To the Editor:
In their recent special article, Appelbaum et al. examine whether there is an ethical duty to reinterpret genetic variants and communicate reclassifications to patients and clinicians.1 After considering ethical arguments on both sides of the issue, they assert that such a duty should be recognized and describe four elements required to fulfill this duty: data storage, initiation of reinterpretation, conduct of reinterpretation, and patient recontact.
The arguments in the article are weighted toward testing for hereditary cancer risk, which is cited by the authors as the most common reason for clinical sequencing. Variant(s) of uncertain significance (VUS) are concerning because they create uncertainty for both patients and clinicians and make it difficult to define appropriate actionable next steps. Appelbaum et al. conclude that patients will benefit from a systematic approach to variant reclassification and call for discussion to reach consensus on the ethical duty to reinterpret, reclassify, and recontact.
We appreciate Appelbaum et al.’s analysis but want to point out additional complexities posed specifically by variant reclassification following genetic carrier screening for reproductive planning. These complexities arise first because of time constraints implicit in carrier screening: Post-test recommendations and next steps are most relevant at or soon after the time of testing—either when considering pregnancy or during pregnancy. In our laboratory’s experience, ~1% of patients undergoing carrier screening receive a reclassification to a clinically actionable result. The interval between initial testing and reclassification ranges from 8 days to 1½ years, with half of patients receiving a reclassification ~7 months or more after initial screening. Although the proportion of clinically significant reclassifications is small, the ethical and logistical issues of reporting them so long after initial testing still apply to many patients. Second, although carrier screening and diagnostic testing share a standard of practice in which benign/likely benign variants are not reported, they differ in that VUS are routinely reported in diagnostic testing but not in carrier screening. Thus, carrier screening is especially subject to patients and providers being surprised with reclassifications from VUS to pathogenic/likely pathogenic when they were not even aware the variants existed.
Before ordering carrier screening, providers discuss the benefits and limitations of the screening with their patients, including that a screening test reduces, but does not eliminate, risk of being a carrier and that results may affect their care. Patient consent to screening occurs within the context of family planning, and the results inform the immediate need for partner screening and follow-up. However, new information obtained late in pregnancy, following delivery, or even many years after the birth of a child, especially if it is neither requested nor expected, no longer serves the original indication for which consent was obtained. Are we acting with nonmaleficence if we share all reclassifications, despite the obvious psychological and emotional costs to presenting them at a time in a couple’s reproductive journey when the revised results cannot be acted upon?
Furthermore, prenatal care providers have expressed concerns that reclassifications can interrupt their workflows and introduce liability issues, such as when pregnancies are late term, partners have not yet been tested, or patients are no longer being followed but may need to know about the reclassified variant for the next pregnancy. Are we acting with beneficence if we disrupt the clinical workflow of clinicians and raise doubts concerning the reliability of medical results?
As with all patient care, health-care providers must include their patients in health-care decisions by weighing the psychological, emotional, clinical, and financial costs against the benefits for the health and well-being of the patients and their families. Adhering to this ideal in patient-centered decision making is difficult when unreported variants are later reclassified as actionable and reported. Furthermore, challenges remain for a laboratory that offers both diagnostic and carrier testing for the same genes. For example, suppose the laboratory had previously reported a VUS in HEXA in a child with infantile neurodegeneration referred for possible Tay–Sachs disease. The laboratory later reclassifies the variant as pathogenic and issues revised reports for all children whose diagnostic test results had included the VUS. What if the same VUS had been seen and not reported in clients undergoing carrier testing? Even if the laboratory had adopted a policy of not reinterpreting VUS during carrier screening, diagnostic testing procedures would have rendered such a policy moot. Should the laboratory suppress this knowledge for carrier screening and not issue an amended report? Further complicating the decision is when the variant is in a gene such as ATM or FH (fumarate hydratase), for which reclassification from VUS to pathogenic following diagnostic testing for cancer risk would have personal health implications that go beyond simply assessing the risk for ataxia telangiectasia or fumarase deficiency in the pregnancy.
In summary, Appelbaum et al. assess the arguments for and against variant reinterpretation, reclassification, and reporting in the course of diagnostic testing and decide that the balance is in favor of doing so. Carrier screening, however, poses additional ethical, medical–legal, and logistical arguments against reporting reclassified variants that apply little if at all to diagnostic testing. In the future, we can imagine a continuum of care model in which genetic information informs management and treatment during all stages of life, and in which variant reclassification throughout someone’s lifetime is expected and routine. However, we are not there yet. It may be advisable to attempt to achieve consensus on when to report reclassified variants by considering categories of testing defined by clinical indication and context rather than by attempting a generic approach for all testing. We believe that the needs and expectations for testing are different in the context of reproductive health and pregnancy management than in other clinical areas and that an approach for dealing with variant reclassification needs to consider those differences until a continuum of care model is available.
Appelbaum PS, Parens K, Berger SM, et al. Is there a duty to reinterpret genetic data? The ethical dimensions. Genet Med. 2020;22:633–639.
N.F., S.A., K.W.A., and R.L.N. are employees and shareholders of Invitae.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Faulkner, N., Aradhya, S., Aradhya, K.W. et al. Correspondence on “Is there a duty to reinterpret genetic data? The ethical dimensions” by Appelbaum et al.. Genet Med (2020). https://doi.org/10.1038/s41436-020-00954-5