CONVECTION-ENHANCED DELIVERY OF CINTREDEKIN BESUDOTOX (INTERLEUKIN13PE38QQR) FOLLOWED BY RADIATION THERAPY WITH AND WITHOUT TEMOZOLOMIDE IN NEWLY DIAGNOSED MALIGNANT GLIOMAS: PHASE

CONVECTION-ENHANCED DELIVERY OF CINTREDEKINBESUDOTOX (INTERLEUKIN-13-PE38QQR)FOLLOWED BY RADIATION THERAPYWITH AND WITHOUT TEMOZOLOMIDEIN NEWLY DIAGNOSED MALIGNANT GLIOMAS: PHASE 1 STUDY OF FINAL SAFETY RESULTS

OBJECTIVE: Cintredekin besudotox (CB), a recombinant cytotoxin consisting of inter-leukin-13 and truncated Pseudomonas exotoxin, binds selectively to interleukin-13Rα2receptors overexpressed by malignant gliomas. This study assessed the safety of CBadministered by convection-enhanced delivery followed by standard external beamradiation therapy (EBRT) with or without temozolomide (Temodar; Schering-Plough,Kenilworth, NJ) in patients with newly diagnosed malignant gliomas.METHODS: After gross total resection of the tumor, two to four intraparenchymalcatheters were stereotactically placed and CB (0.25 or 0.5 μg/mL) was infused for96 hours. This was followed, 10 to 14 days later, by EBRT (5940–6100 cGy, 5 d/wk for6–7 wk) with or without temozolomide (75 mg/m2/d, 7 d/wk during EBRT). Safety wasassessed during an 11-week observation period after catheter placementRESULTS: Twenty-two patients (12 men, 10 women; median age, 55 yr; 21 with glioblas-toma multiforme and one with an anaplastic mixed oligoastrocytoma) were enrolled.None of the patients experienced dose-limiting toxicities in the first two cohorts(0.25 μg/mL CB � EBRT [n � 3] and 0.25 μg/mL CB � EBRT � temozolomide [n � 3]).One patient experienced a dose-limiting toxicity (Grade 4 seizure) in the third cohort(0.5 μg/mL CB � EBRT [n � 6]). Six patients in the final cohort (0.5 μg/mL CB � EBRT �temozolomide [n � 10]) completed treatment, and one patient experienced a dose-limiting toxicity (Grade 3 aphasia and confusion). Four patients were not consideredevaluable for a dose decision and were replaced. CB related adverse events occurringin more than one patient were fatigue, gait disturbance, nystagmus, and confusion. NoGrade 3 to 4 hematological toxicities were observed.CONCLUSION: CB (0.5 μg/mL) administered via convection-enhanced delivery beforestandard radiochemotherapy seems to be well tolerated in adults with newly diagnosedmalignant gliomas. Further clinical study assessment is warranted.

KEY WORDS: Cintredekin besudotox, Convection-enhanced delivery, Malignant glioma

Neurosurgery 61:1031–1038, 2007 DOI: 10.1227/01.NEU.0000280092.91489.08 www.neurosurgery-online.com

NEUROSURGERY VOLUME 61 | NUMBER 5 | NOVEMBER 2007 | 1031

CLINICAL TRIALMichael A. Vogelbaum, M.D., Ph.D.Brain Tumor and Neuro-Oncology Center,The Cleveland Clinic Foundation,Cleveland, Ohio

John H. Sampson, M.D., Ph.D.Department of Neurosurgery,Duke University,Durham, North Carolina

Sandeep Kunwar, M.D.Department of Neurosurgery,University of California, San Francisco,San Francisco, California

Susan M. Chang, M.D.Department of Neurosurgery,University of California, San Francisco,San Francisco, California

Mark Shaffrey, M.D.Department of Neurosurgery,University of Virginia,Charlottesville, Virginia

Anthony L. Asher, M.D.Carolina Neurosurgery and Spine Associates,Charlotte, North Carolina

Frederick F. Lang, M.D.Department of Neurosurgery,MD Anderson Cancer Center,Houston, Texas

David Croteau, M.D.NeoPharm Inc.,Waukegan, Illinois

Kristen Parker, B.S.NeoPharm Inc.,Waukegan, Illinois

Amy Y. Grahn, M.S.NeoPharm Inc.,Waukegan, Illinois

Jeffrey W. Sherman, M.D.NeoPharm Inc.,Waukegan, Illinois

S. Rafat Husain, Ph.D.Center for Biologics Evaluation and Research,United States Food and Drug Administration,Bethesda, Maryland

Raj K. Puri, M.D., Ph.D.Center for Biologics Evaluation and Research,United States Food and Drug Administration,Bethesda, Maryland

Reprint requests:Michael A. Vogelbaum, M.D., Ph.D.,Brain Tumor and Neuro-Oncology Center,Cleveland Clinic Foundation,9500 Euclid Avenue,Cleveland, OH 44195.Email: [email protected]

Received, March 8, 2007.

Accepted, May 14, 2007.

Malignant gliomas (MG) are the mostcommon type of primary braintumor in adults. According to the

Central Brain Tumor Registry of the UnitedStates, it is estimated that there are approxi-mately 10,000 new cases each year (3). Despitetreatment with surgery, radiation therapy, andchemotherapy, the prognosis remains poor,

particularly for glioblastoma multiforme(GBM), with a median survival of 12 to 15months and the best available therapy at theinitial diagnosis (25). Even at initial presenta-tion, infiltrating tumor cells extend at least2 cm away from the radiographic contrast-enhancing mass (26, 30). The infiltratingtumor component in functional tissue has

limited the efficacy of surgery and radiation therapy. Systemicagents are generally ineffective in part because of limiteddrug delivery. Recent advances in alkylator therapy, eithersystemic or intracavitary, have had an encouraging impact,but new avenues of treatment are clearly needed, includingnew delivery methods (2, 25, 29, 31, 32).

The majority of MG cell lines and explants overexpress ahigh number of interleukin (IL)-13 receptors (5, 6). Further-more, detection of messenger ribonucleic acid and protein forthe IL-13 receptor α-2 chain indicates MG specificity and amuch higher expression than in low-grade glioma or non-neo-plastic glia (9, 18). This differential expression provides a spe-cific target for MG therapy.

