Clinical Oncology xxx (2014) 1e13 Contents lists available at ScienceDirect

Clinical Oncology journal homepage: www.clinicaloncologyonline.net

Overview

Imaging and Target Volume Delineation in Glioma G.A. Whitfield *, S.R. Kennedy *, I.K. Djoukhadar y, A. Jackson y * The y

Christie NHS Foundation Trust, Manchester, UK Wolfson Molecular Imaging Centre, University of Manchester, Manchester, UK

Received 2 April 2014; accepted 11 April 2014

Abstract Here we review current practices in target volume delineation for radical radiotherapy planning for gliomas. Current radiotherapy planning margins for glioma are informed by historic data of recurrence patterns using radiological imaging or post-mortem studies. Radiotherapy planning for World Health Organization grade IIeIV gliomas currently relies predominantly on T1-weighted contrast-enhanced magnetic resonance imaging (MRI) and T2/fluid-attenuated inversion recovery sequences to identify the gross tumour volume (GTV). Isotropic margins are added empirically for each tumour type, usually without any patientspecific individualisation. We discuss novel imaging techniques that have the potential to influence radiotherapy planning, by improving definition of the tumour extent and its routes of invasion, thus modifying the GTV and allowing anisotropic expansion to a clinical target volume better reflecting areas at risk of recurrence. Identifying the relationships of tumour boundaries to important white matter pathways and eloquent areas of cerebral cortex could lead to reduced normal tissue complications. Novel magnetic resonance approaches to identify tumour extent and invasion include: (i) diffusion-weighted magnetic resonance metrics; (ii) diffusion tensor imaging; and (iii) positron emission tomography, using radiolabelled amino acids methyl-11C-L-methionine and 18F-fluoroethyltyrosine. Novel imaging techniques may also have a role together with clinical characteristics and molecular genetic markers in predicting response to therapy. Most significant among these techniques is dynamic contrast-enhanced MRI, which uses dynamic acquisition of images after injection of intravenous contrast. A number of studies have identified changes in diffusion and microvascular characteristics occurring during the early stages of radiotherapy as powerful predictive biomarkers of outcome. Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Amino acid PET; glioma; imaging biomarkers; molecular resonance imaging; radiotherapy planning; target volume delineation

Statement of Search Strategies Used and Sources of Information This paper reflects expert opinion and current literature accessed by the authors; no formal search strategy has been defined.

Introduction High-grade [World Health Organization (WHO) grade III and IV] gliomas comprise most malignant primary central nervous system tumours in adults; most are glioblastoma (GBM, WHO grade IV astrocytoma). Low-grade gliomas are much less common in adults and most are WHO grade II Author for correspondence: G. Whitfield, The Christie NHS Foundation Trust, 550 Wilmslow Road, Manchester M20 4BX, UK. Tel: þ44-161-4468095; Fax: þ44-161-446-3265. E-mail address: gillian.whitfi[email protected] (G.A. Whitfield).

gliomas. In 2010, the age-adjusted incidence of primary central nervous system and intracranial tumours was 12 per 100 000 population in England [1]. By tumour type, over the period 2006e2010 in England, 34% were astrocytomas (95% of these high grade, 80% GBM), 3% oligodendrogliomas, 2% ependymomas and 6% other (unspecified) gliomas [1]. This overview focuses on WHO grade IIeIV astrocytic, oligodendroglial and oligoastrocytic tumours. It does not cover WHO grade I gliomas, which are rare in adults, ependymomas and other rare gliomas. Radiotherapy has a major role in the management of WHO grade IIeIV gliomas. The current standard of care for newly diagnosed GBM in patients of good performance status and aged up to 70 years is maximal safe surgical debulking, followed by adjuvant radiotherapy to a dose of 60 Gy in 30 fractions, with concurrent and adjuvant temozolomide chemotherapy [2,3]. For WHO grade III gliomas, the standard of care is maximal safe surgical debulking and radiotherapy, which in the case of tumours with 1p and 19q chromosomal deletion is supplemented with procarbazine,

0936-6555/$36.00 Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clon.2014.04.026

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

2

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

lomustine and vincristine (PCV) chemotherapy either before or after the radiotherapy [4,5]. In recent and ongoing trials in WHO grade III glioma, the radiotherapy dose has usually been 59.4 Gy in 33 fractions [4e6]. For WHO grade II gliomas, when radiotherapy is given, doses of 50.4e54 Gy in 1.8 Gy fractions are standard, as higher doses do not improve progression-free or overall survival [7,8]. Early postoperative radiotherapy of 54 Gy in 30 fractions has a progression-free survival benefit and a benefit in seizure control compared with the same radiotherapy given on tumour progression, but not an overall survival benefit [9]. More recent trials are addressing the role of initial chemotherapy instead of radiotherapy, the role of chemoradiotherapy and the effect of 1p/19q chromosomal deletion on outcomes, but at present observation, surgery, radiotherapy and chemotherapy all have a place in the initial management of grade II gliomas, depending on the patient’s age and symptoms, the histological type, evidence of 1p/19q deletion, the size and the location of the tumour [10]. Here we review current practices in target volume delineation for radical radiotherapy planning for gliomas. We discuss current magnetic resonance imaging (MRI) techniques and developments in imaging that might influence future radiotherapy for gliomas.

Current Practices in Radiotherapy Planning for Gliomas Imaging for Radical Radiotherapy Planning Radical radiotherapy planning should use both an MRI and planning computed tomography for tumour delineation and dosimetry. Computed tomography is acquired with the patient supine in a custom-made immobilisation shell; it typically uses 3 mm slice spacing and is acquired postcontrast. The planning MRI is acquired without the immobilisation device. For high-grade gliomas, the most helpful MRI sequence is the T1-weighted post-gadolinium (T1 gd) sequence. If there is also a low-grade component, a T2weighted or preferably a fluid-attenuated inversion recovery (FLAIR) sequence should also be obtained. For low-grade gliomas, only the FLAIR (or T2-weighted) sequence is needed; however, a T1 contrast-enhanced scan may also be obtained to exclude new contrast enhancement indicative of high-grade transformation. The MRI may be acquired as axial slices or as a three-dimensional volumetric MRI reconstructed into axial slices. The MRI is co-registered with the planning computed tomography and both are accessible as one combined imaging data set on the radiotherapy planning software. Target Volume Delineation for World Health Organization Grade IV Gliomas A standard approach to target volume delineation for WHO grade IV gliomas is to define the gross tumour volume (GTV) as the surgical cavity, plus areas of enhancement on the T1 gd MRI, which represents residual macroscopic

