START Study Could Change How We Look at HIV Treatment, and HIV Itself

Simon Collins
Simon Collins

Have you heard of the START (Strategic Timing of Antiretroviral Treatment) study? In my opinion, it is probably the most important study currently running. It generates strong views, both for and against initiating antiretroviral therapy early, but the results are likely to be surprising.

Let's look at what START will and won't tell us about "when to start" treatment.

As an activist living with HIV, I want decisions about my care to be supported by evidence. Jumping ahead of the data -- even with the best intentions -- has led to treatment guidelines in the past getting it wrong. So, over the last five years, I have been surprised at how passionately some people object to START, already claiming to know the answers.

The big question from START is whether there are better health outcomes from starting antiretroviral therapy with a CD4 count above 500. This is compared to waiting until 350 -- a level at which we already know it is good to start.

No other randomized study had answered this.

And the answer to this question is not obvious. Here's why:

Even though many people already start treatment early and do well, people can reach "normal" life expectancy by just starting above 350. On the other hand, many people start far too late -- and if START shows benefits at very high CD4 counts, this could change guidelines globally.

Some questions in START are about fine tuning: Everyone in the study is lucky enough to have been diagnosed early and has access to antiretroviral therapy.

But START will also define the risks from earlier treatment. This includes direct risks such as side effects, which we expect to be low. It also includes long-term risks such as drug resistance. If earlier treatment has lower adherence rates, this might mean some people lose treatment options without any medical benefit.

The Power of a Large, Randomized Study, and Its Sub-Studies

START is important because it is a very large, randomized study. Between December 2009 and January 2014, more than 4,600 people were enrolled from 36 countries. By not letting people chose whether they start early or late, other factors that affect the results are also likely to be balanced in each group. This includes education, income, motivation, recreational drug use, smoking, country, coinfection and even genetics. The size of the study means that it is large enough for the outcomes to be real rather than by chance.

Just as crucially, the sub-studies in START look at major scientific issues about HIV itself. These may prove even more important than the main question of when to start. Everyone entering the study has a "normal" CD4 count when they join. This means the results will compare the effect of HIV on the brain, on bone health and on lung health to the effect of HIV treatment.

For example, we already know that antiretrovirals reduce bone density by a small amount (in addition to the reduction in density caused by HIV and any reduction caused by aging). If there is a similar negative impact of antiretrovirals in the brain, we need to know this. Currently, it is considered more important to control HIV, but we need data if early treatment is going to be widely recommended. Also, the lung disease called COPD -- principally linked to smoking -- is the third leading cause of death globally. While many doctors hope that early antiretroviral therapy might help, we need to know whether this is true.

All these examples overlap with issues of aging and with mental health, both of which have a complex relationship to HIV. By studying a few thousand people in early infection, START will provide data that could affect millions of people.

Quantifying Individual Health Benefits Versus Public Benefits

It is important that START will quantify individual health benefits compared to risks because many countries, including the U.S., already merge the personal and public benefits of earlier antiretroviral therapy in policies aimed at reducing new infections.

But many of us never risk passing HIV to anyone else. For example, we might not be sexually active, our partner(s) may also be HIV positive or we might always use condoms. These are just three situations that show why the decision to start treatment may often be purely focused on clinical benefits versus risk for the individual. To inform this decision, we need data, not guesswork.

There are lots of theories about why reducing viral load earlier in infection might be good. But, it might not be. Or it might not make much difference, so long as when you do start treatment you take it in the right way. This would be another significant finding. Sometimes people need -- or want -- time before deciding to start.

Prepare to Be Surprised

START will not show the best CD4 count for starting treatment. But, it will quantify how safe and effective earlier treatment can be. It will not show drug safety for antiretrovirals that were not yet approved. But this is true of every study. The big picture for START is to define the risks from uncontrolled HIV when the immune system is still strong. This is not primarily about individual drugs.

The SMART researchers ran another large, randomized study about 10 years ago. Many people then were convinced that taking a treatment break would be safe. This included leading doctors, researchers and activists -- and a large percentage of HIV-positive people.

It was only with a large, randomized study that this belief was overturned. The results from that study -- so strong that they were determined to be conclusive years earlier than anticipated -- surprised nearly everyone. That study was called SMART, and it radically changed the direction of HIV research.

So, no one is helped by second-guessing the results from START. For example, some benefits might have to be offset against some downsides. Or the risk versus benefit balance might be stronger for some people to start earlier and for other people to wait. Although guidelines always try to simplify recommendations, they also emphasize that antiretroviral therapy should be individualized.

Whatever the actual results, the value of START includes the quality of the evidence. It will include a large sample and data set that will help answer many other questions.

Results are expected by early 2017. Until then, people have to use their judgment and preferences when choosing when to start. Current guidelines emphasize the advantages and disadvantages of the various options.

But by World AIDS Day 2016, the study will wrap up the collection of a tremendous amount of data that can go on to inform both individual decisions and public policy. The results from START -- and similar "when to start" studies -- will be a significant achievement that will inform guidelines and care globally. The most useful prediction is to be prepared to be surprised.

Simon Collins is an HIV-positive treatment advocate at i-Base in London. He was a community representative to both the START and SMART studies. He has been on antiretroviral therapy since 1996, which he started with a CD4 count in the single figures.