Outcomes After Reirradiation for Recurrent Pediatric Intracranial Ependymoma

This is a retrospective review of 101 children aged up to 21 years treated with repeat irradiation for recurrent ependymoma. In the setting of disease recurrence, traditional dogma is that re-irradiation can lead to toxicity. In this study, the efficacy and safety of re-irradiation for patients with recurrent ependymoma was evaluated.
Retrospective review of patients treated between 1985 and 2017. Patients with localized, intracranial ependymoma treated with surgery followed by focal irradiation were included. Patients with recurrence were treated with repeat surgery, followed by either focal repeat irradiation (for those with local recurrence) or craniospinal irradiation followed by focal boost RT (for those with any distant recurrence).
For the entire cohort, 5-year overall survival (OS) was 57% and 5-year freedom-from-progression (FFP) was 37%. Patients with distant-only failure and treated with CSI had the best outcomes, with 5-year OS and FFP of 76% and 49%, respectively.
In a multivariable analysis, male sex and anaplastic recurrence were negative prognostic factors for both OS and FFP. Again, those with distant-only failure treated with CSI had the best outcomes, with a hazard ratio (HR) of 0.4 (versus local failure treated with focal RT).Gain of chromosome 1q was associated with poorer survival in the subgroup of patients with distant failure after first RT (HR 3.5). Administering CSI as part of re-irradiation reduces distant-only failures.10-year cumulative incidence of grade ≥3 radiation necrosis was low, at 7.9%.
This was a retrospective study and represents a lower level of evidence as compared with a prospective study. The study had some patients who received CSI for locally recurrent ependymoma. It is unclear why CSI was given for some cases and not for others.
Association of histological grade with outcome in this study may be a proxy for underlying molecular biology. Although the authors had relatively complete data on chromosome 1q gain and C11orf95-RELA fusion, they did not have methylation data to determine if these patients had Group A or B ependymomas.
Retrospective study showed re-irradiation of children with recurrent ependymoma is effective without severe toxicity, particularly those with distant failures who received CSI. Further evaluation of re-irradiation will be done in upcoming prospective studies. (St. Jude re-irradiation study for ependymoma in progress (ClinicalTrials.gov identifier NCT02125786)

Vincristine, irinotecan, and temozolomide in children and adolescents with relapsed rhabdomyosarcoma

Relapsed rhabdomyosarcoma remains a disease with dismal prognosis in need of new treatment options. The addition of temozolomide to vincristine and irinotecan (VIT) in relapsed Ewing sarcoma and in two small studies of patients with relapsed RMS showed promise with improved response rates.
This multicenter retrospective study describes the progression-free survival of 19 patients from 5 tertiary care centers with relapsed RMS who received VIT at first or subsequent relapse. It is the largest series to date of children and adolescents treated with VIT for relapsed RMS.
VIT was used as first, second, third, or fourth line of therapy. Of 19 patients with median age 8 years (range 2-17), 4 received VIT as adjuvant therapy post surgery for local control and were not included in the response analysis. Of the 15 evaluable patients, none had complete or partial response, 5 had stable disease, and 11 had progressive disease. At median follow-up of 8 months, 2 patients were alive without disease, 3 were alive with disease, and 14 patients died or progressive disease. PFS at 3 months was 23%.
Small sample size, limited by quality of information documented/collected due to retrospective analysis, no central radiology/pathology review, interpatient variability in dosing of chemo, no toxicity data recorded. Physicians chose this regimen rather than randomly assigning patients to it so there may be selection bias.
Best response to VIT in relapsed RMS was stable disease, achieved in a quarter of patients. It is unclear from this whether VIT is effective therapy in this disease.

