Glioblastoma (GBM) is the most common primary malignant brain tumor in adults. The chance of cure is very limited due to treatment-refractory disease course with frequent recurrences despite aggressive multimodality management. In this retrospective study, we evaluated treatment outcomes of hypofractionated stereotactic radiotherapy (HFSRT) in the management of recurrent GBM and report our single-center experience.
Twenty-eight patients receiving HFSRT for recurrent GBM between September 2008 and February 2014 were retrospectively assessed. Total radiotherapy dose was 25 Gy delivered in 5 fractions over 5 consecutive days for all patients. High-precision, image-guided volumetric modulated arc therapy was delivered with a linear accelerator using 6-MV photons using the frameless technique. Analyzed prognostic factors were age, gender, Karnofsky performance status (KPS), tumor location, planning target volume (PTV) size, overall survival (OS), progression-free survival (PFS), time interval between completion of treatment with Stupp protocol at primary diagnosis and recurrence.
Median follow-up time was 42 months (range 2-68). Median time interval between primary chemoradiotherapy and HFSRT was 11.2 months (range 4-57.9). Median OS and PFS calculated from reirradiation was 10.3 months and 5.8 months, respectively. Longer interval between initial treatment and recurrence (p = 0.01), smaller PTV size (p = 0.001), KPS ≥70 (p = 0.005) and younger age (p = 0.004) were associated with longer OS on statistical analysis.
HFSRT offers a feasible and effective salvage treatment option for recurrent GBM management. Prognostic factors associated with longer OS in our study were longer interval between initial treatment and recurrence, smaller PTV size, KPS ≥70 and younger age.
Tumori 2015; 101(2): 179 - 184
Article Type: ORIGINAL RESEARCH ARTICLE
AuthorsFerrat Dincoglan, Murat Beyzadeoglu, Omer Sager, Selcuk Demiral, Hakan Gamsiz, Bora Uysal, Cuneyt Ebruli, Mustafa Akin, Kaan Oysul, Sait Sirin, Bahar Dirican
- • Accepted on 08/07/2014
- • Available online on 19/03/2015
- • Published in print on 28/04/2015
This article is available as full text PDF.
Glioblastoma (GBM) constitutes the most frequent primary malignant brain tumor in adults. The chance of cure is very limited due to treatment-refractory disease course with frequent recurrences despite aggressive multimodality management (1). Maximal tumor resection followed by conventional radiotherapy with concurrent and adjuvant temozolomide (TMZ) is the current standard treatment for GBM. The majority of the patients (>90%) with GBM have locally recurrent disease despite aggressive management (2, 3). Salvage treatment options for recurrent GBM include surgery, radiotherapy in the form of stereotactic radiosurgery (SRS), fractionated stereotactic radiotherapy (FSRT) and/or systemic treatment (TMZ, alkylating agents, targeted therapies) (4). Several combinations of chemotherapeutics have been used in the recurrent and progressive disease setting; however, these combinations were found to have limited efficacy which may be attributed to the blood–brain barrier and hydrostatic pressure resulting from abnormal tumor vascularity (5). Studies using alkylating agents for recurrent GBM have conferred 6-month disease-free survival of 5-11 months, and median overall survival (OS) of 13-24% (6-7-8). Efficacy and safety of TMZ in recurrent GBM management has been investigated in many studies (9-10-11). Although some improvement was achieved in progression-free survival (PFS), OS was not affected (9-10-11). Targeted agents have been used as a treatment option for GBMs based on the premise that
Optimal surgical resection for recurrent GBM may be hampered by large lesion size, critical location and unfavorable patient characteristics (poor performance status, older age, with comorbidities), and few patients are amenable to optimal surgery (14, 15).
Reirradiation is a therapeutic option for recurrent GBM; however, treatment-related morbidity is a major concern due to the high radiation doses delivered initially (16). The use of brachytherapy in recurrent GBM is limited by its invasive nature and the related morbidity risk (4-5-6-7-8-9-10-11-12-13-14-15-16-17). Promising treatment outcomes have been achieved with SRS and FSRT for recurrent GBM (4, 18). These relatively new treatment modalities allow the delivery of highly focused radiation to well-defined intracranial targets by exploiting the advantage of rigid immobilization, stereotactic target localization and improved set-up accuracy under image guidance. While reducing normal tissue exposure with tighter treatment margins, these recent technologies provide steep dose gradients around the target which allow the delivery of ablative radiation doses with hypofractionated schedules. In this study, we evaluate treatment outcomes of hypofractionated stereotactic radiotherapy (HFSRT) in the management of recurrent GBM and report our single-center experience.
