|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Phase I Trial Results of Iodine-131–Labeled Antitenascin Monoclonal Antibody 81C6 Treatment of Patients With Newly Diagnosed Malignant GliomasFrom the Departments of Pathology, Medicine, Surgery, Radiology, and Community and Family Medicine, Duke University Medical Center, Durham, NC. Address reprint requests to Ilkcan Cokgor, MD, Department of Medicine, Duke University Medical Center, Box 3624, Durham, NC 27710; email cokgo001{at}mc.duke.edu
PURPOSE: To determine the maximum-tolerated dose (MTD) of iodine-131 (131I)–labeled 81C6 antitenascin monoclonal antibody (mAb) administered clinically into surgically created resection cavities (SCRCs) in malignant glioma patients and to identify any objective responses with this treatment. PATIENTS AND METHODS: In this phase I trial, newly diagnosed patients with malignant gliomas with no prior external-beam therapy or chemotherapy were treated with a single injection of 131I-labeled 81C6 through a Rickham reservoir into the resection cavity. The initial dose was 20 mCi and escalation was in 20-mCi increments. Patients were observed for toxicity and response until death or for a minimum of 1 year after treatment. RESULTS: We treated 42 patients with 131I-labeled 81C6 mAb in administered doses up to 180 mCi. Dose-limiting toxicity was observed at doses greater than 120 mCi and consisted of delayed neurotoxicity. None of the patients developed major hematologic toxicity. Median survival for patients with glioblastoma multiforme and for all patients was 69 and 79 weeks, respectively. CONCLUSION: The MTD for administration of 131I-labeled 81C6 into the SCRC of newly diagnosed patients with no prior radiation therapy or chemotherapy was 120 mCi. Dose-limiting toxicity was delayed neurologic toxicity. We are encouraged by the survival and toxicity and by the low 2.5% prevalence of debulking surgery for symptomatic radiation necrosis.
MALIGNANT GLIOMAS are the most common type of primary brain tumors in adults. Conventional therapy consists of surgery, external-beam radiation, and chemotherapy. Median survival with conventional therapy varies from 40 to 60 weeks.1 The best survival outcome has been seen in patients under the age of 50, with a normal mental status and Karnofsky performance status (KPS)  90%. In such a favorable prognostic group of glioblastoma multiforme (GBM) patients treated with surgery and external-beam radiation therapy (XRT), median survival was 70.4 weeks, whereas median survival was 234 weeks in the anaplastic astrocytoma (AA) population.2 Despite advances in external-beam XRT, the best median survival in all GBM patients treated with surgery and external-beam XRT and chemotherapy with adjuvant radiosurgery is still between 56 and 76 weeks.3 Therapeutic modalities focused to the site of the original tumor have been used in several clinical trials because most CNS tumors recur at or near the site of the origin of the tumor.4 The first studies of brain tumor antibody-targeted therapy involved injections of iodine-125–labeled polyclonal antibody against brain tumors and iodine-131 (131I)–labeled control immunoglobulin. These studies measured the amount of tumor-targeted polyclonal antibodies.5 However, systemically administered radiolabeled monoclonal antibodies (mAbs) have not been effective in the treatment of brain tumors because of (1) the small amount of mAb that will cross the blood-brain barrier, (2) high interstitial fluid pressure in tumors and surrounding normal tissue, (3) lack of mAb specificity and less than optimal binding affinity, and (4) catabolism of the radioactive label.6,7 Our group8,9 and others10 have attempted to overcome these difficulties associated with systemic administration by directly injecting radiolabeled mAb into spontaneous cysts, into surgically created resection cavities (SCRCs), intrathecally, and into tumors. 81C6 antibody is a murine immunoglobulin G2b mAb that binds to tenascin, a tumor-associated extracellular matrix hexabrachion glycoprotein that is ubiquitous in gliomas but also found in melanomas and in breast, lung, and squamous cell carcinomas. mAb 81C6 binds to an epitope within the variably transcribed fibronectin Type III region, an isoform abundantly expressed in gliomas.11 mAb 81C6 was chosen for glioma studies on the basis of its well-characterized reactivity with the target antigen, tenascin, and because of the high levels of target antigen expression in U-251 MG and D-54 MG xenografts and stable target antigen expression in extracellular matrix of the glioma xenografts. Bourdon et al12 first reported radioiodinated antitenascin mAb 81C6 localized preferentially in subcutaneous and intracranial human glioma xenografts in athymic mice and rats. The specificity and high levels of 81C6 localization in glioma xenografts in athymic mice were determined by paired label analysis. Treatment with 131I-labeled 81C6 has resulted in significant tumor growth delay and regression in athymic mice bearing subcutaneous D-54/MG human glioma xenografts and in prolongation of median survival for athymic rats bearing intracranial tumors.8,13,14 Human pilot studies that compared radiolabeled specific with nonspecific mAbs after intravenous administration showed specific localization of radioiodinated 81C6.8 These studies also showed that intra-arterial administration did not result in greater localization compared with intravenous administration.15 Because of the small amount of radiolabeled mAb that accumulates in brain tumors with systemic administrations and, thus, the limited radiation dose to tumor tissue, a direct local or compartmental administration is carried out to deliver high therapeutic doses of radiation without inducing systemic toxicity. We have performed a phase I clinical trial of intrathecally administered 131I-labeled 81C6 mAb in 37 patients who had leptomeningeal neoplasms or brain tumor resection cavities that communicated with the subarachnoid space.16 The maximum-tolerated dose (MTD) was reached at 80 mCi in adults, and the dose-limiting factor was hematologic toxicity. Bigner et al9 reported a phase I study using 131I-labeled 81C6 mAb injected directly into SCRCs of 34 patients with recurrent malignant gliomas who had received prior XRT with or without chemotherapy. That study showed that the MTD was 100 mCi, and dose-limiting toxicity (DLT) was manifested as acute neurologic or hematologic symptoms. Median survival was 53 weeks for all patients and 52 weeks for patients with GBM, significantly higher than in a similar group of patients treated with carmustine (BCNU) wafers.17 We now report phase I trial results of patients with newly diagnosed malignant gliomas with no prior external-beam therapy or chemotherapy who received 131I-labeled antitenascin 81C6 mAb therapy. We conducted this phase I study to determine the MTD, toxicity, and objective responses in newly diagnosed malignant glioma patients. As in our previous recurrent disease studies, the incidence of serious toxicity and re-operation for radionecrosis was low, and there was a significant survival increase compared with outcomes of treatment by conventional therapy.
