What are the key components in the process of setting up a new diagnostic assay as detailed by the ACCE framework?
Analytical validation, clinical validation, clinical utility and considerations of ethical, legal and social implications of the test.
What does the development stage of setting up a new test involve?
Assessment of both diagnostic and technical use of the process to ensure that the measurements obtained are relevant to the diagnostic question(s) and that the analyte(s) can be unambiguously identified (i.e. that there are no confounding factors).
The final stage is to determine whether the performance of the test, in terms of accuracy, meets the required diagnostic standards. This can be done by validation or verification depends on the existence of suitable performance specification that details the expected accuracy of the test
The result of the validation or verification determines whether and how the test will be implemented and set the requirements for performance monitoring (ongoing validation) of the test.
What must be considered when determining whether there is a clinical need for a service?
- What is the incidence (prevalence) of the condition being tested for?
- Determine the nature of the disease (i.e. are the severity and characteristics of the disease severe enough to warrant genetic diagnosis?
- Generally a test will be warranted if it 1) affects a large number of people and/or 2) has a severe affect
- Identify predisposition to a disease and evaluate the risk of developing a condition (carrier risk, reproductive counselling)
- Establish how many samples will be expect per annum
- Is the service already provided by an alternative accredited laboratory with sufficient capacity, or as part of a research programme?
What must be considered when determining whether there is a clinical demand for a service?
- Who is requesting the new service?
- Where would the request come from?
- Is it necessary to introduce this test if another laboratory already provides this test?
- If users need/want the test, would it be a definitive test or a screening test to rule out a condition?
What must be considered when determining the clinical utility of a service?
- Would testing directly affect the type of treatment or identify an effective drug regimen?
- Would it provide a diagnosis for a currently unidentifiable genetic disease, or one for which a test currently does not exist?
- Identification of low penetrate alleles?
- Do environmental factors have a significant contribution to the phenotype?
- Significant evidentiary strength to be clinically valid (similar to finding a CNV of unknown significance on aCGH or the mutations may not be linked by evidence to a specific clinical condition.
- Would testing include predictive and prenatal testing of family members at risk?
- Evaluate how testing would ultimately lead to improved health
What should be considered when determining whether a test is legal/ethical?
- This should particularly be considered with regard to predictive and prenatal testing
- Increased relevance with exome/WGS in terms of the ethical implications of reporting incidental findings and legal challenges associated with data storage and sharing.
- Is there any legal implications?
What should be considered when setting up the technical process of a new test?
- Determine the type of test needed (i.e. molecular, cyto, pyto, sequencing etc.), whether it is a single gene / multi gene disorder and whether there are any hotspot regions (i.e. EGFR in lung cancer)
- Are alternative tests available? Other tests may be needed in conjunction with a test test e.g. mutation analysis vs methylation
- Establish how accurate, reliable, effective, reproducible, specific and sensitive the test would be.
- Consider how best to minimise false positives and false negatives.
- Would the test detect intermediate outcomes? (i.e. mosaics in mitochondrial disorders, premutations in Fragile X)
- Would it provide too much information? (Incidental findings)
- What would be the best target? (DNA, RNA, chromosomes, genes, biomarkers) and is this material easy to obtain?
- Determine the best sample type for testing of the condition with regard to genetic load (i.e. some mutations not seen in blood like solid tumour etc)
When implementing a new test what do you need to consider in relation to current practice?
- What would be the potential laboratory workload for the test?
- Evaluate the level of staff training (clinical and scientific) needed to obtain relevant and required expertise to implement and run the service.
- Consider the time required to set up the service?
What do you need to consider when determining whether a new test will replace a current test?
- Does it lead to an improved turnaround time as well as being more sensitive/specific?
- Does the TAT fit with clinical need?
- How many cells to look at/how much DNA is needed and how does this compare to current requirements?
- Are there any national guidelines or EQA schemes yet?
- Has the test been validated externally? Even if it has it will require some degree of internal validation/verification - how will this be achieved?
- Are positive and negative controls required? And if so are they available?
- Confirmation of result by same test of other test? and how does this impact in other services.
What budget considerations need to be made when implementing a new test?
- Realistic budget needs to be established to enable planning for consumables, equipment and overhead costs. Are additional staff needed etc?
- Are the service costs feasible i.e. would the test be offered at a competitive price, in a competitive turn around time without losing out financially? Will the charged price cover the costs?
- Would funding be needed to accommodate the introduction of a new service or would it be added to existing workload?
- How much is the start up cost?
- What is the recurring cost of on-going testing (number of tests/year and the maintenance of equipment)
- Does a business case meed to be put forward to acquire necessary funds?
- If the result indicates treatment with a particular drug, would finding from the pharmaceutical company be possible?
What are the implications on staff when introducing a new test?
- Need to determine the potential workload and staffing requirements.
- Determine the level of bioinformatics or IT support needed plus data storage requirements
- Health and safety aspects should be considered
- Can it be developed in house or are kits better placed?
- A staffing structure needs to be put in place for the service.
What equipment considerations should be made when implementing a new test?
- Determine which tehcniques ot use for the service
- Determine the requirements for new equipment - how would it be purchased?
- If new equipment needed determine the start up and maintenance cost, running cost and cost per test after implementation
- Does a cheaper test exist? Is there a way to automate the process?
- If new equipment is needed can it be used for other tests in the future?
- Is there suitable space on the premises?
What population factors should be consider when implementing a new test?
- What is the incidence of the disease in the population? (impact on test throughput requirements)
- Does the incidence warrant the implementation of the service?
- Will it be offered as a regional or national service?
- Are some patients more affected than others? (variable expressive, reduced penetrance, pre-mutation, recessive carrier?)
- What is the ethnicity of the population? Is the disorder seen at a higher prevalence in certain ethnicities?
What background information is needed before development of a new test is carried out?
- Clinical aspects including molecular genetics of the condition, inheritance patterns
- Determine the clinical referral and acceptance criteria in agreement with clinicians
- Research the technical aspects including procurement of equipment, assay design, order reagents and determine and obtain positive controls.
What is involved with the Validation of a new test?
- Gain experience of the new test
- Make adjustments to protocols as required
- Write SOPs
- Screen a batch of samples
- test samples in parallel using old/new method and positive and negtaive controls to determine accuracy, repeatability, reproducibility and robustness.
- Undertake short and medium term audits to assess the new test.
- If available enroll in a relevant EQA scheme
- Make sure the test adheres to ISO requirements.
Following development and validation/verification, what is needed to implement the service?
- Ensure all staff working in the service are trained properly and are familiar with the SOP
- Authorise the SOP
- Determine how to manage workloads
- Design worksheets and information folders
- Create reporting templates
- Advertise the service
- Where appropriate submit a gene dossier/form Bets Practice Guidelines
-
Computing5
-
Data12
-
Variant Analysis & Annotation7
-
Service Development16
-
NGS File Formats18
-
NGS16
-
Pathway Analysis & Ontology13
-
Bioinformatics Tools12
-
Software Development22
-
SQL30
-
Command Line32
-
GIT21
-
Good Programming Practice10
-
Databases10
-
OSFA Station Ideas15
-
Bioinformatics Pipelines8
-
SOP Review6
-
NGS Quality19
-
Validation of NGS pipelines9
-
BPG7