Material and Methods
In this retrospective study, outcomes of 28 patients receiving HFSRT for recurrent GBM between September 2008 and February 2014 at Gulhane Military Medical Academy Radiation Oncology Department were assessed. All patients had undergone maximal surgical resection followed by radiotherapy with concurrent and adjuvant TMZ according to Stupp protocol (60 Gy external-beam radiotherapy [EBRT] + TMZ) at primary diagnosis of GBM and were referred to our department for recurrent disease management. Written informed consent to receive HFSRT was obtained from all patients. All patients had Karnofsky performance status (KPS) ≥60 with lesion sizes ranging between 3 and 6 cm. Dexamethasone 16 mg/day was routinely prescribed for all patients during treatment and tapered thereafter. The presence of recurrent disease was decided by an expert multidisciplinary team including radiation oncologists, neurosurgeons and neuroradiologists. The appearance of a new lesion on T1-weighted magnetic resonance imaging (MRI) or ≥25% enlargement of the initial lesion size was considered as recurrence. Reassessment after 1 month was considered when pseudoprogression was suspected. An experienced neuroradiologist evaluated the perfusion MR, diffusion MR and MR spectroscopy images for the differentiation of radionecrosis and tumor progression. Decreased
A thermoplastic mask (Novastereo; Novater, Milano, Italy) fixed to the stereotactic head frame and mouth-bite (Civco, USA) was used for patient immobilization. Planning computed tomography (CT) images with 1.25-mm slice thickness were acquired at the CT simulator (GE Lightspeed RT; GE Healthcare, Chalfont St. Giles, UK) and transferred to the contouring workstation (SimMD; GE, UK). Gross tumor volume (GTV), planning target volume (PTV) and organs at risk (OARs) were contoured on the fused CT and MR images. GTV was defined as the contrast-enhancing lesion, excluding the surrounding edema. PTV was generated by uniformly expanding the GTV by 3 mm. After completion of the delineation procedure, structure sets including the GTV, PTV and OARs were sent to the radiosurgery planning system (ERGO++ planning system, Elekta). Frameless radiosurgery treatment planning was performed with the volumetric modulated arc treatment technique using 6-MV photons and 3- or 5-mm dynamic multileaf collimators (DMLC). High-precision, image-guided volumetric modulated arc therapy was delivered with a linear accelerator using 6-MV photons using the frameless technique. Total radiotherapy dose was 25 Gy delivered in 5 fractions over 5 consecutive days for all patients. Dose was prescribed to the 85%-95% isodose line encompassing the PTV. Accurate set-up verification was performed by matching of reference planning CT images with the kv-CBCT images with the XVI program (version 4.0) using bony anatomy registration. Details of the radiosurgical procedure have been described elsewhere (21-22-23-24).
Follow-up visits were scheduled for all the patients at 2-month intervals for the first year, and at 3-month intervals thereafter, including history and physical examination with assessment of neurological status, KPS and toxicity. In-field recurrence was defined as the presence of >80% of the recurrent tumor within the 95% isodose surface, whereas marginal recurrence was defined as the presence of 20% to 80% of the recurrent tumor within the 95% isodose surface. All other cases were deemed to have out-of-field recurrence (25). Evaluation of complete blood counts, chemistry profile and neuroimaging with contrast-enhanced MRI was routinely performed at each follow-up visit.
Endpoints of the study were OS and PFS. OS and PFS were calculated from the date of HFSRT until the last follow-up visit or death. Analyzed prognostic factors included age, sex, KPS, tumor location, PTV size, OS, PFS, time interval between completion of treatment with Stupp protocol at primary diagnosis and recurrence. Toxicity was graded according to Common Terminology Criteria for Adverse Events, version 4 (CTCAE). OS and PFS curves were generated using the Kaplan-Meier method, measured from the date of HFSRT. Data analysis was performed using p value <0.05 for statistical significance.
