren Bedeutung.

Schlussfolgerung

Wir konnten für die VASARI-Eigenschaften eine prädiktive Vorhersagekraft für PFS und OS nachweisen. Die Vorhersagekraft multivariabler Modelle wurde durch Hinzunahme von klinischen und pathologischen Informationen verbessert.

Schlüsselwörter

Biomarker VASARI Prognostisches Modell Radiomics Semantische Information 

Notes

Funding

The work was funded in part by Deutsches Konsortium für Translationale Krebsforschung (DKTK), Partner Site Munich.

Compliance with ethical guidelines

Conflict of interest

J.C. Peeken, J. Hesse, B. Haller, K.A. Kessel, F. Nüsslin, and S.E. Combs declare that they have no competing interests.

Ethical standards

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975 (in its most recently amended version). Informed consent was obtained from all patients included in the study.

Supplementary material

66_2018_1276_MOESM1_ESM.docx (2.7 mb)
Supplemental figures and supplemental data

References

  1. 1.
    Ostrom QT, Gittleman H, Xu J, Kromer C, Wolinsky Y, Kruchko C et al (2016) CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2009–2013. Neuro-Oncology 18:v1–v75.  https://doi.org/10.1093/neuonc/now207 CrossRefPubMedGoogle Scholar
  2. 2.
    Lacroix M, Abi-Said D, Fourney DR, Gokaslan ZL, Shi W, DeMonte F et al (2001) A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 95:190–198.  https://doi.org/10.3171/jns.2001.95.2.0190 CrossRefPubMedGoogle Scholar
  3. 3.
    Zinn PO, Sathyan P, Mahajan B, Bruyere J, Hegi M, Majumder S et al (2012) A novel volume-age-KPS (VAK) glioblastoma classification identifies a prognostic cognate microRNA-gene signature. PLoS ONE 7:1–9.  https://doi.org/10.1371/journal.pone.0041522 CrossRefGoogle Scholar
  4. 4.
    Pope WB, Sayre J, Perlina A, Villablanca JP, Mischel PS, Cloughesy TF (2005) MR imaging correlates of survival in patients with high-grade gliomas. AJNR Am J Neuroradiol 26:2466–2474PubMedGoogle Scholar
  5. 5.
    Curran WJ, Scott CB, Horton J, Nelson JS, Weinstein AS, Fischbach AJ et al (1993) Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group malignant glioma trials. J Natl Cancer Inst 85:704–710CrossRefPubMedGoogle Scholar
  6. 6.
    Filippini G, Falcone C, Boiardi A, Broggi G, Bruzzone MG, Caldiroli D et al (2008) Prognostic factors for survival in 676 consecutive patients with newly diagnosed primary glioblastoma. Neuro-oncology 10:79–87.  https://doi.org/10.1215/15228517-2007-038 CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Burth S, Kickingereder P, Eidel O, Tichy D, Bonekamp D, Weberling L et al (2016) Clinical parameters outweigh diffusion- and perfusion-derived MRI parameters in predicting survival in newly diagnosed glioblastoma. Neuro-Oncology 18:1673–1679.  https://doi.org/10.1093/neuonc/now122 CrossRefPubMedPubMedCentralGoogle Scholar
  8. 8.
    Putz F, Putz T, Goerig N, Knippen S, Gryc T, Eyüpoglu I et al (2016) Improved survival for elderly married glioblastoma patients. Strahlenther Onkol 192:797–805.  https://doi.org/10.1007/s00066-016-1046-0 CrossRefPubMedGoogle Scholar
  9. 9.
    Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ, Janzer RC et al (2009) Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5‑year analysis of the EORTC-NCIC trial. Lancet Oncol 10:459–466.  https://doi.org/10.1016/S1470-2045(09)70025-7 CrossRefPubMedGoogle Scholar
  10. 10.
    Combs SE, Edler L, Rausch R, Welzel T, Wick W, Debus J (2013) Generation and validation of a prognostic score to predict outcome after re-irradiation of recurrent glioma. Acta Oncol (Madr) 52:147–152.  https://doi.org/10.3109/0284186X.2012.692882 CrossRefGoogle Scholar
  11. 11.
    Kessel KA, Hesse J, Straube C, Zimmer C, Schmidt-Graf F, Schlegel J et al (2017) Validation of an established prognostic score after re-irradiation of recurrent glioma. Acta Oncol (Madr) 56:1–5.  https://doi.org/10.1080/0284186X.2016.1276621 CrossRefGoogle Scholar
