A Common Sense Approach to Radiotherapy Planning of Glioblastoma Multiforme Situated in The Temporal Lobe

A Common Sense Approach to Radiotherapy Planning of Glioblastoma Multiforme Situated in The Temporal Lobe

Int. J. Radiation Oncology Biol. Phys., Vol. 72, No. 3, pp. 900–904, 2008 Copyright Ó 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-...

610KB Sizes 0 Downloads 16 Views

Int. J. Radiation Oncology Biol. Phys., Vol. 72, No. 3, pp. 900–904, 2008 Copyright Ó 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/08/$–see front matter

doi:10.1016/j.ijrobp.2008.01.053

CLINICAL INVESTIGATION

Brain

A COMMON SENSE APPROACH TO RADIOTHERAPY PLANNING OF GLIOBLASTOMA MULTIFORME SITUATED IN THE TEMPORAL LOBE FELIX BOKSTEIN, M.D.,* FELIX KOVNER, M.D.,y DEBORAH T. BLUMENTHAL, M.D.,* ZVI RAM, M.D.,z HAIM TEMPLEHOFF, MS.C.,y ANDREW A. KANNER, M.D.,z AND BENJAMIN W. CORN, M.D.y * Neuro-Oncology Service, y Department of Radiation Oncology, and z Department of Neurosurgery, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel Purpose: Irradiation remains the cornerstone of management for glioblastoma multiforme. The Radiation Therapy Oncology Group and European Organization for Research and Treatment of Cancer advocate encompassing the primary tumor plus a 2-cm margin in the high-dose volume. One shortcoming of this approach is the exposure of critical structures to radiation doses that could exceed organ tolerance. We investigated whether the temporal bone (rather than the aforementioned 2-cm radius) would serve as a barrier to tumor spread when regarded as the anterior margin for temporal lobe lesions. We hypothesized that by using the temporal bone as the radiation field margin, toxicity could be reduced without compromising tumor control. Methods and Materials: Between 2003 and 2007, 342 patients with newly diagnosed glioblastoma multiforme were treated with surgery and primary irradiation at our institution. Of these 342 patients, 50 had lesions confined to the temporal lobe. The clinical target volume included the primary lesion, the area of edema when present, and a 2-cm margin, except in the direction of the temporal bone. Results: Of the 50 patients, 40 were available for evaluation. At a median follow-up of 12.95 months, 8 patients had not yet shown signs of tumor progression, 24 had local failure, 7 had distant or mixed (local plus distant) failure, and only 1 patient had failure in the infratemporal fossa. Conclusions: The results of the study have demonstrated an acceptable level of recurrence when the temporal bone, rather than a 2-cm margin, is used as the anterior border of the clinical target volume. The strategy we have proposed achieves tumor control and respects optic tolerance without resorting to complex, expensive approaches such as intensity-modulated radiotherapy. Ó 2008 Elsevier Inc. Glioblastoma, Radiotherapy planning, Temporal lobe.

lines for RT for both anaplastic astrocytoma and glioblastoma multiforme often advocate a radiation field design that includes the contrast-enhancing tumor plus a 2–3-cm margin (4, 5). In so doing, coverage of $70–80% of malignant cells (within the tumor and adjacent brain parenchyma) can be achieved. Although larger margins (up to and including whole brain RT) can always be sought, irradiation of more normal tissue is likely to increase morbidity, without necessarily improving the rates of survival or local control. During the past decade, several strategies have emerged, including three-dimensional conformal RT and intensitymodulated RT, to achieve dose deposition within the target volume and dose minimization within the surrounding environment. These strategies (6, 7), which have usually set out to exploit the availability of sophisticated technological breakthroughs, have shown limited degrees of success to date. Fundamental to most treatment strategies is the sculpting of

