By Rachel Sachs
Last week, the FDA issued a report presenting 20 case studies of laboratory-developed tests (LDTs) that have or may have harmed patients, in support of its ongoing efforts to impose greater regulatory oversight on LDTs. The report came just before yet another House Energy & Commerce Committee hearing on the issue, and the timing of its release may have been motivated in part by questions FDA officials faced at the last Congressional hearing, about the existing evidence of harm to patients. The report categorizes the 20 case studies into seven groups, organized by the primary problem posed by the LDT: those with a high degree of false positives (the test yields a positive result when the disease is not present), those with a high degree of false negatives (the test yields a negative result when the disease is present), those which yield both false positives and false negatives, those which test for a factor seemingly unrelated to the disease in question, those linked to treatments based on disproven scientific concepts, those which affirmatively undermined the drug approval process, and of course, “other.”
I’ve blogged here about LDTs no less than five times (including about the last House hearing), and longtime readers may recognize a few of the case studies. OvaSure, a blood test which purported to detect the existence of ovarian cancer but which in fact had a very high false positive rate, is likely the most well-known. A false positive from OvaSure may have caused women to have healthy ovaries removed, a major procedure not only requiring a hospital stay and lengthy recovery, but also imposing psychological burdens on women who would not otherwise have wanted to become infertile.
A particularly worrisome example from the false negative category is the Oncotype DX HER2 Breast Cancer test. A significant minority of breast cancer tumors have a mutation causing them to overexpress a protein called HER2, and women with these tumors may be treated with cancer drugs specifically targeted to these mutations. As such, accurate companion diagnostics are essential to properly treat these patients, and the FDA had approved such tests already. But an LDT on the market, the Oncotype DX HER2 test, had a high rate of false negatives. The result was that patients testing negative failed to receive the right treatment for their cancer, putting them at higher risk the cancer would progress and potentially reducing their life expectancy.
Most interesting to patent scholars will be a test in the third category (too many false positives and false negatives): non-invasive prenatal testing (NIPT), also called cell-free fetal DNA testing. I’ve previously blogged about the ongoing patent litigation in this area, but essentially these are tests determining whether a fetus possesses a genetic abnormality not by performing risky procedures like amniocentesis, but simply by drawing blood and examining the small amounts of fetal DNA circulating in the pregnant woman’s bloodstream. The FDA is primarily concerned about the high rate of false positives of these tests, which in some cases caused women to abort healthy pregnancies. However, the tests do appear to have a high false negative rate as well.
There are a number of interesting questions to examine in this report, including the FDA’s choice of “alternatives” for each of the relevant LDTs. It will be interesting to see if these case studies continue to appear either in the FDA’s procession toward a final guidance or in the Congressional discussions about whether a legislative solution is more appropriate.