disease. The presurgical scans, early postoperative MRI and operation note may all help in interpreting the appearances. Treatment-related contrast enhancement usually does not appear for 3e4 days after surgery and therefore postoperative MRI within the first 24e48 h helps to assess the presence of residual tumour. The clinical target volume (CTV) is obtained by applying a uniform (isotropic) expansion to the GTV, most often of 2.0e2.5 cm, to include the highest density of microscopic disease. The CTV should be carefully edited, taking into account anatomical boundaries to tumour spread, such as the skull, tentorium and falx, but bearing in mind that tumour may spread around such boundaries, e.g. via the corpus callosum to the contralateral hemisphere, or via the cerebral peduncles to the brainstem. The CTV to planning target volume (PTV) expansion depends on geometric uncertainties (particularly set-up variation), which may vary among radiotherapy departments; 5 mm is most often used. However, there is no complete consensus on target volumes for WHO grade IV gliomas. Both the European Organisation for the Research and Treatment of Cancer (EORTC) and the US/Canadian Radiation Therapy Oncology Group (RTOG) guidelines (Table 1), developed for use in clinical trials, have been widely adopted. The EORTC method [3] is single phase. The GTV is the surgical cavity plus the T1 gd enhancing volume. The CTV is a 2.0e3.0 cm expansion of the GTV, without the intentional inclusion of peritumoural oedema. The CTV to PTV margin is typically 5 mm. The RTOG method [12] is two phase. In phase 1 the surgical cavity, T1 gd contrast enhancement and peritumoural oedema are treated with a 2.0e2.5 cm expansion to CTV. In phase 2, the surgical cavity plus T1 gd enhancing volume only with a 2 cm expansion to CTV are treated. Current radiotherapy planning margins for high-grade glioma are informed by historic data on patterns of recurrence after radiotherapy on radiological imaging or at postmortem, in which around 80e100% of recurrences occurred within 2 cm of the initial contrast-enhancing tumour [15e17]. As a result, whole brain radiotherapy with a boost to macroscopic disease was superseded by partial brain treatment. Although current margins may not include all microscopic disease, treating larger volumes, which limits the dose that can be given, will not necessarily increase survival for these radio-resistant tumours. Burger et al. [18] and Halperin et al. [19] identified glioma cells within the region of peritumoural oedema, but both also identified glioma cells beyond that. More recently, Farace et al. [20] showed that huge changes in oedema are observed between pre- and postoperative MRI, which may be a consequence of steroid treatment and changes in mass effect and does not support the deliberate inclusion of the T2 abnormality in the CTV. In the modern era of radiotherapy with temozolomide, Minniti et al. [21] analysed 105 patients with recurrent GBM treated to a CTV comprising enhancing tumour plus 2 cm. They also constructed theoretical plans, including peritumoural oedema with a 2 cm margin, and concluded that treating the smaller volumes (without intentional inclusion of oedema) reduced the brain volumes treated to a high dose without a significant increase in the

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

Guideline WHO grade IV glioma EORTC (from EORTC 22981/ 26981/NCIC CE3) [3] RTOG (from RTOG 0825) [12]

WHO grade III glioma EORTC 26053/22054, the CATNON Intergroup trial (non 1p/19q deleted WHO grade III gliomas) [6] EORTC 26951 [5] (anaplastic oligodendroglioma and oligoastrocytoma)

Phases and dose

GTV

CTV

PTV

Single phase (60 Gy in 30 fractions) Phase 1 (46 Gy in 23 fractions)

GTV þ 2e3 cm

Not stated

Phase 2 (14 Gy in 7 fractions)

Surgical cavity þ T1 gd abnormality GTV1 ¼ surgical cavity þ T1 gd abnormality þ T2/FLAIR abnormality GTV2 ¼ surgical cavity þ T1 gd abnormality

CTV1 ¼ GTV1 þ 2 cm (if oedema present) or GTV1 þ 2.5 cm (if no oedema) CTV2 ¼ GTV2 þ 2 cm

PTV1 ¼ CTV1 þ 3e5 mm, depending on centre’s reproducibility. PTV2 ¼ CTV2 þ 3e5 mm

Single phase (59.4 Gy in 33 fractions)

GTV ¼ surgical cavity þ T2/FLAIR abnormality þ T1 gd abnormality

GTV þ 1.5e2 cm, except at anatomic borders, where 7e10 mm is sufficient

CTV þ 5e7 mm

Not specified, but equates approximately to CTV1 ¼ preoperative T2 abnormality þ 2 cm Equates approximately to CTV2 ¼ postoperative T2 abnormality þ T1 gd abnormality þ 1 cm Not specified, but equates approximately to CTV1 ¼ surgical cavity þ T2 abnormality þ 1 cm Equates approximately to CTV2 ¼ surgical cavity þ T1 gd abnormality

PTV1 ¼ preoperative T2 abnormality (or hypodense area on preoperative CT) þ 2.5 cm PTV2 ¼ postoperative enhancing and nonenhancing tumour (on CT or MRI) þ 1.5 cm Phase 1 field ¼ surgical cavity þ T2 abnormality þ 2 cm

GTV þ 1e1.5 cm, except at anatomic borders where 5 mm is sufficient Not specified, but equates approximately to GTV þ 1.0 cm margin

CTV þ 5e7 mm

Phase 1 (45 Gy in 25 fractions)

Phase 2 (14.4 Gy in 8 fractions)

RTOG 9402 [4] (anaplastic oligodendroglioma and oligoastrocytoma)

Phase 1 (50.4 Gy in 28 fractions)

Phase 2 (9 Gy in 5 fractions) WHO grade II glioma EORTC (from EORTC 22033-26033/CE5 trial) [13] RTOG (from 9802) [14]

Single phase (50.4 Gy in 28 fractions) Single phase (54 Gy in 30 fractions)

Surgical cavity þ T2/FLAIR abnormality (corresponding to hypodense area on CT), including any possible CT enhancement. T2 or FLAIR abnormality

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

Phase 2 field ¼ surgical cavity þ T1 gd abnormality þ 1 cm

GTV þ 2 cm margin to block edge

T1 gd, T1-weighted post-gadolinium; FLAIR, fluid-attenuated inversion recovery; CT, computed tomography; MRI, magnetic resonance imaging.

