Acta Neurochir (2011) 153:1761–1767 DOI 10.1007/s00701-011-1073-7
CLINICAL ARTICLE
Stereotactic LINAC radiosurgery for incompletely resected or recurrent atypical and anaplastic meningiomas Mustapha El-Khatib & Faycal El Majdoub & Mauritius Hoevels & Martin Kocher & Rolf-Peter Müller & Hans-Jakob Steiger & Volker Sturm & Mohammad Maarouf
Received: 12 May 2011 / Accepted: 9 June 2011 / Published online: 25 June 2011 # Springer-Verlag 2011
Abstract Background The optimal management of subtotally resected or recurrent malignant meningiomas remains controversial. We evaluated the efficacy of linear accelerator (LINAC) radiosurgery for atypical and anaplastic meningiomas after incomplete resection or treatment of recurrences. Methods Between August 1990 and December 2003, 16 patients with 28 meningiomas WHO II and III were treated by stereotactic LINAC radiosurgery at our institution. The median radiological follow-up was 60.3 months, respectively (range: 7.2–173.9 months). Fourteen tumors in nine patients were classified as WHO II and 14 tumors in seven patients as WHO III. The median surface dose was 14 Gy (range: 10–15 Gy) with a median tumor volume of 4.8 ml (range: 0.51–51.4 ml). Results Clinical condition improved in four patients, remained unchanged in nine and deteriorated in one. Tumor shrinkage was seen in eight of 28 meningiomas and a stable disease in 12. Eight of 28 meningiomas showed local tumor progression. The
Mustapha El-Khatib and Faycal El Majdoub contributed equally to theis study M. El-Khatib : F. El Majdoub : M. Hoevels : V. Sturm : M. Maarouf (*) Department of Stereotaxy and Functional Neurosurgery, University of Cologne, Kerpenerstraße 62, 50937 Cologne, Germany e-mail:
[email protected] M. El-Khatib : H.-J. Steiger Department of Neurosurgery, Heinrich Heine University, Düsseldorf, Germany M. Kocher : R.-P. Müller Department of Radiation Oncology, University of Cologne, Cologne, Germany
overall tumor control rate (TCR) was 84%, 70%, 70% after 3, 5, 10 years. According to grading the corresponding TCR after 3, 5, 10 years was 91%, 81%, 81% for grade II and 77%, 60%, 60% for grade III meningiomas. Overall progression-free survival (PFS) was 74%, 67%, 58% after 3, 5, 10 years. According to grading the PFS after 3, 5, 10 years was 88%, 75%, 75% for grade II meningiomas and 57%, 57%, 43% for grade III meningiomas. Conclusion Our results show the efficacy and safety of LINAC radiosurgery for incompletely resected or recurrent malignant meningiomas with a relatively high local tumor control and low morbidity. Keywords Atypical meningioma . Anaplastic meningioma . Malignant meningiomas . LINAC radiosurgery . Stereotactic radiosurgery
Introduction The World Health Organization (WHO) has classified meningiomas in WHO I = benign, WHO II = atypical and WHO III = anaplastic meningiomas [8]. Histopathological findings for malignancy are increased mitotic activity (WHO II mitotic index of ≥4 and WHO III≥20) or more than two of the following histopathological features: (1) loss of lobular architecture (sheeting), (2) prominent nucleoli, (3) increased cellularity, (4) small cells with high nuclear to cytoplasmic ratios, and (5) foci of spontaneous necrosis. More than 90% are benign lesions (WHO I), whereas atypical meningiomas (WHO II) account for 5–7% and anaplastic meningiomas (WHO III) for 1–3% [8]. Main predictors for recurrence after microsurgery are the extent of resection and grading. Atypical and anaplastic meningiomas are associated with a higher recurrence rate than
1762
benign meningiomas. Even after gross total resection (GTR), the success of microsurgery is not satisfactory, therefore the indication for an adjuvant therapy after GTR is recommended. A recently published study of 108 totally resected atypical meningiomas showed a tumor recurrence rate of 7% after 1 year, 41% after 5 years and 48% after 10 years [1]. We present our long-term experience with 28 WHO II and III meningiomas in 16 patients treated by stereotactic radiosurgery with a modified linear accelerator (LINAC-RS) and report about the degree of tumor control, progression-free survival (PFS), predictors for the outcome, as well as morbidity and mortality.
Patients and methods Between August 1990 and December 2003, 154 patients with 183 meningiomas were treated with LINAC-RS at our institution. Twenty-eight meningiomas in 16 patients showed histological findings of WHO II or III and were included into this retrospective study. Fourteen tumors were classified as WHO II and 14 as WHO III. The median age at the time of radiosurgery was 54 years (range: 31–87 years; males/females=6:10). All patients had undergone one or multiple microsurgical resections before they were referred for LINAC-RS (median, 2 surgeries; range: 1–5). Three patients underwent adjuvant fractionated external beam radiation therapy (EBRT) of the tumor bed at least 2 years before radiosurgery. No patient received any chemotherapy prior to LINAC-RS. The median time from the last surgical resection to LINAC-RS was 21.3 months (range: 8.4– 38.7 months). The inclusion criteria for radiosurgery were: tumor remnant after subtotal resection or recurrence, wellcircumscribed tumor with a diameter ≤3 cm on CTand /or MR images, no considerable brainstem compression, distance to the optic system (optic nerve, optic chiasm) ≥1-2 mm and a Karnofsky index ≥70%.
