The New England Journal of Medicinehas published the first well-documented case of HIV pre-exposure prophylaxis (PrEP) failure despite good medication adherence.
We heard lots of this information at CROI last year, and again I'm impressed at the extraordinary degree of virologic investigation done on a case from clinical practice.
To refresh your memory, here are the critical details from the published case report and the supplementary appendix:
A 43-year-old man began PrEP with TDF/FTC in April 2013, and had multiple negative 4th generation HIV screening tests over the next 21 months.
Pharmacy refill records indicated excellent adherence. (Side note to young clinicians: Refill frequency is an amazingly powerful tool to monitor adherence.)
Prior to diagnosis, he had multiple sexual exposures, including receptive anal intercourse without the use of condoms.
A couple of weeks before his first positive HIV screening test (antigen positive, antibody negative -- Day 0), he developed abdominal pain which waxed and waned over the next 3 weeks. A CT scan demonstrated thickening of the sigmoid colon, ascending colon and rectum. Endoscopy revealed erythematous patches of the sigmoid colon.
Adherence was confirmed by analysis of a plasma sample obtained on Day 0 revealing tenofovir concentrations consistent with recently taking the drug.
HIV RNA peaked at 28,000, and became undetectable with the addition of boosted-darunavir and raltegravir.
His viral isolate had extensive multi-class resistance, including to FTC (M184V), TDF (multiple thymidine-associated mutations), NNRTIs (Y181C), and first-generation integrase inhibitors (92Q).
He's now virologically suppressed on DRV/c, RPV, and DTG.
The supplementary appendix has an excellent figure of the timeline, which I've pasted below:
So what can we learn from this single case?
PrEP is very effective, but it's not 100% protective. I don't think clinicians are claiming 100% effectiveness, but PrEP-takers may be hearing this from various "experts", or may be misunderstanding the data.
Transmitted drug resistance will weaken PrEP. Although all of the mutations in the case are to drugs not in the TDF/FTC combination -- except M184V -- the multiple TAMs could have weakened TDF through cross resistance. Furthermore, FTC could have been transmitted, or alternatively selected by period of viral replication after viremia developed. The source patient's virus was not available for sequencing, unfortunately.
The symptoms of acute HIV infection are highly variable. The supplement to the case reports states, "The patient did not have classic symptoms of acute HIV seroconversion", which is true -- but he did have GI symptoms, which are quite common in acute HIV. We need to remember this variability when following patients on PrEP.
The monitoring strategy recommended by the CDC in the guidelines should be followed. Although it might seem like overkill to monitor for HIV and other STIs every 3 months while receiving PrEP, remember that only those at highest risk for getting HIV should be receiving PrEP to begin with. Clinicians may choose to recommend more frequent monitoring if clinically warranted.
Integrase resistance is rare in clinical practice, and transmitted integrase resistance is even less common. While some might interpret this case report to indicate a need for baseline integrase testing for newly diagnosed individuals, the detection of transmitted integrase resistance in most studies of newly diagnosed patients is less than 1%, and often 0%. Notably, the transmitted virus was still sensitive to dolutegravir (I'd interpret the genotype as fully susceptible).