|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Temozolomide in Malignant Gliomas of Childhood: A United Kingdom Children’s Cancer Study Group and French Society for Pediatric Oncology Intergroup StudyByFrom Christie National Health Service Trust, Manchester; Queens Medical Centre, Nottingham; Royal Hallamshire Hospital, Sheffield; Western General Hospital, Edinburgh; United Kingdom Children’s Cancer Study Group Data Centre, Leicester; and Royal Victoria Infirmary, Newcastle upon Tyne, United Kingdom; Centre Leon Berard and Hopital Neurologique, Lyon; Institut Gustav Roussy, Villejuif; and Institut Curie, Paris, France; and Schering-Plough, Kenilworth, NJ. Address reprint requests to L.S. Lashford, MD, Translational Research and Experimental Therapeutics, The Cancer Research Campaign, 10 Cambridge Terr, London NW1 4JL, United Kingdom; email: llashford{at}crc.org.uk
PURPOSE: To determine the response rate of the malignant gliomas of childhood to an oral, daily schedule of temozolomide. PATIENTS AND METHODS: A multicenter, phase II evaluation of an oral, daily schedule of temozolomide (200 mg/m2 on 5 consecutive days) was undertaken in children with relapsed or progressive, biopsy-proven, high-grade glioma (arm A) and progressive, diffuse, intrinsic brainstem glioma (arm B). Evidence of activity was defined by radiologic evidence of a sustained reduction in tumor size on serial magnetic resonance imaging scans. RESULTS: Fifty-five patients were recruited (34 to arm A and 21 to arm B) and received 215 cycles of chemotherapy. Grade 3/4 thrombocytopenia was the most frequent toxic event (7% of cycles). Prolonged myelosuppression resulted in significant treatment delays and dose reductions (17% and 22% of cycles, respectively). Two toxic deaths were documented and were related to myelosuppression and sepsis in one patient and pneumonia in a second. The overall (best) response rate was 12% for arm A (95% confidence interval [CI], 3 to 28 in the study cohort, and 2 to 31 for eligible patients) and 5% and 6%, respectively, for arm B (95% CI, 0 to 26 in the study cohort, and 0 to 27 for eligible patients). Stabilization of disease was also documented and was most noteworthy for brainstem gliomas, where two patients achieved both radiologic static disease and discontinued steroid medication. CONCLUSION: Despite moderate toxicity, objective response rates to temozolomide have been low, indicating that temozolomide has minimal activity in the high-grade gliomas of childhood.
MALIGNANT GLIOMAS occur relatively infrequently in childhood and account for approximately 5% of new cases of childhood malignancy each year. These tumors are classified primarily by anatomic location (brainstem or other) and secondarily by histologic phenotype. A distinction by anatomic site has arisen because although a full spectrum of glioma phenotypes are seen in brainstem tumors, brainstem location, radiologic appearances, and clinical features are more reliable prognostic factors than histologic appearance or grade.1-3 Eighty percent of patients with a brainstem tumor will present with rapidly progressing, cerebellar, cranial nerve or long-tract signs and radiologic features of a diffuse expanding lesion centered on the pons. The clinical features rarely mimic other conditions and, given the lack of concordance between histology and outcome, patients presenting with classical features no longer routinely undergo biopsy.4 Treatment with radiotherapy is largely palliative and produces a median survival of 8 to 10 months.5-7 Although the number of phase II and phase III studies of chemotherapy in this condition is relatively small, no drug has been identified that alters the ultimately poor outcome. Conversely, the clinical behavior of supratentorial and cerebellar gliomas is much more difficult to predict on the basis of radiologic and clinical characteristics, and prognosis is more closely related to histologic phenotype and grade. Studies in adult patients clearly demonstrate that lower grade and presence of oligodendroglial elements predict for better prognosis,8 and although there are few clinical studies in pediatric tumors, they also support a view that grade is an important prognostic feature.9,10 After an initial, extensive resection, radiotherapy is widely accepted as first-line treatment for the majority of high-grade gliomas in both pediatric and adult practice. However, despite radical treatment, long-term survival rates remain poor (30% to 40% overall survival for grade 3 and 10% for grade 4 tumors), and although there is evidence that a range of chemotherapeutic agents have modest activity in high-grade gliomas, chemotherapy has failed to make a major impact on outcome.11 Thus, clinical research continues to concentrate on the investigation of new agents, with the monofunctional alkylating agent temozolomide (Temodal; Schering-Plough, Kenilworth, NJ) exciting most recent interest. Temozolomide is a prodrug that is rapidly degraded at physiologic pH to a reactive and cytotoxic methylating derivative, monomethyl 5-triazeno imidazole carboxamide. Principal cytotoxic lesions involve the O6 and to a lesser extent N7 alkylation of guanine with subsequent aberrant repair of the methyl adduct.12 The drug is stable at low pH, is completely absorbed after oral administration, and rapidly