Irinotecan and temozolomide in combination with dasatinib and rapamycin versus irinotecan and temozolomide for patients with relapsed or refractory neuroblastoma (RIST-rNB-2011): a multicentre, open-label, randomised, controlled, phase 2 trial.

BACKGROUND: Neuroblastoma is the most common extracranial solid tumour in children. Relapsed or refractory neuroblastoma is associated with a poor outcome. We assessed the combination of irinotecan-temozolomide and dasatinib-rapamycin (RIST) in patients with relapsed or refractory neuroblastoma. METHODS: The multicentre, open-label, randomised, controlled, phase 2, RIST-rNB-2011 trial recruited from 40 paediatric oncology centres in Germany and Austria. Patients aged 1-25 years with high-risk relapsed (defined as recurrence of all stage IV and MYCN amplification stages, after response to treatment) or refractory (progressive disease during primary treatment) neuroblastoma, with Lansky and Karnofsky performance status at least 50%, were assigned (1:1) to RIST (RIST group) or irinotecan-temozolomide (control group) by block randomisation, stratified by MYCN status. We compared RIST (oral rapamycin [loading 3 mg/m2 on day 1, maintenance 1 mg/m2 on days 2-4] and oral dasatinib [2 mg/kg per day] for 4 days with 3 days off, followed by intravenous irinotecan [50 mg/m2 per day] and oral temozolomide [150 mg/m2 per day] for 5 days with 2 days off; one course each of rapamycin-dasatinib and irinotecan-temozolomide for four cycles over 8 weeks, then two courses of rapamycin-dasatinib followed by one course of irinotecan-temozolomide for 12 weeks) with irinotecan-temozolomide alone (with identical dosing as experimental group). The primary endpoint of progression-free survival was analysed in all eligible patients who received at least one course of therapy. The safety population consisted of all patients who received at least one course of therapy and had at least one post-baseline safety assessment. This trial is registered at ClinicalTrials.gov, NCT01467986, and is closed to accrual. FINDINGS: Between Aug 26, 2013, and Sept 21, 2020, 129 patients were randomly assigned to the RIST group (n=63) or control group (n=66). Median age was 5·4 years (IQR 3·7-8·1). 124 patients (78 [63%] male and 46 [37%] female) were included in the efficacy analysis. At a median follow-up of 72 months (IQR 31-88), the median progression-free survival was 11 months (95% CI 7-17) in the RIST group and 5 months (2-8) in the control group (hazard ratio 0·62, one-sided 90% CI 0·81; p=0·019). Median progression-free survival in patients with amplified MYCN (n=48) was 6 months (95% CI 4-24) in the RIST group versus 2 months (2-5) in the control group (HR 0·45 [95% CI 0·24-0·84], p=0·012); median progression-free survival in patients without amplified MYCN (n=76) was 14 months (95% CI 9-7) in the RIST group versus 8 months (4-15) in the control group (HR 0·84 [95% CI 0·51-1·38], p=0·49). The most common grade 3 or worse adverse events were neutropenia (54 [81%] of 67 patients given RIST vs 49 [82%] of 60 patients given control), thrombocytopenia (45 [67%] vs 41 [68%]), and anaemia (39 [58%] vs 38 [63%]). Nine serious treatment-related adverse events were reported (five patients given control and four patients given RIST). There were no treatment-related deaths in the control group and one in the RIST group (multiorgan failure). INTERPRETATION: RIST-rNB-2011 demonstrated that targeting of MYCN-amplified relapsed or refractory neuroblastoma with a pathway-directed metronomic combination of a multkinase inhibitor and an mTOR inhibitor can improve progression-free survival and overall survival. This exclusive efficacy in MYCN-amplified, relapsed neuroblastoma warrants further investigation in the first-line setting. FUNDING: Deutsche Krebshilfe.

Distinct pattern of genomic breakpoints in CML and BCR::ABL1-positive ALL: analysis of 971 patients.

