Metabolic complications of HIV infection at CROI 2010 (Feb 16-19)
University of Pennsylvania. Philadelphia.
This year the Retrovirus conference took place in San Francisco. It was nice to get out of the East Coast and the 2 feet of snow in the front yard. The weather was good, the meeting was interesting, I saw friends from all over the place…, it is difficult to ask for more in a scientific meeting. Due to the lack of blockbuster presentations about new antiretroviral drugs, the focus of this years meeting seemed to be on the complications of HIV and its treatment and TB/hepatitis co-morbidities. In many ways this year Retrovirus meeting reminded me of the lipodystrophy meetings that I usually attend in the fall. I will summarize in this article what I thought were the main themes related to the complications of HIV and its treatment and the most important take home messages for the clinician.
Inflammation and cardiovascular risk
This has been the year of inflammation in the complications section of CROI. The field is very confusing, and I think it is reasonable to take a broad view before going into too many details. This is what we know as facts: uncontrolled HIV infection is associated with high levels of inflammation, measured as increased levels of CD4 and CD8 cells expressing markers such as CD38 and HLA-DR, as well as a bunch of soluble inflammatory markers in plasma. Successful antiretroviral treatment decreases markedly markers of inflammation, but in a significant proportion of patients treatment does not completely normalize those markers.
This residual inflammation has been blamed for almost anything that you can think of, including the lack of CD4 response in some patients (also called immune discordance), accelerated aging, increased cardiovascular risk, stiffness of the blood vessels, progressive osteopenia/osteoporosis, immune senescence, renal impairment etc. etc. The data supporting all of this, and what is most important, the ultimate role that this residual inflammation plays in clinical outcomes remains at best thin. Most of the data presented so far are associations in cross-sectional studies of limited size, comparing patients with HIV infection to patients without HIV infection. These studies are not well suited to address causality. Obviously longitudinal data is stronger, but we have only data on the effects of therapy in some or many of those markers, but much less data linking these markers with clinical endpoints.
The reader should not be surprised about the prominent role inflammation has been taken over the last couple of years in our field. Many other fields like cardiovascular disease and rheumatology they have been pursuing the central role of chronic inflammation in the pathogenesis of "their" diseases for years. At the end, and when the dust settles, I think inflammation will be proven to be in the causative pathway of some of the "badness" (non-AIDS events) that occur to our patients in spite of successful antiretroviral therapy, but I suspect it will not be the whole story. I also suspect that the overwhelming enthusiasm for this trendy theme will fade, as it usually happens in our field.
One small, but fascinating study, because of the unexpected results, evaluated the use of maraviroc in fully suppressed individuals but immunological non responders22. The use of maraviroc was associated with decreased immune activation as evidenced by reduced %CD38+, % HLA-DR+/CD38+, and improvement in markers of apoptosis, however was not associated with the hypothesized increase in CD4 counts. The study was small and non controller, but goes against the prevailing wisdom that persistent residual inflammation is responsible of the lack of CD4 increases in this group of patients.
Sessions 33 (Progression of Atherosclerosis: Role of Inflammation and T Cell Activation) belonged to Priscilla's Hsue and her group at UCSF. In a series of posters, they showed an association of markers of inflammation and endothelial dysfunction and carotid atherosclerosis 1-2. In another poster they showed that earlier initiation of antiretroviral therapy was associated with improved stiffness in blood vessels3. Longitudinal data is always stronger than cross-sectional studies. This data suggest that HIV itself, maybe through inflammation, is a significant contributor to the increased cardiovascular risk associated with HIV infection. Prolonged exposure to HIV viremia and inflammation might be the explanation why patients with more advanced disease are more prone to have these complications, even after controlling HIV replication.
The INSIGHT network is conducting a clinical trial [START] to try to answer the question if the initiation of early antiretroviral treatment would affect the progression of cardiovascular risk and other non classically associated AIDS events that are seen earlier in the course of HIV infection and that the initiation of antiretroviral treatment does not seem to prevent completely. The answer to this question is probably yes, but we need good data. START, although very heavily criticized by some, might help us quantify the benefits of early antiretroviral therapy, and tell as how many patients need to be treated to prevent one of these non-AIDS events. In the meantime it seems prudent and reasonable to consider earlier initiation of antiretroviral therapy if the resources are unlimited, something that unfortunately is not the case in the developing world (and even in our beloved country).
In a different session (Session 37, Long-term Complications: Hearts and Bones), another presentation from the UCSF group linked persistent inflammation and the dramatic progression of atherosclerosis at the level of the bifurcation of the carotid4. Other groups, outside the San Francisco region were not able to reproduce the dramatic results of this group5. This could be due to methodological differences between the 2 groups (multicentre groups outside San Francisco tend to measure the more reproducible internal carotid and not the very challenging and highly variable carotid bulb), or something specific to the San Francisco UCSF population.