Intravenous immunoglobulin with prednisone and risk-adapted chemotherapy for children with opsoclonus myoclonus ataxia syndrome associated with neuroblastoma (ANBL00P3): a randomised, open-label, phase 3 trial

This is the first randomized clinical trial in patients with opsoclonus-myoclonus-ataxia syndrome associated with neuroblastoma. Patients affected by OMA have an excellent prognosis for OS, but face a high risk of long-term neuro-cognitive sequelae, including developmental delay, speech and language deficiencies, motor sequelae, ataxia, psychiatric symptoms. The study aims at assessing the response to treatment with prednisone and risk-adapted chemotherapy, and at understanding the role of IVIG.
This was a randomized, phase 3 trial run by the Children's Oncology Group. Children up to 8 years old, with newly diagnosed, biopsy proven neuroblastoma or ganglioneuroblastoma, with OMA were randomized to receive prednisone, risk adapted chemotherapy +/- IVIG. The randomization was stratified by risk assignment.
The severity of OMA and response to treatment was evaluated using the Pike and Mitchell score.
53 patients were enrolled between 2004 and 2013. 62% were female. The response rate was 81% in the IVIG group (21 of 26 patients), and 41% (11 of 27) in the non-IVIG group (odds ratio 6·1, 95% CI 1·5–25·9, p=0·0029). Median follow up for patients with no progression of OMA was 6.1 years. Toxicity was acceptable, with only one death, occurring in a high-risk patient post bone marrow transplant.
The trial was designed before two important pieces of information became available in the field: 1. the efficacy of dexamethasone pulses is comparable to prednisone, but with much more limited toxicity; 2. role of rituximab in the treatment of OMA. Therefore, the results of this trial may be a bit outdated. Nevertheless, the trial answers an important question regarding the role of IVIG. The lack of information on the long term follow up and neuro-cognitive outcomes is also a limitation of the study.
IVIG significantly improve the response rate in patients with OMA and neuroblastoma or ganglioneuroblastoma treated with prednisone and risk-adapted chemotherapy. Their impact on long term neurocognitive outcome, and their role in patients treated with regimens including rituximab remain unknown.

Results of methotrexate-etoposide-ifosfamide based regimen (M-EI) in osteosarcoma patients included in the French OS2006/sarcome-09 study

The article reports the outcome of patients with osteosarcoma, treated in the French OS2006 study with methotrexate, etoposide, ifosfamide pre-operatively and risk adapted chemotherapy post-operatively (M-EI for good responders, M-AP for poor responders). OS2006 was a randomized trial, exploring the effect of zoledronate on survival. The results of the randomization have previously been reported. The group now reports the survival and safety of the total study population.
The French group has previously compared methotrexate/doxorubicin to methotrexate/etoposide-ifosfamide pre-operatively. Given the comparable efficacy, M-EI has become standard of care in France. Post-operatively, patients who are high-risk (either poor responders, metastatic or unresectable) receive MAP, while low-risk patients receive M-EI. This paper reports survival results on the study population of OS2006, including patients who had refused the randomization.
Only 134/374 patients received post-operative treatment as per protocol (89/221 standard-risk patients and 45/153 high-risk). In the study population, 213/409 patients, including 187/324 with localized disease, received neither doxorubicin nor cisplatin first-line treatment (pre- or postoperatively). Median follow up was 4.8 years. The 3- and 5-year EFS were 63% (95% CI, 58-68%) and 56% (95% CI, 51-62%), respectively. The 3- and 5-year OS were 81% (77-85%) and 71% (66-76%), respectively.
The authors suggest that the chemotherapy protocol proposed had the same efficacy as MAP, but with more limited long-term side effects. Unfortunately, this is not a randomized trial and there is no comparison group, so it is not adequate to answer the question. Also, a high proportion of cases was given chemotherapy not according to the protocol. While these results are promising, the role of EI in osteosarcoma treatment remains to be clarified.
Treatment of osteosarcoma patients with M-EI preoperatively and risk stratified post -operative chemotherapy (MAP for high-risk patients, or M-EI for standard-risk patients) is safe and translates to survival rates comparable to those obtained with MAP. Further, randomized trials are necessary to finally define the role of EI in the treatment of osteosarcoma (perhaps identifying groups of patients, which by age, or stage of disease, could be more likely to benefit from EI).