Patient Eligibility and Treatment Eligible patients had a diagnosis of supratentorial primary malignant glioma and were candidates for surgical resection and Rickham reservoir placement. Patients with infratentorial tumors or diffusely infiltrating gliomas, gliomas with ventricular access, subependymal spread, or multifocal tumors were not eligible. Each patient’s biopsy was evaluated for immunoreactivity with 81C6 mAb and an affinity-purified rabbit polyclonal antitenascin antibody before treatment; levels of reactivity ranged from weak to strong. Positive immunoreactivity for tenascin was required and was demonstrated by using either fresh or paraffin-embedded tissue stained positively with 81C6 or affinity-purified polyclonal rabbit antitenascin serum, respectively. Eligible patients were at least 3 years of age at the time of entry into the study. A KPS of at least 50% was necessary at the time of entry. Pregnant patients, lactating patients, and patients with iodine allergy were ineligible. Concurrent systemic chemotherapy was not allowed. Other requirements have been described previously.9 Patients underwent gross total tumor resection, and at the discretion of the neurosurgeon, any surrounding edema-involved tissue was removed. A Rickham reservoir was put in place. A magnetic resonance imaging (MRI) scan with contrast was performed within 48 hours after surgery to measure any residual tumor. The residual tumor (contrast enhancement) could not extend more than 1 cm beyond the margins of the resection cavity. A flow study with technetium-99m (99mTc)-labeled albumin was performed to confirm that a sealed resection cavity was present. The Rickham reservoir was used for the administration of 99mTc-labeled albumin and 131I-labeled 81C6. MRI images obtained immediately after surgery were used as a baseline for comparison with further MRI scans. An increase in the radiographic enhancing rim was defined as when the rim-enhancing volume increased more than the baseline MRI. The baseline MRI scans showed no disease or minimal residual disease (contrast enhancement). Patients with leakage of 99mTc-labeled albumin from the SCRC or who had a resection cavity that communicated with the subarachnoid space were not eligible for treatment. A baseline [18F]fluorodeoxyglucose positron emission tomography (18FDG PET) scan was obtained before treatment. Before treatment and 30 to 120 days after treatment, all patients were tested with a double antibody radioimmunoassay or an enzyme-linked immunoabsorbent assay for circulating antibodies to 81C6 murine (mu81C6), human/mouse chimeric (ch81C6), and single-chain Fv (sFv) antibody constructs.9 Starting at least 48 hours before 131I-labeled 81C6 mAb administration, patients were treated orally with a daily dose of four drops of potassium iodine solution in water or juice and 75 µg of liothyronine sodium (Cytomel; SmithKline Beecham, Philadelphia, PA) to decrease the radioactive iodine accumulation in the thyroid. The reservoir was accessed under sterile conditions with a 25-gauge butterfly needle. Cystic fluid was removed to a volume of 6 mL when possible. The 131I-labeled 81C6 was injected into the reservoir in a volume of 6 mL or less. The reservoir and catheter were flushed after the mAb injection. Patients were placed on radiation isolation until the whole-body retention of 131I was less than 30 mCi, as determined with a cross-calibrated radiation survey meter. After this level was reached, radionuclide imaging with a gamma camera was performed to document the biodistribution of 131I activity in the whole body. Patients were discharged after a posttreatment MRI was performed. Patients were treated on a dose escalation protocol to determine the MTD. The initial dose was 20 mCi of 131I on 10 mg of 81C6. Doses were escalated in 20-mCi increments on a fixed amount of 81C6 (10 mg). There were three to six patients per dose level. To maintain optimal immunoreactivity, 20 mg of 81C6 was used with doses of 100 mCi or greater. This investigation was approved by the Duke Investigational Review Board. Informed consent approved by the Duke Institutional Review Board was obtained from each subject or the subject’s guardian.
Antibody Production and Labeling
Toxicity Determinations DLT was defined as grade 3 or 4 nonhematologic toxicity or major hematologic toxicity consisting of more than 28 days of either an absolute neutrophil count of less than 500 cells/mL or a platelet count of less than 20,000 platelets/mL. If one or two out of three patients developed toxicity at a radioactive dose level, an additional three patients were treated at that level.