  12. 12.
    Muth C, Rubner Y, Semrau S, Rühle P‑F, Frey B, Strnad A et al (2016) Primary glioblastoma multiforme tumors and recurrence. Strahlenther Onkol 192:146–155.  https://doi.org/10.1007/s00066-015-0926-z CrossRefPubMedGoogle Scholar
  13. 13.
    Combs SE, Rieken S, Wick W, Abdollahi A, von Deimling A, Debus J et al (2011) Prognostic significance of IDH-1 and MGMT in patients with glioblastoma: one step forward, and one step back? Radiat Oncol 6:115.  https://doi.org/10.1186/1748-717X-6-115 CrossRefPubMedPubMedCentralGoogle Scholar
  14. 14.
    Reifenberger G, Hentschel B, Felsberg J, Schackert G, Simon M, Schnell O et al (2012) Predictive impact of MGMT promoter methylation in glioblastoma of the elderly. Int J Cancer 131:1342–1350.  https://doi.org/10.1002/ijc.27385 CrossRefPubMedGoogle Scholar
  15. 15.
    Weller M, Felsberg J, Hartmann C, Berger H, Steinbach JP, Schramm J et al (2009) Molecular predictors of progression-free and overall survival in patients with newly diagnosed glioblastoma: a prospective translational study of the German Glioma Network. J Clin Oncol 27:5743–5750.  https://doi.org/10.1200/JCO.2009.23.0805 CrossRefPubMedGoogle Scholar
  16. 16.
    Park JK, Hodges T, Arko L, Shen M, Dello Iacono D, McNabb A et al (2010) Scale to predict survival after surgery for recurrent glioblastoma multiforme. J Clin Oncol 28:3838–3843.  https://doi.org/10.1200/JCO.2010.30.0582 CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Hammoud MA, Sawaya R, Shi W, Thall PF, Leeds NE (1996) Prognostic significance of preoperative MRI scans in glioblastoma multiforme. J Neurooncol 27:65–73CrossRefPubMedGoogle Scholar
  18. 18.
    Diehn M, Nardini C, Wang DS, McGovern S, Jayaraman M, Liang Y et al (2008) Identification of noninvasive imaging surrogates for brain tumor gene-expression modules. Proc Natl Acad Sci USA 105:5213–5218.  https://doi.org/10.1073/pnas.0801279105 CrossRefPubMedPubMedCentralGoogle Scholar
  19. 19.
    Stummer W, Reulen HJ, Meinel T, Pichlmeier U, Schumacher W, Tonn JC et al (2008) Extent of resection and survival in glioblastoma multiforme: identification of and adjustment for bias. Neurosurgery 62:564–576.  https://doi.org/10.1227/01.NEU.0000297118.47076.5E CrossRefPubMedGoogle Scholar
  20. 20.
    The Cancer imaging Archive (2014) VASARI research project. https://wiki.cancerimagingarchive.net/display/Public/VASARI+Research+Project. Accessed 29 Dec 2016Google Scholar
  21. 21.
    Mazurowski MA, Desjardins A, Malof JM (2013) Imaging descriptors improve the predictive power of survival models for glioblastoma patients. Neuro-Oncology 15:1389–1394.  https://doi.org/10.1093/neuonc/nos335 CrossRefPubMedPubMedCentralGoogle Scholar
  22. 22.
    Gutman DA, Cooper LAD, Hwang SN, Holder CA, Gao J, Aurora TD et al (2013) MR imaging predictors of molecular profile and survival: multi-institutional study of the TCGA glioblastoma data set. Radiology 267:560–569.  https://doi.org/10.1148/radiol.13120118 CrossRefPubMedPubMedCentralGoogle Scholar
  23. 23.
    Rao A, Rao G, Gutman DA, Flanders AE, Hwang SN, Rubin DL et al (2016) A combinatorial radiographic phenotype may stratify patient survival and be associated with invasion and proliferation characteristics in glioblastoma. J Neurosurg 124:1008–1017.  https://doi.org/10.3171/2015.4.JNS142732 CrossRefPubMedGoogle Scholar
  24. 24.
    Zhou H, Vallières M, Bai HX, Su C, Tang H, Oldridge D et al (2017) MRI features predict survival and molecular markers in diffuse lower-grade gliomas. Neuro-Oncology 19:862–870.  https://doi.org/10.1093/neuonc/now256 CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Gittleman H, Lim D, Kattan MW, Chakravarti A, Gilbert MR, Lassman AB et al (2017) An independently validated nomogram for individualized estimation of survival among patients with newly diagnosed glioblastoma: NRG Oncology RTOG 0525 and 0825. Neuro-Oncology 19:669–677.  https://doi.org/10.1093/neuonc/now208 CrossRefPubMedGoogle Scholar
