- Email: [email protected]
Preirradiation Ifosfamide, Carboplatin, and Etoposide for the Treatment of Anaplastic Astrocytomas and Glioblastoma Multiforme
Archives of Medical Research 31 (2000) 186–190
ORIGINAL ARTICLE
Preirradiation Ifosfamide, Carboplatin, and Etoposide for the Treatment of Anaplastic Astrocytomas and Glioblastoma Multiforme: A Phase II Study Enrique López-Aguilar,* Ana Carolina Sepúlveda-Vildósola,** Hugo Rivera-Márquez,* Fernando Cerecedo-Díaz,* Isidro Hernández-Contreras,*** Guillermo Ramón-García,**** Jaime Diegopérez-Ramírez***** and Enrique Santacruz-Castillo****** *Departamentos de Oncología, **Pediatría, ***Radiología, ****Patología, *****Neurocirugía, ******Radioterapia, Hospital de Pediatría, Centro Médico Nacional Siglo XXI, Instituto Mexicano del Seguro Social (IMSS), México, D.F., Mexico Received for publication February 12, 1999; accepted November 15, 1999 (99/015).
Background. Central nervous system (CNS) tumors are the second most common pediatric tumors. Astrocytomas represent 35% of all CNS tumors in children. Traditional treatment of anaplastic astrocytoma (AA) and glioblastoma multiforme (GM) consisting of surgery-radiotherapy-chemotherapy with nitrosoureas has resulted in a survival rate of 26% at 1 year. Neoadjuvant chemotherapy has proven good results in the treatment of other solid tumors. Chemotherapy with ifosfamide, carboplatin, and etoposide (ICE) permits synergism among the different drugs and sensitizes the tumor to radiotherapy. Our objective was to evaluate the efficacy, security, and survival rate of postoperative chemotherapy with ICE in pediatric patients with AA or GM. Methods. Phase II study. We evaluated 11 children with AA or GM who had received no prior treatment. A magnetic resonance image (MRI) study of the tumor was made after surgery to evaluate residual tumor and routine laboratory analysis. Chemotherapy with carboplatin, ifosfamide and etoposide was given every 3 weeks for four courses. MRI studies were repeated after the second and last courses and laboratory analyses were carried out before each course to evaluate toxicity. Each patient then received hyperfractionated radiotherapy and a final MRI was done at the end of the treatment. Results. Sixty percent of the patients had partial response, 30% complete response after two courses, and 60% of CR after four courses. Supratentorial and infratentorial tumors had a good response to chemotherapy. Brainstem tumors had an initial response after two courses and then increased in size. AA was the tumor with the greatest reduction of residual tumor after treatment. Overall and free survival at 53 months was 70%. To date, three patients have died secondary to tumoral progression. There have been no relapses in the seven patients with a CR. Conclusions. Postoperative chemotherapy with ICE reduces the tumor size and increases the survival rate of pediatric patients with malignant astrocytomas with minimal toxicity. Brainstem responded poorly to treatment. © 2000 IMSS. Published by Elsevier Science Inc. Key Words: Astrocytoma, Glioblastoma multiforme, Ifosfamide, Carboplatin, Etoposide.
Introduction Address reprint requests to: Dr. Enrique López Aguilar, Hospital de Pediatría, Centro Médico Nacional Siglo XXI, IMSS, Av. Cuauhtémoc 330, Col. Doctores, CP 06720 México, D.F., Mexico. Tel.: (⫹525) 627-6900, ext. 3525; FAX: (⫹525) 761-2594.
Tumors of the central nervous system (CNS) constitute the largest group of solid neoplasms in children, and are the second cause of cancer in this group age (1). Astrocytomas are the most common form of primary intracranial tumor.
0188-4409/00 $–see front matter. Copyright © 2000 IMSS. Published by Elsevier Science Inc. PII S0188-4409(00)00 0 5 1 - 5
López-Aguilar et al. / Archives of Medical Research 31 (2000) 186–190
However, survival of malignant tumors has changed little from survival reported in the early 1970s. In combined series of adult and pediatric patients, the median survival rarely exceeds 12 months, especially in high-grade astrocytomas such as anaplastic astrocytomas and glioblastoma multiforme, in which median survival duration is 40–52 weeks (2– 5). Trials conducted by the International Society of Pediatric Oncology and the Children’s Cancer Group in the late 1970s and early 1980s showed a possible benefit in survival for patients treated with adjuvant chemotherapy. Favorable results were reported from a pilot study using the eight-drugs-in1-day therapy for children with newly diagnosed brain tumors, including high-grade astrocytomas (6). Nevertheless, responses to chemotherapy have not had significant impact on the outcome of these tumors. Although the use of preirradiation or neoadjuvant chemotherapy offers an opportunity to evaluate the potential benefit of new drug combinations, few such studies have been performed in children. Limited data are available from contemporary reports of single-institution experiences with malignant gliomas. These series are made up of patients diagnosed over periods exceeding 20 years. Al-Mefty analyzed a group of children with newly diagnosed high-grade gliomas, 80% of whom were treated with a variety of neoadjuvant chemotherapy regimens, and compared these results with those of other single- and multiinstitutional pediatric studies, with a discrete increment of survival in the group treated with neoadjuvant regimens (7). To improve the survival of these patients, we have initiated a phase II study with preirradiation ifosfamide, carboplatin, and etoposide (ICE) for the treatment of these malignant tumors. Each drug has been used before as a single agent, or in combination, for the treatment of recurrent brain tumors and in phase II studies of childhood brain tumors with different response rates (8–10), but the combination of these three drugs has never been attempted for the treatment of childhood brain tumors. The most serious limitation of brain tumor chemotherapy is the blood brain barrier (BBB), which restricts the passage of chemotherapy agents. However, the presence of the BBB is not absolute. It is usually disrupted in most brain tumors and does not function normally in the area surrounding the tumor (11–14). We report on a prospective phase II trial of preirradiation ifosfamide, carboplatin, and etoposide in newly diagnosed patients with high-grade astrocytomas. Our objectives include assessment of tumor response rate to ICE before irradiation, the assessment of acute and subacute toxicity of preirradiation ICE, and the determination of disease control interval and survival following combined modality therapy.
