all the information, none of the junk | biotech • healthcare • life sciences

Immuneering, Led by Young CEO and Mentor, Aims to Pick Which Cancer Drugs Should Work

(Page 2 of 3)

whether a drug will actually work or not.

“Nobody that we know of is taking this kind of approach,” Carpenter says. “Maybe it’s because I was a computer scientist in the early days, but I love the precision of the mathematical modeling. This is the way to go to understand how drugs interact with the immune system, so we can make better predictions on which drugs will work.”

Immuneering is starting out by trying to solve one of the classic prescribing problems for oncologists. Patients with kidney cancer or melanoma that has spread through the body can get Novartis’ interleukin-2 (Proleukin), a drug that offers about a one-in-10 chance of providing a long-lasting remission, while putting patients through nasty side effects that often end up putting them in an intensive care unit, Carpenter says. The drug costs $60,000, so insurers understandably would like to see some solid evidence it will work before they start writing reimbursement checks.

The Immuneering model looks at all kinds of variables in blood and tissue samples to get a handle on the probability a patient will be one of those success stories. It looks at ligand proteins on the surface of a type of immune cell known as Natural Killer cells. The Immuneering test examines whether a Natural Killer cell has characteristics that make it likely to be activated by a drug, but it accounts as well for whether the tumor has some of the sneaky traits that undermine the tumor-killing ability of Natural Killer cells. It also takes into account other factors, such as how many inflammatory proteins are in the blood with the potential to be activated against tumors, Carpenter says. It might even be useful for predicting how patients will respond to treatment for autoimmune diseases, in which the immune system goes haywire and starts attacking healthy tissues. The research underpinning all this has its roots in the MIT lab of Doug Lauffenburger.

This whole concept still needs proof, and the proof is going to require some money. About $2 million in venture capital, to be precise, Carpenter says. That amount is needed to pay for a study of 60 patient samples, which looks backward in time to see whether this test would have successfully predicted who responded, and who didn’t, to a course of interleukin-2 treatment. The study has already received approval by the patient safety monitoring board of Kaiser Permanente, Carpenter says.

A future study, a prospective one that makes predictions upfront and then follows patients over time to see how accurate those predictions really are, will be done later to pass tougher muster with the statisticians.

If that test and subsequent studies are successful, Carpenter says he envisions building … Next Page »

Single Page Currently on Page: 1 2 3 previous page

By posting a comment, you agree to our terms and conditions.

Comments are closed.