Liver transplantation for primary hepatic malignancies of childhood: The UNOS experience

Factors associated with patient and graft survival following liver transplantation for children/adolescents with primary hepatic malignancies were analysed. The aim was to identify features associated with better outcomes from transplant. The United Network for Organ Sharing (UNOS) database was reviewed which is a resource containing organ transplant data. Five year patient and graft survival as well as independent predictors of survival were assessed.
All pediatric patients who underwent liver transplant between 1987 and 2012 were included. 574 patients were identified - the majority of whom had hepatoblastoma (HB, 70%), followed by hepatocellular carcinoma (HCC, 15%), and then other primary liver tumors (15%).
Five-year overall survival was 73% - HB did much better than HCC (76% versus 63% overall survival). In the most recent time period, 3-year survival was even better with 85% for HB. The most common cause of death was recurrent or metastatic disease in 57% of deaths.
Features associated with reduced survival were higher PELD scores, hospitalization at the time of transplant, ICU at the time of transplant, and people transplanted under earlier regimes.
Large database study - difficult to identify granular details such as PRETEXT stages, changes in chemotherapy regimes, choices of transplant medications postoperatively etc. Also no information on primary transplant or rescue transplant - potentially overall survival could be even higher for primary transplants.
76% of HB patients who undergo transplant are alive at 5 years post transplant, as are 63% of pediatric HCC patients. HCC and medical condition at the time of transplant were independent predictors of graft failure/death.

A pediatric brain tumor consortium phase II trial of capecitabine rapidly disintegrating tablets with concomitant radiation therapy in children with newly diagnosed diffuse intrinsic pontine gliomas

Outcome for children with diffuse intrinsic pontine gliomas (DIPG) remains dismal. Radiation has been shown to increase the pharmacodynamic effect of capecitabine. The primary aim of this study was to estimate PFS for children with DIPG treated with combination of capecitabine and RT.
Children 3-17 years of age with newly diagnosed DIPG without prior experimental treatment. Histology was not required. Capecitabine was initiated within 24 hr of the initiation of RT and was administered over 20 weeks. Conventional or conformal RT was administered to all patients once daily 5 days a week in 180 cGy/day fractions to a total dose of 5,580 cGy.
Thirty-five children with DIPG, all eligible, were enrolled from February 2010 to May 2011 and combines with data from 10 patients from a phase 1 study. 25 patients completed the planned protocol. 3 withdrew from treatment. Therapy was well tolerated. There was earlier progression with the capecitabine group, with 6 and 12-month PFS of 33.7% and 7.2% compared to 54.7% and 15.6% in the historical control, respectively. There was no difference in OS. Shorter PFS hypothesized to be pseudoprogression in response to therapy.
Non-randomized study, so reason for shorter PFS is unclear. Regardless, this study design remains the standard for evaluating new agents in children with DIPG since the outcomes in children with DIPG have not changed over time.
This combination therapy failed to improve the outcome in children with DIPG.

Phase II, Open Label Randomized, Multicenter Trial (HERBY) of Bevacizumab in Paediatric Patients with newly diagnosed high-grade glioma

This paper presents the results of the HERBY study – a Phase II, randomized multicentre trial adding bevacizumab to temozolomide and radiotherapy in pediatric patients with high-grade gliomas. This trial was spurred by the success of bevacizumab in adult HGG.
This was a multi-center randomized trial comparing the two treatment regimes. The primary end-point was EFS (progression, recurrence, or death) as assessed by central radiology review blinded to the treatment that the patient received.
121 patients were enrolled. The EFS did not differ significantly between the two treatment groups (Radiotherapy and temozolomide 11.8 months, Radiotherapy and temozolomide plus bevacizumab 8.2 months, p = 0.13).
In overall survival adding bevacizumab did not reduce the risk of death. Death occurred in 28 patients (50%) in the radiotherapy and temozolomide group and 33 patients (55%) radiotherapy and temozolomide plus bevacizumab group.
Importantly there were more serious adverse events in the bevacizumab group compared to the radiotherapy and temozolomide group without bevacizumab.
The trial wasn't powered to detect an effect in genetically-defined subgroups such as in tumors with histone mutations.
Adding bevacizumab to radiotherapy and temozolomide did not improve EFS in pediatric patients with high-grade glioma. This is disappointing and different from the experience in previously reported adult trials using bevacizumab in adult high-grade glioma. This highlights the biological difference between adult and pediatric gliomas and the importance of conducting paediatric specific studies.