187
We included 11 consecutively admitted children who had a newly diagnosed high-grade astrocytoma (AA or GM) who were operated on at our hospital, and whose diagnosis was histologically confirmed by a pathologist (kappa 0.68) using the Kernohan classification system. All had measurable residual disease after initial surgery. Informed consent was obtained from the children’s parents before enrollment and the study was approved by the local Ethics and Investigation Committees (AB/06/95). Initial evaluation of disease and inclusion in the study was carried out within the first postoperative week and included magnetic resonance image (MRI) of the brain (performed 72 h after surgery), lumbar CSF cytology, complete blood cell count including platelets, and renal function evaluation. After initial surgery, every patient received chemotherapy with carboplatin (400 mg/m2/day on day 1, diluted in a 0.9% NaCl solution for 1 h) followed by ifosfamide (2 g/m2 on days 1, 2, and 3, diluted in a 0.9% NaCl solution for 1 h) and etoposide (100/mg/m2/day on days 1, 2, and 3 diluted in a 0.9% NaCl solution for 1 h), every 3 weeks for four courses. Each patient also received antiemetics (ondasetron 8 mg/m2/dose every 8 h) initiating simultaneously with chemotherapy treatment until 24 h after terminating chemotherapy, hyperhydration (3,000 mL/m2/day dextrose 5% 2:1 normal saline), and a bladder protector (MESNA 500 mg/m2 every 6 h on days 1, 2, and 3, given simultaneously with ifosfamide). Twenty-eight days after finishing the fourth course of chemotherapy, they all underwent hyperfractionated radiotherapy (total dose of 54 Gy, given in two fractions of 1.1 Gy per day during 25 days). Follow-up studies included MRI after the second and last courses of chemotherapy and after 4 weeks from completing radiotherapy (RT). To evaluate toxicity, renal function tests and a complete blood cell count were done before each course. Response to ICE was graded on the basis of neuroimaging findings by a radiologist unaware of the treatment given to the patient (kappa 0.65), comparing measurable disease in the primary site after surgery and after two and four courses of chemotherapy as well as after radiotherapy. Response was categorized as follows: complete response (CR) was complete resolution of all radiographic evidence of tumor; partial response (PR) was at least a 50% reduction of the product of maximal perpendicular diameters at the plane of greatest tumor area by MRI with stable or improving neurologic function; stable disease (SD) was less than a 50% reduction in tumor area or no more than a 25% increase with stable or improving neurologic function, and progressive disease (PD) was more than a 25% increase in tumor area. Statistical analyses included measures of middle tendency and spread, percentages, Student’s t test, ANOVA, and actuarial survival analysis.
Patients and Treatment The study was conducted at the Hospital de Pediatría, Centro Médico Nacional Siglo XXI in Mexico City between December 1994 and October 1999.
Results Eleven patients were initially included. One abandoned treatment before completing two courses of chemotherapy
188
López-Aguilar et al./ Archives of Medical Research 31 (2000) 186–190
prior to evaluation of response to treatment and was excluded from the study. Mean age was 10.4 years (range 4– 14) with a male-female relation of 2.3:1 (Table 1). According to the site of initial presentation, six tumors (60%) were supratentorial, two (20%) infratentorial, and two (20%) in the brainstem. According to the histology, seven patients (70%) had anaplastic astrocytoma and three (30%), glioblastoma multiforme. After two courses of chemotherapy, 60% of patients had partial response and 30%, complete response. After four courses of chemotherapy, six patients were categorized as CR (60%), two as PR (20%), and two as PD (20%). Following radiotherapy, one additional patient achieved CR (Table 1). According to the tumor site, tumor reduction following four courses of chemotherapy was 80% and 100% for supratentorial and infratentorial tumors, respectively. Brainstem tumors had an initial response to treatment but involved progressive disease after four courses. According to the histology, patients with anaplastic astrocytomas presented a reduction of 62% after two courses, 75% after four courses, and 81% after radiotherapy (p ⫽ 0.02). Patients with glioblastoma multiforme had tumor burden reductions of 24, 24, and 27%, respectively (p ⫽ 0.28). Overall and disease-free survival was 70% at 53 months after diagnosis (Figure 1). According to histopathology, survival for anaplastic astrocytomas was 68% at 53 months and 66% at 47 months for glioblastoma multiforme. According to tumor location, there was 100% survival at 53 months for infratentorial tumors, 66% for supratentorial tumors, and 0% at 11 months for brainstem tumors. There was minimal toxicity during the 40 courses of chemotherapy. Hemoglobin values were within normal range in 97% of the courses and only one patient required hemotransfusion. Grade 3 neutropenia occurred in 10% of the
courses, and no patients required hospitalization for neutropenia and fever. Platelets and creatinine had no significant toxicity during the 40 courses of chemotherapy. Three patients have died to date. One was a patient with a supratentorial tumor (patient no. 1), who had PR to four courses of chemotherapy and PD during radiotherapy, and died 6 months after diagnosis. Another patient (patient no. 4) had a brainstem tumor with PD after four courses of chemotherapy and died 11 months after diagnosis. The last patient (patient no. 8), who also had a brain stem tumor, had no response to chemotherapy, presented tumor progression after radiotherapy, and died 8 months after diagnosis. Six patients had CR after four courses of chemotherapy and one more after radiotherapy. They are considered cured at present and lead normal lives for their age (they attend school, are independent from their parents, and have normal neuropsychological evaluations). At present, all have been in follow-up for more than 3 years without relapses or progression.
