Journal of Cancer Research and Therapeutics Medknow Publications on behalf of the Association of Radiation Oncologists of India (AROI) ISSN: 0973-1482 EISSN: 1998-4138 Vol. 7, Num. 1, 2011, pp. 52-57

Journal of Cancer Research and Therapeutics, Vol. 7, No. 1, January-March, 2011, pp. 52-57

Original Article

Fractionated stereotactic radiosurgery for the treatment of meningiomas

Carley Bria¹, Rodney E Wegner¹, David A Clump¹, John A Vargo¹, Arlan H Mintz², Dwight E Heron¹, Steven A Burton¹

¹ Department of Radiation Oncology, University of Pittsburgh Cancer Institute, Pittsburgh, USA
² Department of Neurological Surgery, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA

Correspondence Address: Dwight E Heron, Department of Radiation Oncology, UPMC Shadyside Hospital, 5150 Center Avenue, #545, Pittsburgh, PA 15232, USA, herond2@upmc.edu

Code Number: cr11011

PMID: 21546743

DOI: 10.4103/0973-1482.80462

Abstract

Keywords: Benign, radiation therapy, radiosurgery

Background

Benign intracranial neoplasms are confined to the cranial vault, with a minimal likelihood of metastasizing. However, progressive neurological decline can occur secondary to the location and minimal growth can have effects similar to those seen in malignant neoplasms. There are several different types of benign brain tumors, with the most frequent being meningioma, which accounts for about one-third of the benign intracranial tumors. ^[1]

Meningiomas arise from arachnoidal cells in the meninges, the membranes that surround and protect the central nervous system. ^[2] In 2007, the World Health Organization (WHO) published the updated meningioma grading criteria. Grade 1, considered benign based on its pathological characteristics, is the most common type of meningioma (70 - 85%). Grade 2 is considered atypical, while grade 3 tumors are referred to as anaplastic, with the ability to invade the brain and metastasize. ^[1]

Benign tumors can be managed in several different ways. The most basic and conservative approach is observation with serial imaging. However, due to concerns that minimal growth may lead to irreversible neurological effects, medical intervention is often discussed. This is often the case when the tumor causes pain, discomfort, or other symptoms that diminish the quality of life. Treatment options for meningioma include observation, surgical resection, radiation therapy or a combination approach involving multiple modalities. ^[2] Depending on the location of the tumor, surgical intervention may not be possible, based on the risks involved in the procedure. For example, if the tumor is located at the skull base, attempts at resection of the tumor may pose too big a risk to the surrounding structures including the cranial nerves. In this case, radiation therapy may be the preferred treatment modality. Radiation therapy can be delivered in several different manners, including conventional external beam radiation (EBRT), single fraction stereotactic radiosurgery (SRS), or fractionated stereotactic radiosurgery (FRS) delivered over two to five fractions. SRS or FSR is often preferable compared to the standard fractionated radiotherapy, not only from the standpoint of elapsed days of therapy, but also due to the ability to deliver a more effective and conformal dose of radiation to the tumor. ^[3] Additionally, SRS / FRS may minimize the adverse radiation effects, such as the cognitive decline, by decreasing the overall dose to the normal, uninvolved brain.

Materials and Methods

Patient Characteristics

A review of our records revealed 73 consecutive patients with meningioma, who underwent Cyber Knife ^TM stereotactic radiosurgery at our institution between January 2001 and July 2010. All the patients gave informed consent to participate in this retrospective study, which was approved by the Institutional Review Board. No patients were excluded after identification. The median patient age was 59 years (range: 18-89 years). Fifteen patients (20%) were male and 58 (80%) were female. Fifty-two patients (71%) had undergone previous treatment to the tumor site with the remaining 21 patients (29%) undergoing SRS / FSR as their primary treatment modality. Fifty-one patients (70%) had undergone previous resection and six patients (8%) had received previous EBRT. Three patients (4%) had previous GammaKnife^® SRS and two patients (3%) were previously treated with linac-based SRS at an outside facility, with no data available for review. Four patients (5%) had multiple intracranial locations targeted and treated. The median time from any previous treatment to the SRS / FSR at our institution was 15.9 months (range, 1.4 - 354.8). The CyberKnife ^TM Radiosurgery System (Accuray, Sunnyvale, CA) was used to treat all 73 patients in this study. Fifty-two patients presented with baseline symptoms, which included aphasia (3%), ataxia (3%), diplopia (1%), facial pain (1%), gait change (1%), headache (32%), hearing loss (6%), hemianopsia (1%), hemiparesis (7%), memory loss (4%), paresthesia (12%), ptosis (3%), vestibular symptoms (8%), and vision loss (23%). The remaining 21 patients (29%) were asymptomatic at presentation.

