Population Genomic Screening for Three Common Hereditary Conditions: A Cost-Effectiveness Analysis

We enjoyed reading the article by Guzauskas et al modelling the cost-effectiveness of population-based genomic screening for hereditary breast and ovarian cancer syndrome, Lynch syndrome and familial hypercholesterolemia, all of which are designated tier 1 genomic applications by the Centers for Disease Control and Prevention [1]. We commend the authors on the excellent analysis. Modeling is a balance between simulating practices supported by randomized control trial data versus those advocated by strong recommendations or even expert opinion for surveillance and prevention per clinical management guidelines. If the latter were further incorporated in the reported analysis - specifically prevention strategies for ovarian, endometrial, pancreatic and prostate cancer in Lynch syndrome and pancreatic and prostate cancer in hereditary breast and ovarian cancer syndrome, we suspect the reported ICERs would be lower (more cost effective) than currently reported, and may influence age cut offs for screening thresholds [2,3]. With emerging new data on further cancer risks associated with hereditary syndromes and evidence-based prevention strategies, and the decreasing cost of genetic testing, population based genomic screening for these syndromes will likely become even more cost-effective in the future [4,5].

References

1. Guzauskas GF, Garbett S, Zhou Z, Schildcrout JS, Graves JA, Williams MS, Hao J, Jones LK, Spencer SJ, Jiang S, Veenstra DL, Peterson JF. Population genomic screening for three common hereditary conditions : a cost-effectiveness analysis. Ann Intern Med. 2023;176(5):585-595. doi:10.7326/M22-0846.

2. National Comprehensive Cancer Network Guidelines for Genetic/Familial High-Risk Assessment: Colorectal. Available from: https://www.nccn.org/professionals/physician_gls/pdf/genetics_colon.pdf [Accessed 30 May 2023].

3. National Comprehensive Cancer Network Guidelines for Genetic/Familial High-Risk Assessment: Breast, Ovarian, and Pancreatic. Available from: https://www.nccn.org/professionals/physician_gls/pdf/genetics_bop.pdf [Accessed 30 May 2023].

4. ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). Identifier: NCT05419011, Testing a Combination of Vaccines for Cancer Prevention in Lynch Syndrome. Available from: https://clinicaltrials.gov/ct2/show/NCT05419011 [Accessed 30 May 2023].

5. Wetterstrand KA. DNA Sequencing Costs: Data from the NHGRI Genome Sequencing Program (GSP) Available at: www.genome.gov/sequencingcostsdata. Accessed 30 May 2023.

We read with interest the recent paper by Guzauskas and colleagues on population genomic screening for Hereditary Breast-Ovarian Cancer (HBOC), Lynch syndrome (LS), and Familial Hypercholesterolemia (FH). They found that screening for these conditions using a restricted panel of causative genes is likely cost-effective in US adults aged ≤ 40 years, provided the cost of testing is relatively low and preventive interventions are accessible [1]. Targeted approaches to identify individuals with HBOC, LS, and FH belonging to high-risk families miss a substantial proportion of those with pathogenic variants. Therefore, public health-based screening programs are increasingly viewed as a possibility that deserves serious attention. Cost-effectiveness studies are a crucial step in this direction, but the potential harms and unintended consequences of this type of screening should also be considered. Moving from targeted testing to population screening will markedly increase the complexity of results interpretation. Moreover, early detection and preventive strategies for HBOC, LS, and FH, including e.g., increased surveillance, prophylactic surgery, and pharmacotherapy, are not risk-free. Adjusting psychologically to a positive mutation status and its consequences can prove especially challenging for individuals without a family history of the conditions screened for [2]. This and other potential harms, such as screening-related distress, genetic discrimination, false positives, false reassurance, overdiagnosis, overtreatment, undertreatment, inequitable access to preventive interventions, and other health disparities, require that screening candidates should be preemptively informed about the implications of results, both for themselves and their relatives [3]. To what degree the benefits and harms of screening and subsequent interventions for HBPC, LS, and FH should be discussed to enable informed decision-making remains to be understood [3]. Studies on ethical questions associated with population screening programs suggest that individuals offered screening should be informed about the related uncertainties, the content and form of information should be developed with the public and other stakeholders through participative approaches, and information should be layered and balanced [4]. Pre-test genetic counseling has a solid, long-standing tradition in clinical genetics. It has relied on dialogue, education, and informed decision-making to prepare and support individuals faced with adjusting to a genetic diagnosis [5]. Ethically sound integration of population genomic screening for HBOC, LS, and FH into routine healthcare will require that novel, ad hoc models of informed consent be developed, clarifying what constitutes the essential information that should be discussed, and ensuring that the non-genetic professionals who may be offering such screening are appropriately trained.

