Faculty Bibliography

Background: CNS tumors are the second most common cancer in children and the leading cause of mortality due to disease. New therapies are desperately needed. Specific Aims: The primary aim of this phase I study was to determine the safety and feasibility of administering HLA-A2 restricted, tumor associated antigenic peptides with Montanide ISA-51 VG to children with refractory CNS tumors. Secondary aims were to evaluate immune response to the vaccine and tumor response. Methods: Each vaccine consisted of HLA-A2 restricted peptides targeting epitopes on the tumor associated antigens Her2, Trp2, EphA2 and gp100 mixed with Montanide ISA-51 VG as the immune adjuvant. The neoantigen KLH was given with the first vaccine as a control. Patients received the vaccines divided into 2 subcutaneous injections on weeks 1, 4 and 7. Immune responses induced by the vaccine were evaluated by tetramer and intracellular cytokine staining. Results: 15 patients, females=8, median age 12 years (range 7- 20 y) were treated between August 2009 and May 2012. Diagnoses included pilocytic astrocytoma=1, low grade glioneuronal tumor=1, pilocytic/pilomyxoid tumor=2, anaplastic astrocytoma= 3, DIPG=2, radiation-induced glioblastoma=1, and ependymoma=3. Two patients had unbiopsied presumed low grade astrocytomas. One patient with an ependymoma was removed after only 2 immunizations because of progressive disease. 14 pts received all 3 vaccines. Several patients had grade 1 local skin reactions at the injection sites. No patients had grade 2 or higher adverse reactions related to the vaccine. Analysis of immune response shows induction of T cell responses to the tumor associated antigens. Impressively, most patients evaluated so far had detectable T cell responses to gp100 and Her-2 post vaccination. Furthermore, both antibody and T cell responses to the control antigen KLH were detected in most patients. Five of 6 patients with low grade astrocytomas have had stable disease for a median of 24 months (range 6-36 mo). Three patients with anaplastic astrocytomas have stable disease for 16, 24, and 24 months. Conclusions: Vaccine therapy using tumor associated antigenic peptides with Montanide ISA-51 VG was well tolerated. Despite being heavily pre-treated, these children were able to mount both humoral and adaptive immune response. Stable disease was seen in children with refractory low grade and high grade gliomas

Prognosis of diffuse intrinsic pontine gliomas (DIPGs) remains poor. Failure has been predominantly local, with leptomeningeal dissemination (LD) occurring in 4-33% of patients in pre-MRI era series. Routine craniospinal imaging after initial treatment may reveal other relapse patterns relapse. Sixteen consecutive pediatric patients with DIPG treated between 2006 and 2009 were retrospectively reviewed. Treatment regimens, recurrence patterns, survival, and pathologic diagnosis were recorded. Fourteen patients received involved-field radiotherapy to 54 Gy, and two patients received craniospinal irradiation for LD at presentation. Neuraxis MRI was performed at diagnosis and at 4 month intervals following radiotherapy. Fifteen patients have had progression of disease (median progression-free survival 5.0 +/- 1.2 months), and 13 patients have died (median survival 9.0 +/- 1.4 months). Local failure occurred in 12 patients (75%). LD occurred in nine patients (56%). LD was present at diagnosis in three patients, after initial staging and treatment in six patients, and during autopsy in two patients. Median overall survival was 12.0 +/- 3.3 months without LD and 8.0 +/- 2.1 months with LD (P = 0.059, log rank test). Median progression-free survival was 9.5 +/- 3.9 months without LD and 3.0 +/- 2.1 months with LD (P = 0.012, log rank test). The high incidence of LD probably reflects liberal use of spine MRI surveillance. All patients should undergo routine craniospinal imaging at diagnosis and follow-up. Central nervous system prophylaxis should be considered in future clinical trials

To evaluate the Quality of Life (QoL) and Social-Emotional/Behavioral functioning of survivors treated on the 'Head Start I' protocol at participating medical centers across the United States between 1991 and 1997. Parents of 25 of 27 (92.5%) patients completed the Child Health Questionnaire PF-50 (CHQ), to assess their child's QoL, along with the Behavior Assessment System for Children (BASC), to assess their social-emotional/behavioral functioning, after a mean follow-up of 5.7 years (range 13-96 months). Nineteen (76%) of the 25 parents subsequently completed the same instruments after a mean of 11.6 years (range 90-181 months), three (12%) patients died of disease and three (12%) were lost to follow-up. Mean Physical and Psychosocial QoL Summary Scores on the CHQ at Time 1 (T1) and Time 2 (T2) were within the normal range. Of the ten individual means for CHQ subscales, nine were within normal limits with the exception of Parental Emotional Impact at T1 and General Health at T2. Additionally, mean scores on the BASC at T1 and T2 were within normal limits for Externalizing and Internalizing Behaviors as well as Adaptive Skills. Serial analyses between T1 and T2 revealed non-significant changes over time with the exception of decreased General Health on the CHQ. The lack of QoL and Social-Emotional/Behavioral deficits suggests that the Head Start I protocol, involving intensive induction followed by myeloablative chemotherapy and autologous hematopoietic cell rescue in order to avoid or delay cranial irradiation, provides encouraging pilot data warranting continued monitoring

Purpose: To conduct a retrospective analysis of our collective experience over a seven year period of 35 patients with CNS GCT either referred directly or consulted upon for management, assessed from first recurrence. Method: Eleven germinoma patients and 24 mixed malignant GCT (MMGCT)patients were accrued. Of the 11 germinoma patients, seven received chemotherapy initially followed by irradiation, two received irradiation only and two received chemotherapy only. Following recurrence, two germinoma patients received re-induction chemotherapy followed by irradiation, and nine received re-induction chemotherapy followed by myeloablative chemotherapy with autologous hematopoietic cell rescue (HDCx+AuHCR) with subsequent irradiation in five. Of the 23 MMGCT patients, 21 received chemotherapy initially followed by irradiation, one received chemotherapy only (including HDCx+AuHCR) and another received irradiation only. Following recurrence, 22 of 23 MMGCT underwent attempts at remission re-induction with one or more chemotherapy regimens, with or without additional irradiation; one underwent total resection of tumor followed by irradiation only. Results: All 11 recurrent germinoma patients achieved minimal residual disease (MRD) with the first re-induction regimen. Ten of 11 (91%) survive without further recurrence at a mean of 3.5 years following diagnosis of recurrence. Nineteen of 23 MMGCT patients (83%) achieved MRD status; of these, three received further irradiation only, while 16 underwent HDCx+AuHCR either preceded or followed by irradiation in 12. Fourteen of 23 (61%) recurrent MMGCT patients survive without disease at a mean of 4.0 years following diagnosis of first recurrence. Conclusion: In this, the largest cohort of recurrent CNS GCT patients yet reported, most patients undergoing chemotherapy-containing treatment (especially germinoma patients) can be successfully salvaged, provided they achieve MRD status with re-induction chemotherapy. The benefit of consolidation with HDCx+AuHCR seems clear for MMGCT patients, and may obviate the need for further irradiation for germinoma patients, thereby avoiding the late sequelae of CNS re-irradiation