In any case, these measurements are supposed to be surrogate markers of subsequent cerebrovascular events. The UCSF group has been showing for years alarming rates of progression of atherosclerosis of the carotid level. Those rates should be associated with dramatic increases in the number of cerebrovascular and cardiovascular events in the HIV infected population; however that has not been the case. In a presentation last year, the Kaiser Permanente group presented data that suggest the rates of myocardial infarction and cerebrovascualr events is decreasing in patients with HIV infection and is approaching the levels of the general population, probably due to a combination of several factors, including better antiretroviral drugs, a more aggressive management of traditional cardiovascular risk factors and the more aggressive use of statins.
Managing traditional risk factors is not easy in HIV infected individuals. Quitting smoking is difficult, as it was shown last year by Karen Tashima, but if your patient is able to discontinue smoking, the cardiovascular risk will decrease quickly, as it was shown by the DAD cohort this year6. Triglyceride levels, as in the general population, are associated with a modest increase in cardiovascular risk, but clearly not to the level of cholesterol increases7. The clinician should focus in reaching ATP III cholesterol goals in patients with HIV infection to decrease as much as possible the increased vascular risk.
Vitamin D. Got Milk?
Vitamin D was also a big star in this meeting. There was a themed discussion of posters related to this topic. (Session 17. Got Milk? Vitamin D Deficiency Prevalence and Associations) . Groups from the US8, Italy 9 and Switzerland 10 presented data on the high prevalence of low vitamin D levels in their respective countrymen living with HIV infection. The prevalence, however, does not seem to be very different form the general population. The reasons for this unbelievably high prevalence of low vitamin D levels are probably more related to environmental factors than HIV related factors: lack of dairy products in the western diet (normally supplemented with vitamin D) and lack of outdoor and sun exposure and exercise, which for millennia has been the main source of vitamin D.
Vitamin D levels have been associated with a myriad of other complications including osteoporosis, increased cardiovascular risk, increased risk of metabolic syndrome, diabetes, insulin resistance, immune dysfunction and senescence.
What are the clinical effects of these low levels of vitamin D in the HIV infected population? A poster from Tanzania linked the lower levels of vitamin D to complications of HIV in women11, and a US study linked hypovitaminosis D to an increased risk, of all things, of bacterial vaginosis12.
We know that low vitamin D levels are very frequent in our HIV infected population. What we do not know is how to correct them. We are not alone in this problem, as nutritionists do not agree on which should be the recommended daily dose of vitamin D. Everybody knows that the recommended 400 units daily dose is insufficient, and new recommendations will probably elevate that figure to at least 1000 or even 2000 units per day. Vitamin D supplementation is generally safe (although too much vitamin D may lead to hypercalcemia).
Physicians taking care of HIV infected patients should consider checking a level of 25 hydroxy vitamin D and if those levels are below 30 ng/mL consider supplementing the diet of their patients.
There has been some preliminary data13 that suggested some drugs, particularly efavirenz may decrease the levels of vitamin D after the initiation of antiretroviral therapy by interfering with its metabolism. The clinical significance of these decreased levels is still questionable. I am sure that in the near future we will have more information regarding this particular issue.
Kathy Mulligan and I chaired a session of the long term complications of HIV and its treatment (Session 37, Long-term Complications: Hearts and Bones). I have already commented on some of the cardiovascular findings of this session. The other half of it was devoted to bones.
Two studies confirmed the previous observation the patients with HIV infection are at increased risk of fragility fractures14. This should not come to a surprise to anyone, knowing the high prevalence of osteopenia and osteoporosis in our patients. As the HIV infected population ages, and the risk of falls increases, one should expect an increased fracture rate. This was observed by the CDC in the SUN study15 and in the aging VA cohort16. The WIHS cohort did not observe an increased risk of fractures yet17, but this was probably due to the fact that this cohort is younger and mostly premenopausal women, and as such, which a lesser risk of falls and fractures. The fact that the WIHS cohort includes only women (from Jules: and a high percent of African-American women) might also explain this lower risk of fracture because of the un-expected and still unexplained, lower prevalence of osteopenia and osteoporosis among HIV infected women when compared to HIV infected men (exactly the opposite of the general population).
The initiation of antiretroviral therapy is associated with a significant amount of bone loss, as was shown by Grace McComsey when she presented the ACTG study A5224s18. In that study, both the initiation of ABC/3TC [Epzicom] and TDF/FTC was associated with bone loss (between 3 and 5% of the total bone mass). This bone loss was most significant with the use of tenofovir. After 48 weeks the bone loss tended to stabilize or even to improve. The clinical significance of that bone loss might be minimal at the individual patient level (as the ultimate clinical impact will depend on what was the bone mineral density in that particular patient at the time of initiation of antiretroviral therapy), but at the population level it may have a significant impact in the long run, increasing the risk of fractures as the patient population ages. (from Jules: I recall in 5224 and other studies as well bone loss has been seen in patients at baseline naïve to HAART) How can we prevent this initial bone loss? At this point we don't know. The ACTG is developing a study evaluating the use of vitamin D for prevention of the initial bone loss associated with the introduction of antiretroviral therapy and hopefully this study will provide data on how to prevent the initial bone loss associated with antiretroviral treatment.
