Geneticist Helen Hobbs’ work on coronary heart disease (CAD) led to the development of PCSK9 inhibitors – the most powerful cholesterol-lowering drugs to hit the market since statins. These drugs fight the PCSK9 protein, which prevents “bad” low density lipoprotein (LDL) cholesterol from being removed from blood. In the mid-2000s, Dr. Hobbs, director of the Eugene McDermott Center for Human Growth and Development at Dallas’ UT Southwestern Medical Center, found that a mutation in the PCSK9 gene, present mainly in African Americans, suppressed LDL levels. Consequently, it protected carriers from CAD. Importantly, people with two copies of this mutation had no side-effects of very low LDL, such as loss of adrenal function. A key innovation in Dr. Hobbs’ approach was to look for a rare gene variation with a large impact on cholesterol. Most scientists then were carrying out genome-wide association studies (GWAS), which look for common gene variants. Unfortunately, this strategy had mainly identified gene variants with a small impact on CAD risk. Dr. Hobbs’ approach bore fruit. Through research on the 3500-strong Dallas Heart Study cohort, she and colleague Jonathan Cohen discovered rare mutations in the PCSK9 gene. Eventually, drugmakers Amgen and Regeneron developed the PCSK9 inhibitors Evolocumab and Alirocumab, respectively, which mimic this mutation’s effects. In an interview in Hyderabad, where she spoke at the TNQ Distinguished Lectures in the Life Sciences, Dr. Hobbs discussed the way ahead. Excerpts:
How did the Dallas Heart Study begin?
The question we were trying to answer was if elevated LDL is necessary for heart disease. You can’t ask that question by looking at Mendelian disorders (in which defects in a single gene trigger heart disease), because everyone who has these disorders is sick. We were looking for mutations that would protect people from heart disease. Such healthy people wouldn’t be in the clinic. So, we developed the Dallas Heart Study. One of the premises was that there were going to be low-frequency or rare variations associated with major changes in lipids.
Our population was designed to maximise the probability of finding such rare variations. It was a multi-ethnic cohort, in which half the individuals were of African descent, while 15% were Hispanic, and the rest were of European ancestry. Africans, being the most ancient population, are the most genetically diverse. Our approach of looking for rare variations was really key to our findings. As a scientist, you are always looking for things that have large effects, since they are easier to study.
If looking for rare gene variants with large effects could be such a powerful approach, why was everyone else doing the opposite at that time?
Honestly, I couldn’t figure it out. Maybe my approach had something to do with my medical background. Maybe it was because I had worked on Mendelian disorders. In some ways, I benefited from being in Texas, away from the epicentre where everyone was doing the same thing. With our approach, everywhere we went, we found things. PCSK9 is the biggest story. But that’s not the only one. We also found variants in a gene called ANGPTL4, which lowered plasma triglycerides. Mutations in ANGPTL3 were found to lower levels of both cholesterol and triglycerides, and antibodies to this protein are now being developed into a drug as well. What people hadn't known until then was how riddled healthy individuals are with rare mutations that have major effects. What we did then has now become routine. Companies do it all the time today.
One of the problems in genetics is “missing heritability”. GWAS that look for common gene variants to explain disease risk have only been able to explain a small part of disease inheritance. Do you think rare-gene variants explain this gap in heritability?
There are definitely many more common variants than there are rare. But for an individual who has a rare variant, the rare variant can mean everything. That’s the problem.
In our first experiment, we found that people with low HDL (“good” cholesterol) have a major gene defect which is contributing to it.
While this was going on, a woman in France had identified families that had very high cholesterol due to a new gene PCSK9. She found two missense mutations in this gene (missense mutations alter the make-up of a protein coded for by a gene; in contrast, nonsense mutations stop the production of the protein). Individuals with the missense mutations had high plasma levels of LDL. To figure out what the PCSK9 mutations did, three groups over-expressed the mutant forms of the PCSK9 gene in the livers of mice. And in those mice, the LDL receptor, which removes LDL from the blood, disappeared. The mice became hypercholesterolemic (developed high cholesterol). We thought, if you didn’t have PCSK9, you would have a lot of LDL receptors on the cell’s surface, and low LDL. So, we went to the blacks and the whites in the Dallas Heart Study, and we sequenced the individuals who had LDL less than the 5th percentile. We found three sequence variations. One missense mutation was in whites, associated with a modest reduction in LDL. But there were two nonsense mutations, which introduced stop codons in individuals of African descent. Those stop codons are a gift to geneticists, because they are almost invariably a loss-of-function mutation. This means that they kill the (PCSK9) protein.