INTRODUCTION Radiotherapy has been a cornerstone in the treatment of malignant gliomas for nearly four decades (1). Modern, prospective randomized trials (2) have continued to demonstrate a survival advantage for radiotherapy (RT) compared with supportive management. Although the number of long-term survivors of this disease is few, the number of those who live >2 years is growing. Moreover, the improvement in outcome has been attributed to new combinations of RT and chemotherapy (3), which could prove to be associated with more late-term morbidity than the use of either modality alone. Such patients could be susceptible to late sequelae from treatment and could therefore benefit from therapeutic approaches that minimize long-term toxicity. Because of the uncertainty in defining the true geographic extent of high-grade gliomas relative to their computed tomography or magnetic resonance imaging appearance, guide-

American Society of Clinical Oncology, Chicago, IL, June 2007. Conflict of interest: none. Received Nov 13, 2007, and in revised form Jan 15, 2008. Accepted for publication Jan 23, 2008.

Reprint requests to: Felix Bokstein, M.D., Neuro-Oncology Service, Tel-Aviv Sourasky Medical Center, 6 Weizman St., Tel Aviv 64239, Israel. Tel: (+972) 524-26-6532; Fax: (+972) 369-74-789; E-mail: [email protected] Presented, in poster format, at the 43rd Annual Meeting of the 900

RT planning in temporal GBM d F. BOKSTEIN et al.

901

radiation fields in deference to the likely patterns of tumor spread. For some, including the major international groups studying gliomas in the context of cooperative trials (e.g., the Radiation Therapy Oncology Group [RTOG] and European Organization for Research and Treatment of Cancer [EORTC]), this intuitive approach presumes that natural barriers exist that impede the spread of tumor cells. If the latter assumption (which has never been verified) could be confirmed, physicians could reduce the diameter of the margin of treatment, which is usually added to the gross disease as defined on the imaging studies. The present report represents a review of patients diagnosed with glioblastoma multiforme arising exclusively in the temporal lobe. The goal of this study was to determine whether such tumors can be managed with RT strategies that presume that the temporal bone offers a natural barrier to tumor cell infiltration, thereby exposing smaller volumes of normal tissue to high radiation doses. METHODS AND MATERIALS Treatment selection All patients were evaluated in a multidisciplinary forum consisting of neurosurgeons, neuro-oncologists, and radiation oncologists after a pathologic diagnosis of glioblastoma multiforme was established. The recommendations of this forum called for the delivery of definitive RT, often in conjunction with temozolomide (following the publication of Phase III data supporting its use) (3). In most patients who were treated with chemoradiotherapy, temozolomide (Schering-Plough, Kenilworth, NJ) was administered concomitantly at a dose of 75 mg/m2/d, given 7 d/wk from the first day of RT until the last day of RT. After a 4-week break, the patients received up to six cycles of adjuvant temozolomide on a 5-day schedule every 28 days. In a subset of patients (n = 9), temozolomide was only administered after the conclusion of RT (standard 5-day schedule of 200 mg/m2 every 28 days) (8, 9).

Fig. 1. Typical conformal portal arrangement for glioblastoma multiforme arising in temporal lobe. Note, clinical target volume (CTV) included 2-cm margin for microscopic extension in all directions except when natural barrier (e.g., temporal bone) was presumed to exist. GTV = gross tumor volume. which included the contrast-enhancing lesion without edema plus a 2.5-cm margin, received 14 Gy. A clinical judgment was made to modify the planning target volume to exclude sensitive structures by presuming that ‘‘natural barriers’’ would impede the contiguous spread of tumor cells. As such, the traditional 2–2.5-cm margin was not added in the direction of the temporal bone.

RT design Fractionated focal RT at a dose of 2 Gy/fraction was administered once daily for 5 consecutive days each week (Sunday through Thursday to conform to the patterns of work attendance in Israel). The treatment was delivered for a 6-week period until a total dose of 60 Gy was reached. In general, treatment planning was done with the aid of the RTOG guidelines (Fig. 1). The RT plan was devised on a dedicated three-dimensional planning system (CMS, St. Louis, MO). Conformal RT was delivered with linear accelerators possessing a nominal energy of $6 MV. Intensity-modulated RT strategies had not yet been adopted during the study period at our institution. Quality assurance was provided by individual case reviews that were conducted on a weekly basis within the department. The gross tumor volume for both the initial volumes and conedown volume were determined from the postoperative magnetic resonance imaging (MRI) scans (unless only preoperative scans were available). In accordance with most modern RTOG trials, the initial clinical target volume included the contrast-enhancing region and surrounding edema, if present, demonstrated on the imaging studies, plus a 2.0-cm margin. The initial clinical target volume received 46 Gy in 23 fractions. Pursuant to this first treatment phase, the tumor volume (clinical target volume 2) for the cone-down treatment,

