DNA evidence used to prevent costly, traumatic medical errors

The same DNA tests used to catch criminals are now being used to prevent doctors from operating on the wrong patients.

Indianapolis-based Strand Analytical Laboratories, which was started a decade ago to run DNA tests on crime scene evidence, now spends most of its time helping solve false positives on cancer biopsies.

The company, started by former Marion County Prosecutor Scott Newman, did not set out to switch its focus. However, six years ago, one of the company’s leaders saw a news report about a patient who had unnecessary prostate surgery. His biopsy was switched with that of another man who did have cancer.

Such events are surprisingly common, and the effects can be severe.

In 2010, a Boston man sued his hospital after he became incontinent after prostate surgery that revealed he did not have cancer. After his surgery, he learned the pathologist had switched his specimen with that of another man.

A Pittsburgh man left impotent by prostate surgery learned 10 days after surgery that there had been a mix-up of his specimen with that of another man. He won a multimillion lawsuit against his healthcare providers.

In 2008 a Long Island woman died after having surgery to remove both of her breasts, although it turned out that her samples had been interchanged with that of another woman.

“We realized what we do on a forensic basis every day could be easily adapted to the clinical setting to solve this problem,” said Travis Morgan, chief financial officer. “It was really sort of by accident.”

Today, the Know Error system, which the company devised to test specimens that come back positive to make sure they belong to the putative patient, has been used on about 175,000 patients who have undergone breast and prostate biopsies. About 6 percent of all prostate biopsies and 1 percent of all breast biopsies use the Know Error system, Morgan said, and the company would like to see that expand.

Such testing now far outpaces the lab’s forensic science work. Crime tests make up only about 10 percent of Strand’s work and the company has grown from about 20 to 55 employees in a 12,000-square-foot space near the airport.

A study that will appear in the April issue of the Journal of Urology documents just how common and costly such mix-ups, known in science-speak as specimen provenance complications, are. Some medical malpractice law firms even specialize in such cases.

The most recent study, done by Strand Diagnostics scientists in conjunction with others, finds that such mistakes happen in about 2.5 percent of all prostate biopsies, leading to misdiagnoses in about .57 percent, or for about 1 in every 200 patients. That results in an estimated $879.9 million in wasted medical treatment costs, of which $694.8 million comes in medical legal expenditures.

“The conclusion is that it saves the system significant amounts of money,” Morgan said.

Nor does this take into account the psychological toll – both for the patient who may undergo surgery unnecessarily as well as the patient whose correct positive cancer diagnosis may be delayed for weeks as the specimen switch gets sorted out.

While Dr. Michael Koch, chairman of the department of urology at the Indiana University School of Medicine, has only had about two of 3,000 to 4,000 of his prostatectomy patients turn out not to have cancer – and both of those had precancerous changes – he said that may be because the academic medical center where he works does all its biopsy in-house.

To prevent mistakes, the IU lab will not run the same type of biopsy twice in a row to make sure that no tissue remaining from the previous test can contaminate the next sample. But some commercial labs specialize in one type of biopsy.

“If you have a lab that only processes prostate specimens it makes sense to do this,” Koch said. “I think, if it was your own biopsy, if there was a 2 to 3 percent chance it’s wrong, you would be willing to pay a little more to make sure there was no error.”

While other biopsy mix-ups have been known to happen — last year for instance, a Florida woman had a portion of her rectum removed after her specimen was switched in the lab — Strand focuses on the two most common types of cancer, Morgan said. Breast and prostate cancers make up 29 percent of all new cancer diagnoses. The second leading types are less than half as frequent.

Just about every week, the Strand tests turn up an error, Morgan said. Errors can include contaminated samples or samples that simply get switched and they happen in every one of the 230 or so labs for which the company runs tests, he said.

A kit helps the providers start tracking the sample from the moment a biopsy is taken, using the forensic chain of custody procedures. That includes a cheek, or buccal, DNA swab plus all of the supplies used for biopsies. The buccal swab goes off to Strand while the biopsy heads to the lab.

Once the lab completes the biopsy, if it’s positive, they send off a sample from that to Strand, which then checks that against the reference DNA sample just as they would with crime scene evidence, Morgan said. If the two do not match, Strand lets the doctor know immediately there’s a problem so they can stop any planned therapy until the error is resolved.

Because Strand does so many of these tests, they have streamlined the process so it only adds about two or three days to the biopsy-to-result time.

“It doesn’t matter how meticulous you are … you just cannot reduce the error rate to zero,” Morgan said. “This is a safety net that allows you to appreciate the errors that do make it through all of those quality improvement steps and recognize them before a patient is harmed.”

Call Star reporter Shari Rudavsky at (317) 444-6354. Follow her on Twitter: @srudavsky.