Research Article | Open Access
Assessing Response Using -MIBI Early after Interstitial Chemotherapy with Carmustine-Loaded Polymers in Glioblastoma Multiforme: Preliminary Results
Introduction. Early signs of response after applying wafers of carmustine-loaded polymers (gliadel) are difficult to assess with imaging because of time-related imaging changes. -sestamibi (MIBI) brain single-photon emission tomography (SPET) has reportedly been used to reveal areas of cellularity distinguishing recurrent neoplasm from radionecrosis. Our aim was to explore the role of MIBI SPET in assessing response soon after gliadel application in glioblastoma multiforme (GBM). Methods. We retrospectively reviewed the charts on 28 consecutive patients with a radiological diagnosis of GBM who underwent MIBI SPET/CT before surgery (with intracavitary gliadel placement in 17 patients), soon after surgery, and at 4 months. The area of uptake was selected using a volume of interest that was then mirrored contralaterally to obtain a semiquantitative ratio. Results. After adjusting for ratio at the baseline, the effect of treatment (gliadel versus non-gliadel) was not statistically significant. Soon after surgery, however, 100% of patients treated with gliadel had a decreased ratio, as opposed to 62.5% of patients in the non-gliadel group . The difference between ratios of patients with radical versus partial resection reached statistical significance by a small margin . Conclusions. These data seem to suggest that the MIBI ratio could be a valuable tool for monitoring the effect of gliadel early after surgery.
Interstitial chemotherapy with carmustine-loaded polymers (aka “gliadel wafers,” MGI Pharma, Bloomington, MN, USA) has been recognized  as a valid therapeutic option in newly diagnosed [2–5] gliomas and a debatable option in recurrent [6, 7] high-grade gliomas (HGG).
Early signs of response after applying gliadel wafers are very difficult to assess using magnetic resonance imaging (MRI) because of the imaging changes occurring with time at wafer level and in the adjacent brain tissue , which can easily be mistaken for an abscess. MRI of early changes (especially a month after wafer implantation) is unable to clearly distinguish between postoperative changes and residual neoplasm and provides no information for predicting survival or progression-free interval. Pseudoprogression has also been reported, using MRI, within the first two months after gliadel placement [9, 10].
On the other hand, -sestamibi (MIBI) brain single-photon emission tomography/computed tomography (SPET/CT) can reveal areas of cellularity in high-grade gliomas , thus distinguishing recurrent neoplasm from radionecrosis after radiotherapy [12, 13]. MIBI (a substrate for P-glycoprotein and the multidrug resistance protein) has also been claimed to predict survival and early response or resistance to chemotherapy [14, 15].
To our knowledge, there are currently no data on the use of MIBI SPET/CT after applying gliadel at any time (and early after surgery in particular). The object of this study was thus a retrospective review of the charts on 17 patients with histologically proven glioblastoma multiforme (GBM) who underwent SPET/CT using MIBI (before surgery, soon after surgery, and 4 months after gliadel application), the well-known Stupp protocol (radiotherapy plus concomitant and adjuvant temozolomide) , and gliadel implantation. We also reviewed the charts on 11 comparable patients not implanted with gliadel and compared their outcomes with those of the gliadel group.
We retrospectively reviewed a total of 28 consecutive patients (20 males and 8 females, median age 61.5 years, range 39–84 years) with a radiological (MRI) diagnosis of GBM, who had SPET/CT using MIBI at the nuclear medicine unit of the University of Padova before undergoing surgery. All patients were treated surgically for GBM using an MRI neuro-navigational method, with neurophysiological monitoring where necessary. The diagnosis of GBM was confirmed at histology in all cases.
Intracavitary gliadel wafers were inserted in 17/28 patients (13 males and 4 females; median age 58 years, range 39–69 years) who formed the “gliadel group,” while the other 11/28 patients (7 males and 4 females; median age 67 years, range 51–84 years) served as controls (“non-gliadel group”). The controls were cases of multifocal tumour, or disease extending across the corpus callosum, or requiring a large opening in the ventricle, or patients who refused gliadel wafer placement.