Neuroblastoma Patients’ KIR and KIR-Ligand Genotypes Influence Clinical Outcome for Dinutuximab-based Immunotherapy: A Report from the Children`s Oncology Group

The use of dinutuximab-based immunotherapy has improved the outcome for patients with high-risk neuroblastoma in a phase III clinical trial (ANBL0032). Dinutuximab is an anti-GD2 monoclonal antibody whose activity is facilitated by NK cells. KIRs on NK cells highly polymorphic and, similar to HLA haplotypes, may have an effect on cellular immune response. This secondary analysis of ANBL0032 data investigated the association between KIR/KIR-ligand genotypes and clinical outcome in high risk neuroblastoma patients randomized to immunotherapy or no immunotherapy.
A total of 226 high-risk neuroblastoma patients were enrolled into ANBL0032 in 2010 and were randomized to immunotherapy or no immunotherapy; 176 patients had DNA available for genotyping for KIR/KIR-ligands (88 immunotherapy; 86 patients no immunotherapy).
Patients with the KIR/KIR-ligand genotypes “all KIR-ligands present”, “inhibitory KIR2DL2”, and “inhibitory KIR3DL1” were associated with improved event-free and overall survival with >5 years follow-up if patients had received immunotherapy versus no immunotherapy. The “complementary KIR/KIR-ligand” genotypes did not improve outcome in patients receiving immunotherapy compared to isotretinoin alone.
Prior studies on KIR/KIR-ligand genotyping showed conflicting results on the benefit of different KIR/KIR-ligand genotypes on outcome in neuroblastoma, therefore validation by independent clinical studies is needed.
High-risk neuroblastoma patients with certain KIR/KIR-ligand genotypes benefited from immunotherapy in this study. Validation of these findings is necessary, before KIR/KIR ligand genotyping might help to stratify future patients to immunotherapy treatment and avoid toxic side effects in patients who might not benefit from immunotherapy.

Efficacy of Larotrectinib in TRK Fusion-Positive Cancers in Adults and Children

Fusions involving the NTRK genes have been reported in a number of different cancers and are driver events when they occur. The fusions lead to a constitutively activated tyrosine kinase that acts through the classically proliferative MAP/ERK and PI3K pathways. The fusions occur rarely in some common cancers but much more frequently in a few rare tumors including Infantile fibrosarcoma, salivary gland carcinoma, and secretory breast carcinoma. This is a phase 1/2 trial of Larotrectinib - a selective inhibitor of TRK proteins.
A phase 1 trial of children and adults and a phase 2 trial of adults are both included in this publication. An ongoing phase 2 trial including children will be reported elsewhere. These single-arm trials gave either liquid or pill forms of Larotrectinib to patients on an ongoing basis until they reached a dose-limiting toxicity or had progression. Primary outcome was overall response rate by RECIST criteria.
55 patients were treated including 12 patients under 15 years of age (all children had infantile fibrosarcoma or other soft tissue sarcomas). Of all patients, the overall response rate was 75% and after a median of 9 months follow up, 86% of responders had undergone definitive surgery or continued to take Larotrectinib. Toxicity profile was limited with no DLTs and the most common grade 3/4 adverse event being increased liver enzymes, anemia, and fever. In a secondary analysis, the authors sequenced the NTRK gene in patients with drug resistance and found mutations that would likely inhibit Larotrectinib binding.
Although this is a phase I/II study, patients were selected for their good performance score, and the excellent results here mean that this drug will receive approval without a phase III study.
This is an effective drug for the small number of patients with NTRK fusions. This study is actually a poster child of how rare tumors can become attractive for pharmaceutical development if they are included in so-called basket trials. A larger societal limitation is the cost of sequencing all tumors. If NTRK fusions are found rarely in some common cancers (such as melanoma or colon cancer) then how many patients need to have expensive screening in order to find one patient who qualifies for what will likely be an exorbitantly expensive drug? Who will cover the costs and how do we ensure equal access?