Pharmacokinetics and Dosimetry Radiation-absorbed dose estimates for bone marrow were based on the activity in whole blood as a function of time after administration. A reduction factor of 0.3 was used to account for the difference in activity in blood and bone marrow.21 The dose contribution from the SCRC to bone marrow was also considered. The range of absorbed doses as a function of depth from the SCRC interface and in normal brain tissue was considered in evaluation of the imaging, normal tissue toxicity (radionecrosis), and tumor control during treatment with 131I-labeled 81C6 mAb. Using the gadolinium-enhanced T1-weighted axial MRI images of the patient’s head obtained within 48 hours after surgery, we generated a three-dimensional reconstruction of the head and SCRC. The volume of the SCRC was calculated by using current image analysis software (VoxelView 2.5.4; Vital Images, St Paul, MN). This volume was then used to estimate the initial activity concentration in the SCRC at the time of administration, where a uniform activity concentration was assumed. Depth-dose calculations of the SCRC interface, 2-cm–thick margin, and normal brain were then carried out.20 The average absorbed dose for brain tissue located at the cavity interface, the 2-cm–thick margin surrounding the SCRC, and the whole brain was calculated.
Patient Characteristics Forty-two newly diagnosed patients were treated between July 1993 and March 1998. The median follow-up was 3.5 years. The patients’ characteristics are listed in Table 1. Two of the 42 patients were initially treated in this study and then retreated on a protocol for patients with recurrent tumor who had received external-beam XRT. Fourteen of the patients were women, and 28 patients were men. The mean age was 47 years, with a range of 19 to 71 years. Thirty-two patients had GBM, three had AA, five had anaplastic oligodendroglioma (AO), and two had mixed oligodendroglioma (MO). All patients had a KPS between 60% and 100%. Patients were treated with an initial 131I-labeled 81C6 mAb dose of 20 mCi, with increases of 20-mCi increments up to 180 mCi. None of the AO and MO patients were treated with external-beam XRT or chemotherapy. However, one patient with an initial diagnosis of MO had a biopsy after 131I-labeled 81C6 therapy, and the biopsy result was AA; thus, this patient was treated with external-beam therapy and chemotherapy. There are 16 patients presently alive and being observed. Twenty-six patients have died.
Toxicity In this dose escalation study, the number of patients treated with 131I-labeled 81C6 at each dose level was as follows: three patients were treated at the 20-mCi administration level, three at the 40-mCi level, five at the 60-mCi level, six at the 80-mCi level, seven at the 100-mCi level, seven at the 120-mCi level, three at the 140-mCi level, seven at the 160-mCi level, and one at the 180-mCi level. No significant hepatic, renal, thyroid, or other organ toxicities were observed. One patient had acute hemiparesis, which responded to increased corticosteroid therapy, at the 100-mCi administered dose; because of this toxicity at 100 mCi, we treated another three patients at 80 mCi. They developed no toxicities. Consequently, we increased the dose to 100 mCi and continued escalation in 20-mCi increments. Six patients developed irreversible DLT, with a median time to toxicity of 4 months. This consisted of persistent and irreversible neurologic symptoms, including severe memory loss, cognitive deficits, hemiparesis, speech disturbances, and gait ataxia, which did not improve with high-dose corticosteroids. These six patients were treated at the following administered activities: one at 120 mCi, two at 140 mCi, two at 160 mCi, and one at 180 mCi. The average absorbed dose to the 2-cm shell from radiolabeled mAb therapy for these six patients was 46 Gy (range, 37 to 55 Gy); the total absorbed dose to the 2-cm shell (including that from XRT) was 106 Gy (range, 97 to 115 Gy). Four patients have developed reversible neurotoxicity that was non-DLT. These patients were treated as follows: two at 100 mCi, one at 120 mCi, and one at 160 mCi. The average absorbed dose to the 2-cm shell from the radiolabeled mAb was 41 Gy (range, 32 to 52 Gy). The average total absorbed dose to the 2-cm shell (including that from XRT) was 101 Gy (range, 90 to 115 Gy). Therefore, a total of 10 patients developed neurotoxicity (six irreversible DLT and four reversible non-DLT). Because of the delayed neurotoxicity that occurred in two of the three patients who were treated at the 140-mCi level and in two of the seven patients who were treated at the 160-mCi dose level, we established the MTD as 120 mCi of 131I-labeled 81C6 (20 mg). At the MTD of 120 mCi, we estimate an optimal absorbed dose of 42 Gy to the 2-cm shell. With regard to non-DLT, a total of seven patients developed hematologic toxicity. Five of them developed grade 3 hematologic toxicity, and two developed grade 4 hematologic toxicity, requiring both platelet transfusions and Neupogen (Amgen, Inc, Thousand Oaks, CA) injections. None of them developed major hematologic toxicity, and none of these hematologic toxicities were dose limiting. PET profiles were obtained in all 42 patients studied. The PET profiles from these patients did not provide any relevant information regarding neurotoxicity, that is, distinguishing radiation necrosis from tumor progression.
