The failure of Incyte’s (INCY) Phase 3 clinical trial of epacadostat combined with Keytruda, Merck’s breakthrough checkpoint inhibitor, for people newly diagnosed with melanoma was a big disappointment to the companies, the pharmaceutical industry, and patients. This negative trial prompted some to question whether combination immunotherapies have a future. Not me. I believe that immunotherapy and combination therapy are here to stay.
The questions we should be asking are what went wrong here (and with similar failures), and how do we fix it?
The answers lie with precision medicine and biomarkers.
Before the advent of precision medicine, doctors and clinical trialists had little insight into cancer patients’ biological makeup. Patients were often put through grueling — and expensive — treatment protocols without knowing how their tumors would react. And once we saw that a medication worked (or failed), we often had no idea why.
While we still have a long way to go, the introduction of precision medicine has ushered in an era of greater insight. What I mean by “precision medicine” is using individuals’ genetic and molecular makeups to diagnose or treat their disease. This is vastly different from the traditional method — a one-size-fits-all approach in which treatment is designed for the average person and does not take into account the differences between individuals.
We now have greater understanding of how a specific patient will react to a specific medication, and why. Instead of treating patients blindly, modern drug developers can use biomarkers to match the disease with the drug’s mechanism of action to focus on a precisely — and often narrowly — defined group of patients. Failing to take advantage of this medical advancement results in disappointing, expensive trial failures. Big companies may be able to afford these failures, but small ones like mine cannot. And no patients can afford such failures — their lives literally depend on the success of these therapies.
While immunotherapy has shown great promise, the currently available drugs work for only a minority of patients. The goal of the epacadostat/Keytruda trial, and more than 1,100 combination trials like it, was to see if combining an existing drug with a new therapy would improve survival.
Epacadostat (INCB024360) is an investigational, highly potent, and selective oral inhibitor of the indoleamine 2,3-dioxygenase (IDO1) enzyme. In preclinical studies, the combination of epacadostat and immune checkpoint inhibitors was shown to be effective against cancer.
Beyond Incyte’s failed combination trial, other proof-of-concept trials are ongoing in patients with unresectable or metastatic melanoma, non-small-cell lung cancer, kidney cancer, bladder cancer, and squamous cell carcinoma of the head and neck. In these studies, epacadostat combined with the CTLA-4 inhibitor ipilimumab, or the PD-1 inhibitors Keytruda or Opdivo, improved response rates compared to the checkpoint inhibitors alone.
I’m not clear about the biomarker strategy used in the epacadostat/Keytruda trial. Merck’s embrace of biomarkers for Keytruda has been responsible for its success in the market. In 2015, Keytruda received accelerated FDA approval for patients with non-small cell lung cancer and high expression of the programmed death-ligand 1 (PD-L1) biomarker. The use of biomarkers to select patients may be responsible for pushing Keytruda ahead of Bristol-Myers Squibb’s checkpoint inhibitor, Opdivo.
While Bristol-Myers Squibb opted for a broad patient population in its Opdivo trial for advanced non-small-cell lung cancer, Merck employed a precision-medicine and pro-biomarker strategy for Keytruda in its trial for the same disease, testing patients before the trial and allowing into it only those patients who expressed a predefined level of PD-L1. Merck’s trial succeeded, while Bristol-Myers Squibb’s trial failed.
It’s important to remember that not all biomarkers are created equal. Even with a biomarker program in place, it’s important to ask: Is the right biomarker being used?
Most biomarkers in use today are tumor specific, because our mindset is still largely focused on the type of cancer — the tumor type — an individual has. The first question that often gets asked when learning someone has cancer is, “What kind of cancer is it?” But the view of cancer is changing. For example, two women may be diagnosed with breast cancer. But one, who has HER2-positive cancer, will get a different treatment than one who has HER2-negative cancer.
For determining a treatment plan, the type of cancer isn’t the only relevant question. In the era of immunotherapy, the patient’s immune system is equally relevant for identifying successful treatment protocols. There are an increasing number of new cancer drugs in development — think larotrectinib, entrectinib, and others — that target genetic or cellular mutations, regardless of the origin of the cancer. Biomarkers need to follow suit.
Some of the most well-known biomarkers are tumor specific: PD-L1 for kidney, lung, bladder cancer and melanoma; BRCA for breast cancer; BRAF for melanoma; and HER-2 for breast cancer. But to get the full understanding of what treatment is best for an individual patient, we also need to focus on immune-specific biomarkers such as regulatory T cells, natural killer cells, myeloid derived suppressor cells, and others.
Clinical trials sometimes fail. That’s a reality all biotech and pharmaceutical companies face. When trials fail because a drug wasn’t as effective or as safe as expected, it’s disappointing. But it’s a tragedy when a good drug fails because of faulty trial design. Those failures are the most difficult to swallow because they have failed patients who may never get a chance to benefit from a therapy that may have made a difference for them.
In the immunotherapy era, smart use of tumor and immunologic biomarkers will help better match therapies to patients. Precision medicine will give us greater insight into how individuals will respond to treatments. We need to use both to make sure we get it right for cancer patients.