These two nonsense mutations were only seen in black people. And they were associated with about a 40% reduction in LDL cholesterol. We wanted to know what it meant to be born with such a mutation, and to have low LDL throughout one’s life. If LDL is an important factor in heart disease, this mutation should really change one’s risk of developing heart disease. So, we went to the only biracial study in the world — the ARIC (Atherosclerosis Risk in Communities) Study — and asked a simple question. Did people who hadn’t had a heart attack and were not on lipid-lowering agents have the mutations we found? Among black people in this study, we found that the mutation lowered the risk of heart disease by almost 90% — higher than what we had found in the Dallas Heart Study.
We knew that we had made a significant observation. If you have low LDL for years, you are not going to get heart disease. Atherosclerosis starts early and develops gradually.
You have talked previously about the need to lower cholesterol earlier in life. Should people start taking statins earlier then?
The best way to lower cholesterol is through diet, but it must start at an early age.
In 2015, the U.S.’s Dietary Guidelines Advisory Committee said dietary cholesterol may not be a “nutrient of concern for overconsumption”, given that there wasn’t enough evidence to show its impact on LDL. What did you think of this?
I don’t believe it. Of course, cholesterol is a nutrient of concern, and its effects are compounded by saturated fat. The benefits of lowering your cholesterol are indisputable now. We have so many studies. But the problem is that you cannot take 50-year-olds, who have had high cholesterol their whole life, lower it, and eliminate risk of heart disease. Atherosclerosis is already in their arteries.
If you look at the statin trials, a large percentage of those who get treated still have a cardiac event. Statins are not a cure; it’s not like giving antibiotics for an infection. You have to start young. Either you have to consume a different diet, which gets you on a different trajectory, or you have to start statins at a younger age. What age that should be is still disputed, because there haven’t been clinical trials. It’s hard to start people on a drug and follow them for 30-40 years.
PCSK9 inhibitors are very promising, but their use is greatly limited by their high cost. Do you think the cost will come down in the future?
It can only come down. It already has. When they first came out in the market, they were priced too high at $14,000 per year. It really was a mistake. Then they did a cost benefit analysis, and brought the price down to about $5,000 a year. Now, a lot more people are getting the drugs. The hope is that competition will eventually bring down the price even further.
One reason for the high cost of Evolocumab and Alirocumab is that they are monoclonal antibodies. Are there any alternative strategies that target the PCSK9 protein, but are less expensive or easier to use?
There is Inclisiran, a small interfering RNA, which is being tested in humans. Because PCSK9 is made in the liver, you can inhibit it at the level of messenger RNA. Inclisiran, which does this, requires an injection every six months. This is terrific because one of the major problems with statins is compliance; people stop taking them. An injection every six months is much easier to adhere to. People are also thinking about CRISPR-Cas9 strategies, but as everyone knows, that’s not going to happen tomorrow.
In your ongoing work, you have identified genetic mutations (in PNPLA3 and TM6SF2) for fatty liver disease (FLD). Is there scope for therapeutics here?
I think it’s pretty clear that there is.
Again, this disease is burgeoning in frequency, and is huge in India. When we started, almost nothing was known about its pathogenesis, except that when people are obese or have diabetes, they have a higher incidence of fatty liver disease.
However, we saw in the Dallas Heart Study that 50% of people who were obese did not have FLD. This suggests that there might be genes involved in the propensity to deposit fat in liver. In this case, we did a GWAS, and found that the Hispanics had a common sequence variation associated with high triglyceride content.
How big a problem is the under-representation of certain ethnicities in genetic research? Most genome-wide association studies have involved Europeans. What opportunities are we missing because of this?
I think the opportunities are huge. We need to study more groups, especially Indians. It is an incredibly genetically diverse population, which has poorly understood susceptibilities to coronary heart disease and to diabetes at a lower body mass index.
Studying different races is very critical. In PCSK9’s case, if we had not had blacks in our population, we would never have discovered the gene mutations that led to new cholesterol-lowering therapy. In the FLD study, if we didn't have Hispanics, we would never have discovered PNPLA3, which is currently the most important risk factor for fatty liver disease.