penetrates the CNS,13 making it an ideal candidate for study in patients with CNS tumors. Preclinical studies supported further evaluation in CNS tumors, as both in vitro and in vivo xenograft studies indicated that the drug might be active in glioma and ependymoma.14,15 The first clinical studies in adult patients were undertaken under the auspices of the Cancer Research Campaign (CRC), and in pediatric patients by the United Kingdom Children’s Cancer Study Group (UKCCSG). The dose-limiting toxicity in both adults and children was myelosuppression, which occurred at a total dose of 1.25 g/m2 fractionated over 5 days.16,17 Both phase I studies suggested some activity against relapsed high-grade gliomas, and consequently malignant gliomas were targeted for further study. The CRC rapidly undertook a phase II study of temozolomide in adult patients with recurrent or progressive high-grade glioma,18 and as soon as the safety profile of temozolomide had been established for pediatric patients, a multicenter, intergroup, phase II study was initiated in relapsed malignant gliomas of childhood. The objective of the study was to determine the efficacy of temozolomide, administered orally once a day for 5 days in patients with biopsy-proven, high-grade, supratentorial or cerebellar high-grade gliomas that had recurred after conventional treatment (arm A), and in diffuse intrinsic brainstem gliomas progressing after radiotherapy (arm B).
Eligibility and Recruitment Patients aged between 3 and 18 years were eligible for inclusion in the study, provided they had clear evidence of relapsed high-grade glioma (World Health Organization [WHO] grade III/IV, designated arm A); or progressive, diffuse, intrinsic brainstem glioma after prior standard radiotherapy (designated arm B). Eligibility was verified by an independent panel of reviewing pathologists (arm A), or in the case of arm B by a panel of neurologic and pediatric radiologists. On arm A, definition of relapse centered on unequivocal evidence of the appearance of one or more new lesions at sites consistent with the natural history of the disease, or by disease progression in a pre-existing, previously treated lesion. Progression was documented by evidence of tumor growth on two serial, gadolinium-enhanced magnetic resonance imaging (Gd-MRI) scans. As the correlation between disease progression and radiologic changes in the brainstem are more difficult to interpret, progressive disease at this site was defined by sustained neurologic deterioration at 8 weeks after completion of radiation therapy, or by any subsequent neurologic deterioration. The primary aim of the study was to determine an objective response rate to temozolomide. Consequently, important additional eligibility criteria included the presence of measurable disease. For the purpose of the study, this was defined as an enhancing, bidimensional lesion documented on Gd-MRI. Minimum size criteria were set at 8 x 8 mm, and to avoid compounding effects of previous treatment on assessment of response, all patients were to have completed radiotherapy at least 6 weeks previously, and in the case of chemotherapy by at least 4 weeks before entering the study. Where patients had received a nitrosourea as part of their last chemotherapy regimen, the interval between last treatment and study entry was extended to 6 weeks. Response was determined for arms A and B, both by a reduction in size of measurable lesions and by durability of that reduction on serial scans. Hence, complete response (CR) consisted of independently verified, complete disappearance of all detectable disease, determined by two observations not less than 4 weeks apart; partial response (PR) was determined by a decrease of at least 50% of the sum of the products of the largest perpendicular diameters over the same time span; and static disease (SD) was determined by either an intermediate reduction in tumor size (ie, less than 50%) or a small increase in size (less than 25%). Progressive disease was determined by at least a 25% increase in tumor size or appearance of a new lesion. To ensure that study objectives could be met and to safeguard the interests of this vulnerable study population, all patients were expected to survive at least 9 weeks from study entry and to have recovered from the toxic effects of previous therapy. These criteria were reinforced by a study requirement that patients were on a stable dose of steroids for at least 7 days before entering the study and had an adequate performance status as determined by the Karnofsky or Lansky play scale.19,20 An adequate performance status (PS) was defined by a PS of 2 for patients on arm A and a PS of 3 for patients on arm B. Exceptions were made for arm B if the sole reason for allocation to a PS of 4 was immobilization and use of a wheelchair. Written, informed consent was obtained from the child’s legal guardian or parent, and evidence of assent was obtained from the child. All subjects were entered subject to ethical review and approval of the protocol according to national and local practice.
Study Conduct A Gehan21 two-step model was used in which the lowest limit of activity considered to be of clinical interest was set at 20%. The probability of rejecting a treatment with a response rate of at least 20% was determined to be less than 5%.