Discussion Recent decades have been characterized by significant changes in the survival of cancer patients. Nevertheless, CNS tumors, especially high-grade astrocytomas, have seen fewer relevant advances because fewer new treatment strategies have been evaluated. Surgery is frequently unable to reduce the tumor burden due to the great infiltrative capacity of these tumors over the neural tissue. Traditional treatment with postoperative radiotherapy has not been able to significantly increase the survival rate of these patients in that intratumoral isodoses are frequently sublethal to neoplasic cells because of the great tumor burden that usually persists after surgery. Previous ex-
Table 1. Patient characteristics and response to ICE protocol
Patient 1 2 3 4 5 6 7 8 9 10 p value
Age (years)
Sex
Tumor location
Histology
Tumor volume following surgery
13 11 14 8 12 11 4 12 14 5
Male Male Female Male Female Male Male Male Male Female
Supratentorial Supratentorial Infratentorial Brainstem Supratentorial Supratentorial Supratentorial Brainstem Infratentorial Supratentorial
AA AA AA AA GM AA AA GM AA GM
37.9 38.5 8.0 20.0 3.0 14.6 3.6 18.02 4.40 1.69
Response and tumor volume after two courses of chemotherapy
Response and tumor volume after four courses of chemotherapy
Response and tumor volume post Rt
Present status (no. of months of follow-up)
15.6 (PR) 9.6 (PR) 4 (PR) 5.6 (PR) 0.8 (PR) 6.4 (PR) 0 (CR) 18.5 (PD) 0 (CR) 0 (CR) 0.01
7.5 (PR) 7 (PR) 0 (CR) 10.8 (PD) 0 (CR) 0 (CR) 0 (CR) 18.5 (PD) 0 (CR) 0 (CR) 0.006
D 0 (CR) 0 (CR) 15.4 (PD) 0 (CR) 0 (CR) 0 (CR) 18.5 (PD) 0 (CR) 0 (CR) 0.01
D DF (53) DF (53) D DF (47) DF (44) DF (41) D DF (40) DF (39)
Note: Tumor volume is measured as cm2. AA: Anaplastic astrocytoma; D: death; GM: glioblastoma multiforme; DF: disease-free; PR: partial response; CR: complete response; PD: progressive disease.
López-Aguilar et al. / Archives of Medical Research 31 (2000) 186–190
Figure 1. Overall and disease-free survival at 48 months of patients with malignant astrocytomas treated with chemotherapy with ICE.
periences with these patients show that the effect of RT administered after surgery did not completely control the residual tumor and that the tumor presented greater progression. This was reflected by the poor survival rate of 26% at 12 months in our Hospital, similar to other reports (12,15–17). The role of chemotherapy in malignant gliomas remains uncertain. Although meta-analysis of patients treated with various chemotherapeutic regimens suggests that such treatments may be associated with little improvement in survival, multiple problems in the design and execution of such large retrospective analyses render acceptance of conclusions difficult. Finlay et al. (18) demonstrated that survival of patients with grades III and IV astrocytomas treated with adjuvant chemotherapy with vincristine, lomustine, and prednisone was 28 and 16% at 5 years, respectively. While this regimen was suggested to improve the survival of patients with newly diagnosed high-grade astrocytomas, the number of patients was small and close examination reveals that the results rest only on a few long-term survivors. However, for many years it was considered that adjuvant chemotherapy was of little benefit in the treatment of these patients basically because it was thought that few anti-neoplastic agents could pass the blood brain barrier. These agents, it was believed, had to be liposoluble in order to reach the tumor. Also, the cytotoxic effect initially found in cell cultures was minimal. Only nitrosoureas demonstrated some effect in the treatment of these tumors, the reason that most chemotherapy schemes use nitrosoureas. Recently, it has been proven that the blood brain barrier is dysfunctional in areas surrounding the tumor, and that not only liposoluble agents can reach the tumor (11). Kovnar et al. (19) have adopted a policy of using neoadjuvant treatment in most newly diagnosed patients. In the current series, 80% of patients received preirradiation chemotherapy. The regimen most commonly used, cisplatin and etoposide, has been previously reported by these authors and was associated with stable disease in seven and objective responses in only two of 12 patients. The regimen with high-
189
dose cyclophosphamide and thiotepa has also been previously reported (20). There are to date limited data from other investigations testing high-dose chemotherapy in the 20 preirradiation setting. These data, however, do not show better responses than other regimens with adjuvant chemotherapy. A concern associated with the use of neoadjuvant chemotherapy is that the inherent delay in the initiation of radiation therapy may have an adverse impact on survival. Despite the low rate of objective response to neoadjuvant chemotherapy reported by Heideman (20), the median survival is similar to that reported in other series of patients (21,22). This suggests that the delay in radiation therapy associated with neoadjuvant chemotherapy may not significantly alter the survival of children with malignant gliomas, and supports the use of such studies to identify effective regimens in this generally refractory group of tumors (23). In this study, we administered the following new scheme of chemotherapy that includes two agents that have proven a cytotoxic effect in vitro over these tumor cells: ifosfamide and carboplatin, and etoposide, which enhances this latter drug’s effect. We changed the sequence of treatment (RT– QT to QT–RT) to attempt to decrease the residual tumor burden after surgery. With this new approach, we found a reduction in tumor size in most patients (60%), with a mean of 55% after two courses, 66% after four courses, and 72% after RT. We found that the histology of the tumor, as well as its location, influences response to treatment. Infratentorial tumors and anaplastic astrocytomas had a good response in contrast to multiform glioblastoma and brainstem tumors, which had little or no response to chemotherapy or radiotherapy. We consider this phenomena to be mainly due to the almost impossible surgical access of these tumors, and also probably to a reduced concentration of the drugs inside the tumor caused by poor brainstem blood flow. No other factors were found to influence the tumor’s response to treatment. Nevertheless, we think that final conclusions cannot be made due to the small amount of patients in each group. With this new sequence, we found a survival rate of 70% for all tumors, significantly greater than that previously reported. Patients with infratentorial tumors have a survival of 100%, but the sample size is too small to draw conclusions. Even though we found a very poor survival rate for patients with brainstem tumors, we found an improvement in their quality of life during treatment. There have been no relapses during follow-up. Finally, it is important to mention that minimal toxicity was noted with this treatment, and it was found to be greatly inferior to that observed with chemotherapy with nitrosoureas. The number of patients in this report is small, especially after the report is stratified by histology or location, and further studies must be conducted to confirm our results. Additionally, this new chemotherapy regimen must be compared with traditional treatment in a phase III study.
