Table 3.

Lessons learnt during the assessment of relevant MSAC applications

Theme Specific aspect	Why this aspect was a challenge for HTA	How challenges have been addressed by MSAC (to date)	Examples
Health condition
Proposed population too broad and heterogenous	Difficult to establish link between specific genotype and disease/risk (clinical validity), no ‘Star Performer’ gene(s)	Redefined (sub)populations for testing based on pathological and clinical characteristics (see Fig. 3)	1476 1504
Less common conditions	Limited clinical and/or epidemiological evidence as required by MSAC approach	Used evidence-based funding recommendations for clinically related populations to support funding for health conditions with less evidence (see Fig. 3)	1411 1504
Natural history of condition not known or not specified	Difficult to define health outcomes for individuals who do not receive testing	Shortened time horizon to reduce uncertainty (e.g. focus on avoidance of diagnostic odyssey)	1449
Pathways of care
Insufficient consideration of how and when an individual would be considered for testing	Unclear if proposed test would replace or be used in addition to current tests	Sought input from a wider range of stakeholders, and/or referred to relevant, current evidence-based guidelines to develop funding advice	1492 1504
	Place in diagnostic pathway unclear, especially with regard to use of triaging tests such as tumour histopathology
Pre-test probability of a positive result set at 10% by MSAC	Threshold challenged by Sponsors, due to perceived difficulties in quantifying this probability for many health conditions	Has not been a barrier to funding approvals for testing in affected individuals, in most cases the pre-test probability has been > 10%, and funding has been approved in specific instances where it has been < 10% but there is a high unmet need and high clinical utility	1411 1449 1504 1534 1573 1598
Insufficient consideration of who would request the test (referral for testing)	Broadening the type of health professionals who can request the test might improve access (see below) but might also lower diagnostic yield	Balance between facilitating access and optimising diagnostic yield considered on a case-by-case basis, and type of health professionals specified in the MBS item	All
Technology
Test performance	Analytic validation against a reference standard is not always possible	Assume 100% sensitivity and specificity	All
	Different types of testing methodologies (see below) associated with a different diagnostic yield	Engagement with stakeholders and experts to identify the testing methodology most likely to be used in practice in Australia, and the relative clinical importance (by condition) of different diagnostic yields	1534 1573
Type and range of proposed molecular techniques	Wide range of different techniques used in different combinations, with rapid evolution of new techniques	Assumed most efficient testing approach will be used, so funding approval (generally) does not specify technique(s)	1554
	Potential for incidental findings is increased with the use of WES and WGS (see below)	MBS item only specifies WES or WGS if absolutely required	1476
Rapid emergence of knowledge regarding pathogenic variants	Not feasible (evidence or resources required to undertake HTA) to re-assess a test every time a new variant is identified	Developed MBS item for re-interrogation of whole exome or genome sequence data, and circumstances when this MBS item should apply	1476
Effectiveness
Clinical utility	Insufficient information regarding the impact of test results on clinical management	Relied on clinical assumptions if appropriate, or if too uncertain shortened time frame of analysis to period before and immediately after the test	1449 1534
	Insufficient information regarding how change in clinical management impacts health outcomes
‘Star Performer’ genes	Concept was useful for targeted gene panels but was not feasible/appropriate for large gene panels, chromosome analysis, WES or WGS	Relied on management of phenotype (irrespective of specific genes) to inform assessment	1449 1476 1492 1533 1598
Diagnostic yield	Varies based on patient selection and not always available for the proposed population for testing	Engagement with stakeholders and experts to identify the relative clinical importance, by condition, of different diagnostic yields Redefined the population for testing to align with the available evidence, or assumed transferability of evidence from one population to another	1476
Non-health outcomes	Supporting restoration of ‘reproductive confidence’ as a measure of test impact, without placing a value on avoiding the birth of an affected child	Expressed value in terms of diagnostic yield to avoid any perception that judgements were being made regarding the nature of reproductive decisions made	1476 1492 1533
