Targeted radioimmunotherapy for embryonal tumor with multilayered rosettes

Abstract

Purpose:

We explored the use of intraventricular ¹³¹I-8H9 targeting B7-H3 in patients with ETMR.

Methods:

Patients were enrolled in an IRB approved, phase 1, 3+3 dose escalation trial. Patients with CNS disease expressing the antibody target antigen B7-H3, were eligible. We report on a cohort of 3 patients with ETMR who were enrolled on the study. Three symptomatic children (ages 14 months, 3 and 3.5 years) had large parietal masses confirmed to be B7-H3-reactive ETMR. Patients received 2 mCi ¹³¹I-Omburtamab as a tracer followed by 1 or 2 therapeutic ¹³¹I-Omburtamab injections. Dosimetry was based on serial CSF, blood samplings and region of interest (ROI) on nuclear scans. Brain and spine MRIs and CSF cytology were done at baseline, 5 weeks after ¹³¹I-Omburtamab, and approximately every 3 months thereafter. Acute toxicities and survival were noted.

Results:

Patients received surgery, focal radiation, and high dose chemotherapy. Patients 1 and 2 received ¹³¹I-Omburtamab (80 and 53 mCi, respectively). Patient 3 had a local recurrence prior to ¹³¹I-Omburtamab treated with surgery, external beam radiation, chemotherapy, then ¹³¹I-Omburtamab (36 mCi). ¹³¹I-Omburtamab was well-tolerated. Mean dose delivered by ¹³¹I-Omburtamab was 68.4 cGy/mCi to CSF and 1.95 cGy/mCi to blood. Mean ROI doses were 230.4 (ventricular) and 58.2 (spinal) cGy/mCi. Patients 1 and 2 remain in remission 6.8 years and 2.3 years after diagnosis, respectively; patient 3 died of progressive disease 7 month after therapy (2 years after diagnosis).

Conclusions:

¹³¹I-Omburtamab appears safe with favorable dosimetry therapeutic index. When used as consolidation following surgery and chemoradiation therapy, ¹³¹I-Omburtamab may have therapeutic benefit for patients with ETMR.

Introduction

Embryonal Tumor with Multilayered Rosettes (ETMR), C19MC Amplified is a new category of embryonal tumors within the 2016 World Health Organization (WHO) Classification of Tumors of the Central Nervous System (CNS). Previously called Embryonal Tumor with Abundant Neuropil and True Rosettes, Ependymoblastoma, and Medulloepithelioma, these tumors exhibit similar molecular characteristics and clinical behavior.¹

ETMR contains histopathologic features of neuroblastoma and ependymoblastoma, having undifferentiated neuroepithelial cells, neuropil, and ependymoblastic rosettes with a gene fusion between TTYH1 and C19MC and amplification of the microRNA cluster C19MC, identified by FISH analysis of the 19q13.42 locus.² Histologically similar tumors without C19MC alteration are classified as ETMR, Not Otherwise Specified. ETMR demonstrates immunohistochemical reactivity of the RNA binding protein, LIN28A².

Approximately 70 cases of ETMR have been reported, presenting in children less than 4 years old and having an aggressive clinical course with average overall survival (OS) of 12 months, rarely beyond 30 months^{1, 3–6}. Despite standard surgery, chemotherapy, and radiotherapy (RT), relapse rates remains high. Salvage chemotherapy, bevacizumab, autologous stem cell rescue, reresection and re-irradiation have all been tried. In one case series 18% of 38 patients with ETMR had CNS metastasis at diagnosis, necessitating treatment for distant disease, which has led to the use of craniospinal irradiation (CSI) as a potentially curative modality^{6, 7}. Due to the high risk of neurocognitive impairment that generally discourages physicians from prescribing CSI to patients under the age of 3 (which is precisely the age group of ETMR), other modalities of addressing CNS metastatic disease must be sought.⁸