  26. 26.
    Szopa W, Burley TA, Kramer-Marek G, Kaspera W (2017) Diagnostic and therapeutic biomarkers in glioblastoma: current status and future perspectives. Biomed Res Int.  https://doi.org/10.1155/2017/8013575 PubMedPubMedCentralCrossRefGoogle Scholar
  27. 27.
    Narang S, Lehrer M, Yang D, Lee J, Rao A (2016) Radiomics in glioblastoma : current status, challenges and potential opportunities. Transl Cancer Res 5:383–397.  https://doi.org/10.21037/tcr.2016.06.31 CrossRefGoogle Scholar
  28. 28.
    Peeken JC, Nüsslin F, Combs SE (2017) “Radio-oncomics”: the potential of radiomics in radiation oncology. Strahlenther Onkol 193:767–779.  https://doi.org/10.1007/s00066-017-1175-0 CrossRefPubMedGoogle Scholar
  29. 29.
    Kickingereder P, Radbruch A, Burth S, Wick A, Heiland S, Schlemmer H‑P et al (2016) MR perfusion-derived hemodynamic parametric response mapping of bevacizumab efficacy in recurrent glioblastoma. Radiology 279:542–552.  https://doi.org/10.1148/radiol.2015151172 CrossRefPubMedGoogle Scholar
  30. 30.
    Piroth MD, Holy R, Pinkawa M, Stoffels G, Kaiser HJ, Galldiks N et al (2011) Prognostic impact of postoperative, pre-irradiation 18F-fluoroethyl-l-tyrosine uptake in glioblastoma patients treated with radiochemotherapy. Radiother Oncol 99:218–224.  https://doi.org/10.1016/j.radonc.2011.03.006 CrossRefPubMedGoogle Scholar
  31. 31.
    Leitzen C, Wilhelm-Buchstab T, Schmeel LC, Garbe S, Greschus S, Müdder T et al (2016) MRI during radiotherapy of glioblastoma. Strahlenther Onkol 192:481–488.  https://doi.org/10.1007/s00066-016-0983-y CrossRefPubMedGoogle Scholar
  32. 32.
    Dickerson E, Davenport MS, Syed F, Stuve O, Cohen JA, Rinker JR et al (2017) Effect of template reporting of brain MRIs for multiple sclerosis on report thoroughness and neurologist-rated quality: results of a prospective quality improvement project. J Am Coll Radiol 14:371–379.e1.  https://doi.org/10.1016/j.jacr.2016.09.037 CrossRefPubMedGoogle Scholar
  33. 33.
    Grossmann P, Gutman DA, Dunn WD, Holder CA, Aerts HJWL (2016) Imaging-genomics reveals driving pathways of MRI derived volumetric tumor phenotype features in glioblastoma. BMC Cancer 16:611.  https://doi.org/10.1186/s12885-016-2659-5 CrossRefPubMedPubMedCentralGoogle Scholar
  34. 34.
    Gutman DA, Dunn WD, Grossmann P, Cooper LAD, Holder CA, Ligon KL et al (2015) Somatic mutations associated with MRI-derived volumetric features in glioblastoma. Neuroradiology 57:1227–1237.  https://doi.org/10.1007/s00234-015-1576-7 CrossRefPubMedPubMedCentralGoogle Scholar
  35. 35.
    Rios Velazquez E, Meier R, Dunn WD, Alexander B, Wiest R, Bauer S et al (2015) Fully automatic GBM segmentation in the TCGA-GBM dataset: prognosis and correlation with VASARI features. Sci Rep 5:16822.  https://doi.org/10.1038/srep16822 CrossRefPubMedPubMedCentralGoogle Scholar
  36. 36.
    Meier R, Porz N, Knecht U, Loosli T, Schucht P, Beck J et al (2017) Automatic estimation of extent of resection and residual tumor volume for patients with glioblastoma. J Neurosurg.  https://doi.org/10.3171/2016.9.JNS16146 CrossRefPubMedGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  • Jan C. Peeken
    • 1
    • 4
  • Josefine Hesse
    • 1
  • Bernhard Haller
    • 3
  • Kerstin A. Kessel
    • 1
    • 3
    • 4
  • Fridtjof Nüsslin
    • 1
  • Stephanie E. Combs
    • 1
    • 2
    • 4
  1. 1.Department of Radiation Oncology, Klinikum rechts der IsarTechnical University Munich (TUM)MunichGermany
  2. 2.Institute of Innovative Radiotherapy (iRT), Department of Radiation Sciences (DRS)Helmholtz Zentrum MünchenNeuherbergGermany
  3. 3.Institut for Medical Statistics and EpidemiologyTechnical University Munich (TUM)MunichGermany
  4. 4.Deutsches Konsortium für Translationale Krebsforschung (DKTK), Partner Site MunichMunichGermany

Personalised recommendations