Treatment training manual

8.18 Key studies in HIV research

These are just a few examples from thousands of important studies. Reading abstracts and following research takes time to get used to the format and the language – but is an exciting way to learn.

All these studies changed expert opinion and clinical practice. Well- conducted research was essential for care to improve and change.

1. First reports of symptoms

Masur H et al. NEJM, Vol 305:1431-1438. Dec 10, 1981. Number 24. abstract/305/24/1431

One of 3 papers and an editorial from the New England Journal of Medicine in December 1981.

An example of why case reports from insightful doctors can be so important.

2. AZT: ACTG 019 and Concorde studies

Learn the history of the first approved drug…

In 1987, AZT was approved in the US, with the first study (called ACTG 019) stopped early because so many more people died in the placebo group. ACTG 019 had many problems including being very short term, and perhaps that symptoms were managed differently in each arm.

This was also at a terrible time when more than 50,000 people in the US had been diagnosed and about 10,000 people had died. This was a time when there were no treatments. This led to AZT being recommended in the US for everyone with a CD4 count under 500.

But four years later, Concorde showed no clinical benefit from using a single drug in everyone except for the people who very already very ill.

Although the Concorde study began in 1988, ‘preliminary results’ were only reported in a letter to the Lancet in April 1993.

Volberding et al. NEJM abstract/322/14/941

Concorde Trial. Lancet. 1993 Apr 3;341(8849):889-90. Not available online .

As these papers are not online, see: NEJM editorial on ACTG 019: full/329/5/351

After Concorde article in BMJ, 1993: fcgi?artid=1677009

3. PI-based triple combination

Gulick RM et al. IAS World AIDS Conference, 1996, Vanvouver. Abstract Th.B.931. thb931.html (link no longer active)

As the aegis database is no longer online and the IAS early conferences are not yet archived, this link is to the same study published in the New England Journal of Medicine in 1997.

The first triple-combination studies including protease inhibitors showed:

  • Using three drugs was essential.
  • Viral load tests could measure the immediate effect of treatment (rather than having to count infections and deaths or wait months for CD4 changes).

Despite this, numerous studies after 1996 still included treatment groups using one (monotherapy) or two (dual therapy) arms. Activists protested to the companies that these studies were no longer ethical.

This is an example of how early results, presented at a conference rather than in a medical journal, changed clinical practice and guidelines.

4. Resistance, replication and viral load

Richman D et al. Full text publication form Journal of Virology.

Reducing viral load to undetectable (less than 50 copies/mL) stops HIV from developing and mutating, and makes the likelihood of developing resistance very low.

A key paper to understand how treatment works.

5. Using ritonavir as a booster: better drug levels with lower doses

Merry C et al. AIDS (15 March 1997): 11(4):F29-F33.

At a time when early protease inhibitors included high doses and pill counts, complex and difficult side effects, and a higher risk of not working, independent researchers at Liverpool University showed better and more effective ways to use some of these drugs.

They showed that one drug (ritonavir) had the potential to boost other drugs (saquinavir in this case, but also indinavir).

This also meant telling drug companies that they should be selling lower doses to get a better effect. Instead of needing 600 mg of saquinavir, three times a day (1800 mg a day) or 600 mg ritonavir twice a day (1200 mg a day) using one to boost the other, meant lower doses, and achieved higher and more stable drug levels.

The manufacturer’s of all these drugs fought against these changes.

6. Combination therapy in pregnancy – transmission approaching zero

Beckerman K et al. World AIDS Conference, Geneva, 1998. Abstract 459. (PDF 130 Kb))

This was probably the most important study at the World AIDS Conference in Geneva in 1998, yet it received hardly any press coverage.

Using combination therapy during pregnancy went against the mainstream caution to only use AZT, plus C-section for delivery.

But combination therapy reduced transmission to near zero and it still took years for this to become standard-of-care in Western countries.

7. New standards of care: efavirenz

Staszewski S et al.