Twenty-eight patients received HFSRT for recurrent GBM between September 2008 and February 2014. Out of the total 28 patients, 17 patients were male (60.7%), and 11 patients (39.3%) were female. Median age was 55.6 years (range 38-76). Median KPS was 80 (range 60-100). Median follow-up time was 42 months (range 2-68). All patients had received 60 Gy radiotherapy with concurrent and adjuvant TMZ according to Stupp protocol at primary diagnosis of GBM.
Type of surgery at primary diagnosis included total resection in 8 patients (28.6%), subtotal resection in 17 patients (60.7%) and biopsy in 3 patients (10.7%). Tumor localization was in the frontal lobe in 11 patients (39.3%), in the parietal lobe in 9 patients (32.1%), in the temporal lobe in 5 patients (17.9%) and in the occipital lobe in 3 patients (10.7%). Median volume PTV was 36.5 mL (range 14-70.4) for recurrent disease.
Median OS and PFS calculated from reirradiation was 10.3 months and 5.8 months, respectively (
Longer interval between initial treatment and recurrence (p = 0.01), smaller PTV size (p = 0.001;
Overall survival (OS) from the date of hypofractionated stereotactic radiotherapy (HFSRT) stratified by planning target volume (PTV).
Overall survival (OS) from the date of hypofractionated stereotactic radiotherapy (HFSRT) stratified by Karnofsky performance status (KPS).
Overall survival (OS) from the date of hypofractionated stereotactic radiotherapy (HFSRT) stratified by age.
Multivariate analysis revealed smaller PTV size (p = 0.002) and KPS (p = 0.003) as prognostic factors affecting OS. Lesion location, extent of surgery at primary diagnosis and sex were not associated with OS.
Among the prognostic factors, only smaller PTV size was associated with longer PFS on statistical analysis. Median PFS was 8.3 months and 13.4 months for patients with PTV ≥30 mL and PTV <30 mL, which was statistically significant (p = 0.003).
Eight patients had moderate headache and nausea after HFSRT. Resolution of these symptoms was achieved with steroid medication. Three patients had grade 2 radionecrosis and surrounding edema detected with perfusion MRI and MR spectroscopy. These 3 patients also received steroid medication and responded well to this treatment. No patients had treatment-related neurological deterioration. Out of the total 28 patients receiving HFSRT, 4 patients (14.2%) underwent surgery for tumor progression, and histopathological assessment confirmed re-recurrent GBM in these patients. Adjuvant TMZ use after salvage HFSRT was temporarily suspended in 4 patients due to hematological toxicity. Out of these 4 patients, 2 patient had grade 3 thrombocytopenia, 1 patient had grade 2 anemia and 1 patient had grade 3 neutropenia.
In this study, we assessed the use of salvage HFSRT in the management of recurrent GBM in patients initially treated with multimodality treatment including surgery and radiotherapy with concurrent and adjuvant TMZ.
Several studies have assessed the utility of stereotactic reirradiation for high-grade gliomas (18, 26-27-28-29-30-31-32-33-34-35-36-37) (
Studies of stereotactic reirradiation for patients with recurrent glioblastoma (GBM) and anaplastic gliomas
|Authors (ref.)||Patients (no.)||Reirradiation modality||Concomitant systemic chemotherapy||Median total dose/ dose per fraction||Median OS Median PFS (months)||Interval (months)*||KPS|
|BVZ = bevacizumab; CHT = chemotherapy; FSRT = fractionated stereotactic radiotherapy; HFSRT = hypofractionated stereotactic radiotherapy; KPS = Karnofsky performance status; NR = not reported; OS = overall survival; PFS = progression-free survival.|
|*Time interval between primary treatment to recurrence.|
|†Study includes patients with anaplastic glioma and GBM.|
|Gutin et al (27)†||25||HFSRT||BVZ||30Gy/6Gy||OS: 12.5 PFS: 7.3||15||80|
|Fokas et al (28)||53||HFSRT||No CHT||30Gy/3Gy||OS: 9 PFS: NR||12||70|
|Henke et al (29)†||31||HFSRT||No CHT||20Gy/5Gy||OS: 10.2 PFS: NR||NR||NR|