190
López-Aguilar et al./ Archives of Medical Research 31 (2000) 186–190
The results we present are very encouraging and are, as far as we are aware, the best results reported to date in a cohort of malignant glioma patients treated with chemotherapy. We expect to increase the sample size to make final conclusions and to subsequently improve this new approach.
11. 12.
13.
References 1. Silverberg EE. Cancer statistics. Cancer 1981;31:13. 2. Dinapoli RP, Brown LD, Arusell RM. A phase III comparative evaluation of BCNU and carmustine with radiation therapy for high grade gliomas. J Clin Oncol 1993;11:1316. 3. Ganju MB, Jenkins RB, O’Fallon JR, Scheitauer BW, Ransom DT, Katsmann JA. Prognostic factors in gliomas. Cancer 1994;74(3):920. 4. Burger PC, Green SB. Patient age, histologic features and length of survival in patients with glio-blastoma multiforme. Cancer 1987;59:1617. 5. Marchese MJ, Chang CH. Malignant astrocytic gliomas in children. Cancer 1990;65:2771. 6. Pendergrass TW, Milstein JM, Geyer IR. Eight drugs in one day chemotherapy for brain tumors. J Clin Oncol 1987;5(8):1221. 7. Al-Mefty O, Al-Rodhaun HF. Factors affecting survival of children with malignant gliomas. Neurosurgery 1987;20:416. 8. Gentet JC, Doz F, Bouffet E, Plantaz D, Rochè H, Tron P, Kalifa CH, et al. Carboplatin and VP-16 in medulloblastoma: a phase II study of the French Society of Pediatric Oncology ( SFOP ). Med Pediatr Oncol 1994;23:422. 9. Chastagner P, Sommelet OD, Kalifa CH, Brunat MM, Zucker JM, Domeocq F. Phase II study of ifosfamide in childhood brain tumors. Med Pediatr Oncol 1993;21:49. 10. Kovnar EH, Kellie SJ, Horowitz ME, Sanford RA, Longston RK, Mulhern JJ, Jenkins EC, et al. Preirradiation cisplatin and etoposide in the
14. 15.
16. 17.
18.
19. 20.
21. 22. 23.
treatment of high risk medulloblastoma and other malignant embryonal tumors of the central nervous system. J Clin Oncol 1990;8:330. Long DM. Capillary ultrastructure and the blood brain barrier in human malignant brain tumors. J Neurosurg 1970;32:127. Neuwelt EA, Howieson J, Frenkel EP. Therapeutic efficacy of chemotherapy with drug delivery enhancement by blood brain barrier modification in glioblastoma. Neurosurgery 1986;19:573. Levin VA, Freeman-Dove M, Landahl HL. Permeability characteristics of brain adjacent to tumors in rats. Arch Neurol 1975;32:785. Hirano HA, Matsui T. Vascular structures in brain tumors. Hum Pathol 1975;6:611. Grossman SA, Wharam M, Sheidler V. BCNU and cisplatin followed by radiation in poor prognosis patients with high grade astrocytomas. (Abstract). Proc Am Soc Clin Oncol 1992;11:149. Levin VA. Chemotherapy of primary brain tumors. Neurol Clin 1985;3:855. Fine HA, Dear KB, Loeffler JS. Meta-analysis of radiation therapy with and without adjuvant chemotherapy for malignant gliomas in adults. Cancer 1973;71:2585. Finlay JL, Boyett JM, Yates AJ. Randomized phase III trial in childhood high grade astrocytomas comparing vincristine, lomustine and prednisone with the eight drugs in one day regimen. J Clin Oncol 1995;13:112. Kovnar E, Horowitz M, Kun L. Carboplatin and etoposide in the treatment of childhood malignant glioma. Neurology 1991;41:410. Heideman RL, Douglas EC, Krance RA. High dose chemotherapy and autologus bone marrow rescue in newly diagnosed pediatric malignant gliomas. J Clin Oncol 1993;11:1458. Phuphanich S, Edwards M, Levin V. Supratentorial malignant gliomas in childhood. Ann Neurol 1987;22:355. Kleihues P, Burger PC, Scheithorver BW. The new WHO classification of brain tumors. Brain Pathol 1993;3:255. Heideman RL, Kuttesch J, Gajjar AJ, Walter AW, Jenkins JJ, Li Yulan, Sanford RA, Kun LE. Supratentorial malignant gliomas in childhood. Cancer 1997;80(3):497.