3

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

Table 1 European Organization for Research and Treatment of Cancer (EORTC) and Radiation Therapy Oncology Group (RTOG) target volume delineation guidelines for World Health Organization (WHO) grade IIeIV gliomas. Note that EORTC 26951 specified the planning target volume (PTV) directly, whereas RTOG 9402 and RTOG 9802 specified only the radiotherapy fields. These approaches have been superseded in modern radical radiotherapy planning, which requires the gross tumour volume (GTV), clinical target volume (CTV) and PTV to be specified as per International Commission on Radiation Units and Measurement terminology [11]

4

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

risk of marginal recurrences. Wick et al. [22] conducted a longitudinal analysis of MRI recurrence patterns in 63 patients with unilateral GBM in the EORTC 22981/26981/NCIC CE3 trial, of whom 30 were treated with radiotherapy and the remainder with radiotherapy plus temozolomide. A recurrence was considered distant if at least 50% of the tumour mass was outside the borders of the initial T1enhancing tumour plus a 2 cm margin. Overall, 20% of recurrences were distant, with no demonstrable difference in size of recurrence, or distance from preoperative tumour location between the two arms. Recently, some groups have reported retrospective series using smaller margins with comparable proportions of in and out of field failures as would be expected with conventional margins [23e25]. However, because of the inclusion of peritumoural oedema with a small margin of 5e10 mm in the first phase CTV (although this is not clearly stated by Paulsson et al. [25]), it is difficult to compare these margins with margins based on the EORTC approach excluding oedema. Therefore, as yet the available data do not support margin reduction. Recent surveys have shown significant variability in radiotherapy planning for high-grade glioma [26,27]. In the UK survey [26], 31/46 respondents routinely used MRI/ computed tomography co-registration. As GTV, 27/46 delineated the contrast-enhancing tumour, 12/46 the tumour on preoperative imaging and 1/46 the tumour plus peritumoural oedema; 6/46 did not state their method. For the PTV, 41/46 used the GTV plus 2.0e3.0 cm, 2/46 GTV plus only 4e5 mm and 2/46 included peritumoural oedema. However, 18/46 always or sometimes extended the PTV to include peritumoural oedema. A single phase was used by 30/46, whereas 10/46 used two phases (the first to 40e50 Gy and the total to 60 Gy) and 6/46 used either approach. Most (42/46) routinely outlined at least one organ at risk. If organ at risk doses exceeded tolerance, 18/42 reduced the total dose, 27/42 compromised the PTV and 10/ 46 accepted doses above tolerance; 6/46 did not specify their action. With increasing use of intensity-modulated radiotherapy (IMRT), this variability may reduce, as many centres will probably move to a single phase approach and use dose inhomogeneity to achieve organ at risk sparing where desired. Target Volume Delineation for World Health Organization Grade III Gliomas WHO grade III gliomas are a heterogeneous group. In the past, ‘high-grade gliomas’ were often treated as a single entity. In 2007, Burnet et al. [28] reported that patients with grade III histology but imaging features of GBM have a prognosis equivalent to that of GBM and should be treated as such. More recently, in the RTOG 9402 trial of neoadjuvant PCV, a median survival of 14.7 years was shown for optimally treated 1p/19q deleted grade III gliomas, but only around 2.7 years for non-co-deleted grade III gliomas [4]. It therefore seems likely that particular grade III glioma subtypes may benefit from different approaches to target volume delineation; however, there is currently no hard evidence for this. WHO grade III gliomas that have the

radiological appearance of GBM, with extensive contrast enhancement and necrosis, might best be treated exactly as for GBM, in which case we would argue that the T2/FLAIR abnormality need not intentionally be included in the target volumes. On the other hand, for WHO grade III gliomas that lack or have only scant contrast enhancement, the T2/FLAIR abnormality as well as any T1 gd enhancement should be included in the GTV. Objective evidence for the GTVeCTV margin around the T2/FLAIR abnormality is lacking, but arguably it should not be less than the 1.0e 1.5 cm that would be used for WHO grade II gliomas and probably need not be as large as the 2.0e3.0 cm margin that would be used around the T1 gd enhancing tumour in GBM. In addition, we would suggest that it would be prudent to ensure that the T1 gd enhancing tumour is covered with a GTVeCTV margin of 2.0e2.5 cm. To achieve this, it may be helpful to delineate separately a GTVT2/FLAIR and a GTVT1gd, which together make up the GTV, but can be grown by their respective margins and combined to create the CTV. Again the CTVePTV expansion depends on departmental geometric uncertainties and would most often be 5 mm. The RTOG and EORTC approaches are shown in Table 1. The differences between EORTC [5,6] and RTOG [4] approaches partly reflect the era in which trials were initiated, as the earlier protocols did not use the now ubiquitous International Commission on Radiation Units and Measurement (ICRU) terminology. The protocol for the ongoing CATNON trial is the most helpful, as it reflects modern radiotherapy planning techniques [6]. In future, analysis of patterns of failure according to grade III subtype should help refine the target volumes. Target Volume Delineation for Low-grade Gliomas (World Health Organization Grade II) For WHO grade II gliomas, the GTV is the surgical cavity plus the MRI FLAIR or T2 abnormality. The CTV is obtained by expanding the GTV isotropically by 1e1.5 cm and then editing this, taking into account anatomical boundaries, as explained above. The CTV to PTV expansion is usually 5 mm. For low-grade gliomas, the EORTC and RTOG guidelines are closely aligned (Table 1). The EORTC 22033-26033/CE5 trial [13] recruited from 2005 to 2010 and used MRI, whenever possible co-registered for treatment planning. The RTOG 9802 study [14] recruited from 1998 to 2002 and used MRI, although not modern GTV, CTV and PTV terminology as defined by the ICRU [11]. The question of treatment volumes for low-grade glioma is important, as larger volumes might worsen late effects, such as neurocognitive deficits [29,30]. However, although timing and doses of radiotherapy for low-grade glioma have an evidence base from randomised trials, target volumes are based on patterns of failure from retrospective series and trials. We mention here two of the randomised trials that analysed recurrences. The NCCTG/RTOG/ECOG study [8] of low- versus high-dose radiotherapy, which recruited from 1986 to 1994, initially used computed tomography and later MRI. Both arms received 50.4 Gy in 28 fractions with radiation fields including the preoperative tumour volume

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

with a 2 cm margin, whereas the high-dose arm received a boost of 14.4 Gy in eight fractions with fields covering the preoperative tumour volume with a 1 cm margin. In modern ICRU terminology, assuming a 5 mm CTV-PTV margin, and a further 5 mm from the edge of the PTV (95% isodose) to the field edge, the CTV was the pre-operative GTV plus 1 cm in phase 1 and the pre-operative GTV in phase 2. The 5 year progression-free survival was 55% in the low- and 52% in the high-dose arms; after a median follow-up of 6.4 years, 114/203 patients had progressed. In the 65 patients for whom data were available, in both arms 92% of failures were within the radiation field, 3% within 2 cm of the field and 5% more than 2 cm from the field. The EORTC 22845 trial [9] of early versus delayed postoperative radiotherapy recruited from 1986 to 1997 and only specified computed tomography imaging. It delivered 45 Gy to the preoperative computed tomography volume plus 2 cm followed by 9 Gy to the preoperative computed tomography volume plus 1 cm. For the 154 patients assigned early radiotherapy, the 5 year progression-free survival was 55%; at a median followup of 7.75 years, 94 of these patients had a recurrence, of which four were outside the field and five in a borderline area. Thus, even using computed tomography imaging, which may have underestimated tumour extent, using GTVeCTV margins of approximately 1 cm, in these trials more than 50% of patients were progression free at 5 years, and of the failures, around 90% occurred within the radiation fields. It will be informative in due course to see the pattern of failure data from more recent trials using coregistered postoperative MRI and modern treatment planning methods.