Cintredekin besudotox (CB) is a recombinant cytotoxincomposed of human IL-13 and a truncated form of Pseudo-monas exotoxin A (PE38QQR) (6, 8). CB mediates cytotoxicityby the inhibition of protein synthesis and apoptosis leading tocell death (13). It is highly cytotoxic to IL-13 receptors in vitroand in vivo expressing cells with a 50% inhibitory concentra-tion of 0.1 to 30 ng/mL (7, 9, 16). The efficacy of CB seems tobe enhanced by both radiation (in vitro) and temozolomide(Temodar; Schering-Plough, Kenilworth, NJ) (in vitro and invivo) (Puri, unpublished data) (11). Phase 1 evaluation of CBin recurrent MG using intraparenchymal administration aftertumor resection revealed a maximum tolerated infusate con-centration (MTIC) of 0.5 μg/mL. Dose-limiting toxicities(DLTs) consisted of symptomatic imaging changes radi-ographically and histopathologically with an inflammatoryand necrotic process of tumor-infiltrated and possibly nor-mal brain parenchyma. CB and procedure-related adverseevents were primarily limited to the central nervous system.Total flow rates of 0.750 mL/h using two to three cathetersand an infusion duration of 96 hours were well tolerated in aPhase 1 evaluation of recurrent MGs (13). The latter and MTICwere used in all subsequent CB clinical studies.

Tissue distribution of macromolecules into brain paren-chyma interstitial space can be facilitated by convection-enhanced delivery (CED), a regional delivery technique withcatheters placed directly in the target tissue and using a contin-uous pressure gradient during periods of hours to days cir-cumventing the blood-brain barrier. Preclinical studies haveshown a clinically significant, reproducible, and homogeneousdistribution of molecules of various sizes (1, 4, 15, 17, 19, 20).CED of therapeutic agents in recurrent MGs, including CB, hasshown promise in preclinical studies and early clinical develop-ment (10, 14, 16, 22, 23, 27, 28).

CB as part of the initial treatment, along with standard exter-nal beam radiation therapy (EBRT) concurrently with temo-zolomide, may be a potentially beneficial addition to the over-all MG initial treatment. Therefore, the safety and feasibility ofthis approach were examined in this Phase 1 study. The goal ofthis study was to assess the safety of administration of CBbefore radiochemotherapy. The maximum dose to be evalu-ated was 0.5 μg/mL, based on Phase 1 results in recurrent MG(14). Because this study was designed and initiated before thecompletion and publication of a study that documented the

survival benefit of concurrent and adjuvant temozolomide forthe treatment of newly diagnosed GBMs, it was designed toinclude cohorts of patients who received EBRT alone after treat-ment with CB (25).

PATIENTS AND METHODS

Cintredekin BesudotoxThe full sequence encoding CB was developed by Raj K. Puri, M.D.,

Ph.D. (Tumor Vaccines and Biotechnology Branch, Division of Cellularand Gene Therapies, Center for Biologics Evaluation and Research,Food and Drug Administration) and incorporated into a plasmid atAdvanced BioScience Laboratories (Kensington, MD) and later atDiosynth (Morrisville, NC). Escherichia coli cells were transformed withthe plasmid containing CB gene sequence, and the Master Cell Bankand Working Cell Bank were prepared (Charles River, Malvern, PA).The protein expressed after induction was purified from inclusion bod-ies under current good manufacturing practices.

PatientsThe clinical study was open to adults with histologically confirmed

newly diagnosed resectable supratentorial MGs (World HealthOrganization Grade III or IV), including GBMs, anaplastic astrocy-tomas, and mixed anaplastic oligoastrocytomas. A KarnofskyPerformance Scale score of 70 or more was required. Patients had ade-quate baseline organ function as assessed by laboratory studies.Patients must have undergone a gross total resection of the solid con-trast-enhancing components of the tumor more than 1.0 cm in diame-ter and must have been able to have catheters placed within 14 days ofthe resection. Patients were excluded if they had signs of impendingherniation, multifocal disease, subependymal, or leptomeningealspread, and if they received any previous antitumor treatment (otherthan corticosteroids), including any investigational agents. Patientssigned an institutional review board-approved written informed con-sent before enrollment. The study was approved by the institutionalreview boards of all participating centers. The study was conducted insix centers in the United States, and patients were enrolled betweenJuly 2004 and May 2006.

Study Design and TreatmentThe primary objective of the study was to determine the safety and

tolerability of CB in conjunction with EBRT with and without temo-zolomide. The secondary objective was to evaluate time to diseaserecurrence or progression and overall survival. All patients had tumorresection followed by intraparenchymal placement of two to fourcatheters in areas at greatest risk for residual infiltrating tumor, such asregions displaying hyperintense signal abnormality on T2-weighted,fluid-attenuated inversion recovery images, or any residual solid con-trast-enhancing disease. Open-ended, barium-impregnated silicon infu-sion catheters with a 1.0-mm inner diameter and a 2.0-mm outer diam-eter were used (Vygon Neuro, Valley Forge, PA). Catheter placementoccurred within 14 days after tumor resection using postoperative mag-netic resonance imaging (MRI) scans for stereotactic placement plan-ning and catheter positioning guidelines (24). Computed tomographicscans were performed within 24 hours after catheter placement to eval-uate catheter positioning before starting the infusion. If the computedtomographic scan revealed that any catheter was not properly posi-tioned, repositioning or removal was recommended. CB was thenadministered by CED for 96 hours at a fixed total infusion rate of

1032 | VOLUME 61 | NUMBER 5 | NOVEMBER 2007 www.neurosurgery-online.com

VOGELBAUM ET AL.

0.750 mL/h divided by the number of functional catheters on an inpa-tient basis. To control a potential mass effect caused by the volume ofinfusate, patients were maintained on high-dose dexamethasone dur-ing and after CB infusion.

Fractionated EBRT was to begin approximately 10 to 14 days afterthe completion of CB infusion. Fractions of 180 to 200 cGy/day weredelivered 5 days per week to a total dose of 5940 to 6100 cGy. A total ofapproximately 4600 cGy was delivered to a tumor volume consisting ofeither the T2-hyperintense signal abnormality with a 2-cm margin or,if no abnormality was present, the contrast-enhancing component witha 2.5-cm margin. The conedown target volume was to include the con-trast-enhancing component (without edema) plus a 2.5-cm margin andtreated with an additional 1400 cGy.