Simulation and Treatment Planning

Each patient in this study was comfortably positioned on the CT simulation table while a custom thermoplastic mask was created by the staff. A thin-slice high resolution CT with intravenous contrast was obtained while the patient was immobilized. Next, the acquired images were transferred to the treatment planning workstation and fused with a pretreatment thin-slice, contrast-enhanced, SPGR MRI, utilizing commercially available fusion software. The tumor volume and critical structures were manually delineated by a team of radiation oncologists, neurosurgeons, and medical physicists. Dose volume histograms were calculated for the tumor volume and nearby critical structures in order to select the optimal treatment plan. A plan was considered ideal if it delivered ≥ 95% of the prescribed dose to the tumor volume, without an excess dose to any surrounding organs at risk. After choosing the optimal treatment plan, radiosurgery was delivered to the 80% isodose line over one to five fractions on non-consecutive days. Prescription dose, fraction size, and fraction number were all chosen based on tumor location, prior treatment (EBRT, surgery, or both), radiographic tumor features (such as edema), and any existing tumor-related symptoms.

Follow-up

Patients were seen every six months after SRS / FSR in a multidisciplinary clinic. A complete neurological examination was performed and pertinent imaging was reviewed. Radiographic follow-up was based on a thin slice, gadolinium-enhanced MRI that was likewise obtained every six months following treatment. Tumor response was assessed and classified as progressive, stable or partial / complete response. A partial response or progressive disease was defined as a 2 mm change in any dimension of the tumor on imaging, corresponding to a clearly visible change.

Statisitical Analysis

We collected and analyzed the following patient / tumor characteristics, to determine if they affected the following outcomes: age, sex, prior treatment, tumor grade, ki-67 index, prescription dose, maximum dose, minimum dose, conformity index, number of fractions, and treatment volume. The outcomes reviewed included local control, overall survival, and disease-specific survival; all of which were defined from the time of completing SRS / FSR. Survival curves and median survival were calculated using the Kaplan-Meier method. ^[4] Factors affecting survival and local control were determined using the Cox proportional hazards model. ^[5] All statistical tests were carried out using SPSS version 18.0 (SPSS, Chicago, IL).

Results

There were a total of 73 patients with a total of 78 meningiomas included in this study; sixty patients (82%) with WHO Grade 1 meningioma, eleven patients (15%) with WHO Grade 2 meningioma, and two patients (3%) with WHO Grade 3 meningioma. The Ki-67 pathology index was recorded for 39 (53%) of the cases, while the remaining 34 cases were not documented. In 14 instances (36%), the Ki-67 index was less than or equal to 5%. Sixteen cases (41%) had indices of greater than 5% and less than 20%, and six tumors (23%) had Ki-67 readings of greater than or equal to 20%. The locations of the meningiomas are shown in [Table - 1]. The median prescription dose was17.5 Gy (range, 6 - 27 Gy) and the median maximum radiation dose was 22.38 Gy (range, 7.5 - 33.5 Gy). The median tumor volume was 5.54 cm ³ (range, 0.11 - 83.00 cm ³ ). The median number of fractions was three with a range of one to five. Thirty-four patients (47%) were treated in a single fraction. In total, 39 patients (53%) were treated with a multiple fraction regimen. One patient (1%) was treated in two fractions, 18 patients (25%) were treated in three fractions, and 20 patients (27%) were treated with over five fractions. The median conformity index was 1.28 (range, 1.07 - 1.79). [Figure - 1] shows a typical SRS plan for a patient with a meningioma.