We explored the use of radiolabeled intraventricular immunotherapy into the cerebrospinal fluid (CSF) for children with high risk or recurrent CNS malignancies on an institutional Memorial Sloan Ketterig Cancer Center (MSKCC) Institutional Review Board (IRB) study NCT00089245. Omburtamab is a murine IgG1 monoclonal antibody highly expressed on several CNS tumors but with limited normal tissue expression⁹. Omburtamab binds to the tumor antigen B7-H3 (CD276), an inhibitory ligand for natural killer cells and T cells¹⁰. ¹²⁴I-Omburtamab was demonstrated to be safely delivered via Convection Enhanced Delivery to patients with Diffuse Interstitial Pontine Glioma in a Phase I dose escalation trial without any Grade 4 Adverse Events or deaths¹¹. We assessed the role of intraventricular ¹³¹I- omburtamab for patients with ETMR.

Materials and methods

Patients with ETMR were enrolled on IRB approved study 03–133 (NCT00089245). Informed consent was obtained from patients or guardians. Eligibility criteria included B7-H3 reactivity by immunohistochemistry, stable neurologic status, no obstructive or symptomatic hydrocephalus, absolute neutrophil count >1000/μL, platelet count >50,000/μL, serum bilirubin <3.0mg/dL, and serum creatinine <2mg/dL, and adequate CSF flow, which was evaluated by pretreatment 111-indium diethylene triamine pentaacetic acid studies in Ommaya catheters.

Pre- and post-treatment evaluation included a detailed history, physical exam, complete blood count, comprehensive profile, thyroid function tests, and CSF for total protein, glucose, cell count, cytology. Magnetic resonance imaging (MRI) studies of the brain and spinal cord with and without gadolinium occurred within 3 weeks before and 1 month after ¹³¹I- omburtamab, followed by every few months.

Radiolabeling

Omburtamab clinical batches were tested for purity requirements, including absence of nucleic acids, murine viruses, bacteria, fungi, mycoplasma and pyrogens. Radiolabeling was performed at the MSKCC Radiochemistry and Molecular Imaging Probes Core Facility in compliance with the requirements of the Food and Drug Advisory (FDA) Investigational New Drug (IND) using iodination of ~1 mg of the antibody with up to 50 mCi of ¹³¹I (FDA IND #9351)¹². The final drug product was quality control tested to ensure conformance to the acceptance specifications defined in the Chemistry, Manufacturing, and Controls section of the IND. Sterility testing by media inoculation was performed post-release.

Dosimetry estimates

Patients received an imaging dose of 2 mCi ¹³¹I-8H9, followed by CSF and blood sampling, and 3 single-photon emission computed tomography (SPECT) scans over 48 hours to assess dosimetry. Distribution and activity concentrations of radioactivity in the craniospinal axis, and radiation doses to normal tissues were determined. Radiation exposure of CSF, ventricles, spinal cord, normal brain and blood was based on the assumption of complete local absorption of the ¹³¹I beta radiation. Measured aliquots were counted to estimate the time-dependent activity concentrations and to yield the decayed area under the curves, representing the cumulative activity concentrations in the blood and CSF as previously described.¹³

¹³¹I-Omburtamab therapy

Patient were premedicated with oral SSKI drops, Cytomel, acetaminophen and low dose decadron prior to injections. In the presence of a Radiation Safety Officer, Nuclear Medicine and Pediatric Oncology clinician, patients received 50-to-80 mCi ¹³¹I- omburtamab based on age and corresponding CSF volume. Patients had CNS (head and upper spinal cord) imaged by ¹³¹I-8H9 scintigraphy approximately 3–6 hours, 18–24 hours and 44–49 hours after dosimetry dose injection of the first cycle. Whole body scans were performed at the same time post injection. Patients were observed several hours after injection, daily for 2 more days, then weekly for 3 weeks. Toxicity was defined by the Common Terminology Criteria for Adverse Events (CTCAE), Version 3.0, observed over 5 weeks. MRI of the brain and the spine was performed at week 5 after treatment dose. Neurocognitive function and quality of life testing occurred starting at approximately 3 months after the last treatment dose.