Temozolomide was administered daily concurrent with EBRT. Thepatients were to receive the same dose of temozolomide throughoutEBRT. The dose was based on the patient’s body surface area calculatedat the start of EBRT. Temozolomide was orally administered to thepatient, in a fasting state (at least 1 h), once per day, beginning on thefirst day of EBRT at a dose of 75 mg/m2/d and continuing for 7 daysper week during the EBRT period. Dose reduction to 50 mg/m2/d wasallowed if any Grade 3 toxicity occurred at 75 mg/m2/d.

Dose escalation was performed to determine the safety of CB inconjunction with EBRT alone or with temozolomide up to the knownMTIC in recurrent MG when administered intraparenchymally (14).Patients were first enrolled in cohorts with CB concentration levels of0.25 and 0.5 μg/mL followed by EBRT without temozolomide. OnceCB � EBRT safety was sequentially confirmed at these CB concentra-tions, cohorts of patients with CB concentration levels of 0.25 and 0.5μg/mL followed by EBRT with temozolomide were sequentiallyenrolled. Cohorts of three evaluable patients were used, and doseescalation was performed if zero of three patients had DLT. An addi-tional three patients were enrolled if one of three patients had devel-oped DLT. DLTs were defined, in the presence of supportive care, asany Grade 3 or 4 toxicity that was possibly, probably, or definitelyrelated to the study drug. MTIC was defined as the dose level belowthat causing DLT in two to six patients. De-escalation to a CB concen-tration of 0.125 μg/mL was permitted if 0.25 μg/mL was dose limiting.A total of six patients were to be treated at the MTIC or maximal con-centration planned. Each cohort was observed for 11 weeks afteradministration of the study drug to include complete EBRT and, ifapplicable, temozolomide treatment to capture unacceptable toxicitybefore the next cohort was enrolled. Cohorts with 0.5 μg/mL CB �EBRT and 0.25 μg/mL CB � EBRT � temozolomide were enrolledsequentially without waiting for completion of the 11-week observa-tion period for the 0.5 μg/mL CB � EBRT cohort (Fig. 1). All patientswere allowed to receive additional antitumor therapies (temozolo-mide or other) after the follow-up evaluation conducted at Week 11.

Patient AssessmentsAll patients underwent screening evaluations within 14 days before

catheter placement, including medical history, concomitant medica-tions, complete neurological examinations, Karnofsky PerformanceScale score, Mini-Mental Status Examination, and laboratory assess-ments. Immunohistochemistry for IL-13 receptor α-2 chain was per-formed on unstained tissue slides with a goat antibody (R&D System,Minneapolis, MN) using a validated assay (Pathway Diagnostics Corp.,Malibu, CA). A semiquantitative assessment considering only patho-logical elements was performed by a board-certified pathologist: nostaining (0), weak staining (1�), moderate staining (2�), and strongstaining (3�). No staining and 1� were considered nonsignificant (neg-ative), whereas 2� and 3� were significant (positive). MRI scans were

obtained at screening, within 48 hours of catheter placement, 11 weeksposttreatment, every 8 weeks for 12 months, and then every 12 weeksuntil recurrent or progressive disease. Additional ancillary imagingtechniques were used as needed when imaging changes were sugges-tive of treatment effect. All patients were followed for survival. Follow-up clinical assessment, including concomitant medications, completephysical and neurological examinations, Karnofsky Performance Scalescore, and Mini-Mental Status Examination, was performed along witheach follow-up imaging study. Laboratory evaluations, includinghematology and chemistry, were performed weekly during EBRT forpatients receiving temozolomide.

Statistical ConsiderationsPatients were considered evaluable for dose escalation if they

received at least 90% of the planned doses of CB and EBRT and, ifapplicable, at least 90% of the planned doses of temozolomide andcompleted the 11-week observation period or experienced a DLTbetween the start of CB treatment and the end of the observationperiod. Patients were considered evaluable for safety if they receivedany dose of CB. Adverse events were coded using the MedicalDictionary for Regulatory Activities and graded according to theCancer Therapy Evaluation Program Common Terminology Criteriafor Adverse Events, Version 3.0. For each adverse event, causality attri-bution was determined for CB, catheter, EBRT, and temozolomide.Treatment-emergent adverse events were captured. MRI changesrelated to CB were evaluated using a five-point grading system previ-ously published (21).

RESULTS

Patient DemographicsTwenty-two patients were enrolled at six study sites in the

United States. All had catheters placed and received CB.Eighteen of the 22 treated patients received 90% or more of theintended dose of CB, EBRT, and, if applicable, temozolomide.Patient demographics and disposition are summarized inTable 1. IL13 receptor α-2 chain immunohistochemistry assess-

NEUROSURGERY VOLUME 61 | NUMBER 5 | NOVEMBER 2007 | 1033

CONVECTION-ENHANCED DELIVERY OF CINTREDEKIN BESUDOTOX

FIGURE 1. Dose escalation decision flow chart. CB,cintredekin besudotox; EBRT, external beam radiationtherapy; TMZ, temozolomide.

ments were available for 19 patients; 17 patients had speci-mens with positive staining, and two patients had specimenswith negative staining. No patients were lost to follow-up.

Safety and Toxicity

Enrollment and Dose-limiting ToxicitiesThe intraparenchymal concentration of 0.25 μg/mL and

EBRT with or without temozolomide was well tolerated with-out any DLTs. Three patients were enrolled in each cohort. Theintraparenchymal concentration of 0.5 μg/mL and EBRT regi-men was associated with two DLTs, one in the cohort withouttemozolomide and one in the cohort with temozolomide. Sixpatients were enrolled in the cohort of 0.5 μg/mL CB and EBRTwithout temozolomide, and 10 patients were enrolled in thecohort of 0.5 μg/mL CB and EBRT with temozolomide.