The median follow-up for this cohort was 16.1 months (range, 1.5-98.0). Follow-up MRI was available for all 73 patients. Local failure was documented in 11 patients (14 lesions total). The median time to failure after SRS / FSR was 12.2 months (range: 1.5-98.0). Of note, six of fourteen lesions failed before the six-month follow-up (two Grade 1, two Grade 2, and two Grade 3 lesions). On Cox regression, only WHO grade was found to be a significant predictor for local failure. We also analyzed local control according to tumor volume (0-3.0 cm ³ , 3.1-9.9 cm ³ , and ≥ 10 cm ³ ), but did not find any statistically significant difference. The actuarial local control of WHO Grade 1 meningiomas at 12 and 24 months was 95 and 86%, respectively. Five (45%) of the 11 patients with Grade 2 meningioma experienced local failure and both patients with Grade 3 meningioma experienced local failure. The 12-month actuarial local control for Grade 2 and Grade 3 meningioma was 71 and 0%, respectively [Figure - 2].

No patients experienced ≥ Grade 3 acute toxicity based on the Common Terminology Criteria for Adverse Events (CTCAE). Only one patient experienced late Grade 3 toxicity in the form of ataxia. In patients that had tumor-related symptoms prior to treatment, there was a subjective improvement in 31 of 52 patients (60%), following radiosurgery.

At the time of the last follow-up, 51 of the 73 patients were currently alive with disease (70%), 12 were alive without disease (16.44%), six had died from the disease (8.22%), and four had died of other causes (5.48%). Six of the 13 (46%) Grade 2 or 3 meningioma patients had died, five due to the meningioma and one due to metastatic prostate cancer. The remaining seven are still alive with the disease. Only one patient with a WHO grade 1 meningioma was determined to have died from the meningioma. The remaining three deaths were a result of thrombosis, sepsis, and an unknown cause. The actuarial overall survival for the entire cohort at 12 months and 24 months was 97 and 89%, respectively. None of the factors analyzed were found to be significant predictors of overall survival. In terms of disease-specific survival, the actuarial rates were 98.5 and 90% at 12 months and 24 months, respectively [Figure - 3]. None of the factors analyzed were found to be significant predictors of a disease-specific survival.

Discussion

The management of meningiomas can range from observation to combined therapy, including surgery and radiation. Observation is a viable option when the tumor is small, if the patient is asymptomatic, of advanced age or in poor health. Sughrue and colleagues concluded that few patients with initial tumor diameters of less than 2 cm went on to develop new or worsened symptoms over a median follow-up period of 4.6 years. ^[6] Patients with tumors greater than an initial size of 3 cm went on to develop new or worsened symptoms 17% of the time. Therefore, observation in the early course of treatment for small asymptomatic meningiomas is a reasonable option in select instances.

Surgical resection is another approach often used to treat meningiomas. Total resection has been shown to be much more successful than partial resection. Simpson originally defined the extent of resection using five grades. ^[7] In 1985, Mirimanoff et al., found that in 145 patients, local recurrence after gross total resection alone was 7% after 5 years, 20% after 10 years, and 32% after 15 years. ^[8] More recently, Stafford and colleagues found that the recurrence rate following surgery in 465 patients was 12% after 5 years and 25% after 10 years. ^[9] If the lesion is in a critical region of the brain, complete resection may not be feasible and oftentimes results in a subtotal resection. Furthermore, surgical resection comes with inherent risks. Neurological deficits can result and occur in anywhere between 2 and 30% of the cases, depending upon the location of the tumor and the extent of the resection. ^[10] As discussed earlier, only 20 - 32% of the seemingly completely resected tumors recur with long-term follow-up, and if they do, they do so at a slow rate. ^[8] The rate of recurrence increases following subtotal resection, with a local failure rate of 39% after 5 years and 61% after 10 years. ^[9] In these cases, resection is usually followed by radiation therapy.

