1
The Right Drug, the Right Patient, the Right Time: Foundations of Precision Medicine

Eight-year-old Caleb Nolan faced an uncertain future when he was born. At 3 weeks old he was diagnosed with cystic fibrosis (CF), an inherited, devastating, incurable genetic disease that causes a buildup of mucus in various organs, including the lungs, pancreas, liver, and intestines. This results in poor weight gain, infertility, and chronic lung infections that can lead to respiratory failure. While antibiotics are used to treat infections, many CF sufferers eventually require a lung transplant, and few used to live beyond the age of 50. However, Caleb now lives a full life and will probably die of old age instead of CF [1].

CF is caused by an abnormality in the CF transmembrane regulator (CFTR) gene that prevents the normal movement of chloride ions across membranes and currently afflicts about 30,000 children and adults in the United States and 70,000 people worldwide [2]. As with all people, CF patients have two copies of the CTFR gene, one from each parent, but for them both copies are harmfully mutated. There are people who are "carriers" for the mutation who have one normal and one mutant copy, and while they do not have symptoms of CF, they can pass the gene to their children. At this moment, there are about 10 million carriers in the United States.

One of the most common mutations causing CF is called the deltaF508 deletion mutation, which can be detected using a number of molecular techniques. Akin to an error in the blueprints for building a cabinet that misses a shelf support, this means that the CTFR protein made from the blueprint of the mutated CTFR gene is defective due to a deletion at the 508th place along the protein code (see Figure 1.1). This kind of mutation can be tested by DNA amplification-making many copies of parts of someone's DNA CFTR gene and looking for the mutations that cause CF. There are three ways to assess CF: "carrier screening" of parents-to-be or pregnant women determines their CFTR gene status and helps families adequately prepare for the results. Testing is also done on amniocentesis samples to directly assess CF status in the unborn child. Finally, newborns are screened in all 50 states to assess CF status [4].

Did You Know?

DNA stands for deoxyribonucleic acid, and it is the hereditary material in humans and virtually all other organisms. The information that DNA carries is stored as a "code" that is made of four different chemicals called bases-adenine (A), guanine (G), cytosine (C), and thymine (T). The sequence of the bases dictates how that organism is made and, although there are around three billion bases in humans, the sequence is 99.9% identical.

DNA bases pair up, with adenine (A) pairing with thymine (T) and guanine (G) pairing with cytosine (C). A sugar molecule and a phosphate molecule also attach to each base to create a nucleotide, and these nucleotides are then arranged into the DNA double helix (see Figure 1.5). The DNA is arranged into 46 structures called chromosomes that are arranged into 23 matching pairs, which include one sex pair. A female has 2 X's as the sex pair, and a male has an X and a Y. The entire collection of 46 chromosomes is called a karyotype.

DNA is made of building blocks called nucleotides, and as a highly dynamic and adaptable molecule, it can suffer many types of mutations. Some are harmless, some are helpful, and some can harm the DNA and, ultimately, the organism. These mutations can arise by chance or through environmental factors such as exposure to chemicals and can occur in somatic cells (nonreproductive cells) or germ cells (reproductive cells such as sperm and egg). Diseases like cancer largely affect somatic cells, while mutations in germ cells lead to inherited diseases like cystic fibrosis. Various types of mutation exist. There are nucleotide substitutions, where one nucleotide is swapped for another; insertions and deletions (indels), where nucleotides are added or deleted; and frameshift mutations, which are insertions or deletions of more than one nucleotide that result in the complete alteration of the sequence of a protein [6].

Figure 1.1 The CFTR defect in cystic fibrosis.

Source: Data from Pothier et al. [3].

However, CF sufferers have new hope for treatment. This is due to a revolutionary treatment approach that was approved by the FDA in 2012. The drug Kalydeco is the first to target the underlying genetic cause of the disease. Kalydeco is one of a new generation of medicines that are specifically tailored to treat a disease based on the genetic makeup of an individual. In the case of Kalydeco, rather than just treating the symptoms of the disease, the drug acts on the gating defect associated with the defective CFTR protein, helping to open up the blocked chloride channels (see Figure 1.2). This allows for a clearing of the mucus buildup from the inside out. These drugs are part of a new age in medicine, "precision medicine," which strives to provide the right treatment for the right patient at the right time.

Figure 1.2 Kalydeco's mechanism of action. The drug acts to open up the protein gate that regulates the movement of chloride ions between the outside and the inside of the cell.

Source: Adapted from Kalydeco [7].

The Origins of Precision Medicine

The 1990s was a golden age for the pharmaceutical industry. This was the original "blockbuster" era, when the focus was on developing broad-spectrum drugs for large "primary care" indications like high cholesterol, asthma, and depression. Adopting a "one-size-fits-all" approach meant that companies could generate billions of dollars in sales by targeting the largest patient populations, spending hundreds of millions of dollars on marketing campaigns, and having large sales forces focusing on doctors. Even drugs that were fourth, fifth, or sixth to market could achieve stellar sales performance using this approach, with insurance companies willing to cover these products.

Drug development strategies used by many companies during the 1990s were crude and rudimentary compared with those commonly utilized today. When hunting for a new multibillion-dollar drug, companies screened huge libraries of compounds against a target to look for a suitable candidate that was then tested in clinical trials. The companies searched without assessing whether certain people would be nonresponders or would have an adverse reaction. In some cases, researchers didn't even fully understand the compound's mechanism of action.

This often resulted in little or no therapeutic benefit and, in some cases, caused serious side effects and even death. Today, less than half of people prescribed an antidepressant achieve remission with the first therapy [8], while patients treated for asthma, type II diabetes, arthritis, and Alzheimer's all have differing responses to their medications [1] that can lead to limited therapeutic outcomes or serious side effects. Overall, it is estimated that many of the leading drugs in the United States today only benefit between 1 in 25 and 1 in 4 patients [9].

Warfarin, a common blood thinner, can cause major bleeding and death due to the fact that patients respond to the drug in different ways, driven by genetic variations on a person-to-person basis. However, an analysis of a number of independent studies published in September 2014 showed that dosing of warfarin based on an individual's genetics could reduce major bleeding episodes by over 50%, pointing toward a personalized approach to treatment with the drug [10].

The rigid drug development approach of the past has resulted in the termination of a number of compounds late in the clinical trial process, as companies fail to find a therapeutic signal or, worse still, uncover a major safety issue. On many occasions, this is due to the heterogeneous nature of the patient populations that are used in the trials, with researchers never fully understanding the genetic, environmental, or lifestyle factors that can influence drug response in these large population-wide studies [9].

The impact of this approach to drug development was highlighted in 2005 when Tysabri, an immunosuppressant drug used to successfully treat multiple sclerosis, was removed from the market following three cases of a rare neurological condition called progressive multifocal leukoencephalopathy (PML). Two patients subsequently died [11].

Tysabri was returned to the market in 2006 with a black box warning-a statement of serious risks required by the FDA-and a risk management plan. As part of this, the drug's manufacturer, Biogen Idec, worked with a lab to develop a test that helped stratify patient risk based on the specific presence in the patient's body of the John Cunningham (JC) virus, which can cause PML in patients with compromised immune systems. Biogen had therefore developed a precision approach to treatment with Tysabri.

These examples highlight the complex, multifactorial nature of drug response. Often, biomarkers or molecular pathways that scientists think are involved in disease turn out only to be associated with the disease, rather than to be the root cause (see Figure 1.3). The result can be billions of dollars in wasted R&D spend and drugs that either have limited therapeutic benefit or, in...