All articles

Who tells the family about a reclassification?

A classification changes years after the report. The laboratory knows, the clinic has moved on, and the family is making decisions on the old answer.

A woman had a variant reported as likely pathogenic in a cardiomyopathy gene when she was twenty-eight. Her two sisters were tested for it; one carried it and began annual surveillance; the other did not and was discharged. Six years later the variant is downgraded to uncertain significance after population data showed it to be more common than a disease-causing variant could be.

The laboratory that reported it has the downgrade in its database. The cardiologist who ordered it has retired. The sister who was discharged has had a child, and nobody has told anyone in the family that the result they organised their lives around no longer stands.

What the field knows

Reclassification is a routine event. Mersch and colleagues (2018), reviewing a large hereditary cancer testing series, found that a notable share of variants reported over a decade were later reclassified, most often from uncertain significance towards benign, and that the volume of amended reports grew with the testing volume behind it. Harrison and Rehm (2019), examining reclassification data in ClinVar, found that variants classified as likely pathogenic were downgraded often enough that the term's implied confidence deserved scrutiny, and that the pattern differed by gene and by the era of the original classification.

The professional bodies have addressed the duty that follows. The ACMG points to consider on re-contact (David et al., 2019) set out that laboratories, clinicians and patients each hold part of the responsibility, that the laboratory's duty is usually discharged by issuing an amended report to the ordering clinician, and that the clinic's capacity to re-contact is limited by the systems it has. The European Society of Human Genetics recommendations (Carrieri et al., 2019) describe re-contact as desirable where a change is clinically significant, and acknowledge that no service has the resources to do it for every change, so that a triage of which changes warrant it is needed.

The reanalysis guidance (Deignan et al., 2019) distinguishes the case where a laboratory re-evaluates a variant it reported from the case where a clinician asks for a reanalysis, and notes that the first is largely driven by the laboratory's own review of its reported variants against new evidence. In practice that review depends on the laboratory keeping a register of what it reported and checking it against each new release of the evidence.

What the guidance cannot settle is the mechanics. A duty that rests on three parties, one of whom has retired, one of whom has moved house and one of whom has no system for finding either, is a duty that is discharged by accident or not at all.

Where practice breaks

The laboratory's knowledge stays in the laboratory. An amended report is generated, if the laboratory has a reported-variant register and reviews it. It is sent to the ordering clinician's address of record, which may be a department, a predecessor or a retired individual. There is no acknowledgement loop, so the laboratory does not know whether anyone read it.

The clinic receives the amendment as a document rather than an event. It is filed to the patient's record. Nothing links it to the sisters, who are separate patients, in separate records, possibly in separate hospitals. The cascade that the original result triggered has no reverse gear.

The decisions taken on the result are not visible to anyone revisiting it. The scientist who downgrades the variant does not know that a sister was discharged on the strength of it, and so cannot flag the downgrade as urgent for that reason. The classification is reviewed as a variant; the consequences were incurred as a family.

Finally, the burden of asking falls on the family. The guidance encourages patients to check back, and some do. A sister who was told she was in the clear has no reason to.

Upgrades are rarer than downgrades and worse when missed. A variant of uncertain significance that becomes likely pathogenic turns a family that was told nothing could be concluded into a family with a surveillance need, and the relatives who were never tested because the result was uncertain are now the people who most need to be found.

How Tulparus handles it

Every variant reported from the workbench is held in a reported-variant register with its classification, the evidence release it was classified against, the case it came from and the clinician who signed it. When a new evidence release is loaded, each reported variant is checked against it, and a change in classification, or in the evidence that supported the classification, opens a review item on the original case.

The case keeps the downstream record. When a result leads to testing of relatives, the workbench records the cascade: which relatives were tested for which variant, with what result and what action followed, as linked cases under the family. A reclassification review therefore shows the scientist who else is affected and how, and the item is marked for priority when a relative was discharged or placed under surveillance on the result.

Notification is a tracked action rather than a sent document. The review item names the people to be told, the ordering clinician and the clinicians of the cascade cases, and the workbench records who was notified, when, and whether the notification was acknowledged. If the ordering clinician's account is closed, the item is routed to the service's designated successor. The family is contacted by the clinic, in the clinic's own words, and the workbench records that it happened.

The amended report is rendered from the updated case, with the earlier report kept as issued, so the clinician who makes the call can show the family both documents and the dated reason for the change. For the family in the opening case, the discharged sister would have been a named item on the downgrade's review, and the question of who tells her would have had an answer with a name on it.

What you can do tomorrow

The register and the cascade record can be kept by hand, and most of the value is in keeping them at all.

  • Keep a register of reported variants with the classification and date, and review it against each new release of the classification archive, or at a fixed interval if that is all you can manage.
  • When a result triggers testing of relatives, write the relatives and their outcomes into the proband's case note, so that a reclassification review can see them.
  • Agree with the referring clinics a named route for amended reports that does not depend on an individual, and ask for an acknowledgement.
  • Tell families at the time of the original result that classifications can change, and give them a way to ask. It does not discharge the duty, and it catches some of what the system misses.

Sources

  1. Mersch J, Brown N, Pirzadeh-Miller S, et al. Prevalence of variant reclassification following hereditary cancer genetic testing. JAMA, 2018.
  2. Harrison SM, Rehm HL. Is 'likely pathogenic' really 90% likely? Reclassification data in ClinVar. Genome Medicine, 2019.
  3. David KL, Best RG, Brenman LM, et al. Patient re-contact after revision of genomic test results: points to consider, a statement of the American College of Medical Genetics and Genomics (ACMG). Genetics in Medicine, 2019.
  4. Carrieri D, Howard HC, Benjamin C, et al. Recontacting patients in clinical genetics services: recommendations of the European Society of Human Genetics. European Journal of Human Genetics, 2019.
  5. Deignan JL, Chung WK, Kearney HM, Monaghan KG, Rehder CW, Chao EC. Points to consider in the reevaluation and reanalysis of genomic test results: a statement of the American College of Medical Genetics and Genomics (ACMG). Genetics in Medicine, 2019.

One downgrade, three sisters.See who the case says to tell.

A demo takes 20 minutes, screen shared, on synthetic cases. Bring the question from one of your own.