In the gliadel group, 11/17 patients underwent gross-total resection, while resection was partial in 6/17 because intraoperative monitoring showed that eloquent cortical areas were involved. In the non-gliadel group, 7/11 had gross-total resections and 4/11 had partial resections. Six of the 17 patients in the gliadel group were classified as first diagnoses of HGG and 11/17 as recurrent cancers; all the non-gliadel controls were first diagnoses.
SPET/CT using MIBI started 15 min after intravenous injection of 740 MBq of MIBI (Bristol-Myers Squibb Medical Imaging, Brussels, Belgium). MIBI preparation and quality control were according to the manufacturer’s instructions. A dual-head γ-camera (Infinia Hawkeye; GE Healthcare, Chalfont St. Giles, UK) equipped with a crystal 3/8ths of an inch thick and a low-energy, high-resolution collimator was used to acquire SPET/CT. Emission data were acquired first (“head first,” supine position, step-and-shoot mode, 45 s/step— matrix size—energy window of KeV and scatter correction window % keV). The transmissive CT (2.5 mA, 140 kV) field of view was selected on emission images, which were reconstructed (using an algorithm for resolution recovery and scatter correction) at a Xeleris workstation using 4 iterations for ordered subset expectation maximisation (OSEM-10 subsets) and 20 iterations for maximum likelihood expectation maximisation (MLEM). A 3D Butterworth post-filter was applied (cut-off frequency 0.65 cycle/cm, order 10) to each study. Attenuation was corrected using transmissive images. An average effective dose of 0.1 mSv was estimated for CT head scanning using Infinia Hawkeye . An effective dose of mSv/MBq was estimated for -MIBI in resting individuals .
SPET/CT using MIBI was performed(i)before surgery in all cases;(ii)early after surgery in 27/28 patients (17/17 gliadel group; 10/11 non-gliadel group), after a mean 1.16 months (min 0.67, max 2.17 months); 1 non-gliadel control patient refused the procedure at this time point and was not considered for the purposes of the statistical analysis;(iii)at 4 months in 19/28 patients (10/17 gliadel group; 9/11 non-gliadel group), after a mean of 3.77 months (min 2.43, max 5.13 months); 9/28 cases were not acquired at this time point.
Four patients (2 in each group) were not considered in the statistical analysis because(i)MIBI was impossible to interpret in one patient (gliadel-group) due to the patient moving excessively (even when the acquisition was repeated);(ii)the interval between presurgical MIBI and surgery was too long in one patient (gliadel group), that is, 138 days versus a mean of 8.96 days (min 1 day; max 61 days);(iii)one neoplasm (non-gliadel group) was not visible on presurgical MIBI, probably due to resolution limits, so any acquisition early after surgery became pointless;(iv)one periventricular neoplasm (non-gliadel group) revealed only a faint uptake and it was impossible to draw VOIs without including physiological activity in the choroid plexus.
A summary of the statistically analysed data is given in Table 1.
FD: first diagnosis; RD: recurrent disease; BS: before surgery; AS: after surgery.|
2.3. Image Analysis
Transaxial, sagittal, and coronal views (attenuation-corrected (AC) and nonattenuation-corrected (NAC) images) were obtained for each study and analysed qualitatively and semiquantitatively.
For each study and time period, the area of MIBI uptake was selected (on AC images) using a volume of interest (VOI) that was then corrected manually on each transaxial slice to ensure that only specific uptake was included (excluding any activity in the skull or choroid plexus, for instance). The resulting “tumoural VOI” was then mirrored contralaterally on healthy brain to obtain the following ratio (an estimate of the lesion’s “cellularity”), which is a dimensionless quantity:
2.4. Surgical Technique and Sample Collection
Bis-chloroethylnitrosourea (BCNU) wafers were implanted after surgical removal of the neoplasm and intraoperative pathological confirmation of HGG. An accurate haemostasis was achieved before implanting the wafers in all cases. The surgical cavity was checked to see if there was any communication between the surgical cavity and the cerebral ventricles: if so, the wafers were only inserted when this defect was smaller than 10 mm . The number of wafers inserted depended on the size of the surgical cavity, but it was never more than 8 (mean 6; min. 4, max. 8).