This link is to the results from the efavirenz 006, published in the New England Journal of Medicine in 1999,

This study was presented at CROI in 1999 (LB16) and six months earlier at the World AIDS Conference in Geneva. CROI abstracts are no longer online and the Geneva abstract didn’t include the results in the summary.

This study was important for comparing a new HIV drug to the best standard-of-care triple combination. It was the first study to separate results for people who started with a high viral load (over 100,000 copies/mL).

It was also the first study that established the importance of intention-to-treat analysis (ITT) for the results. This meant results included everyone who started the study when reporting the percentage of people who did well.

Previous studies presented percentages of people who did well out of people who still stayed in the study – which made treatment seem much better than it actually was.

8. Biopsy studies showing d4T and AZT cause changes in fat cells

Hammond et al. 7th Lipodystrophy Workshop, Dublin, 2005. Abstract 2.

This careful fat biopsy study showed which nucleosides cause fat loss on a cellular level.

9. The SMART study, treatment interruptions and immune inflammation. El-Sadr W et al. 13th CROI, 2006. Abs LB106. Abstracts/28085.HTM (link no longer active)

One of the most important studies since 2000 was presented at CROI in 2006. When the study started is was very common for people to take a break in treatment if their CD4 count was high on treatment. This was thought to be safe.

As the CROI abstracts are no longer online, this link is to the paper published in the New England Journal of Medicine later in 2006.

The SMART study was able to show the risks from interrupting treatment because it was a large study with a randomised design.

Linking risk of serious complication to detectable viral load, it found that HIV treatment protects against heart, liver and kidney disease.

The same researchers later ran the START study looking at earlier treatment.

10. HIV rebounds after ten years of undetectable viral load. Chun TW et al. AIDS, 2010.

Case of a person who started very early ART who had undetectable viral load for ten years. There was so little HIV that researchers needed to check 1.7 billion CD4 cells before finding one with HIV. In a research setting – and against his doctors advice – this person stopped ART – and viral load still rebounded (even though it took a few months).

Ref: Chun T-W et al. Rebound of plasma viremia following cessation of antiretroviral therapy despite profoundly low levels of HIV reservoir: implications for eradication. AIDS 19 October 2010. dos: 10.1097/QAD.0b013e328340a239.

11. START study: benefits of ART for everyone

Large randomised study looking in 4500 people with CD4 counts above 500 who either started immediate ART or waited until the CD4 count dropped to 350. This study immediately changed treatment guidelines in all countries. It showed that ART reduces the risk of serious health complications even at high CD4 counts

Q&A on the START results:

Ref: Lundgren J et al. Initiation of antiretroviral therapy in early asymptomatic infection. NEJM (20 July 2015). DOI: 10.1056/NEJMoa1506816.

12. PARTNER studies: ART stops HIV transmission. JAMA 2016 and Lancet 2019.

A large observational study in almost 900 serodifferent couples, where one partner is HIV positive and the other is HIV negative. The positive partners were on ART with undetectable viral load and the couple were already not using condoms.

After 58,000 times when couples had sex without condoms, there were no HIV transmissions. These results (first presented at CROI 2014) were groundbreaking. The study included both gay and straight couples.

Q&A on the PARTNER study:

The PARTNER 2 studies continued from 2014 to 2018 to prove that ART was just as protective for gay as well as straight couples. The full compiled results were published in the Lancet in 2019.


Rodger AJ et al for the PARTNER study group. Sexual activity without condoms and risk of HIV transmission in serodifferent couples when the HIV-positive partner is using suppressive antiretroviral therapy. JAMA, 2016;316(2):1-11. DOI: 10.1001/jama.2016.5148. (12 July 2016). Full free access.

Rodger AJ et al. Risk of HIV transmission through condomless sex in serodifferent gay couples with the HIV-positive partner taking suppressive antiretroviral therapy (PARTNER): final results of a multicentre, prospective, observational study. Lancet 2019; published online May 2. (Open access)

Last updated: 29 May 2017.