HAMA
Dosimetry Results Retention of 131I-labeled 81C6 mAb in the SCRC and whole body was different. A large fraction of the activity remained in the SCRC, and lower activity concentrations were detected in the whole body. The time-activity curves for the whole body and the SCRC were described by either monoexponential or biexponential functions. The time-activity clearance for the whole-body activity, excluding the SCRC, follows the functional form of a serial, two-compartment model.20 Fig 1A presents the time-activity data and fitted curves for the SCRC and whole body for a patient who received an administered activity of 180 mCi.
The estimated average residence time of 131I-labeled 81C6 mAb in the SCRC and whole body was 79 hours (range, 10 to 113 hours) and 36 hours (20 to 55 hours), respectively. The activity in blood was characterized by an exponential uptake phase followed by an exponential clearance phase. Figure 1B presents the 131I-activity concentration in blood and the protein-associated activity as a function of time for the same patient. A trichloroacetic acid precipitation assay in blood samples showed that more than 90% of the 131I activity remained bound to protein 72 hours after administration. Overall, patients remained in radiation isolation for a period of 3 to 8 days. The average absorbed dose and average absorbed dose per unit administered activity to the whole body were 0.42 Gy (range, 0.09 to 0.87 Gy) and 1.1 x 10-4 Gy/MBq (range, 3.9 x 10-5 to 1.7 x 10-4 Gy/MBq), respectively. The estimated average absorbed dose to red marrow was 0.42 Gy (range, 0.08 to 0.89 Gy), and the average absorbed dose per unit administered activity was 1.1 x 10 Gy/MBq (range, 9 x 10 to 1.5 x 10 Gy/MBq). The average SCRC volume was 21 cm3 (range, 2 to 81 cm3). The average absorbed dose from 131I-labeled 81C6 mAb to the cavity interface and to a 2-cm–thick region surrounding the cavity interface was 1,435 Gy (range, 45 to 9,531 Gy) and 32 Gy (range, 2 to 59 Gy), respectively. The average absorbed dose per unit administered activity to the cavity interface and to the 2-cm–thick region surrounding the cavity interface was .42 Gy/MBq (range, .06 to 2.1 Gy/MBq) and 1 x 10 Gy/MBq (range, 2 x 10 to 2 x 10 Gy/MBq), respectively. These variations in absorbed dose were primarily caused by large differences in cavity size among patients and secondarily caused by variation in the SCRC residence time. A regression analysis based on absorbed dose per unit administered activity to the cavity interface and cavity volume showed a relationship of D(Gy/MBq) = 4.14 x V-0.95, where V is expressed in cm3, with a correlation coefficient of r = 0.98. Similarly, the average absorbed dose to 1- and 2-cm–thick margins surrounding the cavity interface was D(Gy/MBq) = 0.73 x V-0.53 and D(Gy/MBq) = 0.028 x V-0.40, respectively. This relationship shows that the absorbed dose to the SCRC is primarily dependent on cavity size and secondarily on residence time. Further details are described elsewhere.19 Ten patients developed neurotoxicity, six irreversible DLT and four reversible DLT. The patients who developed irreversible DLT were treated at the following administered activities: one at 120 mCi, two at 140 mCi, two at 160 mCi, and one at 180 mCi. The average absorbed dose to the 2-cm rim from the cavity interface from 131I-labeled 81C6 mAb was 46 Gy (range, 37 to 55 Gy). The average total absorbed dose to the 2-cm rim (including XRT) was 106 Gy (97 to 113 Gy). The four patients who developed reversible DLT were treated at the following administered activities: two at 100 mCi, one at 120 mCi, and one at 160 mCi. The average absorbed dose to the 2-cm rim from the cavity interface from 131I-labeled 81C6 mAb was 42 Gy (range, 32 to 52 Gy). The average total absorbed dose to the 2-cm rim (including XRT) was 101 Gy (range, 90 to 115 Gy). At the MTD of 120 mCi, the estimated absorbed dose to the 2-cm rim is 42 Gy. There was a statistically significant difference in increase in the contrast-enhancing rim (39 patients) and clinical symptomatology (41 patients) between the group of patients who received a total absorbed dose to the 2-cm–thick cavity margins of less than 90 Gy and the group who received higher doses. The median time for contrast enhancement for these two groups was 45 and 8 weeks, respectively (P = .01). Similarly, the median time to clinical symptomatology for these two groups was 54 and 30 weeks, respectively (P = .04). This localized therapy approach has been used intratumorally in preliminary studies, by Papanastassiou et al22 and Riva et al,23 in which patients underwent multiple injections of 131I-labeled mAbs with cumulative absorbed doses to the cavity interface between 70 and 410 Gy. Furthermore, Hopkins et al24 estimated absorbed doses to the cavity interface as a function of mean cavity radius and antibody-binding fraction for yttrium-90–labeled ERIC-1 and reported a range of absorbed doses to the cavity interface from 20 to 5,000 Gy. The dose estimates reported in our study are in the range of doses reported by Papanastassiou et al22 and Riva et al23 using 131I-labeled mAbs.