Data Analysis
Study Population Fifty-five children with relapsed malignant glioma were registered onto the two arms of the study, 34 on arm A (high-grade astrocytoma [HGA]) and 21 on arm B (brainstem glioma [BSG]) (Table 1). These patients consisted of 25 males and 30 females spanning an age range from 4.2 to 17.5 years. All patients had received radiotherapy as part of their initial treatment plan. This consisted of a median tumor dose of 54 Gy (range, 45 to 72.4 Gy), with one patient who had a biopsy-proven, grade 3, cerebellar primary tumor receiving 50 Gy with supplementary craniospinal irradiation. All but one patient received conventionally fractionated radiotherapy, with one individual with a BSG receiving hyperfractionated radiotherapy at a tumor dose of 72.4 Gy. Twenty-three individuals also received adjuvant chemotherapy at the time of initial treatment. The type of chemotherapy varied, reflecting both local practice and participation in national clinical trials (Table 1).
Fifty-three of 55 patients completed their first 5 days of temozolomide. After central neuropathologic review, eight patients were excluded from arm A on the basis of grade (n = 3), uncertainty surrounding initial diagnosis (n = 2), or inadequate material for review (n = 3). Two patients were also excluded from arm B—a patient who on radiologic review was judged to have a tumor centered on the cerebellum and in whom no material was available for subsequent pathology review, and a second in whom the MRI scans were deemed inadequate for assessment of response. A single patient treated on arm A was subsequently judged to have a WHO grade III pontine tumor and was reassigned to arm B for analysis of response. Thus, 43 (81%) of 53 patients were considered eligible for an assessment of tumor response after review of major eligibility criteria (histopathology and imaging criteria) (Table 1).
Toxicity Despite the relatively low incidence of severe hematologic toxicity, prolonged thrombocytopenia and neutropenia resulted in both delays to treatment and dose reductions. Thus, 22% of cycles were administered at reduced dose and 17% were delayed by 7 days or more (median delay, 10 days; range, 7 to 30 days). Dose reduction occurred relatively early in treatment prescription, with 23% of second cycles administered at a lower dose.
Neurologic Deterioration During Initial Dosing
Efficacy
All 25 patients were originally treated with curative intent and had received age-appropriate radiotherapy doses (median dose, 54 Gy; range, 39.5 to 60 Gy). Eleven patients received chemotherapy as part of their initial treatment, six of whom received a lomustine-containing regimen, and one of whom had received fotemustine. Time from initial diagnosis to study entry was short, with a median time from diagnosis to starting temozolomide of 9.5 months (range, 4 to 43 months). At study entry, seven of these patients had evidence of multifocal relapse and three had a PS of 3. The radiologic review panel found it difficult to accurately identify tumor margins in eight patients (32%). With this caveat, only one patient achieved a PR at course 2. Two additional patients who were judged to have SD at course 2 were late responders (courses 3 and 4). The responses occurred in two patients with unequivocal glioblastoma multiforme (patient nos. 204 and 228), and in an additional patient with a high-grade astrocytic tumor (patient no. 221). Thus, the initial response rate was 4% and the overall response rate was 12% (95% confidence interval [CI], 2% to 31%) (Table 3). The interval between initial diagnosis and study entry for these three responders was 4, 8, and 10 months. None of these patients had previously been treated with a lomustine- or procarbazine-containing regimen. All responses were durable, lasting for 7, 8, and 24+ months, with patient 221 alive and in good health after 2 years of treatment.