Acta Neurochir (2011) 153:1761–1767
We used a standard LINAC-RS system (Elekta SL25 linear accelerator; Philips, Best, The Netherlands; equipped with tertiary cylindric collimators). Our radiosurgery technique has been published elsewhere [9, 16]. Figure 1 demonstrates a treatment plan. Using multiplanar arcs we applied a median tumor surface dose of 14 Gy (range: 10–15 Gy) with a median maximum dose of 25 Gy (range: 16.25–43.6 Gy). In 22 of 28 tumors we used standard round collimators. Since March 2001 we used a computer-controlled micro-multi-leaf collimator (μMLC) with 1.5-mm lamella width (maximal field size 72 × 68 mm, Siemens, Heidelberg, Germany) to shape the 6-MV photon beams. The remaining six patients were treated with this technique. The number of target fields differed between one and four with a median of two fields and a median isodose of 65% (range: 60–80%). The median target volume included in the applied surface dose was 4.8 ml (range: 0.51–51.8 ml). Patients’ characteristics and treatment parameters are summarized in Table 1. Six meningiomas were located at the skull base with infiltration of the cavernous sinus. Most other meningiomas were located adjacent to the falx or parasaggital (14 of 28). The remaining eight were located at the convexity of the brain. Follow-up We evaluated the clinical outcome through regular visits in our outpatient clinics. A median clinical follow-up time of 60.4 months (range: 10.3–175.8 months) was achieved. All 28 meningiomas were also included into the radiological follow-up analysis (median radiological follow-up 60.3 months; range: 7.2–173.9 months). The Macdonald criteria were used for defining response [10]. Statistical analysis We performed a univariate analysis with regard to the potential prognostic factors tumor volume, tumor margin
Fig. 1 Treatment plan showing a cystic lesion (red line) in a T1-weighted axial, coronal and sagittal MRI (isodose: 80% yellow dotted line, 50% blue line)
Acta Neurochir (2011) 153:1761–1767
1763
Table 1 Patients’ characteristics and treatment parameters Number of patients (n) Number of tumors (n) Sex females/males WHO classification Age (years) Previous surgeries (n) Karnofsky Performance Scale (%) Volume (ml) Clinical follow-up (months) Radiological follow-up (months) Tumor surface dose (Gy) Maximum dose (Gy) Number of fields (n) Isodose (%)
16 28 6/10 Grade II 14 Median
Grade III 14 Range
54 2 90 4.8 60.4 60.3 14 25 2 65
31-87 1-5 70-90 0.51-51.4 10.3-175.8 7.2-173.9 10-15 16.25-43.6 1-9 60-80
dose, age, localization, number of surgeries and grading. Kaplan-Meier curves were computed and the log rank test was used to compare PFS and tumor control rate (TCR) between the groups (Figs. 2 and 3).
Results TCR During follow-up, tumor shrinking was seen on MR images in eight of 28 meningiomas (28.6%, two WHO II and six WHO III, Fig. 4), a stable tumor size in 12 (42.8%, ten WHO II and two WHO III) and tumor progression in eight (28.6%, two WHO II and six WHO III). The median time
Fig. 3 Kaplan Meier curve for PFS of 28 tumors
to progression was 24 months. One WHO III meningioma showed progression after 16 years. Five patients revealed a new manifestation during the follow-up time. In four of these patients the treated meningioma showed shrinkage. The remaining patient with a new manifestation showed progression of the treated meningioma. The actuarial TCR for all treated malignant meningiomas was 84%, 70% and 70% after 3, 5 and 10 years. According to the grading the corresponding TCR after 3, 5 and 10 years was 91%, 81% and 81% for WHO II meningiomas and 77%, 60% and 60% for WHO III meningiomas, respectively. The overall PFS was 74%, 67%, 58% after 3, 5, 10 years. According to grading the PFS after 3, 5, 10 years was 88%, 75%, 75% for WHO II meningiomas and 57%, 57%, 43% for WHO III meningiomas, respectively. Clinical response Eight of 16 patients showed a stable clinical status, three an improvement (amelioration in the Karnofsky Performance Scale score of 10% or more and in neurological symptoms) and the remaining five a deterioration of the clinical condition. Four of the five deteriorations were related to tumor progression and one to a new tumor, a hemangiopericytoma, in the sella. Treatment morbidity
Fig. 2 Kaplan Meier curve for TCR of 28 tumors
A treatment related morbidity after LINAC-RS occurred in one of 28 tumors (3.5%). The patient developed a paraparesis due to an increased edema 5 weeks after treatment of a central parasaggital meningioma. Symptoms were managed with corticoid treatment and complete resolution
1764
Acta Neurochir (2011) 153:1761–1767
Fi g. 4 a -d : T 1 - wei gh t ed , gadolinium-enhanced MR image of a 32-year-old male showing a cystic tumor before radiosurgery (a,b) and 25 months after radiosurgery (c,d)
of the edema was achieved within 2 months. There were no other LINAC-RS related side effects. Mortality Three patients died at 1.5 year, 2 years and 14.5 years after treatment. No death was related to LINAC-RS. One patient died of aggressive tumor progression as a non responder to LINAC-RS after 2 years. The initial tumor volume in this patient was very large with 51.8 ml involving functional structures such as orbita, cranial nerves and also infiltrating the skull base. The second patient had multiple meningiomas and had undergone microsurgery twice prior to radiosurgery. The last MRI before death showed local tumor
control of the lesion treated. The patient died of cardio vascular failure after lung embolism 1.5 years after treatment. The last patient had three manifestations of an atypical meningioma with a transformation from WHO II into WHO III during the follow-up. All three meningiomas were treated by LINAC-RS. The first treated meningioma progressed after 14 years. The patient underwent further surgical procedures but finally died of disease progression. Prognostic factors Univariate analysis provided significant differences only with regards to age. TCR was significantly better if age at time of treatment was