Response/Survival Data Further MRI scans were obtained and evaluated at the follow-up during the first year. Among 39 patients assessable with imaging, four patients developed contralateral lesions consistent with multifocal tumor. One of the patients started to show these lesions within 1 month after 131I-labeled 81C6 therapy. All the patients showed some rim enhancement around the SCRC on the initial postoperative scan. These enhancements were contiguous to the cavity margins. From each MRI scan, the volume of the cavity and that of the enhancing rim were calculated. Four out of five AO patients have been stable since their 131I-labeled 81C6 mAb treatment, with no further external-beam XRT or chemotherapy. These patients received an average absorbed dose of 9 to 37 Gy to the 2-cm cyst margin. They declined further treatment. However, if they had shown any signs of recurrence, we would have recommended both external-beam XRT and chemotherapy, which our other patients received. They have not developed any abnormalities to suggest recurrence as indicated by imaging. One AO patient developed a new lesion, detected by MRI and PET, and required re-operation for recurrent tumor 2 years after her therapy. The re-operation biopsy diagnosis remained AO. Both MO patients had progressive disease, which was observed both clinically and by imaging, within the first year. A biopsy that was performed on one patient was interpreted as an AA. Biopsy samples from the contrast-enhancing rim were obtained several months after radiolabeled mAb therapy from a total of 15 patients, and a sample from an autopsy was obtained for one patient. The biopsy results on seven patients showed radiation necrosis only. Five patients showed tumor recurrence only. Results for the remaining three biopsies (along with the autopsy) showed mixed radiation necrosis and tumor. A second biopsy was required after several months on four patients who had only radiation necrosis initially. Three biopsy results indicated radiation necrosis and tumor. The fourth biopsy still showed only radiation necrosis. Only one patient of the 42 required debulking surgery for radionecrosis. No studies were carried out to assess antibody uptake in the recurrent tissue (tumor or necrotic tissue). Because of the long period of time elapsed from 131I-labeled 81C6 treatment, clinical symptomatology and survival were also assessed. Sixteen of the 42 patients are alive. The median survival after 131I-labeled 81C6 mAb therapy for all 42 patients and the 32 patients with GBM was 79 weeks (95% confidence interval, 61 to 106 weeks) and 69 weeks (95% confidence interval, 52 to 106 weeks), respectively. Figure 2 presents a Kaplan-Meier graph for all patients and for GBM patients. The median time to development of therapy-related clinical symptomatology after antibody therapy, based on clinical history and examination, was 38 weeks.
Patterns of Failure Thirty-seven patients were assessable for patterns of failure, which were based on contrast-enhanced MRI images. Current MRI and PET techniques will not distinguish between tumor recurrence and radionecrosis, as indicated by our experience comparing imaging and biopsy results. The contrast-enhancing region as determined from MRI scans was always hypermetabolic on PET, which made it impossible to differentiate between radiation necrosis and tumor recurrence (Fig 3). Therefore, the enhancing regions detected in the MRI scans were considered to be either tissue radionecrosis or tumor recurrence. Only in cases in which a biopsy was obtained along with MRIs were we able to distinguish between radionecrosis and tumor.
The median time to increase in contrast-enhancing rim and clinical symptomatology was 15 weeks (n = 28) and 38 weeks (n = 41), respectively. A significant relationship was found between administered activity of 131I-labeled 81C6 and time to increased contrast-enhancing rim (P = .01); however, no relationship was found between administered activity of 131I-labeled 81C6 and time to development of clinical symptomatology (P = .86). There was a significant relationship between survival and clinical symptomatology (P = .0001) and an increase in contrast-enhancing rim (P = .004). Areas of liquefactive necrosis that were infiltrated by reactive macrophages characterized all 15 of the biopsies after initial 131I-labeled 81C6 therapy. Typically, the viable brain had plump gemistocytic astrocytes along with macrophages. The numbers of macrophages were variable from case to case, presumably a function of biologic factors including time to treatment and proximity to blood vessels. In two cases, second biopsies revealed recurrent malignant glioma, one as a massive, diffuse growth of tumor and the other as perivascular accumulations of mitotically active tumor cells that were positive for glial fibrillary acidic protein.
