Whole genome and transcriptome integrated analyses guide clinical care of pediatric poor prognosis cancers

Abstract

The role for routine whole genome and transcriptome analysis (WGTA) for poor prognosis pediatric cancers remains undetermined. Here, we characterize somatic mutations, structural rearrangements, copy number variants, gene expression, immuno-profiles and germline cancer predisposition variants in children and adolescents with relapsed, refractory or poor prognosis malignancies who underwent somatic WGTA and matched germline sequencing. Seventy-nine participants with a median age at enrollment of 8.8 y (range 6 months to 21.2 y) are included. Germline pathogenic/likely pathogenic variants are identified in 12% of participants, of which 60% were not known prior. Therapeutically actionable variants are identified by targeted gene report and whole genome in 32% and 62% of participants, respectively, and increase to 96% after integrating transcriptome analyses. Thirty-two molecularly informed therapies are pursued in 28 participants with 54% achieving a clinical benefit rate; objective response or stable disease ≥6 months. Integrated WGTA identifies therapeutically actionable variants in almost all tumors and are directly translatable to clinical care of children with poor prognosis cancers.

Fig. 1. Enrollment and tumor diagnosis details for participants in the pediatric POG study.

A Consort diagram depicting patient enrollment and sample progression including patient consent/assent, number of tumor biopsies conducted (includes archival tissue samples), and number of samples with completed WGTA (n = 83) for final study cohort (n = 79) participants; includes one participant with only WGS results). B Age of participants at diagnosis and enrollment. C Alluvial plot of patient tumor types for all successfully sequenced WGTA samples (n = 83) matched to primary biopsy site. D, E Kaplan–Meier plots of event-free and overall survival outcomes from sample sequencing start date for all pediatric POG study cohort participants (n = 79) and by tumor group. P values determined using two-sided log rank tests. CNS Central Nervous System, Heme hematological. Source data are provided as a source data file.

Fig. 2. Germline findings for participants in pediatric POG study.

A Evolution of the germline analysis pipeline in the pediatric POG cohort (left). Number of germline variants per gene reviewed prospectively by the POG Germline and Ethics Working Group during the study. B All pathogenic/likely pathogenic germline variants identified among 98 cancer predisposition genes within the pediatric POG cohort. SNV Single nucleotide variant, LOH loss of heterozygosity, POG Personalized Oncogenomics program. Source data are provided as a source data file.

Fig. 3. Therapeutically actionable somatic findings for participants in pediatric POG study.

A Recurrent genes with therapeutically actionable alterations by variant type, grouped in pathways; each gene was reported in at least three patients; bar plot: gene list on the y-axis with number of patients on the x-axis; bubble plot: variant type of therapeutically actionable alterations (copy number variant, CNV; structural variant, SV; single nucleotide variant/insertions and deletions, SNV/indels; expression outlier). The size of the bubbles is proportional to the number of patients. For gene expression, hollow bubbles indicate high outlier expression, and solid indicate low expression outliers. For CNVs, hollow bubbles are gains, and solid bubbles are losses. B Level of evidence (LOE, 1-5) for therapeutically actionable alterations, color coded by type of variants. C Therapeutically actionable alterations based on support from DNA and/or RNA sequencing data. D Categories of targeted drug therapy discussed at molecular tumor board based on somatic variants, color coded by level of evidence (LOE1-5). Source data are provided as a source data file.

Fig. 4. Whole genome mutation signatures across pediatric tumors.

A Proportion of small somatic mutations attributed to various mutational mechanisms in advanced or metastatic adult (N = 570) and pediatric (N = 81) cancers. Signature exposure was estimated by deconvolution of 96 single base substitution (SBS) signatures described in COSMIC v3.2 and grouped by proposed etiology: homologous recombination deficiency (HRD, SBS3 and −8), APOBEC-mediated mutagenesis (SBS2 and −13), mismatch repair deficiency (dMMR, SBS6, −14, −15, −20, −21, −26 and −44), MUTYH-related base excision repair deficiency (SBS18 and −36), NTHL1-related base excision repair deficiency (SBS30), tobacco exposure (SBS4 and −29), UV exposure (SBS7a, −7b, −7c, −7d and −38), platinum exposure (SBS17b, −31 and −35), and clock-like mutagenesis by spontaneous deamination of 5-methylcytosine to thymine (SBS1, −5 and −40). Boxplots display the median, and upper and lower quartiles with whiskers indicating minima and maxima. Statistics were derived using Wilcoxon rank sum tests. B Estimated contribution of SBS signatures by proposed etiology to pediatric tumors in POG. Tumor types listed in B are as follows: RMS/STS rhabdomyosarcoma/soft-tissue sarcoma, MRT malignant rhabdoid tumor, HGG high grade glioma, DIPG Diffuse intrinsic pontine glioma, ATRT atypical teratoid rhabdoid tumor, CNS central nervous system, ALL acute lymphocytic leukemia, NHL non-Hodgkin’s lymphoma, AML acute myeloid leukemia, AF Ameloblastic fibroma. Source data are provided as a source data file.

Fig. 5. Whole genome tumor mutation burden (TMB) and CIBERSORT immune signatures in pediatric POG patients and adult POG patients (POG570).

Pediatric tumors (N = 72) showed lower TMB and T-cell scores compared to adult tumors (N = 423). A CD8+ T-cell scores on the y-axis and TMB on the x-axis. B Total T-cell scores on the y-axis and TMB on the x-axis. Correlations shown are derived from Spearman tests, and comparisons between adult and pediatric used two-sided Wilcoxon rank sum tests. C CD8+ T-cell score by tumor sub-group. D Total T-cell scores by tumor sub-group. Boxplots display the median, and upper and lower quartiles with whiskers indicating minima and maxima. For panels C and D, sample sizes are as follows: High grade central nervous system (CNS), n = 16; Low grade CNS, n = 4; Benign, n = 8; Bone sarcoma, n = 12; Neuroblastoma, n = 8; Rhabdomyosarcoma/soft-tissue sarcoma (RMS/STS), n = 13; Other solid, n = 11. Source data are provided as a source data file.

Fig. 6. Therapeutic recommendations and outcomes for participants in the pediatric Personalized Oncogenomics (POG) study.

A Clinical action flowchart depicting the number of participants from the final study cohort (n = 79) who received therapeutic recommendations by the molecular tumor board (MTB) (n = 76) and were either treated (n = 28), untreated (n = 40), or under surveillance (n = 8). B Among patients receiving therapy, Swimmer plot depicting full disease trajectory from diagnosis including disease type, response status (CR complete response, PR partial response, SD stable disease, PD progressive disease, NE Not evaluable), and key milestone events (consent date, relapse or progression, date of last follow-up or death). C Event free survival (EFS) from initiation of POG-directed therapy for treated patients and grouped by tumor type. D Overall survival from initiation of POG-directed therapy among treated patients and grouped by tumor type. E EFS grouped by treatment level of evidence (LOE). F OS grouped by treatment LOE. Statistics for outcome were derived using two-sided log rank tests. Source data are provided as a source data file.