An additional three patients achieved a transient stabilization of disease during their first three cycles of treatment. Disease stabilization did not result in steroid withdrawal in two of these patients and was not commented on in the third. Disease progressed in all three patients after three courses of treatment. Arm A: registered patients. As 21% of patients were withdrawn from arm A after review of the eligibility criteria, response rates were also calculated for the total cohort. After this analysis, the response rate did not materially change (response rate, 12%; 95% CI, 3% to 28%) (Table 3). One additional response occurred in a well-defined temporal lobe tumor that had proved difficult to classify, both by the referring pathologist and on subsequent review. Median survival for arm A (eligible patients) was 4.7 months (95% CI, 2.8 to 9.2 months) and 4.8 months for the full cohort (95% CI, 3.7 to 7.5 months). Arm B: brainstem tumors (eligible patients). Eighteen patients, including one patient with a pontine tumor initially treated on arm A, completed one course of temozolomide and are thus assessable for response (Table 3). Fifteen are known to be centered on the pons and one in the midbrain (site unrecorded, n = 2). The median interval from initial diagnosis to study entry was slightly longer than usual for diffuse intrinsic pontine tumors (10 months), and for two patients it was notably prolonged (7.6 and 2.2 years). However, both these patients had diffuse, heterogenously enhancing tumors at the time of clinical progression. Performance status was adequate in the majority, with only six individuals recording a Lansky play score of 3 or less. There were no early responders; however, one patient who initially had static disease achieved a PR after seven courses of temozolomide. The patient went on to receive 12 courses of treatment, steroid support was withdrawn, and the patient survived for a period of 17 months before death from tumor progression. This patient was not unusual in terms of previous history, with an interval of 4 months from diagnosis to study entry. The individual had received an age-appropriate radiation dose to the original tumor volume (54 Gy) but had not received any previous chemotherapy. Imaging characteristics at progression indicated a well-defined lesion involving the pons with a small exophytic component. Stabilization of disease occurred in an additional three patients and steroid use was withdrawn in two. Duration of stabilization was 4, 6, and 38+ months, the latter prolonged remission occurring in a midbrain tumor. Thus, although there are no early responses, the overall (best) response rate is 6% (95% CI, 0% to 27%). If disease stabilization is taken as evidence of response, then the combined PR and SD rate is 22% (95% CI, 6% to 48%). Arm B: registered patients. The results are similar if recalculated for the full cohort (Table 3). Median survival was 6.2 months for arm B eligible patients and 6.4 months for the full cohort.
Survival by Site
Malignant gliomas in childhood share some clinical similarities with their adult counterparts including similar morphology, poor outcome, and relative chemoresistance and radioresistance. Differences include a higher proportion of brainstem tumors and some preliminary data that young children with high-grade tumors at other sites may be chemotherapy curable.22 Notwithstanding the possible influence of young age on outcome, the majority of tumors carry a poor prognosis, which varies with site and grade.23 A poor prognosis can be inferred from radiologic characteristics in BSGs, such that diffuse intrinsic pontine gliomas have an overall survival of approximately 10% to 15%. Brainstem tumors with focal, cystic, or exophytic characteristics carry a better prognosis, as do tumors centered in the midbrain or in the cervicomedullary region.24,25 Although the outcome for supratentorial and cerebellar tumors more closely correlates with histopathologic subtype and grade, the evidence for certain histologic variants such as malignant oligodendrogliomas having a different natural history than anaplastic astrocytomas is lacking in childhood. This study took account of the usual clinical practice of defining a subset of brainstem tumors by radiologic characteristics alone, and thus stratified the patient population into two groups: diffuse, intrinsic BSGs (radiologic diagnosis) and biopsy-proven high-grade malignancy (WHO grade III/IV) arising at other sites. Although there was an expectation that midbrain tumors would be biopsied and assigned to arm A, this was not explicit in the study design. Thus, arm B contains one patient with a diffuse, midbrain tumor. A total of 55 patients received 215 cycles of chemotherapy, and the data significantly extend our knowledge of the safety profile of this drug in childhood malignancy. The data confirm hematologic toxicity as the principal complication of drug administration and was implicated on two occasions with a toxic death. Both of these occurred in patients with brainstem tumors and low PS. In one, the timing of the pancytopenia is atypical and, although attributed to a possible complication of drug exposure, is not adequately explained. The dose and schedule was otherwise well tolerated, but did result in frequent treatment delays and does reductions. These findings differ from experience of temozolomide toxicity in the adult population, where patients experience less toxicity. There are two factors that probably account for the difference. First, adult patients commonly begin therapy with a lower starting dose if they have had any previous chemotherapy. Second, the pharmacokinetic profiles undertaken in the United Kingdom pediatric phase I study indicated significantly grater drug exposure after oral dosing in childhood, with pediatric patients achieving both higher risk peak plasma levels of drug and approximately 40% greater systemic drug exposure as determined by estimates of area under the curve.17 Thus, despite dosing at the phase I maximum-tolerated dose, greater experience