The first DLT occurred in a 38-year-old woman who under-went a gross total resection of a left parieto-occipital mass,histopathologically consistent with GBM, followed by0.5 μg/mL CB and EBRT. The patient experienced a Grade 4convulsive status epilepticus approximately 10 weeks aftertreatment, which resolved within 48 hours with lorazepamand fosphenytoin. The patient had a history of seizures andwas receiving levetiracetam at the time of the event. She hadbeen weaned from phenytoin a few weeks before the event.No neuroimaging changes other than marginal contrastenhancement were observed on MRI scans. The second DLT

occurred in a 55-year-old man who underwent a gross totalresection of a left temporoparietal mass, histopathologicallyconsistent with an anaplastic oligoastrocytoma, followed by0.5 μg/mL CB and EBRT with temozolomide. The patientexperienced a Grade 3 expressive dysphasia and confusionapproximately 1 week after the completion of CB infusion andconsequently before EBRT and temozolomide. A head com-puted tomographic scan without contrast performed shortlyafter symptom onset showed a subcortical hypodense areanear a catheter tract and an infusion area in the left temporallobe (Fig. 2). The expressive aphasia and confusion resolvedwithin a few days with a small increase in the dose of corticos-teroids. Consequently, an MRI scan was not performed untilthe next scheduled follow-up imaging study approximately11 weeks after treatment. That study showed a new subcorti-cal area of cavitation sparing the cortex in the left temporallobe near the infusion area of the left temporal catheter. Thearea of cavitation was associated with predominantly subcor-tical atrophy (Fig. 2). The topography of the abnormality wasnot thought to be consistent with an ischemic infarction butrather possibly related to the CB effect on tumor-infiltratedand normal brain parenchyma.

1034 | VOLUME 61 | NUMBER 5 | NOVEMBER 2007 www.neurosurgery-online.com

VOGELBAUM ET AL.

FIGURE 2. Axial T1-weighted MRI scans with gadolinium and fluid-attenuated inversion recovery at baseline (pretreatment) (A) and 11 weeksafter treatment (B). Catheter tip position overlaid onto the MRI scan fromthe post-catheter placement computed tomographic scan is indicated by thecross. The left temporal area of cavitation with predominantly subcorticalatrophy was visible 11 weeks after treatment.

TABLE 1. Patients’ disposition and demographic summarya

Parameters Categories No. of patients

Catheters placed 22/22 (100%)

Received complete IL13- 21/22 (95.5%)PE38QQR infusion

Received complete 19/22 (86.4%)EBRT course

Received complete 9/13 (69.2%)temozolomide course

Sex Male/female 13/8 (59.1/40.9%)

Age, yr Median (range) 55 (30–76)

Race Caucasian 20 (91%)

Hispanic 1 (4.5%)

Asian 1 (4.5%)

KPS score Mean (range) 86.8 (70–100)

70–80 7 (33.3%)

90–100 15 (64.7%)

Histopathology Glioblastoma multiforme 21 (95.5%)

Anaplastic oligoastrocytoma 1 (4.5%)

Extent of tumor Gross total (�95%) 19 (86.4%)resection Subtotal (�95%) 3 (13.6%)

IL13Rα-2 IHC Positive/negative 17/2 (77/9%)

a IL13, interleukin-13; EBRT, external beam radiation therapy; KPS, Karnofsky PerformanceScale; IHC, immunohistochemistry.

A total of six patients were enrolled in the cohort of0.5 μg/mL CB and EBRT without temozolomide because onepatient experienced a DLT; none of the other patients experi-enced a DLT. A total of 10 patients were enrolled in the cohortof 0.5 μg/mL CB and EBRT with temozolomide because sixpatients were required to confirm safety at this final concen-tration level and four patients were not evaluable for doseescalation. The reasons for non-evaluability included incom-plete CB infusion secondary to reversible mass effect symp-toms (hemiparesis), inability to begin temozolomide becauseof thrombocytopenia, incomplete EBRT and temozolomidetreatment secondary to fatal pulmonary embolism, andincomplete temozolomide treatment secondary to Pneumo-cystis carinii pneumonia. These events were judged to be unre-lated to CB. The patient with thrombocytopenia developedthis event after surgical resection and before catheter place-ment and CB infusion. Reversible mass effect symptomsrelated to local volume infusion, but not drug toxicity, havebeen observed and well characterized (14). No dose-limitingor maximum tolerated dose was identified in this study.

Adverse EventsFrom the first procedure (stereotactic biopsy or cran-

iotomy) to a minimum of 71 days after the completion ofdrug infusion, adverse events were reported for all patients.Most of the reported adverse events originated from the cen-tral nervous system. Table 2 presents an overall summary ofadverse events severity and causal attribution. The mostcommon adverse events regardless of attribution are outlinedin Table 3. Approximately 33% of all adverse events were neu-rological or psychiatric, although fatigue was one of the mostfrequently reported adverse events. The majority (87%) ofadverse events were Grade 1 or 2. Thirteen percent of alladverse events were Grade 3 or higher in severity; 11% wereGrade 3, 1.5% were Grade 4, and 0.5% were Grade 5. The onlyfatal (Grade 5) event was a pulmonary embolism unrelatedto the study drug or catheter. Seventy-five percent of all CB-related adverse events were neurological or psychiatric.Twenty-nine percent were Grade 3 or higher; 21% were Grade

3, and 7% were Grade 4. The most common adverse eventsrelated to CB included fatigue (9%), gait disturbance (9%),nystagmus (9%), and confusion (9%). All other reportedadverse events related to CB occurred in one patient (5%).Fifty-nine percent of all catheter-related adverse events wereneurological or psychiatric, and 9% were infectious. Twenty-four percent were Grade 3 or higher; 21% were Grade 3, and3% were Grade 4. The most common adverse events related tocatheter placement included headache (32%), gait disturbance(9%), and pyrexia (9%). All other catheter-related adverseevents occurred in one patient. There was no relationshipbetween the drug concentration and the severity of adverseevents. No clinical or laboratory abnormalities were sugges-tive of a drug effect on organ functions.