Potential Contributions of Novel Imaging Techniques In recent years, there has been considerable interest in the application of advanced imaging techniques to improve brain tumour treatment in three main areas. First, attempts to better identify tumour cell distribution and to localise tumour invasion are of particular relevance to radiotherapy planning. Second, attempts to identify the relationships of tumour boundaries to important white matter pathways and eloquent areas of cerebral cortex could provide information to reduce normal tissue complications. Finally, early prediction of response to conventional therapy, based either on baseline tumour characteristics or early changes in response to therapy, could potentially be used to modify or adapt the radiotherapy plan. In this section we will give an overview of the rationale behind the application of individual imaging techniques and review the current evidence that they may be of benefit in radiotherapy planning. Improved Imaging of Invasion and Tumour Extent It is well recognised that conventional MRI techniques underestimate the distribution of malignant cells within the brain. On T1-weighted MRI, contrast enhancement reflects only areas of bloodebrain barrier (BBB) breakdown and does

5

not reflect the extent of tumour cell infiltration around the tumour margins. This is highlighted by observations after vascular endothelial growth factor (VEGF) inhibition, where the enhancing tumour volume is markedly reduced early in therapy while T2 and FLAIR images clearly show tumour tissue and tumour progression [31]. FLAIR images identify non-enhancing tumour, but distinction of tumour from very focal oedema and treatment-related changes such as gliosis or leukoencephalopathy is difficult or impossible. Recognition of the limitations of conventional MRI techniques has led to attempts to develop novel MRI approaches to identify tumour extent and invasion. Early work comparing diffusion-weighted MRI metrics showed a close correlation between tumour cell density and fractional anisotropy at the margins of low-grade glioma [32]. In 2007, Price et al. [33] used measurements of the isotropic and anisotropic components of the diffusion-weighted signal to study tumour margins, identifying three separate patterns: (i) a diffuse pattern of abnormality (isotropic exceeds anisotropic measurements in all directions), which was associated with diffuse increasing tumour size over time; (ii) a localised pattern of abnormality where tumour occurred in one particular direction and (iii) a pattern of minimal abnormality seen in some patients and associated with no evidence of recurrence (Figure 1). They concluded that diffusion tensor imaging (DTI) is able to predict patterns of tumour recurrence and could be incorporated into radiotherapy treatment planning [34]. These concepts have been taken forward by a number of groups, leading to the development of mathematical models to predict patterns of tumour invasion. Although these remain experimental in nature, they have the potential to impact significantly on radiotherapy planning in the future [35]. Molecular imaging techniques using positron emission tomography (PET) overcome some of these limitations. Radiolabelled amino acids have been of particular interest for the identification of tumour extent because they show high uptake in biologically active tumour tissue and low uptake in normal brain. The radiolabelled amino acids commonly used in clinical studies are methyl-11C-Lmethionine (MET) and 18F-fluoroethyltyrosine (FET). Although it was originally postulated that amino acid uptake was related to increased proliferative activity in tumour cells, pharmacokinetic analysis using FET showed that increased uptake in glioma is caused by an increase in transport at the BBB [36] that is not dependent on BBB damage. Clinical studies with MET have shown that increased uptake in gliomas without BBB damage can be competitively inhibited by branched-chain amino acids, and uptake is present in most low-grade gliomas despite apparently intact BBBs. In high-grade gliomas, passive diffusion also contributes to amino acid uptake [37]. The rate of uptake is correlated with expression of Ki-67 [38], proliferating cell nuclear antigen [39] and microvessel density [40]. This led to the conclusion that methionine uptake reflects both proliferation potential and angiogenic capability. However, recent work showed that methionine uptake correlates far more closely with cell density than with microvessel density [41]. Despite this, methionine uptake can be low, even

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

6

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

Fig 1. An example of the use of diffusion tensor imaging (DTI) to predict patterns of recurrence. This patient with a glioblastoma developed localised recurrence whose position was predicted by the DTI. The images show a T2-weighted image of the tumour before radiotherapy (A), a map of the isotropic component of the DTI carried out at the same time (B) and a T2-weighted image carried out after tumour progression (C). These images have been co-registered with the isotropic (green) and anisotropic (red) tensor abnormalities. There is a mismatch posteriorly where the anisotropic abnormality is greater than the isotropic abnormality. Imaging showed recurrence with a localised growth pattern in this area (C). Reproduced with permission from [33].

in the presence of significant tumour cell burden and 11Cmethionine uptake and 5-ALA-induced fluorescence have been shown to act as independent indices for tumour cell density [42]. Methionine PET has been extensively studied as a diagnostic modality, a grading tool, a prognostic indicator and as an indicator of tumour extent for radiotherapy planning [43]. Tumour delineation using MET-PET showed that tumour volume definition was improved in 88% of lowgrade and 78% of high-grade tumours [44] and the spatial extent of increased uptake is larger than that seen on MRI in about 70% of cases and equal in the remaining 30% [45]. A number of subsequent studies have supported the suggestion that MET-PET provides a more accurate radiotherapy planning tool than computed tomography or MRI [43,46]. The use of FET is increasingly popular principally because the longer half-life of the 18F label makes the logistics far more acceptable. A number of studies have shown that FETPET, combined with MRI, improves the assessment of glioma for both neurosurgery and radiotherapy planning. Histological correlation studies of stereotactic brain biopsies compared with MRI and FET-PET showed that MRI had a sensitivity of 96% for detecting tumour tissue, but a specificity of only 53%. The combination of MRI and FET-PET yielded a sensitivity of 93% and a specificity of 94% [47]. FETPET has been studied as an indicator of postsurgical residual tumour volume and was found to have a strong prognostic effect on both overall survival and disease-free survival, which was not shown by tumour volume assessed using gadolinium-enhanced MRI [48] (Figure 2). This and other work led to interest in FET-PET for radiotherapy planning. Recent studies [49] have shown that target volume definition for malignant gliomas varies significantly depending on the imaging modality and on the analysis technique used to define the tumour margins [51]. The combination of MRI and FET-PET was shown to avoid large incongruities between standard anatomical MRI and PET demonstration of tumour extent (Figures 3 and 4). However, in terms of clinical outcome, a combination of

MRI and FET-PET adapted dose escalation in GBM with a total dose of 72 Gy based on FET-PET did not lead to any survival benefit, although no increase in acute or late toxicity was identified [52]. More recently, Kinoshita et al. [54] developed a novel image analysis method using matched 18F-FDG and METPET scans to calculate a decoupling score. This score allowed the detection of specimens with a tumour cell density of more than 1000/mm2, with a sensitivity and specificity of 93.5 and 87.5%, respectively, whereas for MET alone the sensitivity and specificity were 87.0 and 87.5%,

Fig 2. KaplaneMeier curves of overall survival in patients with a postoperative tumour volume on 18F-fluoroethyltyrosine-positron emission tomography 25 ml (grey line). Reproduced with permission from [48].