Follow-up The study patients were followed monthly by all members of the neuro-oncology team (neurosurgeons, neuro-oncologists, and radiation oncologists) until death. Interval imaging studies (usually MRI) were performed every 3 months. Functional neuroimaging to prove true tumor progression was done only in cases suspicious for radiation necrosis.

RESULTS Table 1 lists the characteristics of the 50 patients included in this study. Of the 342 patients with newly diagnosed glioblastoma multiforme treated at the Tel Aviv Medical Center between 2003 and 2007, 50 (14.6%) had the epicenter of their tumor in the temporal lobe (35 in the left temporal lobe and 15 in the right temporal lobe). Their median age was 58 years (range, 23–82). In most patients (n = 22), gross total resection was achieved. In 18 patients, subtotal resection was performed, and biopsy only was done in 10 patients. Of the 10 biopsy-only patients, 7 had disease that involved the left insular region. Of the 3 patients with right-sided disease who

902

I. J. Radiation Oncology d Biology d Physics

Table 1. Patient characteristics Patients (n) Age (y) Median Range Extent of surgery Gross total removal Subtotal removal Biopsy only RT Conformal fields WBRT + fractionated RT No treatment Data unknown Chemotherapy Chemoradiotherapy (Stupp et al. [3]) Temozolomide after RT No chemotherapy Data unknown Pattern of progression Local Local and distant Distant only Local and infratemporal fossa No relapse Data unknown Follow-up (mo) Median Range

50 58 23–77 22 18 10 42 3 3 2 22 9 9 10 21 3 2 1 12 11 5.45 1.5–23.4

Abbreviations: RT = radiotherapy; WBRT = whole brain RT.

underwent biopsy only, 2 had multifocal disease, which rendered aggressive resection meaningless. The third patient was 82 years old and it was decided to forego additional resection in deference to his advanced age. After a median follow-up of 12.95 months, 40 of 50 patients were available for evaluation of treatment failure after administration of treatment as outlined in the ‘‘Methods and Materials’’ section. Of the 40 patients, 24 had failure solely within the radiation fields, 2 had failure at the intracranial foci outside the radiation fields, 5 had failed within the

Volume 72, Number 3, 2008

radiation field and also distantly in other parts of the brain. Only 1 patient, however, had treatment failure in a region immediately outside the bones in question. This was demonstrated in the MRI scan (Fig. 2; computed tomography scan with bone windows was not available). This patient had treatment failure within the radiation fields and showed signs of tumor extension into the infratemporal fossa. Her tumor had been subtotally excised before administration of conformal RT without temozolomide. At the last follow-up, 8 patients were still receiving treatment with temozolomide or were continuing with follow-up only and had not yet shown signs of tumor progression.

DISCUSSION The temporal lobe of the brain is located in the middle cerebral fossa. It is delimited from the frontal lobe by the sylvian fissure and merges with the parietal lobe posterosuperiorly and with the occipital lobe posteriorly. The mesial portion of the temporal lobe borders on the cerebral peduncles and optic tract. The inferior and lateral surfaces of the temporal lobe are adjacent to the temporal and sphenoid bones of the calvarium, which serve as a natural barrier and divide the lobe from the temporal fossa laterally and from the infratemporal, as well as the pterygopalatine fossa inferiorly (Fig. 3). Malignant glial tumors originating in the temporal lobe can freely extend into the adjacent tissue of other brain lobes. Unlike meningeal tumors that can extend through the bone foramina or even grow through the bone itself to extrude from the cranial cavity, expansion of malignant glioma to the orbit and bony, structures as well as to the soft tissues outside the cranium, occurs only rarely (10, 11). About 30% of glioblastomas originate in the temporal lobe (12). Within our database, gliomas arising exclusively in the temporal lobe represented approximately 14% of the cases of glioblastoma multiforme encountered in the past 5 years.