CB-related Imaging ChangesMRI assessment for CB-induced neuroimaging changes was

performed on the first follow-up MRI scan approximately11 weeks after CB treatment because the onset of thesechanges have consistently been observed within 4 to 8 weeksof treatment in Phase I studies for recurrent MGs (21). Onlythe highest imaging change grade observed for each patient isreported here. Grade III or IV imaging changes induced by CBwere not observed in any patients enrolled in this study.Seven patients (32%) had Grade II changes, 11 patients (50%)experienced Grade I changes, one patient (4.5%) had nochanges (Grade 0), two patients (9%) had consistent recur-rence, and one patient (4.5%) had missing data. None of theimaging changes observed were symptomatic.

The maximal CB concentration of 0.5 μg/mL evaluatedwith an infusion duration of 96 hours and total flow rate of

NEUROSURGERY VOLUME 61 | NUMBER 5 | NOVEMBER 2007 | 1035

CONVECTION-ENHANCED DELIVERY OF CINTREDEKIN BESUDOTOX

TABLE 2. Overall summary of adverse events severity andcausalitya

Causality attribution

Maximum Any relation- CB-related, Catheter-severity ship, no. (%) no. (%) related, no. (%)

Any grade 204 28 34

Grade 1 108 (53%) 15 (54%) 17 (50%)

Grade 2 69 (34%) 5 (18%) 9 (26%)

Grade 3 23 (11%) 6 (21%) 7 (21%)

Grade 4 3 (1.5%) 2 (7%) 1 (3%)

Grade 5 1 (0.5%) 0 0

a CB, cintredekin besudotox.

TABLE 3. Summary of the most frequently reported adverseevents regardless of causality

No. (%) ofAdverse event

No. (%) ofpatients with

patientsGrade 3 or 4 events

Fatigue 16 (73%) 0

Headache 11(50%) 0

Nausea 9 (41%) 0

Alopecia 9 (41%) 0

Constipation 6 (27%) 1 (5%)

Deep vein thrombosis 5 (23%) 3 (14%)

Dyspepsia 3 (14%) 0

Peripheral edema 3 (14%) 0

Pyrexia 3 (14%) 0

Radiation skin injury 3 (14%) 0

Aphasia 3 (14%) 3 (14%)

Convulsion 3 (14%) 2 (9%)

Confusion 3 (14%) 3 (14%)

Dermatitis 3 (14%) 0

0.750 mL/h using two to four catheters was well toleratedwhen administered before standard radiochemotherapy.

SurvivalEleven (50%) of the 22 patients remain alive, including 10

with GBMs and one with anaplastic oligoastrocytoma, after amedian follow-up period of 44 weeks. Survival ranges from 5to 113 weeks, and progression-free survival ranges from 5 to78� weeks. One patient died before having progressive dis-ease documentation, and six patients (27%) remain progres-sion free at a median follow-up period of 40 weeks. Follow-upcontinues for all patients. Given the small sample of patientsand the ongoing patient follow-up, no correlation could beestablished between immunohistochemistry and survival out-comes at this point.

DISCUSSION

The results of this Phase I study performed in newly diag-nosed MGs suggest that CB followed by EBRT with or with-out temozolomide is safe and well tolerated at similar CBconcentrations as were used in monotherapy of recurrentMGs. Most of the adverse events we observed originatedfrom the central nervous system. Those adverse events werealso typical of events that are expected in a population ofpatients with brain tumors undergoing surgical procedures(2, 24, 29). No significant systemic toxicity was noted, consis-tent with the fact that locoregional delivery results in mini-mal systemic exposure.

Neuroradiographic changes induced by CB in this studywere not as severe as those reported for Phase I studies inrecurrent MG (21). No Grade III and IV imaging changesregarded as a necrotic and inflammatory process involvingtumor-infiltrated and normal brain parenchyma were ob-served. Only Grade I and II imaging changes were observed,and these were always asymptomatic even in eloquent brainregions. We suspect that these changes are probably consistentwith the successful local distribution of the drug. The reasonfor the lower incidence of neuroradiographic changes inducedby CB in newly diagnosed MG compared with recurrent MGmay include the relative immunosuppression induced by cor-ticosteroids and temozolomide because CB may have animmune-mediated mechanism for cytotoxicity, in addition toits potent effect inhibition of protein synthesis (12). The DLTconsisting of symptomatic neuroimaging changes with area ofcavitation may represent an interaction among EBRT, temo-zolomide, and CB because it has not been reported in Phase 1studies for recurrent MGs. Future studies in newly diagnosedMGs should closely monitor these changes.

Although CB efficacy is best enhanced with concurrent radi-ation in vitro, this sequence of administration would be techni-cally difficult (11). One of the main limitations would be com-plications related to the process and timing of wound healing.In addition, the short half-life of CB when administered intrac-erebrally would make it necessary to perform repeated sequen-tial administrations during radiotherapy. Consequently, a

sequential administration with CB first and EBRT with or with-out temozolomide given as soon as possible after enough timeto allow wound healing (at least 10 d) was selected for thisPhase I study. Administration of CB after EBRT is less preferredfor the same reasons outlined above, in addition to theunknown effect of a recently radiated brain on drug distribu-tion after CED.

CONCLUSION

CB 0.5 μg/mL administered via CED before standardradiochemotherapy seems to be safe and well tolerated inpatients with newly diagnosed MGs. Despite its relative com-plexity, the combination regimen was safely administered inthis Phase 1 dose-finding study. Because temozolomide admin-istered concurrently and after EBRT has become the standard ofcare for patients with newly diagnosed GBMs, subsequent effi-cacy studies in newly diagnosed GBMs will include temozolo-mide therapy. Further clinical studies of CB in the patient pop-ulation with newly diagnosed GBMs are warranted.

REFERENCES

1. Bobo RH, Laske DW, Akbasak A, Morrison PF, Dedrick RL, Oldfield EH:Convection-enhanced delivery of macromolecules in the brain. Proc NatlAcad Sci U S A 91:2076–2080, 1994.

2. Brem H, Piantadosi S, Burger PC, Walker M, Selker R, Vick NA, Black K, SistiM, Brem S, Mohr G, Muller P, Schold CS: Placebo-controlled trial of safetyand efficacy of intraoperative controlled delivery by biodegradable polymersof chemotherapy for recurrent gliomas. The Polymer-brain Tumor TreatmentGroup. Lancet 345:1008–1012, 1995.

