CNS Tumours: Exploring Barriers to Registration and Feasibility of Extracting Molecular Characteristics

• Author / Creator

  Eckstrand, Angela C

• Two concerns regarding the collection of data for brain and other central nervous system (CNS) tumours in Canada are under-reporting of non-malignant tumours and the need for improved reporting of clinically relevant molecular information. The first part of our study, addressing under-reporting, has two objectives: 1) evaluate the impact linkage with hospital discharge, as recorded in Discharge Abstract Database (DAD), has on supplementing case ascertainment for brain tumours, and 2) identify potential barriers for initial registration in the Alberta Cancer Registry (ACR).

  Information for all patients with a brain tumour diagnosed and residing in Alberta between 2010 and 2015 were extracted from the ACR. Descriptive statistics were compiled by behaviour and type of registration (originally registered or identified through DAD review). The age-standardized incidence rates and number of cases (observed vs expected) in Alberta were compared to the United States. Phi coefficients (as a measure of correlation) and chi-square tests for the homogeneity of proportions were conducted to examine bivariate relationships of the characteristics of interest. Multiple logistic regression was used to summarize the independent effects on the probability of being identified through DAD review.

  The results show 5% of malignant and 35% of non-malignant brain tumours were identified through DAD review. When comparing observed to expected number of non-malignant cases after DAD review, the ACR ultimately captured about 75% of those expected. Cases identified through DAD were statistically significantly (p ≤ 0.05) associated with patients over 75 years old at diagnosis (OR=2.5), benign behaviour (OR=2.6), location at diagnosis in Northern Alberta (OR=1.5), non-microscopically confirmed tumours (OR=1.3), no visit to a CancerControl Alberta facility (OR=8.7) and certain histological subtypes, including cranial and spinal nerve tumours (OR=1.7). Given the significant impact DAD review had on case ascertainment of non-malignant brain tumours, it is recommended that DAD review continues on an annual basis while other techniques for case ascertainment are explored. Those characteristics identified as potential barriers to registration should be investigated to further identify possible process improvements in Alberta.

  The second part of our study investigates the collection of molecular data for CNS tumours in a new era of personalized patient care. The objectives for this part of the study are: 1) conduct a pilot study to explore the feasibility of electronically extracting molecular parameters for CNS tumours diagnosed in Calgary from CancerControl Alberta’s Electronic Medical Record (EMR); and 2) estimate the prevalence of testing for molecular parameters by histological subtype and year of diagnosis.

  Analysis involved text mining to extract molecular parameter information from the EMR for all invasive CNS cancers from 2010 to 2015 in the Calgary zone. This information was used to calculate the prevalence of availability of molecular parameter test results by histological subtypes and year of diagnosis. To assess the accuracy of the molecular data extracted, we linked cases to the physician databases and calculated percent agreement and Kappa coefficient.

  The results from the pilot study support the feasibility of extracting molecular characteristics for CNS tumours electronically from the EMR. Electronic extraction had a greater chance of missing information, however when information was accessible there was a high percent agreement (~99%) with the physician database. The study also showed Isocitrate dehydrogenase (IDH) testing for gliomas, O6-methylguanine-DNA methyltransferase (MGMT) testing for glioblastomas, loss of heterozygosity of 1p/19q testing for oligodendroglial tumours, and Alpha Thalassemia/Mental Retardation Syndrome X-linked (ATRX) testing for glioblastomas and astrocytic tumours have varied from 2010 to 2015. However, all four biomarkers had over 75% of relevant subtypes with a test result available in the EMR in 2015. These results support the need to formally collect this information with the possibility of using electronic extraction as a feasible solution to decrease workload.

• Subjects / Keywords

  • CNS tumours
  • brain tumours
  • case ascertainment
  • hospital discharge
  • completeness
  • cancer registry
  • EMR
  • molecular markers
  • biomarkers

• Graduation date

  Fall 2018

• Type of Item

  Thesis

• Degree

  Master of Science

• DOI

  https://doi.org/10.7939/R3JD4Q53V

• License

  Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.