BACKGROUND: The BCR::ABL1 is a hallmark of chronic myeloid leukemia (CML) and is also found in acute lymphoblastic leukemia (ALL). Most genomic breaks on the BCR side occur in two regions - Major and minor - leading to p210 and p190 fusion proteins, respectively. METHODS: By multiplex long-distance PCR or next-generation sequencing technology we characterized the BCR::ABL1 genomic fusion in 971 patients (adults and children, with CML and ALL: pediatric ALL: n = 353; pediatric CML: n = 197; adult ALL: n = 166; adult CML: n = 255 patients) and designed "Break-App" web tool to allow visualization and various analyses of the breakpoints. Pearson's Chi-Squared test, Kolmogorov-Smirnov test and logistic regression were used for statistical analyses. RESULTS: Detailed analysis showed a non-random distribution of breaks in both BCR regions, whereas ABL1 breaks were distributed more evenly. However, we found a significant difference in the distribution of breaks between CML and ALL. We found no association of breakpoints with any type of interspersed repeats or DNA motifs. With a few exceptions, the primary structure of the fusions suggests non-homologous end joining being responsible for the BCR and ABL1 gene fusions. Analysis of reciprocal ABL1::BCR fusions in 453 patients showed mostly balanced translocations without major deletions or duplications. CONCLUSIONS: Taken together, our data suggest that physical colocalization and chromatin accessibility, which change with the developmental stage of the cell (hence the difference between ALL and CML), are more critical factors influencing breakpoint localization than presence of specific DNA motifs.

Imatinib treatment and longitudinal growth in pediatric patients with chronic myeloid leukemia: influence of demographic, pharmacological, and genetic factors in the German CML-PAED cohort.

In children and adolescents, impaired growth due to tyrosine kinase inhibitor therapy remains an insufficiently studied adverse effect. This study examines demographic, pharmacological, and genetic factors associated with impaired longitudinal growth in a uniform pediatric cohort treated with imatinib. We analyzed 94 pediatric patients with chronic myeloid leukemia (CML) diagnosed in the chronic phase and treated with imatinib for >12 months who participated in the Germany-wide CML-PAEDII study between February 2006 and February 2021 (clinicaltrials gov. Identifier: NCT00445822). During imatinib treatment, significant height reduction occurred, with medians of -0.35 standard deviation score (SDS) at 12 months and -0.76 SDS at 24 months. Cumulative height SDS change (Δ height SDS) showed a more pronounced effect in prepubertal patients during the first year but were similar between prepubertal and pubertal subgroups by the second year (-0.55 vs. -0.50). From months 12 to 18 on imatinib, only 18% patients achieved individually longitudinal growth adequate to the growth standard (Δ height SDS ≥0). When patients were divided into two subgroups based on median Δ height SDS (classifier Δ height SDS > or ≤-0.37) after 1 year on imatinib therapy, cohort 1 (Δ height SDS ≤-0.37) showed younger age at diagnosis, a higher proportion of prepubertal children, but also better treatment response and higher imatinib serum levels. Exploring the association of growth parameters with pharmacokinetically relevant single nucleotide polymorphisms, known for affecting imatinib response, showed no correlation. This retrospective study provides new insights into imatinib-related growth impairment. We emphasize the importance of optimizing treatment strategies for pediatric patients to realize their maximum growth potential.

Approaches for bridging therapy prior to chimeric antigen receptor T cells for relapsed/refractory acute lymphoblastic B-lineage leukaemia in children and young adults.

The ongoing development of immunotherapies, including chimeric antigen receptor (CAR) T cells, has revolutionized cancer treatment. In paediatric relapsed/refractory B-lineage acute leukaemia antiCD19-CARs induced impressive initial response rates, with event-free survival plateauing at 30-50% in long-term follow-up data. During the interval between diagnosis of relapse or refractoriness and CAR T cell infusion, patients require a bridging therapy. To date, this therapy has consisted of highly variable approaches based on local experience. Here, in an European collaborative effort of paediatric and adult haematologists, we summarise current knowledge with the aim of establishing a guidance for bridging therapy. This includes treatment strategies for different patient subgroups, the advantages and disadvantages of low- and highintensity regimens, and the potential impact of bridging therapy on outcome after CAR T cell infusion. This guidance is a step towards a cross-institutional harmonization of bridging therapy, including personalized approaches. This will allow better comparability of clinical data and increase the level of evidence for the treatment of children and young adults with relapsed/refractory B-lineage ALL until CAR T cell infusion.

Human-Level Differentiation of Medulloblastoma from Pilocytic Astrocytoma: A Real-World Multicenter Pilot Study.

Medulloblastoma and pilocytic astrocytoma are the two most common pediatric brain tumors with overlapping imaging features. In this proof-of-concept study, we investigated using a deep learning classifier trained on a multicenter data set to differentiate these tumor types. We developed a patch-based 3D-DenseNet classifier, utilizing automated tumor segmentation. Given the heterogeneity of imaging data (and available sequences), we used all individually available preoperative imaging sequences to make the model robust to varying input. We compared the classifier to diagnostic assessments by five readers with varying experience in pediatric brain tumors. Overall, we included 195 preoperative MRIs from children with medulloblastoma (n = 69) or pilocytic astrocytoma (n = 126) across six university hospitals. In the 64-patient test set, the DenseNet classifier achieved a high AUC of 0.986, correctly predicting 62/64 (97%) diagnoses. It misclassified one case of each tumor type. Human reader accuracy ranged from 100% (expert neuroradiologist) to 80% (resident). The classifier performed significantly better than relatively inexperienced readers (p < 0.05) and was on par with pediatric neuro-oncology experts. Our proof-of-concept study demonstrates a deep learning model based on automated tumor segmentation that can reliably preoperatively differentiate between medulloblastoma and pilocytic astrocytoma, even in heterogeneous data.