3. Central Brain Tumor Registry of the United States (CBTRUS): StatisticalReport: Primary Brain Tumors in the United States, 1998–2002. Chicago, CentralBrain Tumor Registry of the United States, 2005.

4. Chen MY, Lonser RR, Morrison PF, Governale LS, Oldfield EH: Variablesaffecting convection-enhanced delivery to the striatum: A systematic exami-nation of rate of infusion, cannula size, infusate concentration, and tissue-cannula sealing time. J Neurosurg 90:315–320, 1999.

5. Debinski W, Gibo DM, Hulet SW, Connor JR, Gillespie GY: Receptor for inter-leukin 13 is a marker and therapeutic target for human high-grade gliomas.Clin Cancer Res 5:985–990, 1999.

6. Debinski W, Obiri NI, Powers SK, Pastan I, Puri RK: Human glioma cellsoverexpress receptors for interleukin 13 and are extremely sensitive to a novelchimeric protein composed of interleukin 13 and pseudomonas exotoxin.Clin Cancer Res 1:1253–1258, 1995.

7. Husain SR, Joshi BH, Puri RK: Interleukin-13 receptor as a unique target foranti-glioblastoma therapy. Int J Cancer 92:168–175, 2001.

8. Husain SR, Puri RK: Interleukin-13 receptor-directed cytotoxin for malignantglioma therapy: From bench to bedside. J Neurooncol 65:37–48, 2003.

9. Joshi BH, Plautz GE, Puri RK: Interleukin-13 receptor alpha chain: A noveltumor-associated transmembrane protein in primary explants of humanmalignant gliomas. Cancer Res 60:1168–1172, 2000.

10. Kaiser MG, Parsa AT, Fine RL, Hall JS, Chakrabarti I, Bruce JN: Tissue distri-bution and antitumor activity of topotecan delivered by intracerebral clysis ina rat glioma model. Neurosurgery 47:1391–1399, 2000.

11. Kawakami K, Kawakami M, Liu Q, Puri RK: Combined effects of radiationand interleukin-13 receptor-targeted cytotoxin on glioblastoma cell lines. IntJ Radiat Oncol Biol Phys 63:230–237, 2005.

12. Kawakami K, Terabe M, Kioi M, Berzofsky JA, Puri RK: Intratumoral therapywith IL13-PE38 results in effective CTL-mediated suppression of IL-13Ralpha2-expressing contralateral tumors. Clin Cancer Res 12:4678–4686,2006.

1036 | VOLUME 61 | NUMBER 5 | NOVEMBER 2007 www.neurosurgery-online.com

VOGELBAUM ET AL.

13. Kawakami M, Kawakami K, Puri RK: Intratumor administration of inter-leukin 13 receptor-targeted cytotoxin induces apoptotic cell death in humanmalignant glioma tumor xenografts. Mol Cancer Ther 1:999–1007, 2002.

14. Kunwar S, Prados MD, Chang SM, Berger MS, Lang FF, Piepmeier JM,Sampson JH, Ram Z, Gutin PH, Gibbons RD, Aldape KD, Croteau DJ,Sherman JW, Puri RK; Cintredekin Besudotox Intraparenchymal StudyGroup: Direct intracerebral delivery of cintredekin besudotox (IL13-PE38QQR) in recurrent malignant glioma: A report by the CintredekinBesudotox Intraparenchymal Study Group. J Clin Oncol 25:837–844, 2007.

15. Laske DW, Morrison PF, Lieberman DM, Corthesy ME, Reynolds JC, Stewart-Henney PA, Koong SS, Cummins A, Paik CH, Oldfield EH: Chronic intersti-tial infusion of protein to primate brain: Determination of drug distributionand clearance with single-photon emission computerized tomography imag-ing. J Neurosurg 87:586–594, 1997.

16. Lidar Z, Mardor Y, Jonas T, Pfeffer R, Faibel M, Nass D, Hadani M, Ram Z:Convection-enhanced delivery of paclitaxel for the treatment of recurrentmalignant glioma: A Phase I/II clinical study. J Neurosurg 100:472–479, 2004.

17. Lieberman DM, Laske DW, Morrison PF, Bankiewicz KS, Oldfield EH:Convection-enhanced distribution of large molecules in gray matter duringinterstitial drug infusion. J Neurosurg 82:1021–1029, 1995.

18. Liu H, Jacobs BS, Liu J, Prayson RA, Estes ML, Barnett GH, Barna BP:Interleukin-13 sensitivity and receptor phenotypes of human glial cell lines:Non-neoplastic glia and low-grade astrocytoma differ from malignantglioma. Cancer Immunol Immunother 49:319–324, 2000.

19. Morrison PF, Chen MY, Chadwick RS, Lonser RR, Oldfield EH: Focal deliv-ery during direct infusion to brain: Role of flow rate, catheter diameter, andtissue mechanics. Am J Physiol 277:R1218–R1229, 1999.

20. Morrison PF, Laske DW, Bobo H, Oldfield EH, Dedrick RL: High-flowmicroinfusion: Tissue penetration and pharmacodynamics. Am J Physiol266:R292–R305, 1994.

21. Parney IF, Kunwar S, McDermott M, Berger M, Prados M, Cha S, Croteau D,Puri RK, Chang SM: Neuroradiographic changes following convection-enhanced delivery of the recombinant cytotoxin interleukin IL13-PE38QQRfor recurrent malignant glioma. J Neurosurg 102:267–275, 2005.

22. Patel SJ, Shapiro WR, Laske DW, Jensen RL, Asher AL, Wessels BW,Carpenter SP, Shan JS: Safety and feasibility of convection-enhanced deliv-ery of Cotara for treatment of malignant glioma: Initial experience in 51patients. Neurosurgery 56:1243–1253, 2005.

23. Sampson JH, Akabani G, Archer GE, Bigner DD, Berger MS, Friedman AH,Friedman HS, Herndon JE 2nd, Kunwar S, Marcus S, McLendon RE, PaolinoA, Penne K, Provenzale J, Quinn J, Reardon DA, Rich J, Stenzel T, Tourt-Uhlig S, Wikstrand C, Wong T, Williams R, Yuan F, Zalutsky MR, Pastan I:Progress report of a Phase I study of the intracerebral microinfusion of arecombinant chimeric protein composed of transforming growth factor(TGF)-alpha and a mutated form of the Pseudomonas exotoxin termed PE-38(TP-38) for the treatment of malignant brain tumor. J Neurooncol 65:27–35,2003.