Data availability

De-identified patient data will be shared at the discretion of the sponsor of the trial (YMabs, Therapeutics, Inc.) who is also the holder of the IND. Data in addition to that provided in this report that would be considered for sharing upon request are doses of drug administered, and pertinent radiographic images. No statistical analysis is planned for these 3 subjects of this report.

Patient 1

A 3 year old boy presented with headaches, ataxia and behavior changes. A left parietal mass measured 8.4 x. 6.5 x.7.2 cm with mass effect was emergently resected, confirmed to be ETMR (C19MC status unknown). There was no evidence of metastatic disease. He received 3 cycles of methotrexate, vincristine, etoposide, cyclophosphamide, and cisplatin followed by 3 cycles of consolidation (carboplatin and thiotepa) with autologous stem cell rescue (as per clinicaltrials.gov NCT00002515). Diffuse fungal lung and spleen lesions curtailed the full regimen. During treatment, the patient developed status epilepticus and MR showed local tumor recurrence. Tumor histology showed ganglion cell maturation of ETMR. He was then treated with serial injections of ¹³¹I-3F8 without complication followed by ¹³¹I-Omburtamab. After the 2 mCi ¹³¹I-Omburtamab dosimetry dose, he suffered from rotavirus, parainfluenza and grade 3 transaminitis related to infections so he was temporarily removed from the trial. The patient was treated with CSI and RT to the primary tumor bed (total dose = 54Gy), and resumed ¹³¹I-Omburtamab (2 mCi and 80 mCi) without incident.

Patient 2

A 3 year old girl presented with seizures; MRI brain revealed a large parietal lesion. Following complete resection, pathology confirmed ETMR; cytogenetic report showed amplification of the C19MC/MIR-371–373 region (19q13.42) by FISH analysis. She was treated with multi-agent chemotherapy including doxorubicin, cisplatin, vincristine, cyclophosphamide, etoposide, intraventricular methotrexate and focal proton beam RT (total dose =54Gy)¹⁴. Nine months after diagnosis she received 2 mCi and 50 mCi ¹³¹I-Omburtamab, treatment complicated by grade 4 prolonged thrombocytopenia.

Patient 3

A 3-year old boy presented with seizures from a right parietal tumor. He underwent a complete resection; CSF cytology was unremarkable; pathology showed ETMR, C19MC status unknown. He was treated with focal RT (45 Gy), multiagent chemotherapy (vincristine, etoposide, cyclophosphamide, and cisplatin, and twice with carboplatin and thiotepa with stem cell rescue. He suffered a local recurrence 11 months after diagnosis with further progression through retrieval chemotherapy. A second resection confirmed recurrent ETMR. One month later, he had a third local progression and again underwent subtotal resection and additional focal proton RT (36 CGE). He then received 2 and 50 mCi ¹³¹I-omburtamab. Post therapy MRIs demonstrated continued local progression of disease.

Results

Intraventricular ¹³¹I-Omburtamab was well tolerated in these 3 patients with ETMR.(Table 1). A high therapeutic index to CSF vs blood was achieved in all patents. Mean ROI doses were 230.4 to the ventricles (range 147.8–392 cGy/mCi) and 58.1 (range 43.1–81.7 cGy/mCi) to the spine. (Figure 1). At last follow-up, Patient 1 is 6.8 years after diagnosis and remains in remission; long term complications include cataracts, hypothyroidism, and growth hormone deficiency. Patient 2 is 2.3 years after diagnosis and remains in clinical and radiographic remission. Patient 3 continued to progress through therapy and died 7 months after therapy, 2 years after initial diagnosis.

Table 1.