Radiolabeled mAbs have been administered to patients with CNS malignancy by several nonsystemic routes, including intrathecally, into spontaneously developing cystic cavities, into SCRC, or directly into tumor. The direct intrathecal application of radiolabeled mAbs was first reported by Kemshead et al6 for the treatment of neoplastic infiltration of meninges by leukemia. Riva et al10 reported their 6-year experience with 105 patients treated with intratumoral injections or injection into the SCRC of 131I-labeled antitenascin mAbs BC2 and BC4. Bigner et al16 published phase I studies with intrathecal and SCRC administration of 131I-labeled 81C6 mAb. Phase I trial results of the intrathecal administration of 131I-labeled 81C6 mAb to patients with leptomeningeal neoplasms or primary brain tumor resection cavities with subarachnoid communication showed that the MTD was reached at 80 mCi in adults. The median survival after therapy was more than 42 weeks; long-term follow-up shows seven patients are still alive with a posttreatment survival range of 138 to 324 weeks. The dose-limiting factor in this study was hematologic toxicity. In our phase I study using 131I-labeled 81C6 mAb administered into the SCRC in recurrent malignant glioma patients with prior external-beam XRT, the MTD was reached at 100 mCi.9 DLT was short-term neurologic toxicity. The median posttreatment survival for all patients was 53 weeks, and median survival was 52 weeks for those with GBM. At the time of publication of that article (1998),9 14 patients were alive and 20 were dead. Presently, four patients in that study remain alive, demonstrating posttreatment survivals of 161 to 180 weeks. In the current phase I study, patients with newly diagnosed malignant gliomas were treated after gross total surgical resection with 131I-labeled 81C6 mAb injected into SCRC. No major hematologic toxicity was observed, although seven patients developed grade 3 hematologic toxicity that consisted of neutropenia and thrombocytopenia, all of which responded to therapy with platelet infusions and Neupogen. DLT observed in six patients consisted of delayed neurologic symptoms, including focal motor seizures, hemiparesis, memory loss, and ataxia. Delivery of a highly localized radiation absorbed dose was the goal of this treatment approach. An MTD of 120 mCi was established, which resulted in whole-brain doses of less than 8 Gy and no major hematologic toxicity. Large absorbed radiation dose at the cavity interface and margins led to neurologic toxicity, which was the DLT in this study. Increases in contrast enhancement of the SCRC rim and clinical symptoms occurred earlier in those patients who had high total absorbed doses from external-beam and 131I-labeled 81C6 mAb therapy; these were patients who received doses to the 2-cm–thick margins greater than 90 Gy (P = .01 and P = .04, respectively). Consequently, cavity volume is a relevant parameter in this therapy modality, where lower cavity volume leads to a higher absorbed dose per unit of administered activity to the cavity interface and 2-cm cavity margins. Cavity volume also determines the volume of the 2-cm shell and thus the volume of normal brain that receives a large radiation dose. At the MTD of 120 mCi, the fraction of tumor that will receive an absorbed dose from 131I-labeled 81C6 mAb higher than 80 Gy varies primarily as a function of cavity size and, secondarily, as a function of residence time. For an average cavity size of 21 cm3 and an average residence time of 70 hours, the tumor margin that will receive an absorbed dose higher than 80 Gy is 0.25 cm, and for a cavity size of 10 cm3, the tumor margin will be 0.5 cm. However, the average absorbed dose to a 1-cm–thick margin for cavity sizes between 10 and 30 cm3 ranges between 50 and 96 Gy.
Early studies performed by Walker et al25 showed that the median survival time after treatment for newly diagnosed GBM patients treated with BCNU alone was 18.5 weeks, and with external-beam XRT alone, it was between 36 and 42 weeks. When BCNU and external-beam XRT were combined, the median survival time was 45 weeks. Curran et al2 noted a median survival time of 46 weeks in GBM patients treated with surgery and external-beam XRT and whose KPS was Similar to that observed in patients treated in our phase I clinical protocol for recurrent brain tumors, there was an apparent erosion of the SCRC that may have been caused by the high-dose irradiation. The increase in cavity size with this erosion made interpretation of images as recurrent tumor or radionecrosis difficult. Furthermore, these contrast-enhancing masses were hypermetabolic on PET scans, which precluded use of the PET scans to differentiate radiation necrosis from tumor recurrence. Neither PET nor gadolinium-enhanced MRI images could differentiate the effect of therapeutic radiation from tumor recurrence. Coregistration of MRI and PET images resulted in a good correlation between hypermetabolic regions in PET images with abnormal contrast enhancement regions in MRI images. However, the reverse was not true. Abnormal contrast enhancement regions in MRI images did not correlate with hypermetabolic regions in PET images. Furthermore, this correlation was not specific to tumor recurrence or normal tissue radionecrosis. As an example, Fig 3 shows the coregistration of MRI and PET images for a patient who was treated at the 140-mCi dose level. Biopsy samples obtained from this patient at 24 and 52 weeks after 131I-labeled 81C6 mAb therapy demonstrated frank tissue radionecrosis with macrophage infiltration and no tumor involvement. Of 42 patients, 15 were biopsied, and one was autopsied. Six biopsied patients showed radiation necrosis; nine showed tumor recurrence associated with therapeutic effects. The autopsy also showed tumor recurrence and radiation necrosis. A second biopsy was obtained in four of the six patients; three of them showed tumor recurrence. The overall percentage of HAMA responders (89%, 34 of 38 patients) to mu81C6 and ch81C6 is greater in this treatment protocol than that reported from our previous study9 of 131I-labeled 81C6 in SCRC of recurrent glioma patients. The number of responders to 81C6 sFv (nine of 38 patients), which contains only the variable region, is lower in this group than in the previous recurrent disease group (14 of 