Pre-irradiation Intensive Induction and Marrow-ablative Consolidation Chemotherapy in Young Children with Newly Diagnosed High-Grade Brainstem Gliomas: Report of the “Head-Start” I and II Clinical Trials

Abstract

BACKGROUND:

The dismal outcome in children with high-grade brainstem gliomas (BSG) accentuates the need for effective therapeutic strategies. We investigated the role of intensive, including marrow-ablative, chemotherapy regimens in the treatment of young children with newly-diagnosed high-grade BSG.

METHODS:

Between 1991-and-2002, 15 eligible children less than 10 years of age with a diagnosis of high-grade BSG were treated on “Head-Start” I and II protocols (HSI and HSII). Treatment included Induction with 4-5 cycles of one of three intensive chemotherapy regimens followed by Consolidation with one cycle of marrow-ablative chemotherapy (thiotepa, carboplatin and etoposide) with autologous hematopoietic cell rescue (AHCR). Irradiation was required for children over 6 years of age or for those with residual tumor at the end of Consolidation.

RESULTS:

We had two long-term survivors who were found retrospectively to harbor low-grade glial tumors and thus were not included in the survival analysis. Of the remaining 13 patients, the 1-year event-free (EFS) and overall (OS) survival for these children were 31% (95% CI: 9-55%) and 38% (95% CI: 14 to 63%), respectively. Median EFS and OS were 6.6 (95% CI: 2.7, 12.7) and 8.7 months (95% CI: 6.9, 20.9), respectively. Eight patients developed progressive disease during study treatment (seven during Induction and one at the end of Consolidation). Ten children received focal irradiation, five for residual tumor (three following Induction and two following Consolidation) and five due to disease progression.

CONCLUSIONS:

Children with high-grade BSG did not benefit from this intensive chemotherapy strategy administered prior to irradiation.

INTRODUCTION

Ten to 15% of childhood intracranial tumors are located in the brainstem and the majority of these tumors are gliomas.¹ Eighty-percent of gliomas are high-grade and diffusely involve the pons, known as a diffuse intrinsic pontine glioma (DIPG). The remaining 15-20% are low-grade gliomas and arise primarily in the midbrain and medulla.^2,3 The majority of children are diagnosed between 5 and 10 years of age with an equal predilection for males and females. Children present with a short period of symptoms between 2 weeks to 2 months prior to their diagnosis, typically with gait imbalance, dysmetria, dysarthria, long-tract signs and/or 6^th, 7^th cranial nerve palsies.⁴

DIPGs are one of the only brain tumors diagnosed by characteristic findings seen on magnetic resonance imaging (MRI). These tumors are widely infiltrative showing diffuse expansion of more than 50% of the pons and basilar artery encroachment. They are hypo- or iso-intense on T1-weighted images and hyperintense on T2-weighted MRI, with minimal to no-enhancement with gadolinium.⁵

Prior to the current molecular diagnostic era, surgical biopsy was rarely performed due to the potential for peri-operative complications without providing a significant rationale for treatment assignments.⁴ Recent trials have brought tumor biopsy at diagnosis back to the forefront to molecularly profile these tumors.

Nonetheless, children with high-grade BSG continue to have a dismal prognosis, with a median OS of less than 1 year, and fewer than 20% of children are alive beyond 2-years from diagnosis.^6,7 Radiation therapy is the standard treatment of these tumors, which at best provides temporary neurological and radiographic improvement.^8,9 Various strategies for delivering radiation therapy (i.e.: hypo- or hyperfractionated irradiation +/−sensitizers) have failed to improve OS for these children, at times with unacceptable toxicities.^10–18

The “Head-Start” prospective, multi-center chemotherapy trials have been utilized in infants and children to treat highly aggressive brain tumors in young children with irradiation-avoiding strategies to prevent irradiation-induced deleterious side effects and hopefully improve survival. In this manuscript we report the impact upon survival of two prospective, sequentially conducted “Head-Start” trials on children with high-grade BSG. At the time of initiating this protocol, there were no clinical precedents exploring intensive chemotherapy regimens in DIPG, especially in very young patients. The delay in completion of this manuscript for publication reflected the difficulty in retrospectively retrieving original MRI studies from archival records and several different institutions to perform centralized confirmation that tumors represented true diffuse intrinsic pontine gliomas versus other, more focal or atypical brainstem tumors.

