Implementation of paediatric precision oncology into clinical practice: The Individualized Therapies for Children with cancer program 'iTHER'

Abstract

iTHER is a Dutch prospective national precision oncology program aiming to define tumour molecular profiles in children and adolescents with primary very high-risk, relapsed, or refractory paediatric tumours. Between April 2017 and April 2021, 302 samples from 253 patients were included. Comprehensive molecular profiling including low-coverage whole genome sequencing (lcWGS), whole exome sequencing (WES), RNA sequencing (RNA-seq), Affymetrix, and/or 850k methylation profiling was successfully performed for 226 samples with at least 20% tumour content. Germline pathogenic variants were identified in 16% of patients (35/219), of which 22 variants were judged causative for a cancer predisposition syndrome. At least one somatic alteration was detected in 204 (90.3%), and 185 (81.9%) were considered druggable, with clinical priority very high (6.1%), high (21.3%), moderate (26.0%), intermediate (36.1%), and borderline (10.5%) priority. iTHER led to revision or refinement of diagnosis in 8 patients (3.5%). Temporal heterogeneity was observed in paired samples of 15 patients, indicating the value of sequential analyses. Of 137 patients with follow-up beyond twelve months, 21 molecularly matched treatments were applied in 19 patients (13.9%), with clinical benefit in few. Most relevant barriers to not applying targeted therapies included poor performance status, as well as limited access to drugs within clinical trial. iTHER demonstrates the feasibility of comprehensive molecular profiling across all ages, tumour types and stages in paediatric cancers, informing of diagnostic, prognostic, and targetable alterations as well as reportable germline variants. Therefore, WES and RNA-seq is nowadays standard clinical care at the Princess Máxima Center for all children with cancer, including patients at primary diagnosis. Improved access to innovative treatments within biology-driven combination trials is required to ultimately improve survival.

Keywords: Adolescent; Cancer; Child; Hereditary; Molecular biology; Molecular targeted therapy; Next-generation sequencing; Precision medicine.

Fig. 1

iTHER program workflow and cohort demographics. (A) Detailed iTHER pipeline is depicted. ∗ After consent, samples are processed and sequenced utilising INFORM and/or Máxima pipeline. Curated and prioritised events are discussed at the Molecular Tumour Board to identify molecularly matched treatment options. ∗ Created with BioRender.com. (B) iTHER cohort consisting of 302 samples from 253 patients with relapsed, refractory, or high-risk newly diagnosed paediatric cancer. The innermost ring visualises disease distribution by the three main categories: extracranial solid tumour (Solid tumour), Central Nervous System tumour (CNS) and Haematopoietic Malignancies (HM). The outer ring represents the frequency of samples within each subtype: neuroblastoma (NBL), osteosarcoma (OS), rhabdomyosarcoma (RMS), other sarcoma (Sarcoma other), other solid tumours (Other ST), Ewing sarcoma, Wilms tumour (Wilms), malignant rhabdoid tumour (MRT) and extracranial germ cell tumour (GCT); high-grade glioma (HGG), other CNS tumours (Other CNS), medulloblastoma (Medullo), Ependymoma; acute lymphoblastic leukaemia (ALL), acute myeloid leukaemia (AML), and lymphoma (Lymphoma). (C) Age distribution of the cohort, highlighting stage of disease: primary high-risk disease, primary refractory disease, relapse, and secondary malignancy.

Fig. 2

iTHER sample flow and target priority. (A) Consort flow chart for all 302 samples included. (B) Distribution of the highest priority level target per sample, highlighted per cancer subtype for all subgroups with 4 or more samples included: neuroblastoma (NBL), Ewing sarcoma (Ewing), osteosarcoma (OS), Wilms tumour (Wilms), rhabdomyosarcoma (RMS), other types of sarcomas (Sarcoma other), high-grade glioma (HGG) and medulloblastoma (Medullo); lymphoma (Lymphoma), acute myeloid leukaemia (AML), and acute lymphoblastic leukaemia (ALL).

Fig. 3

Commonly altered druggable genes per cancer subtype. Each bubble cloud represents a subgroup with 4 or more samples harbouring actionable events. Every bubble represents a single potentially targetable gene. Its size indicates the relative frequency within the cancer subtype. The gene is highlighted if priority score is considered very high, high, or moderate, and the event is observed in a minimum of two samples.

Fig. 4

Germline pathogenic variants. (A) All identified germline pathogenic variants in the cohort with a subset of reportable cancer-associated germline variants. (B) Reportable germline pathogenic variants in detail.

Fig. 5

Applied molecularly matched treatments and clinical response. Twenty-one molecularly matched therapies were initiated in 19 patients. Time on treatment is depicted as per the prioritised target. Of note, 2 patients received 2 consecutive treatments each: one patient with neuroblastoma (#) and one patient with rhabdomyosarcoma (ˆ). x axis is time on treatment (weeks). The colour of the bars indicates the priority score of the targeted alteration. Responses are indicated by the symbols. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article).