Machine Learning-Supported Diagnosis of Small Blue Round Cell Sarcomas Using Targeted RNA Sequencing.

Small blue round cell sarcomas (SBRCSs) are a heterogeneous group of tumors with overlapping morphologic features but markedly varying prognosis. They are characterized by distinct chromosomal alterations, particularly rearrangements leading to gene fusions, whose detection currently represents the most reliable diagnostic marker. Ewing sarcomas are the most common SBRCSs, defined by gene fusions involving EWSR1 and transcription factors of the ETS family, and the most frequent non-EWSR1-rearranged SBRCSs harbor a CIC rearrangement. Unfortunately, currently the identification of CIC::DUX4 translocation events, the most common CIC rearrangement, is challenging. Here, we present a machine-learning approach to support SBRCS diagnosis that relies on gene expression profiles measured via targeted sequencing. The analyses on a curated cohort of 69 soft-tissue tumors showed markedly distinct expression patterns for SBRCS subgroups. A random forest classifier trained on Ewing sarcoma and CIC-rearranged cases predicted probabilities of being CIC-rearranged >0.9 for CIC-rearranged-like sarcomas and <0.6 for other SBRCSs. Testing on a retrospective cohort of 1335 routine diagnostic cases identified 15 candidate CIC-rearranged tumors with a probability >0.75, all of which were supported by expert histopathologic reassessment. Furthermore, the multigene random forest classifier appeared advantageous over using high ETV4 expression alone, previously proposed as a surrogate to identify CIC rearrangement. Taken together, the expression-based classifier can offer valuable support for SBRCS pathologic diagnosis.

Impact of age on safety of Busulfan-Melphalan followed by autologous hematopoietic stem-cell transplantation versus standard chemotherapy in the patients of the EURO-E.W.I.N.G. 99 and Ewing 2008 clinical trials.

INTRODUCTION: Ewing sarcoma (ES), is a rare cancer affecting children, adolescents and adults. After VIDE (vincristine-ifosfamide-doxorobucin-etoposide) induction chemotherapy, Busulfan-Melphalan (BuMel) high-dose chemotherapy followed by autologous hematopoietic stem cells transplantation improved outcomes in unfavourable localized ES, but with more toxicities than conventional chemotherapy (VAI: Vincristine-dactinomycin-Ifosfamide). We evaluated whether the risk of acute toxicity associated with BuMel compared to VAI varied according to age in patients recruited in the R2Loc and R2Pulm randomised trials of the Euro-E.W.I.N.G.99 and Ewing-2008 trials. METHODS: We included patients with a localized high-risk disease, or pulmonary or pleural metastasis. We analysed the risk of severe toxicity according to randomised treatment group (VAI versus BuMel) and age group (<12 years, 12-17 years, 18-24 years, ≥25 years). We evaluated the heterogeneity of treatment effects by age group using interaction terms in logistic multivariable models. RESULTS: The analysis included 243 patients treated with VAI and 205 with BuMel. Overall, BuMel was associated with a higher risk of severe acute toxicity than VAI particularly haematological, gastrointestinal, liver, sinusoidal occlusive syndrome, and infections. Severe haematological toxicity and lower general condition were significantly more frequent in younger patients, whatever treatment. We did not observe any significant heterogeneity in terms of the excess risk of severe toxicities associated with BuMel compared to VAI according to age group. CONCLUSION: The excess of acute toxicity associated with BuMel compared to VAI does not vary significantly with age, suggesting the feasibility of BuMel across all age groups.

Finding a balance in reduced toxicity hematopoietic stem cell transplantation for thalassemia: role of infused CD3+ cell count and immunosuppression.