with the compound in childhood indicates that a lower starting does may be necessary for many patients, particularly if temozolomide is to be used in combination with other myelosuppressive drugs. Care was taken to scrutinize the data for neurologic adverse events during initial dosing. The reason for undertaking this analysis was the early report by O’Reilly et al18 of neurologic deterioration in two of 28 patients with large intracranial tumors. The neurologic deterioration was attributed to drug-induced peritumoral edema in patients who had large hemispheric tumors and significantly raised intracranial pressure. In this study, neurologic deterioration occurred in 22% of patients and was documented with equivalent frequency on both arms of the study. With one exception, patients had a predrug PS of 0 to 2. In two patients, clinical deterioration resulted in death. Of course, the temporal relationship between dosing and onset of neurologic symptoms does not necessarily indicate a causal effect and, in two patients and one death, was clearly attributable to disease progression. However, given the requirement for stable neurologic status and adequate PS before study, the potential for further neurologic deterioration should be noted as a possible complication of using this drug and should be taken into account, particularly if considering using this compound in a palliative setting. Given that supratentorial and cerebellar HGA are relatively rare in childhood (and thus there were small numbers of each pathologic subtype in the study), no attempt was made to further define response rates by grade or morphology. However, a rigorous central neuropathologic review of diagnostic material was undertaken to ensure that responses rates were reported only for malignant glioma. In 21% of cases, central review could not confirm a diagnosis of HGA. Reasons were relatively uniformly divided between grade, morphology, and inadequate provision of material for review. The overall rate of attrition is similar to that observed after central review of patient material in a recent large multicenter study investigating the effects of temozolomide in adults with anaplastic gliomas.26 Thirty-two percent of registered patients were ineligible on review, either because of morphologic criteria (half of ineligible patients), grade (37%), or absence of material (13%). Table 2 illustrates the degree of interobserver variation between experienced neuropathologists, and is important given the prognostic significance of oligodendroglial elements in malignant gliomas and influence on chemosensitivity. Other reports have confirmed the difficulty experienced by neuropathologists in distinguishing malignant astrocytes from oligodendrocytes and otherwise unclassifiable cells.27,28 Our results also suggest the possibility of national biases in the designation of tumor type but, given the sample size, are of no clear significance. This issue will be examined more closely in future co-operative studies and will need to take account of the accumulating data on the molecular and cytogenetic events that accompany phenotypic changes.29-31 Given both the high rate of attrition after histopathologic review and the uncertainty surrounding tumor classification, response rates were calculated on the basis of the referring center’s diagnosis (total cohort) and biopsy-proven HGG. At 12%, the overall response rate in this group (arm A) was disappointing, whether analyzed for all patients or after exclusion of ineligible patients. The results appear more promising if stabilization of tumor growth is taken as evidence of efficacy. The rate of objective response (CR + PR) plus stabilization of disease rises to 30% in registered patients and 24% in patients who fulfilled eligibility criteria. However, evidence of efficacy as judged by stabilization of disease should be interpreted with caution, as an important minority of patients registered on arm A included patients with low-grade astrocytoma and otherwise atypical tumors. Moreover, stabilization of tumor growth in patients who fulfilled major eligibility criteria was transient and not accompanied by withdrawal of steroids. These latter issues are reflected in the relatively short median survival (4.2 months) after first course of temozolomide and would be less than recently reported for adult patients entered onto eight phase II trials.32 Aggregated outcomes from these consecutive phase II studies in high-grade astrocytic tumors produced a median survival of 30 weeks. Similar results were obtained in the first CRC evaluation of the compound’s efficacy in adult gliomas, although assessments were largely determined on the basis of clinical rather than imaging criteria.33 Administration of temozolomide resulted in a median survival of 4.6 months for a population of 103 relapsing patients with a spectrum of high-grade gliomas. The objective response rate was estimated as 11%, but considerably more patients achieved disease stabilization, producing a CR + PR + SD rate of 58%. Better results were noted in a recent exploration of the activity of temozolomide in adults relapsing with anaplastic gliomas where rates for objective response and stabilization of disease were both high, reaching 35% and 29%, respectively.26 Both of these studies in adult gliomas contained substantially more patients than reported in this study and thus might be expected to give more confident estimates of the compound’s activity in the HGAs. Although the adult studies do not quote confidence intervals for response rates, an inspection of the data suggests that these will not overlap with the pediatric data, making it more likely that the lower response rates observed in this study are a true observation. BSGs are particularly uncommon in the adult population, and there have been no comparable adult studies in this tumor type. The objective response rate was similar to that documented for arm A and, at 6%, was disappointing. However, there were some prolonged periods of disease stabilization that were associated with