Cerebral blood volume, genotype and chemosensitivity in oligodendroglial tumours

Abstract

Introduction: The biological factors responsible for differential chemoresponsiveness in oligodendroglial tumours with or without the -1p/-19q genotype are unknown, but tumour vascularity may contribute. We aimed to determine whether dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI) could distinguish molecular subtypes of oligodendroglial tumour, and examined the relationship between relative cerebral blood volume (rCBV) and outcome following procarbazine, lomustine and vincristine (PCV) chemotherapy.

Methods: Pretherapy rCBV was calculated and inter- and intraobserver variability assessed. Allelic imbalance in 1p36, 19q13, 17p13, 10p12-15, and 10q22-26 and p53 mutation (exons 5-8) were determined. rCBV was compared with genotype and clinicopathological characteristics (n=37) and outcome following PCV chemotherapy (n=33).

Results: 1p/19q loss was seen in 6/9 grade II oligodendrogliomas, 6/14 grade II oligoastrocytomas, 4/4 grade III oligodendrogliomas, and 3/10 grade III oligoastrocytomas. rCBV measurements had good inter- and intraobserver variability, but did not distinguish histology subtype or grade. Tumours with 1p/19q loss had higher rCBV values (Student's t-test P=0.001). Receiver operating characteristic analysis revealed a cut-off of 1.59 for identifying genotype (sensitivity 92%, specificity 76%). Tumours with high and low rCBV showed response to chemotherapy. The -1p/-19q genotype, but not rCBV, was strongly associated with response, progression-free and overall survival following PCV chemotherapy. Tumours with high rCBV and intact 1p/19q were associated with shorter progression-free and overall patient survival than those with intact 1p/19q and low rCBV or high rCBV and 1p/19q loss.

Conclusion: rCBV identifies oligodendroglial tumours with 1p/19q loss, but does not predict chemosensitivity. The prognostic significance of rCBV may differ in oligodendroglial tumours with or without the -1p/-19q genotype.

Fig. 1

Calculation of rCBV using DSC-MR. CBV is derived from the drop in the T2* signal after injection of a bolus of paramagnetic compound [8, 9]. Using Functool2, the radiological tumour margin was delineated on the T2* sequence axial image and a single circular ROI placed over normal brain mirrored in the contralateral hemisphere (a, b). Negative enhancement integral (NEI) colour maps were generated (c, d) relative to the signal intensity in the superior sagittal sinus using image data between dynamic scans immediately before and after the contrast transient. CBV is proportional to the area under the contrast agent concentration-time curve and was calculated for each ROI by Functool2 using algorithms to integrate ΔR2i values (where ΔR2 is the change in the reciprocal of T2*). rCBV for each tumour was calculated from signal intensity time curves for individual ROIs placed within the delineated tumour region or in the large ROI in the contralateral hemisphere (as illustrated in e and f, arrows scans adjacent to the contrast transient) as described in the Materials and methods. a, c, e Low rCBV (0.73±0.17, mean±SD) in a grade II oligoastrocytoma with intact chromosomes 1p and 19q and; b, d, f high rCBV (4.15±0.87) in a grade II oligoastrocytoma with the −1p/−19q genotype

Fig. 2

Bland and Altman plots illustrating interobserver (a, b) and intraobserver (c) variation of rCBV measurements. The greatest variation is seen at the right of the plot for tumours with the higher rCBV values

Fig. 3

Box plots of rCBV against (a) histopathology subtype (P=0.279*), (b) histopathology grade (P=0.442*), and (c) genotype (P=0.001*) in oligodendroglial tumours. The optimal rCBV cut-off that distinguishes tumour genotype is shown by the horizontal line. *Student’s t-test

Fig. 4

rCBV, molecular genetics and clinical characteristics. rCBV for each tumour is given as the mean±standard deviation of three independent observations. The arrow indicates tumours with rCBV above (high rCBV) and below the cut-off value of 1.59 (low rCBV). CR complete response (disappearance of all tumour, off steroids and neurologically stable or improved), PR partial response (50% or greater reduction in cross-sectional area, steroids stable or reduced, and neurologically stable or improved); MR minor response (>25 to <50% reduction in cross-sectional area, steroids stable or reduced, and neurologically stable or improved); PD progressive disease (25% or greater increase in cross-sectional area or any new tumour on CT/MR images and/or neurologically worse with steroids stable or increased); SD stable disease (all other situations); P primary tumour, R recurrent tumour

Fig. 5

rCBV and outcome following PCV chemotherapy: a, b all patients; c, d patients with primary tumour only; a, c progression-free survival; b, d overall survival.① Patients with tumours with low rCBV (<1.59) and loss of 1p36 and 19q13 (all n=2, primary n=2).② Patients with tumours with high rCBV (>1.59) and loss of 1p36 and 19q13 (all n=15, primary n=12).③ Patients with tumours with low rCBV (<1.59) with intact 1p36 and 19q13 (all n=12, primary n=10).④ Patients with tumours with high rCBV (>1.59) and intact 1p36 and 19q13 (all n=4, primary n=2)