PATIENTS AND METHODS

Patient Selection

Children who were less than 10-years of age with newly diagnosed previously untreated tumors, with radiographic and clinical characteristics of DIPG were enrolled on one of the following non-randomized, prospective, multi-institutional studies, HSI (1991-1997) or HSII (1997-2002).

An MRI of the brain with-and-without gadolinium was required at diagnosis but an MRI of the spine and lumbar cerebrospinal fluid cytology was not mandated. There was no central radiology review prior to enrollment, diagnostic MRI characteristics were defined per institutional radiology review and treating clinician. Chemotherapy was required to begin within 21-days of the diagnostic MRI and/or any definitive surgical diagnostic procedure. Enrollment criteria included no prior exposure to chemotherapy or irradiation and adequate hepatic and renal function.

Surgery and Pathology

Children with presumed BSG were eligible for entry without tumor biopsy provided the tumors demonstrated the characteristic radiological features of DIPG. Patients with largely exophytic BSG or localized to the midbrain or cervico-medullary junction required surgical confirmation of a glioma grade 2-4. The degree of surgical resection was determined by central review of operative reports and of the pre-operative and post-operative MRI reports. If biopsy or resection was performed, tumor pathology was centrally reviewed retrospectively by neuropathologists on the study committee.

Chemotherapy Regimens

The treatment plan included one of these three Induction Regimens followed by one cycle of the Consolidation Regimen. INDUCTION REGIMEN A (HSI) consisted of 5 cycles of chemotherapy at 21-day intervals with cisplatin (3.5mg/kg/dose intravenously (IV) on day 1), cyclophosphamide (65mg/kg/day and mesna on days 2 and 3), etoposide (4mg/kg/day IV on days 2 and 3), vincristine (0.05mg/kg/dose IV on days 1, 8 and 15 of cycles 1 to 3); INDUCTION REGIMEN B (HSII before amendment to Regimen C below) which consisted of 4 cycles of chemotherapy at 28-day intervals with procarbazine (10mg/kg/day orally once daily on days 0, 1, 2, 3 and 4), carboplatin (20mg/kg/day IV on days 3 and 4), lomustine (3.5mg/kg/day orally on days 3 and 4), vincristine (0.05mg/kg/dose IV on days 1, 8 and 15 of cycles 1 to 3); or REGIMEN C (HSII after amendment) which consisted of 4 cycles at 28-day intervals with carboplatin (20mg/kg/day IV on days 3 and 4 or an AUC of 8 using the Calvert Formula dosing¹⁹), temozolomide (6.5mg/kg/day orally once at bedtime on days 1 to 5) and vincristine (0.05mg/kg/dose IV on days 1, 8 and 15 of cycles 1 to 3, max dose 2mg). Patients without progressive tumor growth at the completion of Induction chemotherapy were eligible to proceed with one cycle of the CONSOLIDATION REGIMEN which consisted of carboplatin (calculated using the Calvert formula for a desired area under the curve of 7mg/ml/min on days −8 through −6 to a maximum of 500 mg/m2/day), thiotepa (300mg/m²/day or 10mg/kg/day on days −5 through −3), and etoposide (250mg/m²/day on days or 8.3mg/kg/day −5 through −3). (See Table 1)

Table 1.

Treatment-Induction and Consolidation Chemotherapy Regimen

Part 1: INDUCTION REGIMEN
Regimen A (HSI)	5-cycles at 21 days interval of: Cisplatin 3.5mg/kg IV on day 1 Etoposide 4mg/kg/day IV on days 2 and 3 Cyclophosphamide 65mg/kg/day and mesna on days 2 and 3 Vincristine 0.05mg/kg IV on days 1, 8 and 15 of cycles 1 to 3
Regimen B (HSII before amendment)	4-cycles at 28 days interval of: Procarbazine 5mg/kg orally once daily on days 0, 1, 2, 3 and 4 Lomustine 3.5mg/kg orally on days 3 and 4 Carboplatin 20mg/kg IV on days 3 and 4 Vincristine 0.05mg/kg IV on days 1, 8 and 15 of cycles 1 to 3
Regimen C (HSII after amendment)	4-cycles at 28 days interval of: Carboplatin at AUC of 8mg/ml/min based on pediatric Calvert formula31 IV on days 1 and 2 Temozolomide 6.5mg/kg orally once at bedtime on days 1 to 5 Vincristine 0.05mg/kg IV on days 1, 8 and 15 of cycles 1 to 3
Part 2: CONSOLIDATION REGIMEN
Enrolled in HS I and II trials	Carboplatin at AUC of 7 mg/ml/min based on pediatric Calvert formula31on days −8 through −6 Thiotepa 300mg/M2/day or 10mg/kg on days −5 through −3 Etoposide 8.3mg/kg or 250mg/M2/day on days −5 through −3