steroid withdrawal in two patients. The most durable of these was in a patient with a recurrent midbrain tumor that had previously been documented as a grade II astrocytic tumor. As midbrain tumors are recognized to have a more indolent course,25 a survival analysis by site was performed. The median survival for relapsing patients with pontine tumors was 6.3 months and, with the caveat that the CIs on this estimate are wide (3 to 9.2 months), are slightly better than reported for 100 patients with diffuse pontine tumors who relapsed after initial hyperfractionated radiotherapy.34 There may be a number of explanations for this, including a selection bias that acted to recruit relapsing patients with a slower clinical deterioration. There is some evidence to support this hypothesis, as a median time from diagnosis to progression of 10 months was longer than is usual in diffuse brainstem tumors. Consequently, the conclusion that disease stabilization represents evidence of efficacy in brainstem tumors is premature, but is sufficiently interesting to warrant further study. Both arms suggest that the objective response rates for pediatric gliomas are lower than reported for adult patients. Explanations for lower estimates of response in the pediatric population include differences in the bioavailability of the drug between children and adults and higher levels of intrinsic or acquired resistance functions. The first of these issues has already been commented on, with the UKCCSG phase I pediatric study of temozolomide documenting both higher peak plasma levels of temozolomide and significantly higher systemic exposures of children to drug than that observed in the adult population.17 Given that the incidence of vomiting was no higher in this study, it is reasonable to assume that the apparent differences in efficacy are not because of differences in bioavailability of active compound. Three mechanisms are implicated in resistance to temozolomide. These constitute overexpression of the DNA repair protein O6-alkylguanine-DNA-alkyltransferase (AGT), which removes methyl adducts by means of a stoichiometric and autoinactivating process; and aberrations in mismatch and base excision repair pathways, which lead to a loss of the apoptotic response to cellular injury.35 The expression of AGT in gliomas is generally lower than that observed in other CNS tumors and is thought to be an important factor in the relative sensitivity of HGA to temozolomide and related agents in adult tumors.36 A similar observation has been made for pediatric tumors,37 although in one large study of 110 pediatric CNS tumors it was evident that AGT activity could vary markedly, ranging between 0.34 and 498 fmol/106 cells.38 Significant variation was observed within diagnostic groups and was also influenced by both age of the patient and tumor grade. Thus, given the probable relevance of this resistance mechanism in influencing tumor sensitivity to temozolomide, quantitation of AGT levels should be an important focus in understanding response rate to this drug. Similarly, although mismatch repair deficient glioma cell lines are reported, little is known regarding the prevalence of this phenotype in primary pediatric tumors. These issues are important not only because they help in understanding differences in apparent efficacy of temozolomide, but also because they would have significant influence on the future development of this drug. A number of strategies already exist for improving the therapeutic index of the temozolomide compound, including modulating tumor resistance. Promising approaches include AGT depletion through coadministration of compounds such as O6-benzyl guanine.39 O6-benzyl guanine competes for removal of the toxic methyl adduct produced as a result of treatment with temozolomide, and binds and inactivates AGT. Xenograft studies have shown that although this approach increases the sensitivity of normal tissues to temozolomide exposure, the strategy also increases sensitivity of glioblastomas with high levels of AGT activity to temozolomide administration.40 This strategy has been developed such that a number of investigators have also demonstrated the potential of protecting normal tissues from the anticipated increase in toxic effects.41,42 These are interesting approaches but require a better understanding of the dominant mechanism of tumor resistance to render them clinically useful. In summary, temozolomide has been investigated in two cohorts of patients with relapsing high-grade gliomas of childhood. Despite moderate toxicity, results have been disappointing, with no convincing evidence of activity in either supratentorial and cerebellar HGG or intrinsic brainstem tumors.
The United Kingdom Children’s Cancer Study Group centers are supported through National Health Service research and development funding and the Cancer Research Campaign. The UKCCSG and French Society for Paediatric Oncology gratefully acknowledge the contribution of clinical investigators and patients at the Institut Curie (Dr Francois Doz), Institut Gustave-Roussy (Dr Francois Pein); Nancy (Dr Pascal Chastagner); Marseille (Dr Jean-Claude Gentet); Lille (Dr Christine Baranzelli); Centre Leon Berard, Lyon (Dr Didier Frappaz); Toulouse (Dr Herve Rubie); Queens Medical Centre, Nottingham (Dr David Walker); Bristol Children’s Hospital (Dr Eric Bouffet); The Royal Marsden Hospital, London (Prof Ross Pinkerton); St James Hospital, Leeds (Dr Ian Lewis); Royal Victoria Infirmary, Newcastle (Prof Andy Pearson); the Christie NHS Trust, Manchester (Dr Linda Lashford); and Royal Hospital for Sick Children, Edinburgh (Dr Hamish Wallace). Further pathology review was provided by Dr Gambarelli, Hopitaux de la Timone, Marseille; Dr Lelouche-Tubiana, Hopital Necker-Enfants maladies, Paris; and Dr Ruchoux, Hopital Roger Salengro–CHRU, Lille. Data management and statistical support was provided through the UKCCSG data center, (Ann Elsworth and Claire Weston). The contribution of Dr Margaret Dugan in establishing the study is also gratefully acknowledged.
