Saye Khoo, MD: Takeaways from the DOLPHIN-2 Study

At the Annual Conference on Retroviruses and Opportunistic Infections (CROI 2019), investigators on the DOLPHIN-2 trial presented data that dolutegravir is well-tolerated and achieves more rapid virological suppression prior to delivery compared with efavirenz when initiated during a late stage of pregnancy.

Saye Khoo, MD, professor at the University of Liverpool, and the lead presenter of the DOLPHIN-2 results, spoke to Contagion® in an exclusive interview and explained the findings of the investigation in detail.

Interview transcript: (modified slightly for readability)

Contagion®: What did you observe in the study regarding pregnancy outcomes?

Dr. Khoo: "We did a randomized, control trial in Uganda, in Kampala, and in South Africa at a community health center. We recruited pregnant women in the third trimester and we randomized them to receive the standard-of-care which was an efavirenz regimen, versus dolutegravir together with backbone nucleosides.

The women were well matched at entry and our primary outcome was the undetectable viral load at the time of delivery, as the best proxy for the lowest risk of transmission. We found that there was a difference between arms, so of the women receiving dolutegravir, 74% of them were undetectable at delivery compared to only 43% of mothers who were randomized to receive efavirenz, and this was highly significantly different regardless of all the other baseline parameters. Additionally, we found that that dolutegravir was very well tolerated.

This study was principally designed to look at virological differences between the 2 groups. We were also able to capture adverse outcomes in the mothers and in the infants. Notably, we found 4 stillbirths, all in the dolutegravir arm. We found 8 infant deaths, spread between the dolutegravir and the efavirenz arms. And, we found 3 infant [HIV] transmissions, again all in the dolutegravir arm. So, clearly, these were adverse outcomes that were of great concern to us and so we scrutinized every one of them carefully.

So, taking the 4 stillbirths, these were associated with known risk factors — so 2 stillbirths had infectious causes; maternal syphilis in one case, presumptive tuberculosis in another. And, 2 of the other stillbirths were deaths due to obstetric-related conditions – a ruptured uterus in 1 case and evidence of fetal distress in the other case, so we did not think in that in the case of stillbirths that this was related to the drugs that were used.

In the case of the transmissions all 3 transmissions tested positive for HIV very early on — so 3 days after delivery, 5 days and 11 days after delivery. And, at delivery each of those mothers had a very low viral load, so 2 of the mothers had a viral load that was >50 copies, which would qualify as undetectable, and the third mother had a very low level of 200 copies. So, the early positive test in the infants, coupled with the low viral loads at deliveries, strongly suggested to us that these infant transmissions were in utero transmissions, rather than transmissions at the time of birth. And, clearly in utero transmissions may not be mitigated by drug therapy.

In the case of the infant deaths there were also clear explanations, so 2 cases were extreme prematurity, 4 cases were due to respiratory distress and asphyxia, and 2 cases were due to severe sepsis. In none of those cases was drug thought to be likely to be a cause for [death].

So, in summary, we've shown superiority of dolutegravir over efavirenz in reducing the viral load quickly, which is the best indicator of diminished risk. That said, we found a number of adverse outcomes we do not think those were related to the drugs that were used but rather to the poor overall prognosis. And, it's this particular group of women who present later on in pregnancy tend to engage poorly with health services, tend to have poor access to treatment, tends to come from a different demographic, and other studies have consistently shown that this group of women are very vulnerable to poor outcomes."

This interview is part 2 of a 3 part interview with Dr. Khoo.Part 1 is available here.