ORIGINAL ARTICLE
Preirradiation Ifosfamide, Carboplatin, and Etoposide for the Treatment of Anaplastic Astrocytomas and Glioblastoma Multiforme: A Phase II Study Enrique López-Aguilar,* Ana Carolina Sepúlveda-Vildósola,** Hugo Rivera-Márquez,* Fernando Cerecedo-Díaz,* Isidro Hernández-Contreras,*** Guillermo Ramón-García,**** Jaime Diegopérez-Ramírez***** and Enrique Santacruz-Castillo****** *Departamentos de Oncología, **Pediatría, ***Radiología, ****Patología, *****Neurocirugía, ******Radioterapia, Hospital de Pediatría, Centro Médico Nacional Siglo XXI, Instituto Mexicano del Seguro Social (IMSS), México, D.F., Mexico Received for publication February 12, 1999; accepted November 15, 1999 (99/015).
Background. Central nervous system (CNS) tumors are the second most common pediatric tumors. Astrocytomas represent 35% of all CNS tumors in children. Traditional treatment of anaplastic astrocytoma (AA) and glioblastoma multiforme (GM) consisting of surgery-radiotherapy-chemotherapy with nitrosoureas has resulted in a survival rate of 26% at 1 year. Neoadjuvant chemotherapy has proven good results in the treatment of other solid tumors. Chemotherapy with ifosfamide, carboplatin, and etoposide (ICE) permits synergism among the different drugs and sensitizes the tumor to radiotherapy. Our objective was to evaluate the efficacy, security, and survival rate of postoperative chemotherapy with ICE in pediatric patients with AA or GM. Methods. Phase II study. We evaluated 11 children with AA or GM who had received no prior treatment. A magnetic resonance image (MRI) study of the tumor was made after surgery to evaluate residual tumor and routine laboratory analysis. Chemotherapy with carboplatin, ifosfamide and etoposide was given every 3 weeks for four courses. MRI studies were repeated after the second and last courses and laboratory analyses were carried out before each course to evaluate toxicity. Each patient then received hyperfractionated radiotherapy and a final MRI was done at the end of the treatment. Results. Sixty percent of the patients had partial response, 30% complete response after two courses, and 60% of CR after four courses. Supratentorial and infratentorial tumors had a good response to chemotherapy. Brainstem tumors had an initial response after two courses and then increased in size. AA was the tumor with the greatest reduction of residual tumor after treatment. Overall and free survival at 53 months was 70%. To date, three patients have died secondary to tumoral progression. There have been no relapses in the seven patients with a CR. Conclusions. Postoperative chemotherapy with ICE reduces the tumor size and increases the survival rate of pediatric patients with malignant astrocytomas with minimal toxicity. Brainstem responded poorly to treatment. © 2000 IMSS. Published by Elsevier Science Inc. Key Words: Astrocytoma, Glioblastoma multiforme, Ifosfamide, Carboplatin, Etoposide.
Introduction Address reprint requests to: Dr. Enrique López Aguilar, Hospital de Pediatría, Centro Médico Nacional Siglo XXI, IMSS, Av. Cuauhtémoc 330, Col. Doctores, CP 06720 México, D.F., Mexico. Tel.: (⫹525) 627-6900, ext. 3525; FAX: (⫹525) 761-2594.
Tumors of the central nervous system (CNS) constitute the largest group of solid neoplasms in children, and are the second cause of cancer in this group age (1). Astrocytomas are the most common form of primary intracranial tumor.
0188-4409/00 $–see front matter. Copyright © 2000 IMSS. Published by Elsevier Science Inc. PII S0188-4409(00)00 0 5 1 - 5
López-Aguilar et al. / Archives of Medical Research 31 (2000) 186–190
However, survival of malignant tumors has changed little from survival reported in the early 1970s. In combined series of adult and pediatric patients, the median survival rarely exceeds 12 months, especially in high-grade astrocytomas such as anaplastic astrocytomas and glioblastoma multiforme, in which median survival duration is 40–52 weeks (2– 5). Trials conducted by the International Society of Pediatric Oncology and the Children’s Cancer Group in the late 1970s and early 1980s showed a possible benefit in survival for patients treated with adjuvant chemotherapy. Favorable results were reported from a pilot study using the eight-drugs-in1-day therapy for children with newly diagnosed brain tumors, including high-grade astrocytomas (6). Nevertheless, responses to chemotherapy have not had significant impact on the outcome of these tumors. Although the use of preirradiation or neoadjuvant chemotherapy offers an opportunity to evaluate the potential benefit of new drug combinations, few such studies have been performed in children. Limited data are available from contemporary reports of single-institution experiences with malignant gliomas. These series are made up of patients diagnosed over periods exceeding 20 years. Al-Mefty analyzed a group of children with newly diagnosed high-grade gliomas, 80% of whom were treated with a variety of neoadjuvant chemotherapy regimens, and compared these results with those of other single- and multiinstitutional pediatric studies, with a discrete increment of survival in the group treated with neoadjuvant regimens (7). To improve the survival of these patients, we have initiated a phase II study with preirradiation ifosfamide, carboplatin, and etoposide (ICE) for the treatment of these malignant tumors. Each drug has been used before as a single agent, or in combination, for the treatment of recurrent brain tumors and in phase II studies of childhood brain tumors with different response rates (8–10), but the combination of these three drugs has never been attempted for the treatment of childhood brain tumors. The most serious limitation of brain tumor chemotherapy is the blood brain barrier (BBB), which restricts the passage of chemotherapy agents. However, the presence of the BBB is not absolute. It is usually disrupted in most brain tumors and does not function normally in the area surrounding the tumor (11–14). We report on a prospective phase II trial of preirradiation ifosfamide, carboplatin, and etoposide in newly diagnosed patients with high-grade astrocytomas. Our objectives include assessment of tumor response rate to ICE before irradiation, the assessment of acute and subacute toxicity of preirradiation ICE, and the determination of disease control interval and survival following combined modality therapy.