References

1. Guzauskas GF, Garbett S, Zhou Z, et al (2023) Population Genomic Screening for Three Common Hereditary Conditions. Ann Intern Med 176:585–595. https://doi.org/10.7326/M22-0846

2. Meiser B, Quinn VF, Gleeson M, et al (2016) When knowledge of a heritable gene mutation comes out of the blue: treatment-focused genetic testing in women newly diagnosed with breast cancer. European Journal of Human Genetics 24:1517–1523. https://doi.org/10.1038/ejhg.2016.69

3. Mighton C, Shickh S, Aguda V, et al (2022) From the patient to the population: Use of genomics for population screening. https://doi.org/10.3389/fgene.2022.893832

4. Hoffmann B (2020) Informing about mammographic screening: Ethical challenges and suggested solutions. Bioethics 34:483–492. https://doi.org/10.1111/bioe.2676

5. Resta R, Bowles Biesecker B, Bennett RL, et al (2006) A New Definition of Genetic Counseling: National Society of Genetic Counselors’ Task Force Report The National Society of Genetic Counselors’ Definition Task Force. J Genet Couns 15:. https://doi.org/10.1007/s10897-005-9014-3

We appreciate the thoughtful comments by Battistuzzi and colleagues about the need to consider the complexities and potential harms prior to initiating population genomic screening.  We agree that population genomic screening should be designed with fewer complexities than diagnostic testing.  As recommended previously [1], we assumed variants of uncertain significance (VUS) would not be returned to screened individuals in order to reduce complexity of results interpretation and the return process.  Additionally, we assumed a population genomic screening program would carefully select genes with high evidence for pathogenicity and known penetrance.  The model accounted for many but not all potential harms.  First, we assumed a 1-year disutility of 0.05 in the first model cycle to recognize the potential adverse psychological impact of receiving a positive screening result.  Secondly, we modeled potential harms related to risk-reducing mastectomy and/or risk-reducing salpingo-oophorectomy by incorporating disutilities into the year of the surgery, which reduced the quality adjusted life years gained from cancer risk reduction.  Short duration disutilities representing potential harms related to managing false positive results of surveillance testing were not included.  However, these harms would be found in both the genomic screened and unscreened groups.  Two of the three primary surveillance tests (colonoscopy and serum cholesterol) have few to no false positives as they represent the reference standard in clinical care.  A small proportion (3.9%) of females with a pathogenic variant in a breast cancer gene who selected intensive surveillance would be expected to have breast biopsies from false positive mammogram and breast MRI screening [2]; however, given the small size of that group, we do not think this would have an impact on the overall model result.  Finally, in the published analysis, we evaluated a scenario involving potential harm by modeling false reassurance (individuals who avoid standard screening after testing negative) that would quickly negate the benefits of population screening if present even at low rates.  To avoid false reassurance and minimize adverse psychological effects, we strongly agree that pre- and post-test education and counseling of individuals are needed prior to genomic screening for both cancer related and cardiovascular risks.  Given the scale of implementation required to achieve population-level screening, new economical models for effectively delivering this education and counseling across diverse populations are needed.  

REFERENCES

[1] Prince AE, Berg JS, Evans JP, Henderson G. Genomic Screening of the General Adult Population: Key Concepts for Assessing Net Benefit with Systematic Evidence Reviews. Genet Med. 2015 Jun; 17(6): 441–443.

[2] Warner E. Screening BRCA1 and BRCA2 Mutation Carriers for Breast Cancer.  Cancers (Basel). 2018 Dec; 10(12): 477.

Disclosures:

Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M22-0846.