Part 2 of IVD Hall of Fame
« Come Celebrate IVD Appreciation Month!
The year was 1996, and I was sitting in a meeting room at a hotel in Washington, D.C. It was the annual Retroviral Conference, and the speaker was Dr. John Mellors from the University of Pittsburgh. He was about to present a paper on HIV viral load. And the world of HIV research was about to be turned on its ear.
You see, up until that point, there wasn’t a real good surrogate marker for patient health. You had an antibody test to determine if you were HIV positive, and then there was CD4, which let you know that you were getting sicker. The problem with the latter was that it was usually confirming what was already suspected from a clinical examination. By the time CD4 levels began to drop to a clinically relevant level, the patient was already in rapid decline.
There had been some research into the role of HIV viral load on disease management, but the problem then was that the only reliable way to measure it was by culturing the virus, which was a laborious and very long (28 days) process. By the time you had a viral load measurement by culture, the change in viral load had usually manifested itself by a drop in CD4.
Two companies – Chiron and Roche – had fledgling viral load tests, but the clinical utility had yet to be proven. Then Dr. Mellors opened his freezer and allowed the research team at Chiron to test his large cohort of HIV patients. This was an extremely valuable set of samples, with clinical outcome data for the majority of the specimens, spanning 10 years in some cases. Testing these was a bit of a risk, since they had unfortunately been collected in Heparin tubes. PCR was out due to Heparin interference, but the bDNA test that Chiron had might work, albeit with lower virus levels than expected.
When Dr. Mellors put up the Kaplan-Meier graphs for the patients in the cohort during the ’96 conference, the previously silent conference room was abuzz. For the first time, there was a clear clinical correlation between viral load and prognosis.
Samples had been collected on patients at the time of diagnosis of HIV infection. Those with a relatively low viral load at that time of diagnosis had a slower rate of progression to AIDS, some living for several years post-infection. Those that had a high viral load progressed to AIDS rapidly, and were often dead within months of diagnosis. Median levels of infection led to median survival rates. More telling, some of those patients that progressed to disease had normal CD4 counts, showing that viral load was the more accurate predictor of clinical outcome. The data was - for lack of a better term - textbook Kaplan-Meier.
Everyone in the packed room that day knew what this meant. There was finally a tool for managing patient health that was clinically accurate and could be used in real-time to measure patient health. Pharma now had a diagnostic tool that could be used to measure drug efficacy in clinical trials. Doctors had a tool to determine if a patient had developed resistance to the current medications. The treatment of HIV had been transformed.
For the rest of the meeting, Dr. Mellors was a rock star. There was always a crowd around him, congratulating him and asking him for details on the study. A few months later, the study was reported in Science. More buzz. This set off an arms race amongst the diagnostic companies to push the sensitivity lower. Roche adopted a centrifugation method developed at Chiron (by yours truly) for concentrating the virus, pushing their test to 50 copies / mL. A few months later, Chiron introduced a 50 copy ultrasensitive test of their own. Both tests eventually won FDA approval, though Roche’s was significantly ahead of Chiron’s.
Meanwhile, drug companies raced furiously to show that their drug was better at keeping viral loads below detection. Eventually triple combination drug cocktails solved the problem of resistance, keeping the viral load down to a manageable level. Blood banks eventually adopt a version of the viral load assays to screen the blood supply, as molecular tests could detect the presence of virus weeks earlier than the traditional antibody screening tests. The disease, though not cured, was at least under control. Long-term survival became a reality for thousands of HIV (+) patients.
There have been a lot of significant diagnostic tests in the past, but it’s hard to imagine one that had more of an immediate impact on healthcare than the HIV viral load tests. The pace with which new therapies hit the market, and the subsequent innovations in patient management, would be difficult to imagine without the HIV viral load test. And this is why the test deserves a place in our Diagnostic Hall of Fame.