Fig. 2. Magnetic resonance imaging scans showing (a) initial tumor presentation and (b) recurrent disease for only patient in series with treatment failure outside natural barrier of temporal bone.

RT planning in temporal GBM d F. BOKSTEIN et al.

903

Fig. 3. Depiction of temporal lobe and critical structures within region: (a) temporal lobe; (b) squamous portion of temporal bone; (c) petrous portion of temporal bone; (d) sphenoid wing; (e) clinoid process of the sphenoid bone; and (f) infratemporal fossa.

In the past, treatment planning for malignant glioma was considered to be a matter of intense controversy (4). Specifically, divergent treatment philosophies ranged from moderate doses of whole brain RT at one extreme (13) to dose-intensified RT for limited fields at the other extreme (6). Owing to the fear of neurocognitive damage and with the emergence of improved imaging modalities, such as computed tomography and, especially, MRI (14, 15), the former was abandoned. However, enthusiasm for the latter began to wane when the clinical benefit could not be documented in prospective trials (6). In general, a consensus has emerged (5) to consider the edema as depicted by the T2-weighted or fluid-attenuated inversion recovery abnormality on MRI to be at risk of microscopic tumor extension. This volume is usually targeted to 45–50 Gy (with conventional fractionation) followed by a boost to the gross tumor as represented by the T1-weighted enhancing abnormality on MRI to a total dose of approximating 60 Gy. To support this approach, data on the patterns of failure can be marshaled from correlative evidence from interval follow-up scans (e.g., computed tomography, MRI) or postmortem evaluation (16, 17). The time-honored principle of including the encompassing edema in the initial target volume delineation has recently been questioned by Chang et al. (18). The conclusions of those investigators, who noted that coverage of edema did not alter the patterns of failure, could indicate that other traditional treatment guidelines are worthy of reconsideration. At present, an international protocol to study patients with glioblastoma multiforme has been launched jointly by the RTOG and EORTC. This Phase III clinical trial (RTOG 05-25), which compares conformal RT plus conventionaldose temozolomide with conformal RT plus dose-intense temozolomide, allows the treating radiation oncologist to exercise discretion during treatment planning. Specifically, the

protocol states that ‘‘clinical judgment may be used to modify the planning target volume to exclude sensitive structures such as the optic chiasm, noncranial contents, or anatomic regions in the brain where natural barriers would likely preclude microscopic tumor extension.’’ The results we have reported support this concept in that a very low rate of failure outside the temporal bone was encountered. As such, smaller volume fields can be designed in certain circumstances that would theoretically decrease the risk of selected side effects. This is reassuring, because an important consideration in electing to limit the volume of treatment is the concern for adverse events as the radiation portal is expanded to encompass presumed microscopic foci of disease. To date, we do not have sufficient follow-up to confidently report the rate of late effects among the patients studied. In considering the potential morbidity associated with RT for temporal lobe glioma, the most important organs at risk include the normal brain parenchyma, infratemporal and pterygopalatine fossa, optic apparatus, and cochlea-middle ear complex. For medial temporal tumors in close proximity to the optic apparatus, some physicians might be inclined to limit the dose to approximately 50 Gy to respect optic tolerance. In so doing, optic complications would be statistically negligible, but compromise could result in the rates of local control or survival for a disease that is often presumed to be associated with a dose response at 60 Gy. Even at doses nearing 60 Gy, the true risk of optic complications is probably <5% (19) and might not be a reason to limit the tumor dose. However, physicians and patients who are animated by the fear of such toxicity could choose to rely on the natural barrier function of the temporal bone to treat to 60 Gy with near impunity. Glioblastoma multiforme remains a daunting problem for clinical oncologists. Radiation treatment strategies must continue to seek a balance between enhanced local tumor control

904

I. J. Radiation Oncology d Biology d Physics

and reduced treatment-related toxicity. Although outcome appears to be improving secondary to the combination of new therapies with RT, it is conceivable that as survival improves for individuals treated for malignant gliomas, patient subsets will be at risk of late toxicity at greater rates than reported in historical clinical trials. The results of the present

Volume 72, Number 3, 2008

study suggest that through the judicious application of ‘‘common clinical sense’’ physicians can make compromises on treatment volumes to spare normal structures without sacrificing local control. Future strategies designed for patients with malignant gliomas can build on this modest, but real, benefit.