		Developed MBS items for Reproductive partner testing	1476
	How to value a diagnosis that has no (immediate) clinical utility	Accepted that positive test results can allow access to educational or disability support services and have the potential to become actionable in the future, but only made a qualitative judgement of these benefits, given that MSAC does not take a societal perspective	1476
Safety
Harms associated with test results	Potential to report ‘off-target’ mutations and variants of unknown significance, at initial sequencing and on re-interrogation	Assumed to be appropriately addressed via genetic counselling (see below) and quality programs in laboratories ensuring only pathogenic mutations are issued on molecular pathology reports	1476 1533
	Identification of disorders with no treatments or prevention interventions
Ethics and equity
Funding for pre- and post-test genetic counselling, for probands and for biological relatives	Recognised as integral to patient-centred care, but not coverable by the MBS	Assumed to be delivered by the clinician providing care, and included as a health system cost in the economic evaluations	All
Equitable access for all eligible individuals	Access will be limited if only clinical geneticists can request a test	Balance between equity and appropriateness of testing assessed by MSAC on a case-by-case basis	All
Cost effectiveness
Cascade testing	In circumstances where only biological relatives experience benefits from testing, the economic model still needs to include testing of probands	Concept of ‘co-production of utility’ was implemented, which informs the structure of economic models, and allows exploration of cost-effectiveness in first-, second-, and third-degree relatives	1534 1598
Quality of life	Limited evidence of impact of genetic testing on QoL	CUA attempted, but MSAC relied on CEA if the quality of life data were too uncertain, or, an evidence-based economic evaluation in one population was used to inform the decision for a clinically similar condition with less evidence (see Fig. 3)	1411 1476 1492 1504
	Limited evidence for utility weights for individuals with health conditions other than cancer or cardiovascular disease
Comparing ICERs	Difficult to directly compare cost-effectiveness across different tests when ICERs expressed in different units	Where possible, additional ICERs have been derived based on a relevant measure of diagnostic yield	1492 1504 1533 1573 1598
Budget impact
Number of individuals likely to be eligible for testing	Number of affected individuals who meet the criteria for testing will be larger than the number of individuals with the specific genetic variants of interest	Sought additional clinical input regarding the prevalence of individuals with the characteristics that would make them eligible for the proposed test	1449 1598
	Potential for use in unintended populations	Restrictions on eligibility for testing and on the test referrer	1476
Downstream costs and cost-offsets	Limited evidence on cost consequences of testing	If highly uncertain then budget impact based solely on costs of testing	1504 1534
Legal issues
Potential to detect non-paternity, consanguinity or incest	Outside the usual range of matters considered by MSAC, and represent broader policy issues that cannot be addressed via the MBS	Not resolved	Potentially all
Data storage and privacy
Possible forensic uses of sequence data
Implementation
Cascade testing	Limited evidence of rate of uptake of cascade testing in biological relatives	Assumptions made regarding ‘average’ uptake in families, and alternative scenarios tested in sensitivity analyses	All except 1411
Uptake of risk-reducing strategies in probands and biological relatives	Limited evidence of rates of uptake of different risk-reducing strategies (and age at uptake)	Assumptions made regarding uptake, which are then tested in sensitivity analyses	All except 1411
Health system efficiency	Tension between broad access to cost-ineffective testing versus access to cost-effective testing for individuals at highest risk	Not resolved	1492

CEA, cost-effectiveness analysis; clinical utility; the net health benefit/harm derived from an investigative health technology across all those tested (including true positives, false positives, true negatives, and false negatives); CUA, cost-utility analysis; diagnostic odyssey, the series of investigative services a patients undergoes until they receive a diagnosis; HTA, health technology assessment; ICER, incremental cost-effectiveness ratio; MBS, Medicare Benefits Schedule; MSAC, Medical Services Advisory Committee; QOL, quality of life; star performer; the actionable gene(s) for which the strongest clinical utility and/or cost-effectiveness argument is likely to apply for an affected individual; trio testing, when parents are tested at the same time a child is tested; WES, whole-exome sequencing; WGS, whole-genome sequencing