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

7

Fig 3. Images from a patient with glioblastoma: (A) postoperative magnetic resonance imaging (MRI), (B) 18F-fluoroethyltyrosine-positron emission tomography (FET-PET), (C) volumes of contrast enhancement on MRI (white area) and FET volume (red area) shown together. MRI shows contrast enhancement around the resection cavity. FET-PET shows a small residual tumour that is only partly congruent with the contrast enhancement. Reproduced with permission from [52].

respectively. Reconstructed images (decoupling map) using the decoupling score enabled improved visualisation of glioma extent (Figure 5). These findings show that combination mapping provides more reliable indication of malignant cell extent, which should, theoretically, offer an improved basis for surgical and radiotherapy planning. Imaging Tumour Relationships to White Matter and Grey Matter Structures A considerable amount of work has been published using advanced functional brain imaging methods to identify the

relationships between tumour, eloquent cortex and major white matter pathways. Although it is often impossible to avoid including major white matter tracts in the radiotherapy field, pretherapy mapping may allow modification of treatment planning in an attempt to reduce morbidity. The identification of functional cortical areas using functional MRI (fMRI) has been well described and is relatively established in neurosurgical practice [55]. DTI allows visualisation of major white matter tracts by measuring the anisotropic movement of water [56]. Although a range of methods has been developed to image the distribution and position of white matter tracts, development of a robust

Fig 4. Malignant glioma showing peripheral contrast enhancement around resection cavity on magnetic resonance imaging (MRI). 18F-fluoroethyltyrosine-positron emission tomography (FET-PET) shows metabolically active parafalcine tumour. Modifications of the MRI-based clinical target volume (green) with integration of the FET areas (red). Reproduced with permission from [53]. Note that on the FET-PET image the tumour appears to extend through the falx to the contralateral hemisphere. This probably reflects the relatively poor spatial resolution of PET compared with MRI, suggesting that PET can help identify metabolically active areas, but should complement other imaging modalities with greater spatial resolution. Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

8

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

Fig 5. Glioblastoma showing identification of voxels with a mismatch between methyl-11C-L-methionine (MET) and FDG uptake and the consequent identification of tumour extent on the decoupling map (bottom right). Reproduced with permission from [54].

technique to identify tumour or invasion remains problematic [57]. Several authors have published approaches using fibre tracking, where a seeded point in the whitematter tract is found and contiguous voxels identified on the basis of their diffusion characteristics. Unfortunately, these approaches have significant limitations where the track is disrupted, oedematous, invaded or destroyed [58]. A number of alternative approaches have been described, including identification of specific pathways using

anatomical information combined with DTI data to characterise invasion, oedema or compression [59] (Figure 6). A number of studies have attempted to integrate fMRI and/or DTI into stereotactic radiosurgical planning [60], showing that it is possible to modulate the dose received by major fibre tracks in close physical proximity to the tumour (Figure 7). However, to date there is no independent evidence as to whether this results in any significant change in survival or morbidity.

Fig 6. World Health Organization grade II frontal astrocytoma. (A) T1-weighted inversion recovery (IR) in axial section. (B) Diffusion tensor imaging (DTI)-derived fractional anisotropy (FA) image in axial section. (C) FA image with the projected white matter tract segments identified (green, means FA within normal intervals predicted from comparative dataset of normal controls; red, FA below 0.25 threshold; blue, increased FA above upper limit of prediction interval). The image shows invasion of the anterior part of the internal capsule within the tumour. Reproduced with permission from [59]. Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

9

Fig 7. Stereotactic radiosurgery treatment plans developed without (top panels) and with (bottom panels) the functional structures and fibre pathways considered as organs at risk in the optimisation process, for a patient with World Health Organization grade II astrocytoma plotted in axial, coronal and sagittal planes. The configuration of the beams and the motor cortex (green) derived from functional magnetic resonance imaging are also depicted. In this unconventional case, because the glioma directly abutted the motor cortex, the authors felt the dose could only be modestly reduced from 2100 to 1900 cGy and therefore the biologically equivalent dose was calculated and delivered in three fractions. Reproduced with permission from [60].

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

10

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

Predicting Response to Treatment The ability of imaging biomarkers to provide predictive or prognostic information has been described by many groups. Attempts to predict therapy response may be based on baseline characterisation of tumours or by the detection of early therapy-related changes. Classification of highgrade glioma before treatment using imaging biomarkers of cellular proliferation and using microvascular structure have both been shown to relate to progression-free and overall survival. Changes in the tumour in response to early combination chemoradiotherapy have also been shown to be highly predictive. Imaging Tumour Cell Proliferation A high rate of cellular proliferation is a key feature of gliomas and other malignant tumours. Measurement of proliferation using imaging techniques aims to quantify proliferative activity using uptake of DNA precursors such as thymidine. 18-F thymidine can be used for PET imaging, although transfer across the BBB is slow and tumour uptake   is low. Use of a thymidine analogue (3Ldeoxy-3 L-18F-fluorothymidine) improves the imaging characteristics as the molecule is trapped after it is phosphorylated by thymidine

kinase. It provides high tumour-to-brain uptake ratios and has fewer problems associated with generation of labelled metabolites. Uptake in glioma has been shown to correlate closely with tumour grade and also to provide significant prognostic information [61,62] with tumours showing higher proliferation showing shorter overall survival. Attempts to estimate the rate of cellular proliferation using MRI have also been presented and Ellingson et al. [63] have developed a technique known as CIMPLE (cell invasion, motility, and proliferation level estimates) using serial diffusion MRI scans and a simple mathematical model. They have shown the ability of CIMPLE maps to predict progression-free survival and overall survival in recurrent GBM patients treated with bevacizumab. Although this approach remains experimental, it offers the potential for simple direct estimation of proliferation as a predictive or response biomarker without the need for PET studies. Imaging the Vascular Microenvironment The development of anti-angiogenic therapy led to a rapid uptake of imaging techniques that can provide quantitative estimates of the tumour vascular microenvironment. Most significant among these is the use of dynamic contrast-enhanced MRI, which has become