187
We included 11 consecutively admitted children who had a newly diagnosed high-grade astrocytoma (AA or GM) who were operated on at our hospital, and whose diagnosis was histologically confirmed by a pathologist (kappa 0.68) using the Kernohan classification system. All had measurable residual disease after initial surgery. Informed consent was obtained from the children’s parents before enrollment and the study was approved by the local Ethics and Investigation Committees (AB/06/95). Initial evaluation of disease and inclusion in the study was carried out within the first postoperative week and included magnetic resonance image (MRI) of the brain (performed 72 h after surgery), lumbar CSF cytology, complete blood cell count including platelets, and renal function evaluation. After initial surgery, every patient received chemotherapy with carboplatin (400 mg/m2/day on day 1, diluted in a 0.9% NaCl solution for 1 h) followed by ifosfamide (2 g/m2 on days 1, 2, and 3, diluted in a 0.9% NaCl solution for 1 h) and etoposide (100/mg/m2/day on days 1, 2, and 3 diluted in a 0.9% NaCl solution for 1 h), every 3 weeks for four courses. Each patient also received antiemetics (ondasetron 8 mg/m2/dose every 8 h) initiating simultaneously with chemotherapy treatment until 24 h after terminating chemotherapy, hyperhydration (3,000 mL/m2/day dextrose 5% 2:1 normal saline), and a bladder protector (MESNA 500 mg/m2 every 6 h on days 1, 2, and 3, given simultaneously with ifosfamide). Twenty-eight days after finishing the fourth course of chemotherapy, they all underwent hyperfractionated radiotherapy (total dose of 54 Gy, given in two fractions of 1.1 Gy per day during 25 days). Follow-up studies included MRI after the second and last courses of chemotherapy and after 4 weeks from completing radiotherapy (RT). To evaluate toxicity, renal function tests and a complete blood cell count were done before each course. Response to ICE was graded on the basis of neuroimaging findings by a radiologist unaware of the treatment given to the patient (kappa 0.65), comparing measurable disease in the primary site after surgery and after two and four courses of chemotherapy as well as after radiotherapy. Response was categorized as follows: complete response (CR) was complete resolution of all radiographic evidence of tumor; partial response (PR) was at least a 50% reduction of the product of maximal perpendicular diameters at the plane of greatest tumor area by MRI with stable or improving neurologic function; stable disease (SD) was less than a 50% reduction in tumor area or no more than a 25% increase with stable or improving neurologic function, and progressive disease (PD) was more than a 25% increase in tumor area. Statistical analyses included measures of middle tendency and spread, percentages, Student’s t test, ANOVA, and actuarial survival analysis.
Patients and Treatment The study was conducted at the Hospital de Pediatría, Centro Médico Nacional Siglo XXI in Mexico City between December 1994 and October 1999.
Results Eleven patients were initially included. One abandoned treatment before completing two courses of chemotherapy
188
López-Aguilar et al./ Archives of Medical Research 31 (2000) 186–190
prior to evaluation of response to treatment and was excluded from the study. Mean age was 10.4 years (range 4– 14) with a male-female relation of 2.3:1 (Table 1). According to the site of initial presentation, six tumors (60%) were supratentorial, two (20%) infratentorial, and two (20%) in the brainstem. According to the histology, seven patients (70%) had anaplastic astrocytoma and three (30%), glioblastoma multiforme. After two courses of chemotherapy, 60% of patients had partial response and 30%, complete response. After four courses of chemotherapy, six patients were categorized as CR (60%), two as PR (20%), and two as PD (20%). Following radiotherapy, one additional patient achieved CR (Table 1). According to the tumor site, tumor reduction following four courses of chemotherapy was 80% and 100% for supratentorial and infratentorial tumors, respectively. Brainstem tumors had an initial response to treatment but involved progressive disease after four courses. According to the histology, patients with anaplastic astrocytomas presented a reduction of 62% after two courses, 75% after four courses, and 81% after radiotherapy (p ⫽ 0.02). Patients with glioblastoma multiforme had tumor burden reductions of 24, 24, and 27%, respectively (p ⫽ 0.28). Overall and disease-free survival was 70% at 53 months after diagnosis (Figure 1). According to histopathology, survival for anaplastic astrocytomas was 68% at 53 months and 66% at 47 months for glioblastoma multiforme. According to tumor location, there was 100% survival at 53 months for infratentorial tumors, 66% for supratentorial tumors, and 0% at 11 months for brainstem tumors. There was minimal toxicity during the 40 courses of chemotherapy. Hemoglobin values were within normal range in 97% of the courses and only one patient required hemotransfusion. Grade 3 neutropenia occurred in 10% of the
courses, and no patients required hospitalization for neutropenia and fever. Platelets and creatinine had no significant toxicity during the 40 courses of chemotherapy. Three patients have died to date. One was a patient with a supratentorial tumor (patient no. 1), who had PR to four courses of chemotherapy and PD during radiotherapy, and died 6 months after diagnosis. Another patient (patient no. 4) had a brainstem tumor with PD after four courses of chemotherapy and died 11 months after diagnosis. The last patient (patient no. 8), who also had a brain stem tumor, had no response to chemotherapy, presented tumor progression after radiotherapy, and died 8 months after diagnosis. Six patients had CR after four courses of chemotherapy and one more after radiotherapy. They are considered cured at present and lead normal lives for their age (they attend school, are independent from their parents, and have normal neuropsychological evaluations). At present, all have been in follow-up for more than 3 years without relapses or progression.