24. Sawaya R, Hammoud M, Schoppa D, Hess KR, Wu SZ, Shi W, Wildrick DM:Neurosurgical outcomes in a modern series of 400 craniotomies for treat-ment of parenchymal tumors. Neurosurgery 42:1044–1056, 1998.

25. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ,Belanger K, Brandes AA, Marosi C, Bogdahn U, Curschmann J, Janzer RC,Ludwin SK, Gorlia T, Allgeier A, Lacombe D, Cairncross JG, Eisenhauer E,Mirimanoff RO; European Organisation for Research and Treatment ofCancer Brain Tumor and Radiotherapy Groups; National Cancer Institute ofCanada Clinical Trials Group: Radiotherapy plus concomitant and adjuvanttemozolomide for glioblastoma. N Engl J Med 352:987–996, 2005.

26. Watanabe M, Tanaka R, Takeda N: Magnetic resonance imaging andhistopathology of cerebral gliomas. Neuroradiology 34:463–469, 1992.

27. Weaver M, Laske DW: Transferrin receptor ligand-targeted toxin conjugate(Tf-CRM107) for therapy of malignant gliomas. J Neurooncol 65:3–13, 2003.

28. Weber F, Asher A, Bucholz R, Berger M, Prados M, Chang S, Bruce J, Hall W,Rainov NG, Westphal M, Warnick RE, Rand RW, Floeth F, Rommel F, Pan H,Hingorani VN, Puri RK: Safety, tolerability, and tumor response of IL4-Pseudomonas exotoxin (NBI-3001) in patients with recurrent malignantglioma. J Neurooncol 64:125–137, 2003.

29. Westphal M, Hilt DC, Bortey E, Delavault P, Olivares R, Warnke PC, WhittleIR, Jääskeläinen J, Ram Z: A phase 3 trial of local chemotherapy with

biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients withprimary malignant glioma. Neuro Oncol 5:79–88, 2003.

30. Wilson CB: Glioblastoma: The past, the present, and the future. ClinNeurosurg 38:32–48, 1992.

31. Yung WK, Albright RE, Olson J, Fredericks R, Fink K, Prados MD, Brada M,Spence A, Hohl RJ, Shapiro W, Glantz M, Greenberg H, Selker RG, Vick NA,Rampling R, Friedman H, Phillips P, Bruner J, Yue N, Osoba D, Zaknoen S,Levin VA: A phase II study of temozolomide vs. procarbazine in patientswith glioblastoma multiforme at first relapse. Br J Cancer 83:588–593, 2000.

32. Yung WK, Prados MD, Yaya-Tur R, Rosenfeld SS, Brada M, Friedman HS,Albright R, Olson J, Chang SM, O’Neill AM, Friedman AH, Bruner J, Yue N,Dugan M, Zaknoen S, Levin VA: Multicenter phase II trial of temozolomidein patients with anaplastic astrocytoma or anaplastic oligoastrocytoma atfirst relapse. Temodal Brain Tumor Group. J Clin Oncol 17:2762–2771, 1999.

AcknowledgmentsCB (IL13-PE38QQR) is being developed through a cooperative research and

development agreement between NeoPharm and the laboratory of Raj K. Puri,M.D., Ph.D., at the Food and Drug Administration, Center for BiologicsEvaluation and Research. The views presented in this article do not necessarilyreflect those of the Food and Drug Administration. The clinical studies weresponsored by NeoPharm, Inc. John H. Sampson, M.D., Ph.D., received the fol-lowing support: National Institutes of Health/National Center for ResearchResources K23 RR16065 (Sampson); National Institutes of Health/NationalCancer Institute R01 CA097611 (Sampson); 2P50-NS20023 (Bigner/Sampson);Accelerate Brain Cancer Cure (ABC2) (Sampson); and National Cancer InstituteSpecialized Programs of Research Excellence in Brain Tumors Grants 2P50-NS20023 (Bigner/Sampson). The following people at NeoPharm played animportant role in the conduct of CB Phase I studies and preparation of this arti-cle: Drs. Jeanne Dul, Mahinda Karunaratne, Vincent Shu, Kevin Dahnert, SueForsythe, and Rima Veidemanis. The following people at the U.S. Food andDrug Administration, Center for Biologics Evaluation and Research, played animportant role in the development of IL13-PE38QQR: Bharat Joshi, Ph.D.,Pamela Dover, Koji Kawakami, M.D., Ph.D., Mariko Kawakami, M.D., Ph.D.,and Mitomu Kioi, D.D.S., Ph.D. The following investigators and key personnelparticipated in this study: University of California San Francisco, San Francisco,California (Michael Prados, M.D., Nicholas Butowski, M.D., and Mary Malec);MD Anderson Cancer Center, Houston, Texas (Kenneth Aldape, M.D., JeffreyWeinberg, M.D., Dima Suki, Ph.D., Lamonne Crutcher, Susan Graham); DukeUniversity, Durham, North Carolina (David Reardon, M.D., Allan Friedman,M.D., Henry Friedman, M.D.); Cleveland Clinic, Cleveland, Ohio (DebKangisser, PA-C, Gene Barnett, M.D., Steven Toms, M.D., M.P.H., Robert Weil,M.D., Lilyana Angelov, M.D., David Peereboom, M.D.).

COMMENTS

This is a Phase I study of the combination of cintredekin besudotox(CB) (IL13-PE38QQR) administered by convection-enhanced deliv-

ery in the initial treatment of malignant gliomas before radiotherapy(RT) and temozolomide (TMZ). The regimen appears safe in these22 patients. The study was done to prepare for the use of this combina-tion for the treatment of glioblastoma (GBM) in trials in the adjuvantsetting. These studies were at one time highly anticipated, but nowseem unlikely to happen.

Philip H. GutinNew York, New York

Vogelbaum et al. report the results of a Phase I study to determine thesafety and tolerability of CB in patients with newly diagnosed high-

grade astrocytomas. CB was administered by a convection-enhanceddelivery (CED) followed by standard-of-care external beam radiationtherapy with or without TMZ. Consistent with the results reported forCB used as monotherapy in the treatment of recurrent gliomas, the drugand delivery method seem to be safe and well-tolerated. Importantly, nosignificant systemic toxicities were documented.