Filgrastim (5μg/kg/day) was to be administered subcutaneously 24-hours after the last dose of chemotherapy with each Induction cycle and continued until recovery of the neutrophil count. Induction chemotherapy was to be discontinued if there was evidence of disease progression or if unacceptable toxicity occurred. Bone marrow or autologous peripheral blood hematopoietic cell harvesting was to be undertaken upon recovery from Induction cycles 1 and/or 2.

Extent of disease evaluation was to be performed following completion of two cycles of Induction and following completion of the entire Induction phase of chemotherapy. Any patient who had radiographically stable to better residual tumor was eligible to proceed to Consolidation provided the patient was not corticosteroid dependent. If this criterion could not be met, the patient was considered for off-study therapy. Any patient who developed radiographic evidence of tumor progression at any time was also not eligible to proceed and was considered for off-study therapy.

Seventy-two hours after completion of the Consolidation Regimen, bone marrow or leukapheresed peripheral blood hematopoietic cells (PBHC) were to be re-infused (Day 0). A minimum cell dose of 2 x 10⁶ CD34⁺ cells/kg was required for successful re-infusion. All patients were to receive filgrastim 5μg/kg/day starting 24 hours after hematopoietic cell reinfusion and continued until there was neutrophil engraftment.

After complete recovery from Consolidation, patients were to receive radiotherapy if they were older than 6 years of age at the time of diagnosis and/or had residual tumor at the completion of Induction chemotherapy. Irradiation was to be initiated no earlier than day +42 following hematopoietic cell reinfusion. Patients were to receive focal irradiation to the tumor bed plus a 2.5 cm margin to a total dose not to exceed 59.4 Gy.

After completion of therapy, patients were to undergo brain MRI scans with-and-without gadolinium once every 3 months for the first 2 years, once every 6 months for the next 3 years and annually thereafter.

Response and Toxicity Criteria

Pre-study MRI scans were compared with studies performed after completion of Induction chemotherapy and approximately 6 weeks after Consolidation chemotherapy. Tumor size was estimated as the product of the greatest diameter and its perpendicular. Responses were categorized as follows: progressive disease (PD), greater than 25% increase in tumor size or the appearance of new area of tumor; partial response (PR), greater than 50% decrease in tumor size; minor response (MR), between 25% and 50% decrease in tumor size; complete response (CR), complete disappearance of all previously assessable tumors; stable disease (SD), less than 25% decrease in tumor size Toxicity criteria were graded according to the Common Terminology Criteria for Adverse Events version 3.0.

Study Evaluations

Complete blood cell counts and biochemical screening profiles were to be obtained at prescribed intervals. Audiograms or brainstem auditory-evoked potentials were to be performed before each of the Induction cycles, before Consolidation, at 3 months and 1 year post-Consolidation and annually thereafter.

Supportive Care

Platelet counts were to be maintained above 30,000/μL and hemoglobin was to be maintained above 8.0 g/dL. Febrile neutropenic patients were to be treated with broad-spectrum intravenous antibiotics and antifungal agents when appropriate. Patients were to receive appropriate prophylaxis for pneumocystis jerovecii pneumonia.

Informed Consent

Parents or legal guardians for each child signed an informed consent approved by the Institutional Review Board (IRB) or equivalent committee at each treating center.

Statistical Considerations

The primary aim from these prospective HSI and HSII trials was to determine EFS and OS of these patients from the time of study enrollment and after administration of the intensive chemotherapeutic regimen. Primary endpoints for analysis were: 1) EFS – the time from date of diagnosis to progression, relapse or death from any cause; 2) OS – the time from date of diagnosis to death; and 3) irradiation-free survival – the time from date of diagnosis to death or the receipt of radiotherapy for any reason. These were estimated using the Kaplan-Meier method and presented using corresponding 95% confidence intervals. Patient characteristic and response data were summarized using descriptive statistics.

RESULTS

Patient Characteristics

Between June 1992 and August 2002, a total of 15 children less than 10 years of age with newly diagnosed high-grade BSG were enrolled on either HSI or HSII. See Table 2 for more demographic information.

Table 2.

Demographics of the Study Population (15 patients) including the two long-term survivors.