Discussion Recent decades have been characterized by significant changes in the survival of cancer patients. Nevertheless, CNS tumors, especially high-grade astrocytomas, have seen fewer relevant advances because fewer new treatment strategies have been evaluated. Surgery is frequently unable to reduce the tumor burden due to the great infiltrative capacity of these tumors over the neural tissue. Traditional treatment with postoperative radiotherapy has not been able to significantly increase the survival rate of these patients in that intratumoral isodoses are frequently sublethal to neoplasic cells because of the great tumor burden that usually persists after surgery. Previous ex-
Table 1. Patient characteristics and response to ICE protocol
Patient 1 2 3 4 5 6 7 8 9 10 p value
Age (years)
Sex
Tumor location
Histology
Tumor volume following surgery
13 11 14 8 12 11 4 12 14 5
Male Male Female Male Female Male Male Male Male Female
Supratentorial Supratentorial Infratentorial Brainstem Supratentorial Supratentorial Supratentorial Brainstem Infratentorial Supratentorial
AA AA AA AA GM AA AA GM AA GM
37.9 38.5 8.0 20.0 3.0 14.6 3.6 18.02 4.40 1.69
Response and tumor volume after two courses of chemotherapy
Response and tumor volume after four courses of chemotherapy
Response and tumor volume post Rt
Present status (no. of months of follow-up)
15.6 (PR) 9.6 (PR) 4 (PR) 5.6 (PR) 0.8 (PR) 6.4 (PR) 0 (CR) 18.5 (PD) 0 (CR) 0 (CR) 0.01
7.5 (PR) 7 (PR) 0 (CR) 10.8 (PD) 0 (CR) 0 (CR) 0 (CR) 18.5 (PD) 0 (CR) 0 (CR) 0.006
D 0 (CR) 0 (CR) 15.4 (PD) 0 (CR) 0 (CR) 0 (CR) 18.5 (PD) 0 (CR) 0 (CR) 0.01
D DF (53) DF (53) D DF (47) DF (44) DF (41) D DF (40) DF (39)
Note: Tumor volume is measured as cm2. AA: Anaplastic astrocytoma; D: death; GM: glioblastoma multiforme; DF: disease-free; PR: partial response; CR: complete response; PD: progressive disease.
López-Aguilar et al. / Archives of Medical Research 31 (2000) 186–190
Figure 1. Overall and disease-free survival at 48 months of patients with malignant astrocytomas treated with chemotherapy with ICE.
periences with these patients show that the effect of RT administered after surgery did not completely control the residual tumor and that the tumor presented greater progression. This was reflected by the poor survival rate of 26% at 12 months in our Hospital, similar to other reports (12,15–17). The role of chemotherapy in malignant gliomas remains uncertain. Although meta-analysis of patients treated with various chemotherapeutic regimens suggests that such treatments may be associated with little improvement in survival, multiple problems in the design and execution of such large retrospective analyses render acceptance of conclusions difficult. Finlay et al. (18) demonstrated that survival of patients with grades III and IV astrocytomas treated with adjuvant chemotherapy with vincristine, lomustine, and prednisone was 28 and 16% at 5 years, respectively. While this regimen was suggested to improve the survival of patients with newly diagnosed high-grade astrocytomas, the number of patients was small and close examination reveals that the results rest only on a few long-term survivors. However, for many years it was considered that adjuvant chemotherapy was of little benefit in the treatment of these patients basically because it was thought that few anti-neoplastic agents could pass the blood brain barrier. These agents, it was believed, had to be liposoluble in order to reach the tumor. Also, the cytotoxic effect initially found in cell cultures was minimal. Only nitrosoureas demonstrated some effect in the treatment of these tumors, the reason that most chemotherapy schemes use nitrosoureas. Recently, it has been proven that the blood brain barrier is dysfunctional in areas surrounding the tumor, and that not only liposoluble agents can reach the tumor (11). Kovnar et al. (19) have adopted a policy of using neoadjuvant treatment in most newly diagnosed patients. In the current series, 80% of patients received preirradiation chemotherapy. The regimen most commonly used, cisplatin and etoposide, has been previously reported by these authors and was associated with stable disease in seven and objective responses in only two of 12 patients. The regimen with high-
189
dose cyclophosphamide and thiotepa has also been previously reported (20). There are to date limited data from other investigations testing high-dose chemotherapy in the 20 preirradiation setting. These data, however, do not show better responses than other regimens with adjuvant chemotherapy. A concern associated with the use of neoadjuvant chemotherapy is that the inherent delay in the initiation of radiation therapy may have an adverse impact on survival. Despite the low rate of objective response to neoadjuvant chemotherapy reported by Heideman (20), the median survival is similar to that reported in other series of patients (21,22). This suggests that the delay in radiation therapy associated with neoadjuvant chemotherapy may not significantly alter the survival of children with malignant gliomas, and supports the use of such studies to identify effective regimens in this generally refractory group of tumors (23). In this study, we administered the following new scheme of chemotherapy that includes two agents that have proven a cytotoxic effect in vitro over these tumor cells: ifosfamide and carboplatin, and etoposide, which enhances this latter drug’s effect. We changed the sequence of treatment (RT– QT to QT–RT) to attempt to decrease the residual tumor burden after surgery. With this new approach, we found a reduction in tumor size in most patients (60%), with a mean of 55% after two courses, 66% after four courses, and 72% after RT. We found that the histology of the tumor, as well as its location, influences response to treatment. Infratentorial tumors and anaplastic astrocytomas had a good response in contrast to multiform glioblastoma and brainstem tumors, which had little or no response to chemotherapy or radiotherapy. We consider this phenomena to be mainly due to the almost impossible surgical access of these tumors, and also probably to a reduced concentration of the drugs inside the tumor caused by poor brainstem blood flow. No other factors were found to influence the tumor’s response to treatment. Nevertheless, we think that final conclusions cannot be made due to the small amount of patients in each group. With this new sequence, we found a survival rate of 70% for all tumors, significantly greater than that previously reported. Patients with infratentorial tumors have a survival of 100%, but the sample size is too small to draw conclusions. Even though we found a very poor survival rate for patients with brainstem tumors, we found an improvement in their quality of life during treatment. There have been no relapses during follow-up. Finally, it is important to mention that minimal toxicity was noted with this treatment, and it was found to be greatly inferior to that observed with chemotherapy with nitrosoureas. The number of patients in this report is small, especially after the report is stratified by histology or location, and further studies must be conducted to confirm our results. Additionally, this new chemotherapy regimen must be compared with traditional treatment in a phase III study.