NEUROSURGERY VOLUME 61 | NUMBER 5 | NOVEMBER 2007 | 1037

CONVECTION-ENHANCED DELIVERY OF CINTREDEKIN BESUDOTOX

The authors are to be congratulated for completing this multicenterclinical trial using a novel targeted toxin therapy and locoregionaldelivery method. This study highlights the central role neurosurgeonscontinue to play in the care of patients with malignant gliomas (1). It ishoped that a survival benefit will become apparent for this excitingtherapeutic approach with further testing.

Charles MatoukJames T. RutkaToronto, Canada

1. Kunwar S, Prados MD, Chang SM, Berger MS, Lang FF, Piepmeier JM,Sampson JH, Ram Z, Gutin PH, Gibbons RD, Aldape KD, Croteau DJ,Sherman JW, Puri RK; Cintredekin Besudotox Intraparenchymal Study Group:Direct intracerebral delivery of cintredekin besudotox (IL13-PE38QQR) inrecurrent malignant glioma: A report by the Cintredekin BesudotoxIntraparenchymal Study Group. J Clin Oncol 25:837–844, 2007.

The authors have presented the safety results of a Phase 1 studyusing a tumor-targeted recombinant CB for the treatment of newly-

diagnosed high-grade gliomas administered via a CED system. Thisreport highlights the safety profile of this treatment and an importanttechnique that involves a combination of 3 concepts most likely criticalto the effective treatment of cerebral glioma.

CED enables the tissue distribution of macromolecules into brainparenchymal space, with the possibility that the treatment agent willbe delivered to the infiltrating tumor cells responsible for the failureof present treatments. The use of a recombinant CB composed ofhuman IL13 and a truncated form of pseudomonas exotoxin A builtupon previous experimental studies showing that CB mediates cyto-toxicity by inhibiting protein synthesis and apoptosis leading to celldeath. The authors also note that the efficacy of CB seems to beenhanced by both radiation and TMZ. It is important that novel adju-vant therapies are tested early in the treatment of malignant gliomas,when they have the best chance of success, rather than waiting fortumor recurrence or end-stage disease when the treatment is doomedto failure. The authors are to be congratulated on this exciting study,which opens the possibility for a real advancement in the treatmentof cerebral glioma.

Andrew H. KayeMelbourne, Australia

Vogelbaum et al. present the results of a multicenter Phase 1 studyof CED of the immunotoxin IL13-PE38QQR, a fusion protein of

IL13 and truncated pseudomonas toxin, in 22 patients with newly diag-nosed malignant gliomas. The adjuvant study predominantly includedpatients (21 out of 22) with GBMs who underwent stereotactic catheterinsertion and IL13-PE38QQR CED after image-verified complete resec-tion. Four cohorts of patients were treated: two with conventionalexternal beam RT at two doses of IL13-PE38QQR (0.25 and 0.5 µg/ml)and two with conventional external beam RT and concurrent TMZ,again at two doses of IL13-PE38QQR (0.25 and 0.5 µg/ml). The trialwas associated with minimal toxicity; there was one episode of status

epilepticus occurring weeks after catheter placement and probablyunrelated to CED and one transient episode of dysphasia subsequentlyshown to be due to immunotoxin catheter backflow and intraparenchy-mal injury and subsequent cavitation. Several aspects of this reportmerit commentary.

The trial was designed to assess the feasibility of administering animmunotoxin by CED in an adjuvant setting whereby the experimen-tal therapy was intercalated (neoadjuvant) into the window betweeninitial surgery and RT with or without TMZ. The maximum tolerateddose of IL13-PE38QQR had been established in trials of recurrent GBMat 0.5 µg/ml. Clearly, the trial results indicate adding CED adminis-tered IL13 PE38QQR is both safe and feasible; the primary objectives ofthis Phase I study.

The value of CED remains uncertain. Two previous randomized tri-als (Precise and TransMID) failed to show a benefit relative to beststandard therapy. Clearly, proof of principle regarding the role of up-front treatment with immunotoxin CED will require a randomized trialwith sufficient statistical power to demonstrate benefit. Previous CEDtrials have suggested a major cause of lack of efficacy relative to thetechnical complexities of catheter placement either in tumors or in thebrain surrounding resected tumors. Catheter placement determinesdrug distribution of the infuscate into the brain interstitium and, fre-quently, catheters are malpositioned. At present, there is no easilyapplicable method to quantify or image drug delivery.

Immunotoxin-based CED therapy is by definition a targeted therapy.Trials with targeted agents ideally should be enriched for tumorsexpressing the target. In the present study of 22 patients, IL13 receptorstatus was evaluable in 19 patients of whom 17 were positive.Ostensibly, future IL13-PE38QQR CED trials would be designed as arationale inclusion criteria and include only patients with GBMs thatdemonstrate overexpression of the IL13 receptor.

The neuroradiographic evaluation of patients with GBM treatedwith any local therapy is complicated by regional treatment-relatedenhancement, presumably representing an inflammatory reaction thatcan be difficult to distinguish from a recurrent tumor. This situation isanalogous to the exaggerated reaction to concurrent TMZ and RT seenin some patients with GBM, making the differentiation betweenpseudoprogression and true progression problematic.

This Phase I study is rare in that it was designed and run by neuro-surgeons. The involvement of neurosurgeons in the design and imple-mentation of clinical neuro-oncology studies allows us to provide ourperspective, which is important, especially for trials that involve surgi-cal implantation of devices or drugs in the brain.

Some of the authors have financial and/or advisory relationshipswith NeoPharm, Inc., the company that manufactures and distributesthe toxin evaluated. As federal funding becomes more difficult toobtain, these industry-physician relationships provide a means of con-ducting trials, which are pivotal for advancement of our field.However, there are dangers in these relationships.

Marc C. ChamberlainNeuro-Oncologist

Daniel L. SilbergeldSeattle, Washington

1038 | VOLUME 61 | NUMBER 5 | NOVEMBER 2007 www.neurosurgery-online.com

VOGELBAUM ET AL.