Fig 8. Parametric response map of relative cerebral blood volume (rCBV) colour-coded overlay in high-grade glioma, with (A) pseudoprogression and (C) progressive disease, with corresponding scatter plot analyses (B, D) showing the distribution of rCBV at baseline compared with week 3 of chemoradiation. Reproduced with permission from [68]. A pattern of minimal change in the values of individual voxels as seen in (B) characterised improved outcome, whereas a pattern of significant changes in large proportions of individual voxels as seen in (D) was highly predictive of a poor response. Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

ubiquitous in early phase drug trials of anti-angiogenic agents [64]. This technique uses dynamic acquisition of images after injection of intravenous contrast agent in order to document changes in local contrast concentration over time. These can be fitted to pharmacokinetic models to produce estimates of microvascular features such as fractional blood volume, extravascular extracellular space fraction and contrast transfer coefficient (Ktrans) [65]. Many studies have shown a relationship in gliomas between tumour grade, progression-free and overall survival and the maximal blood volume fraction seen within the tumour. Some workers have also identified a prognostic relationship between Ktrans and overall survival in high-grade tumours, independent of blood volume [66,67]. These biomarkers allow routine quantification of the vascular microenvironment, which is already becoming standard in the early phase trials of molecular therapy and in endotype identification in some cancer applications, particularly in the breast. Early Therapy-induced Changes as Predictive Biomarkers In recent years, a number of studies have identified changes in diffusion and microvascular characteristics during the early stages of radiotherapy as powerful predictive biomarkers of outcome in glioma [68]. These studies are based on the identification of areas within the tumour that show significant increases or decreases in one of a number of biomarkers, which, although significant, are insufficient to result in changes in the overall median or mean values. They have given rise to the concept of the parametric response map, which is constructed by co-registration of scans taken at baseline and early in the radiotherapy course. This allows direct measurements of changes in single voxel values and the identification of areas within the tumour that show significant change in response to treatment (Figure 8). This approach has been presented using both diffusion MRI and dynamic contrast-enhanced MRI in a number of therapeutic settings where the presence of significant change seems to predict progressive disease [69e72]. This approach provides the potential for improved specificity in the detection of specific endotypes based on response prediction to support potential treatment stratification.

Conclusion Radiotherapy planning for WHO grade IIeIV gliomas currently relies predominantly on T1-weighted contrastenhanced MRI and T2/FLAIR sequences to identify the GTV. Isotropic margins are added empirically, usually without any patient-specific individualisation. Novel imaging techniques have the potential to influence radiotherapy planning in the future. Improved definition of tumour extent and routes of invasion could modify the GTV and allow this to be grown anisotropically to a CTV, which better reflects areas at risk of recurrence, for example by having larger margins along white matter tracts that offer little resistance to tumour cell spread. Such imaging could

11

indicate where areas of normal tissue could be safely spared, for example to reduce neurocognitive sequelae of treatment. The identification of tumours with a greater or lesser propensity to invade beyond the visible GTV, a characteristic that may be modified by co-administered novel agents, could allow for patient-specific individualisation of margins. It has been, and continues to be, difficult to make progress in improving outcomes for gliomas. In the last decade, there have been some significant advances, for example as a result of trials for GBM and for 1p/19q deleted grade III gliomas. However, there have been many more negative trials. We are increasingly recognising that gliomas of a given histological type are genetically heterogeneous and may demand different therapeutic approaches. In time, more ‘druggable’ targets will be identified. Novel imaging techniques as potential biomarkers already have an increasing role in early phase drug trials. Advanced imaging techniques could supplement knowledge from clinical characteristics and molecular genetic markers in predicting response to therapy. If response can be predicted robustly before the start of or early in the course of radiotherapy, this could be used to select patients for trials of alternative treatments, such as radiotherapy dose-escalation strategies, including dose painting of high-risk tumour subregions, or the addition of chemotherapy or targeted agents. Much effort will be needed to move the most promising imaging techniques through clinical trials into routine clinical practice. To be effective, such efforts must go hand in hand with exploiting advances in molecular genetics, in drug development and in radiotherapy techniques.

Acknowledgements This work is supported by the CRUK-EPSRC Cancer Imaging Centre in Cambridge and Manchester.

References [1] Cancer Research UK. Brain, other CNS and intracranial tumours incidence statistics. Available at: http://www. cancerresearchuk.org/cancer-info/cancerstats/types/brain/ incidence/. [accessed 27.10.13]. [2] Stupp R, Hegi ME, Mason WP, et al. 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 2009;10(5):459e466. [3] Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352(10):987e996. [4] Cairncross G, Wang M, Shaw E, et al. Phase III trial of chemoradiotherapy for anaplastic oligodendroglioma: long-term results of RTOG 9402. J Clin Oncol 2013;31(3):337e343. [5] van den Bent MJ, Brandes AA, Taphoorn MJ, et al. Adjuvant procarbazine, lomustine, and vincristine chemotherapy in newly diagnosed anaplastic oligodendroglioma: long-term follow-up of EORTC brain tumor group study 26951. J Clin Oncol 2013;31(3):344e350. [6] European Organisation for Research and Treatment of Cancer. EORTC protocol 26053 e 22054. Phase III trial on concurrent and

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

12

[7]

[8]

[9]

[10] [11]

[12]

[13]

[14]

[15] [16]

[17]

[18]

[19]

[20]

[21]

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13 adjuvant temozolomide chemotherapy in non-1p/19q deleted anaplastic glioma. The CATNON intergroup trial. Available at: http://www.eortc.be/protoc/details.asp?protocol¼26053. [accessed 10.11.13]. Karim AB, Maat B, Hatlevoll R, et al. A randomized trial on dose-response in radiation therapy of low-grade cerebral glioma: European Organization for Research and Treatment of Cancer (EORTC) Study 22844. Int J Radiat Oncol Biol Phys 1996;36(3):549e556. Shaw E, Arusell R, Scheithauer B, et al. Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group study. J Clin Oncol 2002;20(9):2267e2276. van den Bent MJ, Afra D, de Witte O, et al. Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial. Lancet 2005;366(9490):985e990. Whittle IR. The dilemma of low grade glioma. J Neurol Neurosurg Psychiatry 2004;75(Suppl. 2):ii31eii36. International Commission on Radiation Units and Measurement. ICRU Report 50: Prescribing, recording and reporting photon beam therapy. Bethesda, MD: ICRU; 1993. Radiation Therapy Oncology Group. Phase III double-blind placebo-controlled trial of conventional concurrent chemoradiation and adjuvant temozolomide plus bevacizumab versus conventional concurrent chemoradiation and adjuvant temozolomide in patients with newly diagnosed glioblastoma. Available at: http://www.rtog.org/ClinicalTrials/ProtocolTable/ StudyDetails.aspx?study¼0825. [accessed 27.10.13]. Fairchild A, Weber DC, Bar-Deroma R, et al. Quality assurance in the EORTC 22033-26033/CE5 phase III randomized trial for low grade glioma: the digital individual case review. Radiother Oncol 2012;103(3):287e292. Shaw EG, Wang M, Coons SW, et al. Randomized trial of radiation therapy plus procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-grade glioma: initial results of RTOG 9802. J Clin Oncol 2012;30(25):3065e3070. Hochberg FH, Pruitt A. Assumptions in the radiotherapy of glioblastoma. Neurology 1980;30(9):907e911. Oppitz U, Maessen D, Zunterer H, Richter S, Flentje M. 3Drecurrence-patterns of glioblastomas after CT-planned postoperative irradiation. Radiother Oncol 1999;53(1):53e57. Wallner KE, Galicich JH, Krol G, Arbit E, Malkin MG. Patterns of failure following treatment for glioblastoma multiforme and anaplastic astrocytoma. Int J Radiat Oncol Biol Phys 1989;16(6):1405e1409. Burger PC, Dubois PJ, Schold Jr SC, et al. Computerized tomographic and pathologic studies of the untreated, quiescent, and recurrent glioblastoma multiforme. J Neurosurg 1983;58(2):159e169. Halperin EC, Bentel G, Heinz ER, Burger PC. Radiation therapy treatment planning in supratentorial glioblastoma multiforme: an analysis based on post mortem topographic anatomy with CT correlations. Int J Radiat Oncol Biol Phys 1989;17(6):1347e1350. Farace P, Giri MG, Meliado G, et al. Clinical target volume delineation in glioblastomas: pre-operative versus post-operative/ pre-radiotherapy MRI. Br J Radiol 2011;84(999):271e278. Minniti G, Amelio D, Amichetti M, et al. Patterns of failure and comparison of different target volume delineations in patients with glioblastoma treated with conformal radiotherapy plus concomitant and adjuvant temozolomide. Radiother Oncol 2010;97(3):377e381.