2.5. Pathological Analysis
Frozen sections were examined intraoperatively to confirm the provisional diagnosis of HGG in all cases before inserting any wafers to rule out patients with metastases or other neoplasms. The final histological diagnosis was subsequently established on formalin-fixed and paraffin-embedded surgical specimens. Fresh surgical specimens were cut into fragments, immersed in 10% buffered formalin, and left for 24 hours. After fixation, specimens were dehydrated in graded alcohol, cleared in xylene, and embedded in paraffin; histology was performed on 3 mm thick sections stained with haematoxylin & eosin.
Additional immunohistochemical stains, such as GFAP, neurofilament, synaptophysin, and the biomarkers MIB1 and p53, were used to reach the correct diagnosis.
Gliadel wafers (MGI Pharma, Bloomington, MN, USA), an FDA-approved medication for newly diagnosed malignant glioma and for recurrent GBM, were used as an adjunct to surgery in 17/28 patients.
Gliadel is a hydrophobic copolymer matrix (wafer) impregnated with BCNU that is inserted in the surgical cavity after tumour resection. The wafers are designed to release the chemotherapeutic agent over a period of 2-3 weeks at a near constant rate (by hydrolysis of the polymers caused by the aqueous environment of the brain).
Four to 6 weeks after the surgical procedure, all patients with a first diagnosis of HGG were administered the “Stupp protocol” with a combination of radiation therapy and temozolomide, followed, after a 4-week rest period, by temozolomide alone. Patients with recurrent GBM received systemic chemotherapy with fotemustine 100 mg/m2 weekly for 3 consecutive weeks followed by a 5-week rest period.
2.7. Statistical Analysis
The ratio (counts/pixel in tumoural VOI)/(counts/pixel in VOI in healthy brain) early after surgery was analysed using a one-way analysis of covariance (ANCOVA) model considering:(i)the baseline ratio as covariate and the treatment factor (gliadel/no gliadel);(ii)the baseline ratio as covariate and the completeness of resection factor (yes/no);(iii)the baseline ratio as covariate and the first diagnosis factor (first diagnosis/relapsing disease) in the gliadel-treated patients alone.
Checks were conducted to ensure that there was no gross violation of the assumptions of normality, linearity, homogeneity of variances, or homogeneity of regression slopes. The results of ANCOVA are presented as raw means and standard deviations with 95% confidence intervals for the least squares adjusted means and for their differences. The percentage of patients with a decreased ratio at 1 month was compared between the gliadel and non-gliadel groups using Fisher’s exact test.
The level of significance was set at 5%. The statistical analysis was conducted with SAS 9.2 for Windows (SAS Institute Inc., Cary, NC, USA).
3. Results and Discussion
First of all, data were analysed considering the completeness of the resection. Based on the surgeon’s intraoperative opinion and CT early after surgery, 15 patients were judged to have undergone a complete resection, while 8 were judged as partial resections. MIBI SPET showed a mean ratio of 5.9 (min 2.42, max 9.39) for complete resections and a mean ratio of 11.74 (min 6.97, max 16.52) for partial resections. After adjusting for ratio at the baseline, statistical comparison (between complete and partial resections) showed a marginally significant difference between the two groups (, ) (Table 2).
