Deep phenotyping and genotyping of chemotherapy-associated neurotoxicity in children on ALL therapy

Wahid, Qurat-ul-Ain (2022) Deep phenotyping and genotyping of chemotherapy-associated neurotoxicity in children on ALL therapy. PhD thesis, University of Glasgow.

Full text available as:
[img] PDF
Download (4MB)


Paediatric ALL is the commonest childhood cancer. Although cure rates are good, treatment is prolonged and toxic. Chemotherapy-induced neurotoxicity remains a significant problem with incidence reported 8-12%. This study broadly aimed to accurately diagnose different forms of neurotoxicity in patients, to be able to identify genetic and environmental risk factors and understand the natural history of different types of neurotoxicity to counsel families, regarding supportive interventions, and decisions on whether to modify treatment.

Exploration of clinical risk-profiling of neurotoxicity and the effects of a neurotoxic event on leukaemia outcome was conducted using SAE reports from the United Kingdom Acute Lymphoblastic Leukaemia (UKALL) 2003 trial. Review of the 276 neurotoxic events identified treatment intensity as the main risk factor for developing neurotoxicity with female sex, increasing age and CNS status having a significant modifying effect.

Next, a large retrospective pooled central neurotoxicity detailed dataset (n=1813) from 14 international study groups (all members of the Ponte di Legno (PdL) consortium) was created and used to deep phenotype and genotype neurotoxic events. The main analysis concentrated on the 3 most common diagnoses: seizures, posterior reversible encephalopathy syndrome (PRES) and stroke-like syndrome (SLS). Initial work identified significant overlap between the diagnostic criteria for SLS and PRES with many cases fulfilling both definitions (i.e., biphenotypic). Using multivariate logistic regression analysis odds ratio values for specific clinical and radiology findings were used to construct a diagnostic scoring system for PRES and SLS which successfully separated the cases into distinct entities.

Deep phenotyping of seizure cases reported in the PdL database showed recurrence, Down syndrome, Capizzi methotrexate, Single intrathecal methotrexate, and late leucovorin rescue was associated with an increased risk of seizures compared to the rest of the neurotoxicity cohort. Deep phenotyping of PRES cases showed younger age group (< 10 years), T-cell and CNS involvement, high vincristine dose protocol and use of oral anticonvulsants were significantly associated with PRES. Finally, SLS cases showed older age, consolidation with high dose methotrexate and High-risk treatment regimen allocation to be associated with SLS. Another important finding, with clinical implications, was that patients with Down syndrome ALL showed a significant association with recurrence of neurotoxicity upon re-exposure to the causative agent.

Following deep phenotyping, cases with defined phenotypes were picked to enter a targeted meta-analysis of existing single nucleotide polymorphism (SNPs) data. A list of candidate SNPs of interest was generated and sent to 3 study groups who then compared allele frequencies between cases (with neurotoxicity) and matched controls (without neurotoxicity). Results were combined and analyzed using METAL software. Although a nominally significant association with some candidate SNPs was observed for all the selected phenotypes no individual SNP remained significant after adjustment for multiple testing.

Finally, using novel in silico analysis of two published datasets an attempt was made to correlate changes in genes associated with clinical episodes of neurotoxicity with methylation patterns of genes in response to methotrexate treatment in vivo. Two genes (PDE4B and ASTN2) were that were consistently hypermethylated in response to all three doses (20, 30 and 50 nM) of methotrexate and mapped with significant genetic association in clinical neurotoxicity in paediatric ALL cases post methotrexate in an in-silico analysis. These results may shed some light on potential genetic predispositions to methotrexate induced neurotoxicity.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry
R Medicine > RJ Pediatrics
R Medicine > RJ Pediatrics > RJ101 Child Health. Child health services
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cancer Sciences
Supervisor's Name: Halsey, Professor Chris and Young, Dr. Robin
Date of Award: 2022
Depositing User: Theses Team
Unique ID: glathesis:2022-83303
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 08 Dec 2022 10:38
Last Modified: 13 Dec 2022 12:19
Thesis DOI: 10.5525/gla.thesis.83303
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year