[22] Wick W, Stupp R, Beule AC, et al. A novel tool to analyze MRI recurrence patterns in glioblastoma. Neuro Oncol 2008;10(6):1019e1024. [23] Dobelbower MC, Burnett III OL, Nordal RA, et al. Patterns of failure for glioblastoma multiforme following concurrent radiation and temozolomide. J Med Imaging Radiat Oncol 2011;55(1):77e81. [24] McDonald MW, Shu HK, Curran Jr WJ, Crocker IR. Pattern of failure after limited margin radiotherapy and temozolomide for glioblastoma. Int J Radiat Oncol Biol Phys 2011;79(1):130e136. [25] Paulsson AK, McMullen KP, Peiffer AM, et al. Limited margins using modern radiotherapy techniques does not increase marginal failure rate of glioblastoma. Am J Clin Oncol 2014;37(2):177e181. [26] Creak AL, Tree A, Saran F. Radiotherapy planning in highgrade gliomas: a survey of current UK practice. Clin Oncol (R Coll Radiol) 2011;23(3):189e198. [27] Ghose A, Lim G, Husain S. Treatment for glioblastoma multiforme: current guidelines and Canadian practice. Curr Oncol 2010;17(6):52e58. [28] Burnet NG, Lynch AG, Jefferies SJ, et al. High grade glioma: imaging combined with pathological grade defines management and predicts prognosis. Radiother Oncol 2007;85(3):371e378. [29] Douw L, Klein M, Fagel SS, et al. Cognitive and radiological effects of radiotherapy in patients with low-grade glioma: long-term follow-up. Lancet Neurol 2009;8(9):810e818. [30] Klein M, Heimans JJ, Aaronson NK, et al. Effect of radiotherapy and other treatment-related factors on mid-term to longterm cognitive sequelae in low-grade gliomas: a comparative study. Lancet 2002;360(9343):1361e1368. [31] de Groot JF, Yung WK. Bevacizumab and irinotecan in the treatment of recurrent malignant gliomas. Cancer J 2008;14(5):279e285. [32] Stadlbauer A, Ganslandt O, Buslei R, et al. Gliomas: histopathologic evaluation of changes in directionality and magnitude of water diffusion at diffusion-tensor MR imaging. Radiology 2006;240(3):803e810. [33] Price SJ, Jena R, Burnet NG, Carpenter TA, Pickard JD, Gillard JH. Predicting patterns of glioma recurrence using diffusion tensor imaging. Eur Radiol 2007;17(7):1675e1684. [34] Jena R, Price SJ, Baker C, et al. Diffusion tensor imaging: possible implications for radiotherapy treatment planning of patients with high-grade glioma. Clin Oncol (R Coll Radiol) 2005;17(8):581e590. [35] Painter KJ, Hillen T. Mathematical modelling of glioma growth: the use of Diffusion Tensor Imaging (DTI) data to predict the anisotropic pathways of cancer invasion. J Theoret Biol 2013;323:25e39. [36] Wienhard K, Pawlik G, Nebeling B, et al. Estimation of local cerebral glucose utilization by positron emission tomography: comparison of [18F]2-fluoro-2-deoxy-Dglucose and [18F]2-fluoro-2-deoxy-D-mannose in patients with focal brain lesions. J Cerebr Blood Flow Metab 1991;11(3):485e491. [37] Roelcke U, Radu E, Ametamey S, Pellikka R, Steinbrich W, Leenders KL. Association of rubidium and C-methionine uptake in brain tumors measured by positron emission tomography. J Neurooncol 1996;27(2):163e171. [38] Chung JK, Kim YK, Kim SK, et al. Usefulness of 11C-methionine PET in the evaluation of brain lesions that are hypo- or isometabolic on 18F-FDG PET. Eur J Nucl Med Mol Imaging 2002;29(2):176e182. [39] Sato N, Suzuki M, Kuwata N, et al. Evaluation of the malignancy of glioma using 11C-methionine positron emission

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026

G.A. Whitfield et al. / Clinical Oncology xxx (2014) 1e13

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49] [51]

[52]

[53]

[54]

[55]

[56]