1. Barkovich AJ, Krischer J, Kun LE, et al: Brain stem gliomas: A classification based on magnetic resonance imaging. Pediatr Neurosurg 16: 73-83, 1990[Medline] 2. Kaplan AM, Albright AL, Zimmerman RA, et al: Brain stem gliomas in children. A Children’s Cancer Group review of 119 cases. Pediatr Neurosurg 24: 185-192, 1996[Medline] 3. Fischbein NJ, Prados MD, Wara W, et al: Radiologic classification of brain stem tumors: Correlation of magnetic resonance imaging appearances with clinical outcome. Pediatr Neurosurg 24: 9-23, 1996[Medline] 4. Albright AL, Packer RJ, Zimmerman R, et al: Magnetic resonance scans should replace biopsies for diagnosis of diffuse brain stem gliomas: A report from the Children’s Cancer Study Group. Neurosurgery 33: 1026-1030, 1993[Medline] 5. Freeman CR, Krischer JP, Sanford RA, et al: Final results of a study of escalating doses of hyperfractionated radiotherapy in brain stem tumors in children: A Pediatric Oncology Group study. Int J Radiat Oncol Biol Phys 27: 197-206, 1993[Medline] 6. Packer RJ, Boyett JM, Zimmerman RA, et al: Outcome of children with brain stem gliomas after treatment with 7800 cGy of hyperfractionated radiotherapy. Cancer 74: 1827-1834, 1994[CrossRef][Medline] 7. Lewis J, Lucraft H, Ghokal A: A UKCCSG study of accelerated radiotherapy for pediatric brain stem gliomas. Int J Radiat Oncol Biol Phys 38: 925-929, 1998 8. Davis FG, Freels S, Grutsch J, et al: Survival rates in patients with primary malignant brain tumors stratified by patient age and tumour histological type: An analysis based on Surveillance Epidemiology and End Results (SEER) data 1973-1991. J Neurosurg 88: 1-10, 1998[Medline] 9. Marchese MJ, Chang CH: Malignant astrocytic gliomas in children. Cancer 65: 2771-2778, 1990[CrossRef][Medline] 10. Wisoff JH, Boyett JM, Berger MS: Current neurosurgical management and the impact of extent of resection in the treatment of malignant gliomas of childhood: A report of the Children’s Cancer Group trial CCG-945. J Neurosurg 89: 52-59, 1998[Medline] 11. DeAngelis LM, Burger PC, Green SB, et al: Malignant glioma: Who benefits from adjuvant chemotherapy? Ann Neurol 44: 691-695, 1998[CrossRef][Medline] 12. Newlands ES, Stevens MF, Wedge SR, et al: Temozolomide: A review of its discovery, chemical properties, pre-clinical development and clinical trials. Cancer Treat Rev 23: 35-61, 1997[CrossRef][Medline] 13. Marzolini C, Decosterd LA, Shen F, et al: Pharmacokinetics of temozolomide in association with fotemustine in malignant melanoma and malignant glioma patients: Comparison of oral, intravenous and hepatic intra-arterial administration. Cancer Chemother Pharmacol 42: 433-440, 1998[CrossRef][Medline]
14. Friedman HS, Dolan ME, Pegg AE, et al: Activity of temozolomide in the treatment of central nervous system tumour xenografts. Cancer Res 55: 2853-2857, 1995 15. Sankar A, Thomas DG, Darling JL: Sensitivity of short term cultures derived from human malignant glioma to the anti-cancer drug temozolomide. Anticancer Drugs 10: 179-185, 1999[Medline] 16. Newlands ES, Blackledge GRP, Slack JA, et al: Phase I trial of temozolomide (CCRG 81045: M&B 39831: NSC362856). Br J Cancer 65: 287-291, 1992[Medline] 17. Estlin EJ, Lashford LS, Ablett S, et al: Phase I study of temozolomide in paediatric patients with advanced cancer. Br J Cancer 78: 652-661, 1998[Medline] 18. O’Reilly SM, Newlands ES, Glaser MG, et al: Temozolomide: A new oral cytotoxic chemotherapeutic agent with promising activity against primary brain tumours. Eur J Cancer 29A: 940-942, 1993[CrossRef][Medline] 19. Karnofsky DA, Abelman NH, Craver LF, et al: The use of nitrogen mustard in the palliative treatment of cancer. Cancer 1: 634-656, 1948[CrossRef] 20. Lansky SB, List MA, Lansky LL, et al: The measurement of performance status in childhood cancer patients. Cancer 60: 1651-1656, 1987[CrossRef][Medline] 21. Gehan EA: The determination of the number of patients required in a preliminary and a disease follow up trial of a new chemotherapy agent. J Chronic Dis 13: 346-353, 1961[CrossRef][Medline] 