In a study conducted in 1975, at the University of California, San Francisco, Wara and Sheline found that using radiation therapy following resection of a meningioma decreased the recurrence rates. ^[11] The recurrence rate after subtotal resection without adjuvant radiation was 74%, while those that received adjuvant radiation therapy had a recurrence rate of 29%. In a study compiled by Goldsmith and Wara in 1994, the outcomes of 140 patients who had undergone radiation therapy following subtotal resection were analyzed. ^[12] Patients received 54 Gy using either the arc or two- or three-field static treatment plans. The 10-year overall survival rate for Grade 1 tumors was 77%. The authors found that the five-year, progression-free survival rate for patients treated before 1980 was 77%, while those treated after 1980 had a progression-free survival rate of 98% - indicating that improvements in imaging and radiation techniques have resulted in improved delivery and outcomes.

Some meningiomas are treated with EBRT alone. EBRT is typically used to treat larger meningiomas that are not amenable to even a subtotal resection. In one series, 101 patients with presumed benign skull base meningiomas were treated with fractionated radiation alone (66%) or after subtotal resection (34%). ^[13] The median follow-up was five years and local control rates for all patients were 95% at five years, and 92% at both 10 and 15 years. In another retrospective series, 41 patients were treated with RT without any surgery. ^[14] RT consisted of SRS, stereotactic radiotherapy, or three-dimensional conformal RT in 22, 11, and 9 cases, respectively. At a median follow-up of five years, local control was achieved in 93% of the cases. Rogers et al., reviewed the literature from several articles that reviewed the outcomes using SRS and EBRT. ^[15] They compared the two modalities and found that the progression-free survival (PFS) rates were almost identical in both groups. SRS had 75 - 100% 5 - 10 year PFS, while EBRT had 80 - 100%. Therefore, location and size were the main factors considered when choosing a treatment modality.

As mentioned, another way to deliver definitive treatment is through SRS. This procedure is often used to treat benign tumors in the central nervous system. SRS is a technique that is capable of delivering extremely focused, high doses of radiation over one to five fractions. The highly focused nature of the treatment allows for a rapid fall-off of the dose, thereby, sparing the surrounding normal brain and preventing potential toxicity. Kondziolka and colleagues retrospectively looked at the medical records of 972 patients (1,045 tumors) treated over an 18-year interval with the Gamma Knife SRS system. ^[16],[17] All patients received one fraction of radiation with a mean marginal dose of 14 Gy. After 10 years of follow-up, the local control for Grade 1, 2, and 3 tumors was 91, 50, and 17%, respectively. The toxicity rate was well within acceptable limits at 7%. This was a landmark study confirming that patients with small-to-medium sized meningiomas can be safely and effectively treated using SRS. A more recent study performed in 2009, by investigators at the Mayo Clinic produced similar results. ^[18] This cohort included 190 patients with 206 distinct meningiomas. The median prescription dose was 16 Gy in a single fraction, using the Gamma Knife SRS system. The median follow-up was 47 months and the disease-specific survival rates at five and seven years were 94 and 92%, respectively. The local control rate at five years was 93% for patients with WHO Grade 1 meningiomas. Thirteen percent of the patients experienced treatment-related complications.

When treating larger tumors, there is concern about the toxic effects of single session radiosurgery treatment, with the toxicity rates sometimes approaching 10%. ^{[16],[17],[18]} A study from the Mayo Clinic reviewed outcomes in 116 patients with Grade 1 meningiomas greater than 10 cm ³ . ^[19] The average tumor volume was 17.5 cm ³ and the average marginal dose was 15.1 Gy. The mean follow-up was an impressive 70 months. Local control was excellent with 92% of the tumors controlled at seven years. Toxicity, however, was significant at 23%, and included complications such as seizures, paresis, diplopia, and infarction. The delivery of the dose over multiple fractions could theoretically reduce this toxicity to the surrounding neurological structures, although most of that data was based on conventional fractionation using daily doses of approximately 2.0 Gy. ^[20] In 2007, a study from a group in Vicenza, Italy, was published on the outcomes of 199 WHO Grade 1 meningioma patients treated between 2003 and 2007, using the Cyberknife ^TM robotic radiosurgery system. ^[21] Patients with tumors less than 8 cm ³ (49 of the 199) were treated with a single fraction to 11 - 12 Gy. Two to five daily fractions were administered to 150 patients who had tumors greater than 8 cm ³ with a dose ranging from 11 to 25 Gy. The median follow-up time was 30 months. No acute or late treatment-related complications were observed. Seventeen percent of the patients showed subjective improvement in symptoms, and after five years, the progression-free survival was 93.56%.

