Genetics testing firm is accused of “hiding vital breast cancer data”

BMJ 2012; 345 doi: (Published 01 November 2012) Cite this as: BMJ 2012;345:e7402
  1. Nigel Hawkes
  1. 1London

A private company that leads the market in genetic testing for breast cancer genetic mutations has refused to share information that could help patients, experts have charged.

Myriad Genetics, based in Salt Lake City, Utah, owns patents that make it the monopoly provider in the United States of tests for BRCA1 and BRCA2, two genetic variants that predispose women to breast cancer. It has conducted nearly a million tests and built up databases that give it unrivalled power to analyse the results, especially when they show mutations whose significance is uncertain.

When Myriad finds such a mutation, it offers free testing to family members, building up the database and enhancing its ability to assess the importance of “variants of unknown significance” (VUS).

In a paper in the European Journal of Human Genetics a team led by Robert Cook-Deegan of Duke University says that since 2005 Myriad has refused to make the databases publicly available.1 As a result, it can assign significance to a far greater proportion of VUS findings than other testing bodies can.

Whereas other testing organisations find VUS mutations in 20-30% of cases, Myriad’s rate is near 3%. This gives it a clear market advantage that will persist long after its patents expire in 2015 (assuming they are upheld by a case now being heard by the US Supreme Court).2

Although Myriad declines to share its data on the grounds that it is proprietary information, it does have access to public databases. Nothing it has done is illegal, but Cook-Deegan and colleagues argue that Myriad’s conduct has drawbacks for patients.

Those who use tests other than Myriad’s run a greater chance of being told that their results cannot be interpreted, and Myriad’s own conclusions are not open to challenge because its database is private.

Cook-Deegan and colleagues argue that Europe needs to think hard about how to deal with the matter as Myriad has declared its intention to increase its marketing effort there. Professor Martina Cornel, chair of the European Society of Human Genetics’ Public and Professional Policy Committee, said: “We are very concerned that such important data is being withheld from those who most need it. Interpreting the variants of unknown significance that may be found on analysing the patient’s genome plays an essential part in being able to provide proper counselling and if necessary, preventive or therapeutic guidance.

“It is vital that progress towards personalised medicine, which holds out so much promise, is not hindered by companies maintaining private genomic databases. Policymakers should take an urgent look at the regulatory and reimbursement issues involved in genomic testing in order for all the data that is essential to understanding the clinical significance of VUS to be made public, to the benefit of patients and healthcare providers alike.”

Cooke-Deegan told Genome Web: “I hope European countries will do what the United States has so far failed to do by requiring laboratories to share their data as a condition of payment for regulatory approval of laboratory services. A national healthcare system could refuse to pay for tests that they cannot independently verify or interpret. Or perhaps Myriad will decide to resume sharing their data—wouldn’t that be nice?”

Myriad is not alone in declining to share its data, Cook-Deegan and his coauthors say. Two other companies, Prevention Genetics, based in Marshfield, Wisconsin, and Medical Neurogenetics of Atlanta, Georgia, follow the same policy, though more than 100 other genetic testing services have agreed to contribute mutation data.

Cook-Deegan and colleagues suggest that journals could exert pressure to make data public, but that has limitations: Myriad’s publications give the sequence of only 118 of more than 1400 mutations it has studied, and publication guidelines apply only to those who wish to publish.

National health systems, insurers, and regulators have stronger weapons, including refusing to pay for tests unless clinically relevant data are shared. Cook-Deegan and colleagues said: “Payers in the USA did not foresee the problems of incomplete access to data and did not put in place policies to ensure independent verification of clinical predictions . . . payers in Europe, South America, Asia and other markets need not be so passive.”

Myriad did not respond to a request for comment.


Cite this as: BMJ 2012;345:e7402