[57]

tomography and proliferating cell nuclear antigen staining. Neurosurg Rev 1999;22(4):210e214. Kracht LW, Friese M, Herholz K, et al. Methyl-[11C]- l-methionine uptake as measured by positron emission tomography correlates to microvessel density in patients with glioma. Eur J Nucl Med Mol Imaging 2003;30(6):868e873. Okita Y, Kinoshita M, Goto T, et al. (11)C-methionine uptake correlates with tumor cell density rather than with microvessel density in glioma: a stereotactic image-histology comparison. Neuroimage 2010;49(4):2977e2982. Arita H, Kinoshita M, Kagawa N, et al. (1)(1)C-methionine uptake and intraoperative 5-aminolevulinic acid-induced fluorescence as separate index markers of cell density in glioma: a stereotactic image-histological analysis. Cancer 2012;118(6):1619e1627. Glaudemans AW, Enting RH, Heesters MA, et al. Value of 11Cmethionine PET in imaging brain tumours and metastases. Eur J Nucl Med Mol Imaging 2013;40(4):615e635. Pirotte B, Goldman S, Dewitte O, et al. Integrated positron emission tomography and magnetic resonance imagingguided resection of brain tumors: a report of 103 consecutive procedures. J Neurosurg 2006;104(2):238e253. Voges J, Herholz K, Holzer T, et al. 11C-methionine and 18F-2fluorodeoxyglucose positron emission tomography: a tool for diagnosis of cerebral glioma and monitoring after brachytherapy with 125I seeds. Stereotactic Functional Neurosurg 1997;69(1e4 Pt 2):129e135. Lee IH, Piert M, Gomez-Hassan D, et al. Association of 11Cmethionine PET uptake with site of failure after concurrent temozolomide and radiation for primary glioblastoma multiforme. Int J Radiat Oncol Biol Phys 2009;73(2):479e485. Pauleit D, Floeth F, Hamacher K, et al. O-(2-[18F]fluoroethyl)-Ltyrosine PET combined with MRI improves the diagnostic assessment of cerebral gliomas. Brain 2005;128(Pt 3):678e687. Piroth MD, Holy R, Pinkawa M, et al. Prognostic impact of postoperative, pre-irradiation (18)F-fluoroethyl-l-tyrosine uptake in glioblastoma patients treated with radiochemotherapy. Radiother Oncol 2011;99(2):218e224. Niyazi M, Geisler J, Siefert A, et al. FET-PET for malignant glioma treatment planning. Radiother Oncol 2011;99(1):44e48. Vees H, Senthamizhchelvan S, Miralbell R, Weber DC, Ratib O, Zaidi H. Assessment of various strategies for 18F-FET PETguided delineation of target volumes in high-grade glioma patients. Eur J Nucl Med Mol Imaging 2009;36(2):182e193. Piroth MD, Pinkawa M, Holy R, et al. Integrated boost IMRT with FET-PET-adapted local dose escalation in glioblastomas. Results of a prospective phase II study. Strahlenther Onkol 2012;188(4):334e339. Rieken S, Habermehl D, Giesel FL, et al. Analysis of FET-PET imaging for target volume definition in patients with gliomas treated with conformal radiotherapy. Radiother Oncol 2013;109(3):487e492. Kinoshita M, Arita H, Goto T, et al. A novel PET index, 18F-FDG11C-methionine uptake decoupling score, reflects glioma cell infiltration. J Nucl Med 2012;53(11):1701e1708. Tie Y, Suarez RO, Whalen S, Radmanesh A, Norton IH, Golby AJ. Comparison of blocked and event-related fMRI designs for pre-surgical language mapping. Neuroimage 2009;47(Suppl. 2):T107eT115. Le Bihan D, Mangin JF, Poupon C, et al. Diffusion tensor imaging: concepts and applications. J Mag Res Imag 2001;13(4):534e546. Castellano A, Bello L, Michelozzi C, et al. Role of diffusion tensor magnetic resonance tractography in predicting the

[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

13

extent of resection in glioma surgery. Neuro Oncol 2012;14(2):192e202. Stadlbauer A, Nimsky C, Gruber S, et al. Changes in fiber integrity, diffusivity, and metabolism of the pyramidal tract adjacent to gliomas: a quantitative diffusion tensor fiber tracking and MR spectroscopic imaging study. Am J Neuroradiol 2007;28(3):462e469. Miller P, Coope D, Thompson G, Jackson A, Herholz K. Quantitative evaluation of white matter tract DTI parameter changes in gliomas using nonlinear registration. Neuroimage 2012;60(4):2309e2315. Pantelis E, Papadakis N, Verigos K, et al. Integration of functional MRI and white matter tractography in stereotactic radiosurgery clinical practice. Int J Radiat Oncol Biol Phys 2010;78(1):257e267. Idema AJ, Hoffmann AL, Boogaarts HD, et al. 3’-Deoxy-3’-18Ffluorothymidine PET-derived proliferative volume predicts overall survival in high-grade glioma patients. J Nucl Med 2012;53(12):1904e1910. Jeong SY, Lim SM. Comparison of 3’-deoxy-3’-[18F]fluorothymidine PET and O-(2-[18F]fluoroethyl)-L-tyrosine PET in patients with newly diagnosed glioma. Nucl Med Biol 2012;39(7):977e981. Ellingson BM, Cloughesy TF, Lai A, Nghiemphu PL, Pope WB. Cell invasion, motility, and proliferation level estimate (CIMPLE) maps derived from serial diffusion MR images in recurrent glioblastoma treated with bevacizumab. J Neurooncol 2011;105(1):91e101. O’Connor JP, Jackson A, Parker GJ, Roberts C, Jayson GC. Dynamic contrast-enhanced MRI in clinical trials of antivascular therapies. Nature Rev Clin Oncol 2012;9(3):167e177. Jackson A, O’Connor JP, Parker GJ, Jayson GC. Imaging tumor vascular heterogeneity and angiogenesis using dynamic contrast-enhanced magnetic resonance imaging. Clin Cancer Res 2007;13(12):3449e3459. Mills SJ, Patankar TA, Haroon HA, Baleriaux D, Swindell R, Jackson A. Do cerebral blood volume and contrast transfer coefficient predict prognosis in human glioma? Am J Neuroradiol 2006;27(4):853e858. Patankar TF, Haroon HA, Mills SJ, et al. Is volume transfer coefficient (K(trans)) related to histologic grade in human gliomas? Am J Neuroradiol 2005;26(10):2455e2465. Tsien C, Galban CJ, Chenevert TL, et al. Parametric response map as an imaging biomarker to distinguish progression from pseudoprogression in high-grade glioma. J Clin Oncol 2010;28(13):2293e2299. Galban CJ, Chenevert TL, Meyer CR, et al. The parametric response map is an imaging biomarker for early cancer treatment outcome. Nature Med 2009;15(5):572e576. Galban CJ, Chenevert TL, Meyer CR, et al. Prospective analysis of parametric response map-derived MRI biomarkers: identification of early and distinct glioma response patterns not predicted by standard radiographic assessment. Clin Cancer Res 2011;17(14):4751e4760. Chiba Y, Kinoshita M, Okita Y, et al. Use of (11)C-methionine PET parametric response map for monitoring WT1 immunotherapy response in recurrent malignant glioma. J Neurosurg 2012;116(4):835e842. Harris RJ, Cloughesy TF, Pope WB, et al. 18F-FDOPA and 18F-FLT positron emission tomography parametric response maps predict response in recurrent malignant gliomas treated with bevacizumab. Neuro Oncol 2012;14(8): 1079e1089.

Please cite this article in press as: Whitfield GA, et al., Imaging and Target Volume Delineation in Glioma, Clinical Oncology (2014), http:// dx.doi.org/10.1016/j.clon.2014.04.026