30 patients). It is not surprising that the overall number of responders in this group is greater; the patients enrolled in this study were newly diagnosed and received 131I-labeled 81C6 before other immunosuppressive therapies, whereas the recurrent patients had been treated with chemotherapy and external-beam XRT before 131I-labeled 81C6 therapy. Nevertheless, both the time to maximum HAMA response and the time to highest titer are comparable in the two studies. It is not clear why fewer patients developed antibody response to 81C6 sFv in this treatment group (nine of 38 patients) than in the previous study (14 of 30 patients). There was no toxicity associated with HAMA response, even in the three patients treated more than once with 131I-labeled 81C6 mAb. Shrieve et al26 and Wen et al27 compared the efficacy of stereotactic radiosurgery and brachytherapy for patients with newly diagnosed GBM who were previously treated with external-beam XRT. Their results showed that radiosurgery plus external-beam XRT improved survival when compared with external-beam XRT only.26,27 The best results obtained were among the 69 patients with GBM who were treated with surgery, external-beam XRT, and a radiosurgery boost.28 The median follow-up time for this group was 58 weeks, and 42% of patients were alive with a median survival of 79 weeks. These results are comparable with those obtained in this phase I clinical trial with 131I-labeled 81C6. However, in the Adessa et al28 study, re-operation to remove symptomatic radiation necrosis was required in 64% of radiosurgery patients and 50% of brachytherapy patients, whereas in our study, the re-operation rate for radionecrosis after 131I-labeled 81C6 mAb therapy was just 2.5%. The Brain Tumor Cooperative Group carried out a phase III randomized prospective trial comparing patients who received brachytherapy (60 Gy), external-beam XRT (60.2 Gy), and BCNU with those who received only external-beam XRT and BCNU.29 The median survival for the patients who received brachytherapy was 64 weeks compared with 52 weeks for those not receiving brachytherapy. Re-operation to remove symptomatic radiation necrosis, recurrent tumor, or both was required for both groups, with rates of 50% for those who received brachytherapy and 42% for those who did not. The total absorbed dose (external-beam therapy and brachytherapy) was approximately 120 Gy. However, this study did not establish a dose-response relationship between survival and total dose. Moreover, 131I-labeled 81C6 mAb therapy generated a sharper dose gradient at the field edges of the cavity interface than that produced by brachytherapy. The average total absorbed dose to the 2-cm shell from 131I-labeled 81C6 mAb and external-beam therapy was 80 Gy (range, 3 to 119 Gy), and the absorbed dose to normal brain tissue was between 3 and 8 Gy. In summary, 131I-labeled 81C6 treatment into the SCRC of newly diagnosed malignant glioma patients with no prior XRT was well tolerated with little toxicity at the MTD level of 120 mCi. We are encouraged by the good median survival of 79 weeks for all 42 patients and 69 weeks for the 32 GBM patients. At this point, we are proceeding with a phase II study at the administered activity level of 120 mCi of 131I-labeled 81C6 to study efficacy and true response rate. Ultimately, we will perform a randomized phase III trial. This article reports the results of a phase I study of dose escalation of 131I-labeled 81C6 in newly diagnosed malignant glioma patients with no prior external-beam XRT or chemotherapy who underwent gross total tumor resection. This phase I study is unique in its focus on such a homogeneous group of patients treated after gross total tumor resection and before external-beam XRT. The data on toxicity and survival cannot be compared with data from other studies because none of the other studies were phase I dose escalation studies in a homogeneous group of patients receiving a similar extent of surgery before treatment. We will conduct a randomized study to compare other local therapies with this adjuvant therapy modality in the near future. In this phase I clinical trial in 42 newly diagnosed malignant glioma patients, the MTD of 131I-labeled 81C6 was reached at 120 mCi. The median survival for all 42 patients and the 32 GBM patients was 79 and 69 weeks, respectively. DLT was delayed neurologic symptomatology. In contrast to radiosurgery and brachytherapy, where up to 64% of patients require re-operation for radionecrosis, only one (2.5%) of 42 patients in this study required re-operation for radionecrosis. Other than the dose-limiting, delayed, severe neurologic toxicity observed above the MTD, the hematologic toxicity and neurologic toxicity were minimal, and all were reversible or controllable with appropriate therapy.
Supported by National Institutes of Health grant nos. NS20023 and CA11898, and by grant no. MO1 RR30 through the General Clinical Research Centers Program, National Center for Research Resources, National Institutes of Health.
1. Walker MD, Alexander E Jr, Hunt WE, et al: Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas: A cooperative clinical trial. J Neurosurg 49: 333-343, 1978[Medline]
2.
Curran WJ, Scott CB, Horton J, et al: Recursive partitioning analysis of prognostic factors in the three radiation therapy oncology group malignant glioma trials. J Natl Cancer Inst 85: 704-710, 1993 3. Shrieve DC, Alexander E III, Black PM, et al: Treatment of patients with primary glioblastoma multiforme with standard postoperative radiotherapy and radiosurgical boost: Prognostic factors and long-term outcome. J Neurosurg 90: 72-77, 1999[Medline] 4. Gaspar LE, Fisher BJ, Macdonald DR, et al: Supratentorial malignant glioma: Patterns of recurrence and implications for external beam local treatment. Int J Radiat Oncol Biol Phys 24: 55-57, 1992[Medline] 5. Mahaley MS Jr, Day ED: Immunological studies of human gliomas. J Neurosurg 23: 363-370, 1965[Medline] 6. Kemshead JT, Papanastassiou V, Coakham HB, et al: Monoclonal antibodies in the treatment of central nervous system malignancies. Eur J Cancer 28: 511-513, 1992 7. Jain RK, Baxter LT: Mechanisms of heterogeneous distribution of monoclonal antibodies and other macromolecules in tumors: Significance of elevated interstitial pressure. Cancer Res 48: 7022-7032, 1988[Medline]
8.