Baseline: MIBI ratio at baseline; EAS: MIBI ratio early after surgery.|
This marginally significant difference could be due to a number of reasons. First of all, the criteria used (surgeon’s opinion and CT very early after surgery) to establish whether or not a resection was complete. More sophisticated methods, such as positron emission tomography/computed tomography (PET/CT), using labelled amino acids (18F-FET, 11C-methionine) or nucleosides (18F-FLT), could have been used, but these tracers (and particularly those with a short half-life) are currently only available at a limited number of units. In addition, HGG is a very aggressive neoplasm that may relapse very soon after surgery, so a significant uptake apparent from the MIBI ratio early after surgery could be genuine neoplasm and not a false-positive finding. Only biopsy could definitively clarify the matter, but this is not done routinely soon after surgery. On the other hand, the mean ratios for partial resections early after surgery were nearly twice as high as for complete resections (11.74 versus 5.9), indicating a greater cellularity in partial resections, as was to be expected.
Then, the effect of gliadel implantation was analysed (Table 3). Although the mean values for the MIBI ratio were significantly higher for patients not treated with gliadel, the effect of the treatment was not statistically significant after adjusting for ratio at the baseline (, ).
Baseline: MIBI ratio at baseline; EAS: MIBI ratio early after surgery.|
It is worth noting, however, that 100% of the patients treated with gliadel had a significantly decreased ratio early after surgery (mean decrease 59.7%; min 18.4%, max 95.5%), while this was true for only 62.5% of patients in the non-gliadel group (mean decrease 73.3%; min 50.8%; max 90.1%) (), while the other 37.5% of the patients in this latter group had a significant increase (mean increase 81.9%; min 64.7%, max 110.7%) in MIBI uptake 1.5 months (early) after surgery, which was a sign of relapsing disease.
Despite the limited number of patients involved, these data seem to suggest a role for the MIBI ratio in monitoring the early effects (within 2 months) of BCNU, which is known to release its chemotherapeutic agent over a period of 2-3 weeks.
When the MIBI ratio at 4 months was compared with the ratio at one month, all but 2 of the patients (one in the gliadel and one in the non-gliadel group) revealed a significant increase in their MIBI ratio despite receiving chemotherapy, meaning that the disease had already begun to relapse in most patients (86.6%).
When the different behaviour of the MIBI ratio after gliadel application in patients with a first diagnosis vis-à-vis those with recurrent disease (Table 4) was analysed, after adjusting for ratio at the baseline, no statistically significant effect of diagnosis (first versus relapsing disease) came to light (, ).
Baseline: MIBI ratio at baseline; EAS: MIBI ratio early after surgery.|
Despite the limited number of patients considered, our data seem to suggest that the MIBI ratio could be a valuable tool for use soon after surgical resections to monitor the effects of chemotherapeutic agents in implanted gliadel wafers. Encouraging preliminary results seem to suggest that MIBI could also be used after gliadel application to rule out any presence of a residual/relapsing neoplastic tissue.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the materials or methods used in this study, the findings, or the publication of this paper.
- J. Perry, A. Chambers, K. Spithoff, and N. Laperriere, “Gliadel wafers in the treatment of malignant glioma: a systematic review,” Current Oncology, vol. 14, no. 5, pp. 189–194, 2007.
- S. Valtonen, U. Timonen, P. Toivanen et al., “Interstitial chemotherapy with carmustine-loaded polymers for high-grade gliomas: a randomized double-blind study,” Neurosurgery, vol. 41, no. 1, pp. 44–49, 1997.
- M. Westphal, D. C. Hilt, E. Bortey et al., “A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma,” Neuro-Oncology, vol. 5, no. 2, pp. 79–88, 2003.
- M. Westphal, Z. Ram, V. Riddle, D. Hilt, E. Bortey, and On behalf of the Executive Committee of the Gliadel Study Group, “Gliadel wafer in initial surgery for malignant glioma: long-term follow-up of a multicenter controlled trial,” Acta Neurochirurgica, vol. 148, no. 3, pp. 269–275, 2006.
- M. G. Hart, R. Grant, R. Gaside, G. Rogers, M. Somerville, and K. Stein, Chemotherapy Wafer for High-Grade Glioma (Review). The Cochrane Collaboration, John Wiley & Sons, New York, NY, USA, 2011.