			Regimen
		Total	A	B	C
Number of Patients		15	6	3	6
Age (years)
	0.0<1.5	1	0	0	1
	1.5<3.0	3	2	1	0
	3.0<6.0	8	4	1	3
	6.0<10.0	3	0	1	2
Sex
	Female	8	3	1	4
	Male	7	3	2	2
Race
	Asian	1	0	0	1
	African	4	2	1	1
	Hispanic	0	0	0	0
	White	9	4	1	4
	Unknown	1	0	1	0

Diagnostic/Imaging Work-up

Of the 15 patients, two underwent a biopsy, and two patients had partial resections. Pathology in three of these four cases was consistent with a high-grade glioma (glioblastoma multiforme (GBM), malignant glioma and anaplastic astrocytoma) confirmed upon central pathological review; one patient however, was ultimately deemed to be a pilocytic astrocytoma upon central review following completion of study therapy and was therefore ultimately not included in the overall survival analysis. The majority of the children were, by institutional report “classic” DIPGs. Six of the 15 patients underwent MRI of the total spine and had no evidence of disseminated disease.

At the time, central review of MR imaging studies was made available following completion of all study therapy. Of the two patients who proved to be long-term survivors, one patient’s tumor was located in the lower pons and upper medulla, and at diagnosis was diffusely enhancing suggestive of a high-grade glioma. It was only after a period of 6 years did we note that the tumor no longer demonstrated any contrast enhancement and was described only by a fusiform enlargement of his lower pons and medulla and was also not included in the final survival analysis. The second patient, who underwent biopsy had a low-grade glioma confirmed on central review with MR imaging findings not typical of a DIPG. The patient who underwent a partial resection at diagnosis of a pathologically confirmed GBM, had MRI findings typical of a high-grade exophytic BSG with irregular gadolinium enhancing areas. The patient with a confirmed brainstem anaplastic astrocytoma had classic DIPG findings on MRI.

Response to Induction Chemotherapy

Eight of the 15 children (53%) completed Induction (Regimen A: n=4 of 6, Regimen B: n=1 of 3, Regimen C: n=3 of 6) the remaining seven (Regimen A: n=2; Regimen B: n=2; Regimen C: n=3) developed progressive disease three of which proceeded directly to irradiation.

Three patients all treated on Regimen C, achieved partial responses. One patient, treated on Regimen A, achieved a minor response. Four patients had stable disease at the end of Induction therapy (Regimen A: n=3, Regimen B: n=1). All three patients from Regimen A with stable disease at the end of Induction discontinued study therapy due to large residual disease and proceeded to receive focal irradiation, one of whom (ultimately deemed ineligible due to pathology of pilocytic astrocytoma) became a long-term survivor.

Response to Consolidation Chemotherapy

Five of the 15 children with BSG proceeded to Consolidation, of whom three maintained stable disease for 5 months, 7 months and one other long-term survivor (ultimately reviewed radiographically as a low-grade medullary tumor). One patient achieved partial response following Consolidation and survived for two years before expiring of tumor progression. One patient developed progressive disease two days into the conditioning regimen and removed from the study.

Toxicity

Toxicities observed were similar to those observed for other children enrolled on HSI and HSII Regimens as previously reported.^4,20 No patients were removed off-study or had dose reductions because of chemotherapy-related persistent grade 3 or 4 toxicities. There were no toxic deaths. One long-term survivor who received craniospinal irradiation (36 Gy) after what deemed to be a recurrence after completion of HSII therapy developed secondary growth-hormone and thyroid deficiency and resides in a group home at 23 years of age as of his last follow up.

Event-Free Survival, Overall Survival and Irradiation-Free Survival

Three of the four patients who completed Induction and Consolidation therapy ultimately developed progressive disease and expired. There were two long-term survivors who were found to have low-grade gliomas and were ultimately excluded from the survival analysis. One of these survivors never received irradiation post-Consolidation and is a long-term survivor (in excess of 14 years) and was ultimately deemed to have a low-grade glioma of the medulla on central imaging review. He developed tumor progression 18 months following completion of Consolidation, and received maintenance treatment with Thalidomide (3 years) and Celebrex (1 year). The one other long term survivor (in excess of 14 years); completed Induction but proceeded directly to irradiation without Consolidation because of a large residual was ultimately deemed ineligible for the study on the basis of central pathology review favoring a pilocytic astrocytoma.

