For virtually every hospital acquired infection (HAI), even multi-drug resistant super bugs, there is a drug that could effectively treat the infection. In the United States, however, almost 100,000 deaths per year are attributable to HAI. The coexistence of these two facts is caused by the time it takes to correctly identify an infection and the drug(s) to which the underlying bacteria are susceptible. While bacteria and their resistance profile can easily be identified through culturing this process takes 48 to 72 hours. This 2-3 day period is much longer than time needed to promote treatment success where “[starting the suitable treatment] the next day is usually too late” according to Accelr8 President David Howson.
With this time lag, doctors are often left to make an educated guess on which medication to choose, guided by previous epidemiology of infections seen in their hospital and hope for the best. Accelr8 Technology Corporation (Alternext: AXK) aims to provide a system that will offer the information physicians need within the required time window. To address this challenge rather than “working to develop a new scientific principle,” the company is “taking what is known and putting it together in a different way.” The company’s BACcel system takes a two step approach, first identifying the bacteria in two hours and then identifying antibiotic resistance over the next four hours – a time period which should vastly improve treatment success.
This vast increase in speed is accomplished because instead of requiring culturing and the associated time it takes bacteria to grow, the BACcel system works on the actual bacterial cells present in the patient’s sample. After a technician performs standard sample cleanup, the patient sample is loaded into up to 32 different compartments within a cassette and placed into the BACcel device. The system then rapidly extracts the bacteria using an electric field, fixes them in place using proprietary coating, and tags their location. With the bacteria fixed and located, they begin to grow and replicate almost immediately, and the system’s camera and microscope can repeatedly interrogate each one. Bacteria are identified through labeling with antibodies and other indicators. After collection of this data, the system provides quantification of bacterial species (in colony forming units per milliliter) within the sample. Upon completion of this step, the particular bacterial infection will have been identified – all within two hours (sample preparation time included).
The second step of the process, which takes an additional four to six hours, involves exposing the bacteria to different antibiotics to identify resistances. Although it is infeasible to test every possible medication, the system works on a rule-out basis, testing a variety of medication classes, specifically those to which the particular bacterial species is known to commonly form resistance. In many cases, the system may identify a particular drug or drugs to which the bacteria are susceptible and provide the doctor with a choice among treatments. Even in cases where it does not identify a specific drug, it will eliminate all but a few options through identification of major classes of drugs that did not affect the bacteria’s growth, giving the physician a very small subset of options from which to choose. According to Howson, core to the system’s value is that “it doesn’t just provide information – it will actually help doctors make decisions.”
Along with the systems in the company’s own lab, Accelr8 also has placed two research units in hospitals at Denver Health Medical Center and Barnes-Jewish Hospital in St. Louis. At Denver Health Medical Center the device is currently being used as part of a study to investigate ways to speed the diagnosis of ventilator-associated pneumonia in ICU patients. Accelr8 is also working to further its technology as it moves towards a commercial device for diagnosis. In these efforts, the company has entered into a development agreement with Becton, Dickinson and Company (BD), a leader in the diagnostic field. The company is also working to improve sensitivity and specificity of tests for each type of bacteria to at least 93%, a level at which the tests become “commercially viable” – it has reached an 85% level on almost all of its tests. The company has already surpassed this target 93% level for MRSA, one of the most common infections, as well as several other bacteria. Howson expects the company’s best path to quickly get to market may be to license out the technology once it surpasses these levels.
Howson has high hopes for the device if it can deliver on identification and resistance profiling within the eight hour time window: “A reduction in empirical treatment failure by 50% and to become the standard of care in the ICU.” With tens of thousands of lives hanging in the balance, emphasizing existing technologies instead of waiting years for a new drug application could dramatically improve the prognosis for patients with drug resistant infections.
