Dynamic contrast enhanced MRI detects early response to adoptive NK cellular immunotherapy targeting the NG2 proteoglycan in a rat model of glioblastoma

PLoS One. 2014 Sep 30;9(9):e108414. doi: 10.1371/journal.pone.0108414. eCollection 2014.

Abstract

There are currently no established radiological parameters that predict response to immunotherapy. We hypothesised that multiparametric, longitudinal magnetic resonance imaging (MRI) of physiological parameters and pharmacokinetic models might detect early biological responses to immunotherapy for glioblastoma targeting NG2/CSPG4 with mAb9.2.27 combined with natural killer (NK) cells. Contrast enhanced conventional T1-weighted MRI at 7±1 and 17±2 days post-treatment failed to detect differences in tumour size between the treatment groups, whereas, follow-up scans at 3 months demonstrated diminished signal intensity and tumour volume in the surviving NK+mAb9.2.27 treated animals. Notably, interstitial volume fraction (ve), was significantly increased in the NK+mAb9.2.27 combination therapy group compared mAb9.2.27 and NK cell monotherapy groups (p = 0.002 and p = 0.017 respectively) in cohort 1 animals treated with 1 million NK cells. ve was reproducibly increased in the combination NK+mAb9.2.27 compared to NK cell monotherapy in cohort 2 treated with increased dose of 2 million NK cells (p<0.0001), indicating greater cell death induced by NK+mAb9.2.27 treatment. The interstitial volume fraction in the NK monotherapy group was significantly reduced compared to mAb9.2.27 monotherapy (p<0.0001) and untreated controls (p = 0.014) in the cohort 2 animals. NK cells in monotherapy were unable to kill the U87MG cells that highly expressed class I human leucocyte antigens, and diminished stress ligands for activating receptors. A significant association between apparent diffusion coefficient (ADC) of water and ve in combination NK+mAb9.2.27 and NK monotherapy treated tumours was evident, where increased ADC corresponded to reduced ve in both cases. Collectively, these data support histological measures at end-stage demonstrating diminished tumour cell proliferation and pronounced apoptosis in the NK+mAb9.2.27 treated tumours compared to the other groups. In conclusion, ve was the most reliable radiological parameter for detecting response to intralesional NK cellular therapy.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Animals
  • Antibodies, Monoclonal / pharmacology*
  • Antigens / genetics
  • Antigens / immunology
  • Brain Neoplasms / genetics
  • Brain Neoplasms / immunology
  • Brain Neoplasms / pathology
  • Brain Neoplasms / therapy*
  • Diffusion
  • Disease Models, Animal
  • Extracellular Fluid / chemistry
  • Extracellular Fluid / drug effects
  • Extracellular Fluid / metabolism
  • Female
  • Gene Expression
  • Glioblastoma / genetics
  • Glioblastoma / immunology
  • Glioblastoma / pathology
  • Glioblastoma / therapy*
  • Image Enhancement
  • Immunotherapy, Adoptive*
  • Injections, Intralesional
  • Killer Cells, Natural / immunology
  • Killer Cells, Natural / transplantation*
  • Magnetic Resonance Imaging
  • Male
  • Molecular Targeted Therapy
  • Proteoglycans / antagonists & inhibitors*
  • Proteoglycans / genetics
  • Proteoglycans / immunology
  • Rats
  • Rats, Nude
  • Tumor Burden / drug effects

Substances

  • Antibodies, Monoclonal
  • Antigens
  • Proteoglycans
  • chondroitin sulfate proteoglycan 4

Grants and funding

This work was supported by grants from The Bergen Medical Research Foundation, The Meltzer Research Fund, The Norwegian Research Council and The Norwegian Cancer Society. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.