For all patients eligible for survival analysis (n=13 of 15 enrolled), the 1-year and 2-year EFS were 31% (95%-CI: 9-55%), 0%, respectively; and 1-year and 2-year OS were 38% (95% CI: 14-63%) and 7.7% (95% CI: 0.4 to 29.2%); respectively, with a median EFS and OS of 6.6 and 8.7 month (Figure 1).

Figure 1.

Kaplan Meier Analysis of OS and EFS

OS: 38% @ 1 year (95% CI: 14 – 63%), 7.7% @ 2 year (95% CI: 0.4 to 29.2%); median survival = 8.7 months (95% CI: 6.9, 20.9)

EFS: 31% @ 1 year (95% CI: 9 – 55%), 0% @ 2 years ; median survival = 6.6 months (95% CI: 2.7, 12.7)

DISCUSSION

The clinical outcome for patients with DIPG remains dismal. Treatment strategies employed over the last three decades for these children have continually failed to prolong their survival. Indeed, with the progressive implementation of advanced imaging techniques, the outcomes of children with DIPG in successive cooperative group clinical trials appears to have worsened, due to exclusion of patients with lower-grade gliomas by MRI criteria.^10,21–23 Focal irradiation remains the only modality that provides transient radiological responses and symptom improvement, but without providing cure. Cohen et al ACNS0126 study demonstrated no apparent survival benefit to the addition of temozolomide having a mean 2-year EFS and OS are 1.7% ± 1.7% and 3.6%± 2.5%, respectively¹⁸.

In recent years, classification of these diffusely infiltrating gliomas has integrated genetic and epigenetic characteristics ²⁴ with clinical-pathological characteristics; these were unknown at the time HSI and HSII were launched, and are therefore outside of the scope of this paper. The role of chemotherapy remains unproven for children with DIPG. The results of our HSI and HSII trials are in concordance with this, and demonstrate that in spite of utilizing the most intensive chemotherapy regimens available, in addition to irradiation, no survival advantage was observed.

We have compared BSG outcomes from our HS trials with other trials where chemotherapy was administered prior to irradiation (Table 3).^{21,22,25–28} In 1993, Duffner et al demonstrated benefit of chemotherapy in delaying irradiation in a small proportion of infants with malignant brain tumors. Frappaz et al delayed irradiation until the time of clinical progression with intensive Induction chemotherapy, including high-dose systemic methotrexate (BSG98-Study). Although the median OS was increased by 8 months, patient quality of life was impaired.²⁸ Gokce-Samar et al compared the BSG-98 strategy with targeted therapy approaches including erlotinib, cilengitide and nimotozumab after stereotactic biopsy and reported an increase in median progression-free survival (3 months versus 8.6 months) for the BSG-98 group and a significant increase in median OS (8.8 months versus 16.1 months, p = 0.0003)²⁹. Other pre-irradiation chemotherapy strategies have failed to show improved EFS and OS.^22,26,27 In addition to the pre-irradiation chemotherapy strategies detailed above, numerous efforts over the years have attempted to use chemotherapy during irradiation, but this also failed to provide any additional survival benefit.^11–18

Table 3.

Response to Pre-irradiation Chemotherapy of Published Studies Including Newly-diagnosed High-grade BSG; Carbo-carboplatin; CCNU-lomustine; CDDP-cisplatin; CPM-cyclophosphamide; Tamox-Tamoxifen; VP-16-etoposide; VCR-vincristine; FHF: focal hyperfractionated; FRT-focal involved-field conventional radiation therapy ranging from 50.4-59.4 GY; RT Radiation therapy; CR-complete responses; MR-minimal responses; OR-total objective response; PD-progressive disease; PR-partial responses; SD-stable disease; m-months; N= total number of radiologically evaluable patients

				Response to chemotherapy
Study	N	Chemotherapy Regimen	Post-Chemotherapy RT	CR	PR	MR	SD	PD	OR%	Median Survival (OS)
Duffner et al.24	8	Reg A: CPM/VCR Reg B: CDDP/VP-16 (AABAAB sequence)		0	0	0	4	4	0	-
Jakacki et al.12	6	4-cycles: CCNU/VCR/Procarbazine	FRT	0	3	1	1	1	50	13m
Kretschmar et al.27	37	4-cycles: CDDP/CPM	FHF 66Gy	0	3	3	20	6	8.1	9m
Doz et al.25	21	2-cycles: Carbo x 3days	FRT+Carbo	0	0	2	6	13	9.5	11m
Jennings et al.28	27	3-cycles: Carbo/VP-16/VCR	FHF 72Gy	0	2	1	12	12	11	-
Jennings et al.28	22	3-cycles: CDDP/VP-16/CPM/VCR	FHF 72Gy	0	1	4	8	9	22.7	-
Frappaz et al.29	23	Cycle-1: Tamox/BCNU/CDDP; cycle-2&3 :high-dose MTX	FRT+Hydroxyurea/tamoxifen	-	-	-	-	-	-	17m
Osorio et al. This study*	15	HS I and II	Focal, no greater than 59.4Gy	0	0	0	7	8	-	8.7m

^*Chemotherapy responses after completion of induction +/−consolidation to reflect response to entire chemotherapy regimen.