We performed a retrospective analysis on 124 patients with transfusion-dependent thalassemia who were registered in the German pediatric registry for stem cell transplantation. All patients underwent first allogeneic hematopoietic stem cell transplantation (HSCT) between 2011 and 2020 and belonged mainly to Pesaro risk class 1-2. Four-year overall (OS) and thalassemia-free survival (TFS) were 94.5% ± 2.9% and 88.0% ± 3.4% after treosulfan-fludarabine-thiotepa- and 96.9% ± 3.1% (P = 0.763) and 96.9% ± 3.1% (P = 0.155) after busulfan-fludarabine-based conditioning. Mixed chimerism below 75% occurred predominantly in treosulfan-based regimens (27.5% versus 6.2%). OS and TFS did not differ significantly between matched sibling, other matched family and matched unrelated donor (UD) HSCTs (OS: 100.0%, 100.0%, 96.3% ± 3.6%; TFS: 96.5% ± 2.4%, 90.0% ± 9.5%, 88.9% ± 6.0%). However, mismatched UD-HSCTs performed less favorable (OS: 84.7% ± 7.3% (P = 0.029); TFS: 79.9% ± 7.4% (P = 0.082)). We generated a scoring system reflecting the risk to develop mixed chimerism in our cohort. The main risk-reducing factors were a high CD3+ cell count (≥6 × 107/kg) in the graft, busulfan-conditioning, pre-conditioning therapy and low-targeted ciclosporin A trough levels. Acute GvHD grade III-IV in treosulfan-based concepts predominantly occurred in patients with UD and reduced GvHD prophylaxis but not in the context of high CD3+ cell doses. Taken together, this information might be used to develop more risk-adapted HSCT regimens for thalassemia patients.

An artificial intelligence-assisted clinical framework to facilitate diagnostics and translational discovery in hematologic neoplasia.

BACKGROUND: The increasing volume and intricacy of sequencing data, along with other clinical and diagnostic data, like drug responses and measurable residual disease, creates challenges for efficient clinical comprehension and interpretation. Using paediatric B-cell precursor acute lymphoblastic leukaemia (BCP-ALL) as a use case, we present an artificial intelligence (AI)-assisted clinical framework clinALL that integrates genomic and clinical data into a user-friendly interface to support routine diagnostics and reveal translational insights for hematologic neoplasia. METHODS: We performed targeted RNA sequencing in 1365 cases with haematological neoplasms, primarily paediatric B-cell precursor acute lymphoblastic leukaemia (BCP-ALL) from the AIEOP-BFM ALL study. We carried out fluorescence in situ hybridization (FISH), karyotyping and arrayCGH as part of the routine diagnostics. The analysis results of these assays as well as additional clinical information were integrated into an interactive web interface using Bokeh, where the main graph is based on Uniform Manifold Approximation and Projection (UMAP) analysis of the gene expression data. At the backend of the clinALL, we built both shallow machine learning models and a deep neural network using Scikit-learn and PyTorch respectively. FINDINGS: By applying clinALL, 78% of undetermined patients under the current diagnostic protocol were stratified, and ambiguous cases were investigated. Translational insights were discovered, including IKZF1plus status dependent subpopulations of BCR::ABL1 positive patients, and a subpopulation within ETV6::RUNX1 positive patients that has a high relapse frequency. Our best machine learning models, LDA and PASNET-like neural network models, achieve F1 scores above 97% in predicting patients' subgroups. INTERPRETATION: An AI-assisted clinical framework that integrates both genomic and clinical data can take full advantage of the available data, improve point-of-care decision-making and reveal clinically relevant insights promptly. Such a lightweight and easily transferable framework works for both whole transcriptome data as well as the cost-effective targeted RNA-seq, enabling efficient and equitable delivery of personalized medicine in small clinics in developing countries. FUNDING: German Ministry of Education and Research (BMBF), German Research Foundation (DFG) and Foundation for Polish Science.

Hodgkin lymphoma: hypodense lesions in mediastinal masses.

Hypodense volumes (HDV) in mediastinal masses can be visualized in a computed tomography scan in Hodgkin lymphoma. We analyzed staging CT scans of 1178 patients with mediastinal involvement from the EuroNet-PHL-C1 trial and explored correlations of HDV with patient characteristics, mediastinal tumor volume and progression-free survival. HDV occurred in 350 of 1178 patients (29.7%), typically in larger mediastinal volumes. There were different patterns in appearance with single lesions found in 243 patients (69.4%), multiple lesions in 107 patients (30.6%). Well delineated lesions were found in 248 cases (70.1%), diffuse lesions were seen in 102 cases (29.1%). Clinically, B symptoms occurred more often in patients with HDV (47.7% compared to 35.0% without HDV (p = 0.039)) and patients with HDV tended to be in higher risk groups. Inadequate overall early-18F-FDG-PET-response was strongly correlated with the occurrence of hypodense lesions (p < 0.001). Patients with total HDV > 40 ml (n = 80) had a 5 year PFS of 79.6% compared to 89.7% (p = 0.01) in patients with HDV < 40 ml or no HDV. This difference in PFS is not caused by treatment group alone. HDV is a common phenomenon in HL with mediastinal involvement.