190
López-Aguilar et al./ Archives of Medical Research 31 (2000) 186–190
The results we present are very encouraging and are, as far as we are aware, the best results reported to date in a cohort of malignant glioma patients treated with chemotherapy. We expect to increase the sample size to make final conclusions and to subsequently improve this new approach.
11. 12.
13.
References 1. Silverberg EE. Cancer statistics. Cancer 1981;31:13. 2. Dinapoli RP, Brown LD, Arusell RM. A phase III comparative evaluation of BCNU and carmustine with radiation therapy for high grade gliomas. J Clin Oncol 1993;11:1316. 3. Ganju MB, Jenkins RB, O’Fallon JR, Scheitauer BW, Ransom DT, Katsmann JA. Prognostic factors in gliomas. Cancer 1994;74(3):920. 4. Burger PC, Green SB. Patient age, histologic features and length of survival in patients with glio-blastoma multiforme. Cancer 1987;59:1617. 5. Marchese MJ, Chang CH. Malignant astrocytic gliomas in children. Cancer 1990;65:2771. 6. Pendergrass TW, Milstein JM, Geyer IR. Eight drugs in one day chemotherapy for brain tumors. J Clin Oncol 1987;5(8):1221. 7. Al-Mefty O, Al-Rodhaun HF. Factors affecting survival of children with malignant gliomas. Neurosurgery 1987;20:416. 8. Gentet JC, Doz F, Bouffet E, Plantaz D, Rochè H, Tron P, Kalifa CH, et al. Carboplatin and VP-16 in medulloblastoma: a phase II study of the French Society of Pediatric Oncology ( SFOP ). Med Pediatr Oncol 1994;23:422. 9. Chastagner P, Sommelet OD, Kalifa CH, Brunat MM, Zucker JM, Domeocq F. Phase II study of ifosfamide in childhood brain tumors. Med Pediatr Oncol 1993;21:49. 10. Kovnar EH, Kellie SJ, Horowitz ME, Sanford RA, Longston RK, Mulhern JJ, Jenkins EC, et al. Preirradiation cisplatin and etoposide in the
14. 15.
16. 17.
18.
19. 20.
21. 22. 23.
treatment of high risk medulloblastoma and other malignant embryonal tumors of the central nervous system. J Clin Oncol 1990;8:330. Long DM. Capillary ultrastructure and the blood brain barrier in human malignant brain tumors. J Neurosurg 1970;32:127. Neuwelt EA, Howieson J, Frenkel EP. Therapeutic efficacy of chemotherapy with drug delivery enhancement by blood brain barrier modification in glioblastoma. Neurosurgery 1986;19:573. Levin VA, Freeman-Dove M, Landahl HL. Permeability characteristics of brain adjacent to tumors in rats. Arch Neurol 1975;32:785. Hirano HA, Matsui T. Vascular structures in brain tumors. Hum Pathol 1975;6:611. Grossman SA, Wharam M, Sheidler V. BCNU and cisplatin followed by radiation in poor prognosis patients with high grade astrocytomas. (Abstract). Proc Am Soc Clin Oncol 1992;11:149. Levin VA. Chemotherapy of primary brain tumors. Neurol Clin 1985;3:855. Fine HA, Dear KB, Loeffler JS. Meta-analysis of radiation therapy with and without adjuvant chemotherapy for malignant gliomas in adults. Cancer 1973;71:2585. Finlay JL, Boyett JM, Yates AJ. Randomized phase III trial in childhood high grade astrocytomas comparing vincristine, lomustine and prednisone with the eight drugs in one day regimen. J Clin Oncol 1995;13:112. Kovnar E, Horowitz M, Kun L. Carboplatin and etoposide in the treatment of childhood malignant glioma. Neurology 1991;41:410. Heideman RL, Douglas EC, Krance RA. High dose chemotherapy and autologus bone marrow rescue in newly diagnosed pediatric malignant gliomas. J Clin Oncol 1993;11:1458. Phuphanich S, Edwards M, Levin V. Supratentorial malignant gliomas in childhood. Ann Neurol 1987;22:355. Kleihues P, Burger PC, Scheithorver BW. The new WHO classification of brain tumors. Brain Pathol 1993;3:255. Heideman RL, Kuttesch J, Gajjar AJ, Walter AW, Jenkins JJ, Li Yulan, Sanford RA, Kun LE. Supratentorial malignant gliomas in childhood. Cancer 1997;80(3):497.