Patient characteristics and results of treatment

Age at diagnosis	Gender	Age at start of 131I-8H9	Number of Relapses Prior to 131I-8H9	Status when treated with 131I-8H9	CSF cGy/mCi	Blood	ROI Ventricle	ROI Spine	Months to Relapse after 131I-8H9	Status at last follow-up
3y	Male	4y 7mo	0	Radio-graphic/cytologic remission	40.9	2.1	170	50.3	-	NED
3y	Female	3y 10mo	0	Radio-graphic/cytologic remission	22.5	3	147.8	81.7	-	NED
1y	Male	2y 8 mo	3	Minimal post-opresidual	179	2.3	392	57.6	1	DOD after local relapse

CSF, cerebrospinal fluid; 131-I-Omburtamab, iodinated 8H9 (Omburtamab); mo, months; OS, overall survival; ROI, region-of-interest; y, years; NED, no evaluable disease; DOD, died of disease.

Figure 1:

SPECT analysis at 4 (A) and 24 hours (B) after injection demonstrating activity of ¹³¹I-Omburtamab throughout the thecal sac.

Discussion

The cure of young children afflicted with ETMR remains a daunting challenge. Radiolabeled monoclonal antibodies targeting micrometastases are a promising therapy for brain tumors.

B7-H3 expression is significantly associated with poor outcome in several cancers, and is uniquely overexpressed in cancers compared to normal human tissues. Here we present mature data of a well-tolerated compartmental radiolabeled anti- B7-H3 monoclonal antibody for an incurable pediatric embryonal tumor. We describe the prolonged survival of 3 children treated with intraventricular ¹³¹I-Omburtamab, 2 of whom remain in clinical and radiographic remission at 6.8 and 2.3 years after completion of treatment. This is in stark contrast to published literature that shows an average OS of 12 months after diagnosis for patients with ETMR^{1, 3–6}.

In this trial, there was no anticipated use of omburtamab in these patients, thus there were no limits of radiation doses of external beam RT before or after radioimmunotherapy. We have previously published the cumulative radiation doses administered to patients with other CNS tumors who received combined external beam RT and radioimmunotherapy. In that group, the radioimmunotherapy radiation is delivered by beta emission targeted 1–2mm. There does not appear to be a cumulative effect with the radiation delivered by combined external beam RT and radioimmunotherapy¹⁵. However we can, as cited, reference the radioimmunotherapy dose delivered to the CSF in cGy / mCi and CSI¹⁵.

We highlight the safety of radioimmunotherapy even in this young patient cohort. Specifically, there is an absence of radionecrosis seen in our three patients. Our prior retrospective analysis of patients with metastatic CNS neuroblastoma and medulloblastoma demonstrated the safety of using intrathecal radioimmunotherapy in patients who previously received radiotherapy without increasing the rate of radionecrosis, and neither of our patients has demonstrated evidence of radionecrosis¹⁵. Patient 1 was noted to have primary hypothyroidism diagnosed several years after completion of treatment. Given that he had 23Gy of radiation exposure to the hypothalamic pituitary axis from RT, it cannot be concluded that radioimmmunotherapy caused hypothyroidism. Neither patient 2 nor 3 had hypothyroidism following radioimmunotherapy. There were no changes on MRI that were attributable to receiving both RT and radioimmunotherapy.

Tumor cell cytotoxicity is attributed to the direct effect of ¹³¹I-radiation, although it is possible that a secondary mechanistic basis for successful eradication of microscopic tumor cells may in part be due to complement activation with the CSF space. Although there was interpatient variability in the dose delivered to the CSF, a high therapeutic ratio was achieved. Determining the optimal cGy delivered to the CSF by to fully eradicate micrometastatic deposits for this patient population and incorporating this into upfront treatment regimens is an ongoing focus of research.

Conclusions

¹³¹I-omburtamab appears safe with favorable dosimetry therapeutic index in ETMR. When used as consolidation following surgery and chemoradiation therapy, ¹³¹I-omburtamab may have therapeutic benefit for patients with ETMR.

Figure 2:

Patient 1 with tumor at diagnosis (A) and at most recent follow up (B).