The application of marrow-ablative chemotherapy in children with high-grade gliomas was based on the premise that higher doses of chemotherapy could overcome the physiological blood-brain-barrier as well as resistance, thereby enhancing both delivery and efficacy of these cytotoxic agents to these tumors.^12,30

There is evidence for a steep-dose response curve of brain tumors to chemotherapy.³¹ Initial studies using single-agent carmustine in high doses followed by marrow rescue in patients with high-grade gliomas reported tumor responses to this agent with longer duration of survival. ^32–34 Subsequent conditioning regimens added thiotepa and etoposide to BCNU; however this combination resulted in significant toxicities and increased treatment-related mortality.³⁵ Therefore, BCNU was subsequently replaced by carboplatin.

Studies have shown that thiotepa readily penetrates the brain parenchyma due to its lipid solubility making it an ideal drug of choice to treat brain tumors.^36–40 Thiotepa has been used in combination with other drugs at marrow-ablative doses in treating patients with recurrent and newly diagnosed high-grade glioma.^41,42 Given the modestly encouraging outcomes of marrow-ablative chemotherapy with thiotepa and etoposide-containing regimens in the treatment of recurrent high-grade astrocytomas in locations other than the brainstem, this was also utilized in patients with high-grade BSG.^43–47 Marrow-ablative chemotherapy has been explored for relapsing or newly diagnosed high-grade BSG either in pre-irradiation or post-irradiation schedules, with the aim of tumor reduction before delivering radiotherapy or as radiotherapy consolidation (Table 4).^{35,43,48–50}

Table 4.

Outcome of Myeloablative Chemotherapy in High-grade BSG from Published Clinical Trials. B-BCNU; Bu-busulphan; C-carboplatin; CPM-cyclophsphamide; DF-disease free; E-etoposide; EFS-event-free survival; N-total number of radiologically evaluable patients; NA-not available; OS-overall survival; RT-radiation therapy; T-thiotepa; m-months; WD-with disease

Recurrent or Refractory High-grade BSG
Study	N	Prior therapy	Conditioning Regimen	Post-RT	Median survival	1-year EFS/OS	2-year EFS/OS	Long-term survivor
Papadakis et al.57	3	RT & chemotherapy	BTE	No	1.1m (0.4-18.7m)	NA	NA	none
Finlay et al.48	6	RT & chemotherapy	TE	No	4.9m (1.6-23.6m)	7%	0%	none
Dunkel et al.55	10	RT & chemotherapy	TE(6); BTE(2); CTE(2)	FHF 72-78 Gy	4.7m (0.1-18.7m)	NA/10±7%	NA/0%	none
Newly-Diagnosed High-grade BSG
Study	N	Induction Chemo	Conditioning Regimen	Pre/Post-RT	Median Survival	1-year EFS/OS	2-year EFS/OS	Long-term survivor
Papadakis et al.57	8	none	BTE	Post RT	14.1m (7.6-46m)	NA	NA	none
Kedar et al.54	6	none	CPM/T	Post: FHF 70-75 Gy	12.5m (77d-24+m)	NA	NA	2 (1 DF/1WD)
Dunkel et al.55	5	none	BTE	FHF 72-78 Gy	11.4m (7.6-17.1m)	NA/50±18%	NA/0%	none
Bouffet et al.56	23	none	BuE	Pre: FRT 50-55 Gy	10m (7.4-12.6m)	7%/7%	0%/7%	none

CONCLUSIONS

The results from the HSI and HSII trials for BSG are disappointing and show similar poor EFS and OS compared with previously reported studies (Table 4). Based on these data, we no longer enroll children with BSG on the “Head-Start” clinical trials (HSIII and HSIV). This therapeutic strategy should not be utilized in children with classical DIPG either at initial diagnosis or at tumor recurrence/progression.