Zalutsky MR, Moseley RP, Coakham HB, et al: Pharmacokinetics and tumor localization of 131I-labeled anti-tenascin monoclonal antibody 81C6 in patients with gliomas and other intracranial malignancies. Cancer Res 49: 2807-2813, 1989 9. Bigner DD, Brown MT, Friedman AH, et al: Iodine-131-labeled antitenascin monoclonal antibody 81C6 treatment of patients with recurrent malignant gliomas: Phase I trial results. J Clin Oncol 16: 2202-2212, 1998[Abstract] 10. Riva P, Franceschi G, Arista A, et al: Local application of radiolabeled monoclonal antibodies in the treatment of high grade malignant gliomas. Cancer Suppl 80: 2733-2742, 1997 11. Wikstrand CJ, Fung KM, Trojanowski JQ, et al: Antibodies and molecular immunology: Immunohistochemistry and antigens of diagnostic significance—Chapter 8, in Bigner DD, McLendon RE, and Bruner JM (eds): Russell and Rubinstein’s Pathology of Tumors of the Nervous System (vol 1, ed 6). London, United Kingdom, Arnold, 1998, pp 251-304 12. Bourdon MA, Coleman RE, Blasberg RG, et al: Monoclonal antibody localization in subcutaneous and intracranial human glioma xenografts: Paired-label and imaging analysis. Anticancer Res 4: 133-140, 1984[Medline]
13.
Lee YS, Bullard DE, Wikstrand CJ, et al: Comparison of monoclonal antibody delivery to intracranial glioma xenografts by intravenous and intracarotid administration. Cancer Res 47: 1941-1946, 1987
14.
Lee YS, Bullard DE, Zalutsky MR, et al: Therapeutic efficacy of anti-glioma mesenchymal extracellular matrix 131I radiolabeled murine monoclonal antibody in a human glioma xenograft model. Cancer Res 48: 559-566, 1988
15.
Zalutsky MR, Moseley RP, Benjamin JC, et al: Monoclonal antibody and F(ab')2 fragment delivery to tumor in patients with glioma: Comparison of intracarotid and intravenous administration. Cancer Res 50: 4105-4110, 1990 16. Bigner DD, Brown M, Coleman RE, et al: Phase I studies of treatment of malignant gliomas and neoplastic meningitis with 131I-radiolabeled monoclonal antibodies anti-tenascin 81C6 and anti-chondroitin proteoglycan sulfate Me1-14 F(ab')2: A preliminary report. J Neurooncol 24: 109-122, 1995[Medline] 17. Brem H, Piantadosi S, Burger PC, et al: Placebo-controlled trial of safety and efficacy of intraoperative controlled delivery by biodegradable polymers of chemotherapy for recurrent gliomas: The Polymer-Brain Tumor Treatment Group. Lancet 345: 1008-1012, 1995[Medline] 18. Food and Drug Administration: Director OoBRaR Center for Drugs and Biologics: Points to Consider in the Manufacture and Testing of Monoclonal Antibody Products for Human Use. Bethesda, MD, Food and Drug Administration, 1987 19. Akabani G, Cokgor I, Coleman RE, et al: Dosimetry and dose-response relationships in newly diagnosed patients treated with iodine-131-labeled anti-tenascin monoclonal antibody 81C6 therapy. Int J Radiat Oncol Biol Phys 46: 947-958, 2000[Medline]
20.
Akabani G, Reist CJ, Cokgor I, et al: Dosimetry of I-131 labeled 81C6 monoclonal antibody administered into surgically created resection cavities in malignant brain tumor patients. J Nucl Med 40: 631-638, 1999
21.
Sgouros G: Bone marrow dosimetry for radioimmunotherapy: Theoretical considerations. J Nucl Med 34: 689-694, 1993 22. Papanastassiou V, Pizer BL, Coakham HB, et al: Treatment of recurrent and cystic malignant gliomas by a single intracavity injection of 131I monoclonal antibody: Feasibility, pharmacokinetics and dosimetry. Br J Cancer 67: 144-151, 1993[Medline] 23. Riva P, Arista A, Tison V, et al: Intralesional radioimmunotherapy of malignant gliomas: An effective treatment in recurrent tumors. Cancer 73: 1076-1082, 1994 (suppl 3)[Medline] 24. Hopkins K, Chandler C, Bullimore J, et al: A pilot study of the treatment of patients with recurrent malignant gliomas with intratumoral yttrium-90 radioimmunoconjugates. Radiother Oncol 34: 121-131, 1995[Medline] 25. Walker MD, Strike TA, Sheline GE: An analysis of dose-effect relationship in radiotherapy of malignant gliomas. Int J Radiat Oncol Biol Phys 5: 1725-1731, 1979[Medline] 26. Shrieve DC, Alexander E III, Wen PY, et al: Comparison of stereotactic radiosurgery and brachytherapy in the treatment of recurrent glioblastoma multiforme. Neurosurgery 36: 275-284, 1995[Medline] 27. Wen PY, Alexander E III, Black PM, et al: Long term results of stereotactic brachytherapy used in the initial treatment of patients with glioblastomas. Cancer 73: 3029-3036, 1994[Medline] 28. Addesa AE, Shrieve DC, Alexander A III, et al: Stereotactic radiosurgery as primary adjuvant treatment for glioblastoma: the JCRT update. Proc Am Soc Clin Oncol 14: 144, 1995 (abstr) 29. Green SB, Shapiro WR, Burger PC, et al: A randomized trial of interstitial radiotherapy (RT) boost for newly diagnosed malignant glioma: Brain Tumor Cooperative Group (BTCG) Trial 8701. Proc Am Soc Clin Oncol 13: 174, 1994 (abstr) Submitted September 29, 1999; accepted June 16, 2000. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