|Vordermark et al (30)†||19||HFSRT||No CHT||30Gy/4-10Gy||OS: 9.3 PFS: 4.6||19||90|
|Fogh et al (31)†||147||HFSRT||Various agents||35Gy/3.5Gy||OS: 11 PFS: NR||8||NR|
|Minniti et al (32)||36||HFSRT||Temozolomide||37.5Gy/2.5Gy||OS: 9.7 PFS: 5||14||70|
|Minniti et al (33)†||54||HFSRT||Temozolomide||30 Gy/6 Gy||OS: 12.4 PFS: 6||15.5||80|
|Lederman et al (34)||18||HFSRT||Paclitaxel||24 Gy/6Gy||OS: 7 PFS: NR||7.8||70|
|Wurm et al (35)†||25||HFSRT||Topotecan||25-30Gy/5-6 Gy||OS:14.5 PFS:10.5||12.8||80|
|Patel et al (36)||10||FSRT||No CHT||36Gy/6Gy||OS: 7.4||14.9||80|
|Sirin et al (26)||19||SRS||No CHT||16Gy/16Gy||OS: 9.3 PFS: 5.7||NR||NR|
|Maranzano et al (18)||13||SRS||No CHT||17Gy/17Gy||OS: 11||9||90|
|Kong et al (37)||65||SRS||No CHT||16Gy/16Gy||OS: 13 PFS: 4.6||NR||80|
|Current study||28||FSRT||No CHT||25Gy/5Gy||OS: 10.3 PFS: 5.8||11.2||80|
Comparable OS rates have been reported in the studies using single fraction SRS or HFSRT for the management of recurrent malignant gliomas (18, 26, 36-37-38). In a study by Maranzano et al, 13 patients were treated with SRS for recurrent GBM. Median OS was 11 months after reradiation (18). Sirin et al from our group reported a median survival of 9.3 months for patients undergoing SRS for recurrent GBM in our department (26). In a study by Patel et al, 36 patients with recurrent high-grade gliomas underwent either SRS or HFSRT. Median survival was 8.5 months in the 26 patients receiving SRS and 7.4 months in the 10 patients receiving HFSRT, without statistical significance (36). Kong et al reported a median OS of 13 months and disease-free survival of 4.6 months with SRS for recurrent GBM (37).
The use of targeted agents in recurrent GBM management is an area of active investigation in recent years. Studies have particularly focused on the use of bevacizumab and its combinations with other systemic agents (27, 38-39-40). Current data show the feasibility of the use of bevacizumab or its combinations with other treatments as a viable therapeutic option in recurrent GBM management (27, 38-39-40), nevertheless, randomized controlled studies may be warranted to draw firm conclusions.
There is no standard algorithm for the management of recurrent GBM, given the scarcity of prospective studies. Stereotactic irradiation techniques have shown significant advances recently with the introduction of image-guided technologies. HFSRT may offer a viable reirradiation alternative in selected patients with recurrent GBM. Particularly patients with larger recurrent GBM lesions located in eloquent brain areas may benefit from HFSRT, considering that HFSRT exploits the advantage of fractionation which allows improved critical organ sparing without compromising target coverage. Clearly, randomized trials are needed to assess comparative feasibility, safety and efficacy of treatment modalities used in the management of recurrent GBM. Nevertheless, HFSRT has proven to be safe and effective for our patients with recurrent GBM, and our results appear to be consistent with the related literature.
In conclusion, HFSRT offers a feasible and effective salvage treatment option for recurrent GBM management. Prognostic factors associated with longer OS in our study were longer interval between initial treatment and recurrence, smaller PTV size, KPS ≥70 and younger age.
- Dincoglan, Ferrat [PubMed] [Google Scholar] 1, * Corresponding Author (firstname.lastname@example.org)
- Beyzadeoglu, Murat [PubMed] [Google Scholar] 1
- Sager, Omer [PubMed] [Google Scholar] 1
- Demiral, Selcuk [PubMed] [Google Scholar] 1
- Gamsiz, Hakan [PubMed] [Google Scholar] 1
- Uysal, Bora [PubMed] [Google Scholar] 1
- Ebruli, Cuneyt [PubMed] [Google Scholar] 1
- Akin, Mustafa [PubMed] [Google Scholar] 1
- Oysul, Kaan [PubMed] [Google Scholar] 1
- Sirin, Sait [PubMed] [Google Scholar] 2
- Dirican, Bahar [PubMed] [Google Scholar] 1
Department of Radiation Oncology, Gulhane Military Medical Academy, Ankara - Turkey
Department of Neurosurgery, Gulhane Military Medical Academy, Ankara - Turkey