How Drugs Are Made: Genetic Data Begins To Pay Off For Pharmaceutical Companies Researching New Treatments

Any day now, two new medicines for treating high cholesterol are expected to earn final FDA approval and hit the market -- Repatha by Amgen and Praluent by Sanofi and Regeneron Pharmaceuticals. Their arrival is being widely celebrated because they are the first new drugs for high cholesterol in decades, promising to lower the amount of bad cholesterol in patients' blood by as much as 60 percent. As a result, the companies that created them are expected to earn a premium over existing medicines.

Both drugs are known as PCSK-9 inhibitors because they target a mutation in the PCSK-9 gene, which causes high cholesterol. The explosion of the genetics field a decade ago raised expectations for the promise of such treatments. Scientists have increasingly realized that far more diseases than previously thought have a genetic link -- obesity, for example, is somehow related to at least 140 locations on the human genome.

Bence Boelcskevy, co-founder of the Drug Development Institute at Ohio State University’s Comprehensive Cancer Center, says about half of a person’s proclivity to any particular disease can be explained through genetics -- and that shouldn't come as a surprise since genetics control many of the body's essential functions through proteins that initiate chemical reactions for everything from digestion to heart rate. “When you think about it, we're nothing but a great big bag of biochemical reactions that are happening billions of times a second -- what could go wrong?" he says jokingly.

Banking On Genetic Insights

To find out what exactly can go wrong when genes get out of whack, the U.S. government got the ball rolling a decade ago with the Human Genome Project. More recently, President Obama requested $215 million to launch a research initiative focused on using genetic insights to create customized medicines and a frenzy of biotech startups have banked their business models on the potential of genetics to deliver groundbreaking treatments.

A handful of gene-based medicines have already earned FDA approval, including Herceptin for breast cancer and Zelboraf for skin cancer. But despite all the excitement, relatively few new genetic treatments have made it to the market since genetics became one of the hottest topics in research.

The reason for the lag is that even though most major pharmaceutical companies have begun using genetic information to identify and evaluate potential new drugs, the process is still messy and companies have been hesitant to pair genetic insights with the tried-and-true habits of evaluation on which they have leaned for decades.

“Genetics is an area that is under tremendous growth right now,” Boelcskevy says. “The road is never straight -- it looks like a pretzel.”

Pharmaceutical companies eager to leverage genetics can do so in one of two ways. The first is to hire teams of researchers to mine case studies and the publicly available genomic data for clues that hint at a potential new treatment. In the case of the PCSK-9 inhibitors, researchers examined a family with high cholesterol and discovered a mutation that was to blame.

Boelcskevy says pharmaceutical companies have increasingly partnered with biotech startups and universities to search for promising candidates in this manner. But casting such a wide net can lead to many dead ends or false positives – just because there is a clear link between a gene and a disease doesn’t mean that a drug built to work on that gene won’t have catastrophic side effects. This frustrating pattern has discouraged many companies from investing as heavily as expected.

That’s why some geneticists suggest another approach -- examining drugs that are already in a pharmaceutical company’s pipeline for genomic links. Today, when researchers examine a drug’s potential, they rely on mouse models or gene expression studies, an imperfect way to gauge how a drug might work in people. These flaws explain why, currently, only one out of every 1,000 potential drug candidates makes it to clinical trials and only one in five of those drugs will be approved.

An early genetic analysis, on the other hand, could provide information about the actual human impact of altering a gene or blocking a receptor. This knowledge could help companies decide whether to proceed with expensive clinical trials.

In fact, Matthew Nelson, head of genetics at GlaxoSmithKline, found in a study published last week to Nature Genetics that potential new drugs with genetic evidence entering the first stage of clinical trials are twice as likely to earn FDA approval than those supported only by animal models or other traditional methods.

“We're not guaranteeing success here, but on average, favoring drugs with genetic evidence will lead to a higher success rate than drugs without genetic evidence,” he says.

Getting At 'The Heart Of The Matter'

Lasse Folkersen, a geneticist at Technical University of Denmark who recently ran an analysis on a dozen drug candidates in the auto-immunology pipeline of a pharmaceutical company called Novo Nordisk, says genes get to the heart of the matter in a way that traditional methods cannot – by answering the question, what causes this disease in humans? Mouse models and all other methods can only suggest a correlation, or association, between a gene and a given condition.

"That's the key word -- causality. That is what everybody is talking about. It is the weakness of human pre-clinical analysis,” he says. “That's what genetic brings to the table because if there's a genetic component, then you have established the causality."

But Folkersen also says pharmaceutical companies have been surprisingly slow to run these analyses on their own pipelines. He blames the traditional and risk-averse culture that persists at many companies.

"Genetics is of interest but not really being implemented yet. There's a lot of doing it the usual way,” he says.

Last fall, Novo Nordisk closed its entire auto-immune division after a drug to treat rheumatoid arthritis failed in phase II of clinical trials.

“There was absolutely no genetic evidence behind that drug,” Folkersen says. “I know that sounds a little bit like -- 'Hey, I told you so.’"

Of course, not every illness has a clear genetic cause. Nelson found that only about 8 percent of approved drugs included in his sample were supported by any genetic evidence – yet all the others worked well enough to earn FDA approval. Meanwhile, many more diseases may have a genetic link that has simply not yet been discovered -- which means that eliminating a potential drug based solely on a lack of evidence is still unwise.

“We have to be very cautious because the absence of an association doesn't mean a gene isn’t involved and it doesn't mean that five years from now we won’t discover an association,” Nelson says.

The price tags for the first-in-class high cholesterol medicines have not yet been announced, but analysts have cautioned that they may cost between $7,000 and $12,000 a year -- far more than previous medicines for the same condition. Meanwhile, companies spend as much as $2.6 billion and 10 years bringing a potential drug through clinical trials. It’s not clear whether the widespread use of genetics could save companies enough money to reduce the cost of new drugs, but it’s unlikely given the rapidly rising prices of new medicines.

Dr. David Gortler, a drug safety expert at Yale University, former FDA medical officer and neuropharmacology expert with FormerFDA.com, points out that patients or insurers still have to pay for an analysis to determine whether the disease they have is truly caused by the gene targeted by a new drug. And since many new genetically-based medicines may only work in a handful of patients, companies may not feel there is sufficient incentive to develop these drugs in the first place.

But John Imig, a researcher who specializes in heart disease at the Medical College of Wisconsin, expects the debut of the PCSK-9 inhibitors will spur companies to invest in this area as scientists continue their search for genetic factors underlying many more of the world’s major diseases.

“I think the pipeline is going to improve, and I think the pharma companies are going to start to use genetics as a tool earlier in their decision-making process,” he says. “I know there were a lot of expectations, but this is science. It's slow, especially with complex diseases such as cardiovascular disease.”

Boelcskevy and Nelson are also confident that pharmaceutical companies will increasingly realize the value of incorporating genetics into the search for new drugs, even if they have been slower to do so than many originally thought.

“The truth is -- this is early. We're so early on this wave,” Boelcskevy says. “People want instant results. This is hugely complex. This is not like going out and finding some paint and dabbling it on a fence.”