REFERENCES 1. Walker MD, Green SB, Byar DP, et al. Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 1980;303: 1323–1329. 2. Keime-Guibert F, Chinot O, Taillandier L, et al. Radiotherapy for glioblastoma in the elderly. N Engl J Med 2007;356: 1527–1535. 3. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987–996. 4. Wallner KE. Radiation treatment planning for malignant astrocytoma. Semin Radiat Oncol 1991;1:17–22. 5. Halperin EC, Burger PC, Bullard DE. The fallacy of the localized supratentorial malignant glioma. Int J Radiat Oncol Biol Phys 1988;15:505–509. 6. Lee SW, Fraass BA, Marsh LH, et al. Patterns of failure following high-dose 3-D conformal radiotherapy for high-grade astrocytomas: a quantitative dosimetric study. Int J Radiat Oncol Biol Phys 1999;43:79–88. 7. Stieber VW, Tatter SB, Lovato J, et al. A phase I dose escalating study of intensity modulated radiation therapy (IMRT) for the treatment of glioblastoma multiforme (GBM). Int J Radiat Oncol Biol Phys 2004;60:s261. 8. Yung WK, Albright RE, Olson J, et al. A phase II study of temozolomide vs. procarbazine in patients with glioblastoma multiforme at first relapse. Br J Cancer 2000;83:588–593. 9. Yung WK, Prados MD, Yaya-Tur R, et al., for the Temodal Brain Tumor Group. Multicenter phase II trial of temozolomide in patients with anaplastic astrocytoma or anaplastic oligoastrocytoma at first relapse. J Clin Oncol 1999;17: 2762–2771.

10. Brandes A, Carollo C, Gardiman M, et al. Unusual nasal and orbital involvement of glioblastoma multiforme: A case report and review of the literature. J Neurooncol 1998;36:179–183. 11. Rainov NG, Holzhausen HJ, Meyer H, et al. Local invasivity of glioblastoma multiforme with destruction of skull bone: Case report and review of the literature. Neurosurg Rev 1996;19: 183–188. 12. Schreiber D, Warzok R. [Localization of brain tumors in autopsy. Part I: Tumors of the temporal lobe]. Zentralbl Neurochir 1981;42:241–250. 13. Salazar OM, Rubin P. The spread of glioblastoma multiforme as a determining factor in the radiation treated volume. Int J Radiat Oncol Biol Phys 1976;1:627–637. 14. Burger PC, Heinz ER, Shibata T, et al. Topographic anatomy and CT correlations in the untreated glioblastoma multiforme. J Neurosurg 1988;68:698–704. 15. Kelly PJ, Daumas-Duport C, Kispert DB, et al. Imaging-based stereotaxic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 1987;66:865–874. 16. Massey V, Wallner KE. Patterns of second recurrence of malignant astrocytomas. Int J Radiat Oncol Biol Phys 1990;18:395–398. 17. Hochberg FH, Pruitt A. Assumptions in the radiotherapy of glioblastoma. Neurology 1980;30:907–911. 18. Chang EL, Akyurek S, Avalos T, et al. Evaluation of peritumoral edema in the delineation of radiotherapy clinical target volumes for glioblastoma. Int J Radiat Oncol Biol Phys 2007; 68:144–150. 19. Parsons JT, Bova FJ, Fitzgerald CR, et al. Radiation optic neuropathy after megavoltage external-beam irradiation: analysis of time-dose factors. Int J Radiat Oncol Biol Phys 1994;30: 755–763.