22. Duffner PK, Krischer JP, Burger PC, et al: Treatment of infants with glioma: The Pediatric Oncology Group experience. J Neurooncol 28: 245-256, 1966 23. Gilles FH, Sobel EL, Tavare CJ, et al: Age related changes in diagnoses, histological features and survival in children with brain tumors 1930-1979: The Childhood Brain Tumour Consortium. Neurosurgery 37: 1056-1068, 1995[Medline] 24. Khatib ZA, Heideman RL, Konvar EH, et al: Predominance of pilocytic histology in dorsally exophytic brain stem tumors. Pediatr Neurosurg 20: 2-10, 1994[Medline] 25. Robertson PL, Muraszko KM, Brunberg JA, et al: Pediatric midbrain tumors: A benign subgroup of brain stem gliomas. Pediatr Neurosurg 22: 65-73, 1995[Medline]
26. Yung WK, Prados MD, Yaya Tur R, et al: Multicentre phase II trial of temozolomide in patients with anaplastic astrocytoma or anaplastic oligoastrocytoma at first relapse. J Clin Oncol 17: 2762-2771, 1999 27. Krouwer HG, van-Duinen SG, Kamphorst W, et al: Oligoastrocytomas: A clinicopathological study of 52 cases. J Neurooncol 33: 223-238, 1997[CrossRef][Medline] 28. Ojeda VJ, al-Moutaery KR: Neuropathological biopsy diagnoses in clinical practice: How thorough and reproducible are they? Pathology 30: 328-329, 1998[CrossRef][Medline] 29. Louis DN: A molecular genetic model of astrocytoma histopathology. Brain Pathol 7: 755-764, 1997[Medline] 30. Von Deimling A, Louis DN, Amman K, et al: Association of epidermal growth factor gene amplification with loss of chromosome 10 in human glioblastoma multiforme. J Neurosurg 77: 295-301, 1992[Medline] 31. Maintz D, Fielder K, Koopman J, et al: Molecular genetic evidence for subtypes of oligoastrocytomas. J Neuropathol Exp Neurol 56: 1098-1104, 1997[Medline]
32. Wong ET, Hess KR, Gleason MJ, et al: Outcomes and prognostic factors in recurrent glioma patients enrolled onto phase II clinical trials. J Clin Oncol 17: 2572-2578, 1999 33. Bower M, Newlands ES, Blehan NM, et al: Multicentre phase II trial of temozolomide in recurrent or progressive high grade glioma. Cancer Chemother Pharmacol 40: 484-488, 1997[CrossRef][Medline] 34. Packer RJ: Brain stem gliomas: Therapeutic options at time of recurrence. Pediatr Neurosurg 24: 211-216, 1996[Medline]
35. Middlemas DS, Stewart CF, Kirstein MN: Biochemical correlates of temozolomide sensitivity in paediatric solid tumour xenograft models. Clin Cancer Res 6: 998-1007, 2000 36. Mineura K, Izumi I, Watanabe K, et al: Influence of O6 methylguanine-DNA alkyltransferase activity on chlorethylnitrosourea chemotherapy in brain tumors. Int J Cancer 55: 76-81, 1993[Medline]
37. Hongeng S, Brent TP, Sanford RA, et al: O6-Methylguanine DNA methyltransferase protein levels in pediatric tumors. Clin Cancer Res 3: 2459-2463, 1997
38. Bobola MS, Berger MS, Ellenbogen RG, et al: O6-Methylgua-nine-DNA methyltransferase in pediatric primary brain tumors: Re-lation to patient and tumor characteristics. Clin Cancer Res 7: 613-619, 2001 39. Friedman HS, Kokkinakis DM, Pluda J, et al: Phase I trial of O6-benzylguanine for patients undergoing surgery for malignant glioma. J Clin Oncol 16: 3570-3575, 1998[Abstract]
40. Kokkinakis DM, Bocangel DB, Schold SC, et al: Thresholds of O6-alkylguanine-DNA alkyltransferase which confer significant resistance of human glial tumour xenografts to treatment with 1,3-bis(2-chlorethyl)-1-nitrosourea or temozolomide. Clin Cancer Res 7: 421-428, 2001
41. Reese JS, Koc ON, Lee KM, et al: Retroviral transduction of a mutant methylguanine DNA methyltransferase gene into human CD34 cells confers resistance to O6-benzylguanine plus 1,3-bis(2 chlorethyl)-1-nitrosourea. Proc Natl Acad Sci U S A 93: 14088-14093, 1996 42. Chinnasamy N, Rafferty JA, Hickson I, et al: Chemoprotective gene transfer II: Multilineage in vivo protection of haemopoeisis against the effects of an antitumour agent by expression of a mutant O6-alkylguanine-DNA alkyltransferase. Gene Ther 5: 842-847, 1998[CrossRef][Medline] Submitted August 21, 2001; accepted July 30, 2002. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