- H. Brem, S. Piantadosi, P. C. Burger et al., “Placebo-controlled trial of safety and efficacy of intraoperative controlled delivery by biodegradable polymers of chemotherapy for recurrent gliomas,” The Lancet, vol. 345, no. 8956, pp. 1008–1012, 1995.
- B. R. Subach, T. F. Witham, D. Kondziolka, L. D. Lunsford, M. Bozik, and D. Schiff, “Morbidity and survival after 1, 3-bis(2-chloroethyl)-1-nitrosourea wafer implantation for recurrent glioblastoma: a retrospective case-matched cohort series,” Neurosurgery, vol. 45, no. 1, pp. 17–22, 1999.
- S. Ulmer, K. Spalek, A. Nabavi et al., “Temporal changes in magnetic resonance imaging characteristics of Gliadel wafers and of the adjacent brain parenchyma,” Neuro-Oncology, vol. 14, no. 4, pp. 482–490, 2012.
- R. R. Colen, P. O. Zinn, S. Hazany et al., “Magnetic resonance imaging appearance and changes on intracavitary gliadel wafer placement: a pilot study,” World Journal of Radiology, vol. 3, no. 11, pp. 266–272, 2011.
- A. Della Puppa, M. Rossetto, P. Ciccarino et al., “The first 3 months after BCNU wafers implantation in high-grade glioma patients: clinical and radiological considerations on a clinical series,” Acta Neurochirurgica, vol. 152, no. 11, pp. 1923–1931, 2010.
- D. Cecchin, S. Chondrogiannis, A. D. Puppa et al., “Presurgical -sestamibi brain SPET/CT versus SPET: a comparison with MRI and histological data in 33 patients with brain tumours,” Nuclear Medicine Communications, vol. 30, no. 9, pp. 660–668, 2009.
- F. P. Le Jeune, F. Dubois, S. Blond, and M. Steinling, “Sestamibi technetium-99m brain single-photon emission computed tomography to identify recurrent glioma in adults: 201 studies,” Journal of Neuro-Oncology, vol. 77, no. 2, pp. 177–183, 2006.
- X. Cheng, Y. Li, Z. Xu, D. Li, and J. Wang, “A meta-analysis of -MIBI SPECT for detection of recurrent glioma after radiation therapy,” Journal of Clinical Neuroscience, vol. 18, no. 3, pp. 307–312, 2011.
- F. P. Le Jeune, F. Dubois, S. Perez, S. Blond, and M. Steinling, “Technetium-99m sestamibi brain SPECT in the follow-up of glioma for evaluation of response to chemotherapy: first results,” European Journal of Nuclear Medicine and Molecular Imaging, vol. 31, no. 5, pp. 714–719, 2004.
- S. B. Perez, F. Le Jeune, F. Dubois, and M. Steinling, “-MIBI brain SPECT as an indicator of the chemotherapy response of recurrent, primary brain tumors,” Nuclear Medicine Communications, vol. 28, no. 12, pp. 888–894, 2007.
- R. Stupp, W. P. Mason, M. J. Van Den Bent et al., “Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma,” New England Journal of Medicine, vol. 352, no. 10, pp. 987–996, 2005.
- L. J. Sawyer, H. C. Starritt, S. C. Hiscock, and M. J. Evans, “Effective doses to patients from CT acquisitions on the GE Infinia Hawkeye: a comparison of calculation methods,” Nuclear Medicine Communications, vol. 29, no. 2, pp. 144–149, 2008.
- International Commission on Radiological Protection, “Radiation dose to patients from radiopharmaceuticals. ICRP publication 80,” Annals of the ICRP, vol. 3, p. 110, 1998.
- A. D. Puppa, M. Rossetto, P. Ciccarino et al., “Carmustine wafer implantation when surgical cavity is communicating with cerebral ventricles: technical considerations on a clinical series,” World Neurosurgery, vol. 76, no. 1-2, pp. 156–159, 2011.
Copyright © 2014 D. Cecchin et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.