Osteosarcoma, Chondrosarcoma, and Chordoma

As part of a special edition on sarcomas in the January edition Number 2, this review article covers osteosarcoma, chondrosarcoma, and chondroma.

Rhabdomyosarcoma, Ewing Sarcoma, and Other Round Cell Sarcomas

This is a review article and part of a special JCO edition on sarcomas (January 2018, number 2) in particular looking at genomics and potential therapeutic targets.

Outcome and Prognostic Factors in Stage III Favorable-Histology Wilms Tumor: A Report From the Children's Oncology Group Study AREN0532

This is a report on the outcomes of children with stage III favorable-histology Wilms tumours treated on the most recently closed COG trial AREN0532. The goal of the prospective therapeutic cohort trial was to identify clinical and biologic variables that affect the prognosis of patients with stage III favorable-histology Wilms tumor, as well as to maintain an EFS >85% and OS >95%.
588 children with stage III disease were included whether they had upfront nephrectomy or delayed nephrectomy. Any children with evidence of metastases or with LOH of 1p AND 16q or with anaplastic histology were excluded. This was a single-arm trial with central review of imaging and pathology. All children received Regimen DD4A (vincristine, doxorubicin, and dactinomycin) as well as flank or abdominal radiation depending on the local extent of disease.
Overall 4-yr EFS was 88% and 4-yr OS was 97%. Both were improved from NWTS5, the last published North American Wilms tumor trial; 116/535 children had neoadjuvant chemotherapy; 7 children with neoadjuvant chemotherapy had completely necrotic disease - none relapsed; 7 children with neoadjuvant chemotherapy had blastemal-predominant disease - five relapsed.
A combination of positive lymph nodes and LOH of 1p OR 16q conferred worse outcome. Conversely, children without nodal metastases and no LOH of 1p or 16q had exceptionally good outcomes. Most relapses occurred early (within 2 years of diagnosis) and the majority were metastatic relapses.
This was a single-arm observational trial so comparisons to other outcomes are difficult. It is especially difficult to compare to SIOP outcomes even in children who received neoadjuvant chemotherapy because the radiation doses were different and some parts of the chemotherapy regimens differed as well. Children with high-risk post-operative histology (i.e. blastemal predominance) did not have their therapy intensified unlike in SIOP. There is also likely stage migration affecting the results of this trial as all patients with lung nodules on CT and those with LOH of 1p AND 16q were excluded in this trial but not from NWTS5. There was a trend noted toward a higher rate of delayed nephrectomy, resulting in some patients who may have had stage I or II disease being upstaged due to biopsy.
There were overall good outcomes using the AREN0532 protocol of DD4A with radiation therapy, including EFS 88% and OS 97%. DD4A chemotherapy and 10-11 Gy of flank/abdo radiation after upfront nephrectomy is still the standard of care for stage III disease in North America. However, this trial indicates that there is no difference in outcome if neoadjuvant chemotherapy is used and so this is a viable option for surgically difficult cases. An association between lymph node positivity and LOH status and EFS was found to be significant. New risk groups may have also been identified based on lymph node metastases and LOH of a single chromosome. In particular, the group with no nodal metastases and no LOH may be able to have a reduction in therapy. Finally, all clinicians treating Wilms tumours should carefully consider intensification of therapy in cases of blastemal-predominance.