A small study by scientists from a clinical genetic testing company found that at-home genetic testing kits misidentified risk factor genes 40 percent of the time – reporting that customers carried possibly disease-causing DNA sequence variations when in reality, the sequences were normal.

The results are a sobering reminder that direct-to-consumer (DTC) genetic analyses, as fun as they may be in telling us about our ancestry, are not accurate enough to provide medical diagnostic information.

As described in the journal Genetics in Medicine, the team reviewed sequence data from DTC tests performed on 49 individuals. In these cases, either the DTC company’s own report or that of a third-party interpretation service flagged the individual’s sequences as being potentially dangerous, and thus, the data was referred to Ambry Genetics for a thorough analysis. Among the 49 subjects, the DTC reports had identified a total of 26 pathogenic variants in genes for cancer, connective tissue disorders, cystic fibrosis, and familial Mediterranean fever.

DTC companies are not allowed by the Food & Drug Administration to market their services as diagnostic tests, yet some companies will give customers the raw data from analyses of selected genes associated with diseases. This data can then be further interpreted by third-party companies, often for a fee.

However, there are two big problems with this: First, DTC companies don’t actually sequence the entire genes associated with say, cancer or Parkinson’s disease. Instead, their tests look for the presence of single nucleotide alterations, called SNPs, that have been previously linked to the development of a disease. Second, DTCs don’t test within all the genes linked to these conditions, and variations in one alone may be safe.

Furthering the problem, the third-party interpretation companies often rely on public databases of gene SNPs to identify high-risk variants, but the classifications by these online tools of which SNPs are dangerous are known to be highly inaccurate. Often, what is flagged as a high-risk variation is actually benign or of currently unknown significance to our health.

For their investigation, the Ambry scientists performed advanced sequencing confirmation tests on DNA samples from the 49 patients and compared their results to the DTC data from each. Their analysis showed that only 60 percent of the individuals actually had the abnormal gene sequence that their DTC test had identified. 

Moreover, the DTC reports and third-party interpretations associated with the 49 cases had classified eight variants in five genes (including the breast and ovarian cancer risk genes BRCA1 and BRCA2) as being causative of disease, whereas Ambry and other professional laboratories consider seven of these to be harmless and one to be of unknown significance.

In light of these findings, the authors worry that many DTC consumers are being presented with misleading information that could cause undue anxiety over their health.

“For these reasons, medical providers should order confirmatory genetic testing from an experienced clinical diagnostic laboratory to guide patients’ medical care,” they write. Once this has been performed, it must "be interpreted by a qualified health-care professional in the context of several other factors, such as personal and family medical history.”