A study from the group at Stanford reviewed the outcomes after FRS for cranial base tumors (including multiple benign histologies) in order to determine if a reduction in toxicity could be achieved. ^[22] Over an interval of nine years, 34 patients with large benign tumors were treated with multisession SRS (two to five sessions). The median prescription dose was 24 Gy over a median of three fractions. The median follow-up was 31 months and in that time they were able to modestly conclude that multisession radiosurgery appeared to not only be effective (100% local control and in 21% of patients, the symptoms were improved without complications), but also safe. However, longer follow-up was needed given the nature of the benign intracranial neoplasms, to fully determine the safety and efficacy of this technique.

Over four years, Iwai et al., looked at a small group of seven patients with large petroclival and cavernous sinus meningiomas, who were treated using two-staged Gamma Knife SRS sessions. ^[23] The median prescription dose was 9 Gy per session (range 8 - 12 Gy), and all patients were followed for a median of 39 months. Six patients (86%) experienced local control, three of those six showed improvement of symptoms, and only one patient progressed. There was no reported toxicity related to the treatment. Pendl et al., also looked at multistage treatments using the Gamma Knife. ^[24] This particular cohort consisted of 12 patients who had meningioma and they were treated with a median dose of 12 Gy per session (range 10 - 25 Gy). The mean follow-up period was 27.9 months. Six patients showed improvement in the neurological symptoms, four remained stable, and two experienced new cranial nerve deficits. The authors of this study concluded that staged Gamma Knife SRS could be used as an alternative to single fraction SRS when treating large, benign, cranial lesions.

The results of the present study are very similar to those discussed earlier, with one- and two-year local control rates of 95 and 86% for Grade 1 meningiomas. The number of patients in our series with Grade 2 and 3 meningiomas was limited, but our outcomes were still similar to those seen in the other SRS series. Furthermore, we were able to document an improvement in tumor-related symptoms in 60% of the patients, although this was based on a subjective evaluation by the patient and treating physician. Future research in this area should incorporate additional objective quality of life surveys to help better quantify the treatment effect on the existing symptoms. One shortcoming of the present study was the relatively small sample size compared to the other series in the published literature. The other drawbacks of this study included its retrospective nature and the inherent biases present in such studies. In addition, interpreting our results could be somewhat difficult and indirect considering the heterogeneous nature of our cohort. Seventy-one percent of the patients in our study had previous treatment (surgery, SRS at an outside facility, or EBRT). However, despite this previous treatment and the relatively larger size of the treated lesions (median volume : 5.54 cm ³ ), SRS was still able to be safely delivered, with only one episode of late Grade 3 toxicity. The most significant shortcoming in our series was the relatively short follow-up (16 months) given the benign nature of the majority of tumors treated in this group. Interestingly, six of the 14 lesions (two of which were Grade 1) that failed did so at a follow-up interval of less than six months, highlighting the need for a fairly rigorous follow-up following completion of treatment. Regardless, a longer follow-up will be needed to confirm the high local control rates and low toxicity reported here.

Conclusions

In conclusion, SRS (both single session and FRS) for the treatment of meningioma is an effective and safe means to provide local control. In addition, treatment of the tumor can often lead to an improvement in the existing symptoms with minimal toxicity.

References

Copyright 2011 - Journal of Cancer Research and Therapeutics

The following images related to this document are available:

Photo images