The use of thymidine analogs, particularly d4T, and to a lesser degree AZT, has been associated with the development of limb lipoatrophy. During this meeting, Grace McComsey presented the results of A522419 , demonstrating that the use of the "modern" NRTIs (abacavir and tenofovir), is not associated with the development of this complication. In this substudy of the larger ACTG 5202 study, whose virological results were also presented in this meeting, the use of both NRTIs was associated with similar increases of the beloved subcutaneous fat, and also increases of the not so loved (because of it cardiovascular repercussions) visceral fat. Atazanavir/ritonavir led to greater gain in limb fat and trunk fat than efavirenz. Lipoatrophy, even the mild protocol-defined form, occurred in 16% of the participants and was not significantly different between TDF/FTC and ABC/3TC or between EFV and ATV/r. (from Jules: the lipoatrophy seen at 16%, we don't know what that is due to: it could be reasons other than HAART or HIV although it could be due to HIV but additional potential reasons could include aging, diabetes, genetics)
Less than 20% of the participants in the study developed lipoatrpphy according to the very strict definition used in the protocol (a loss of more than 10% of the subcutaneous fat). The reader should be aware that this is indeed a very strict definition. The median amount of fat at baseline of a regular size individual is around 7 kg, a loss of 10% of that mass is a loss of 700 g, which really does not put anybody in the range of what clinicians and patients consider severe lipoatrophy. Lipoatrophy becomes clinically apparent when the absolute subcutaneous fat is in the range of 3-4 kg [a loss of approximately 50% of the total subcutaneous adipose tissue).
The only intervention so far that has been proven effective in the prevention of lipoatrophy has been avoiding the use of thymidine analogs, or stopping them once a problem developed.
Uridine supplementation was proposed in the past as a potential intervention to replenish the dNTP supply and reverse the mitochondrial toxicity thought to be responsible for the development of lipodystrophy. Uridine supplementation was evaluated in a small trial in Finland. The ACTG completed a large randomized trial that evaluated NucleomaxX19, a uridine supplementation product that has been successful in treating some of the mitochondrial congenital diseases. Unfortunately the use of this product was not effective for the treatment of established lipoatrophy. It was also very poorly tolerated among patients with HIV infection (less than 50% of the participants were able to tolerate NucleomaxX for 48 weeks). After 48 weeks there were no improvements in the amount of subcutaneous fat, either by intention to treat or in the on treatment analysis.
The message for the clinician, (mainly for the clinician in the developing world) is to avoid the use of thymidine analogs if at all possible.
Tenofovir has been associated in rare cases with kidney failure and the clinician should monitor the kidney function in all the patients receiving this medication. EuroSIDA presented data on the risk of chronic renal failure and the use of particular antiretroviral regimens20. Not surprisingly, the use of tenofovir (and more unexpectedly, atazanavir, indinavir and lopinavir) was associated with a slight increase of chronic renal failure (a relative risk of 1.1 to 1.2), which was statistically significant because of the massive sample size of the study.
How relevant is this for the individual patient? To tell you the truth, not much, as the individual risk of CKD is approximately 1% if the patient never received tenofovir and increased to 2.4% after 4 years, with traditional kidney disease risk factors playing also a significant role in that risk.
I think there are three lessons to learn from these massive studies looking at the risk of complications with specific antiretroviral medications. The first is that there is no perfect drug; that each of the medications that we administer to our patients has the potential for toxicity, and that patients should be monitored frequently for these toxicities, and switched to alternative regimens should they appear. Second, that at the individual level these increases in risk are in most of the cases minor and do not outweigh the benefits of antiretroviral treatment, particularly when the events are rare, as in this case; and third, that one should use with extreme caution when using large clinical databases and not attribute causality at any particular association because the potential of channeling bias –some medications are prescribed more frequently in some particular population, that may be a greater risk for a particular outcome.
Gilead continues its love affair with the kidney (cidofovir, adefovir, tenofovir and now GS-9350 and counting). It seems that the company has a curse, and that every drug that they try to does something to the kidneys (I am sure that they do not find the comment amusing). Cal Cohen presented the results of 2 studies, showing the impressive antiviral activity of the Quad pill -tenofovir/FTC/elvitegravir and GS-9350) when compared to Atripla21, and GS-9350 as a booster agent with atazanavir. GS-9350-, also known now as cobicistat, is a new P450 inhibitor (like ritonavir) that boosts the levels of drugs metabolized through CYP3A4. One of the unexpected side effects of the drug is that increases the serum creatinine concentration modestly, although it seems that it does not decrease creatinine clearance when this parameter is measured directly. Cal Cohen presented pretty convincing evidence suggestive that cobicistat prevents active tubular secretion of creatinine, by a similar mechanism that other drugs like trimethoprim and cimetidine do. At this point is still unclear how significant this problem is going to be, but it has the potential of complicating the use of the new Quad tablet, as the clinician will have difficulty ascribing increases in creatinine after treatment initiation to tenofovir or to cobicistat. I think we will have to wait until the planned phase 3 trials are completed before deciding if this will affect the use of this compound or not.
1. Inflammation Is Associated with Endothelial Dysfunction among Individuals with Treated and Suppressed HIV Infection. Priscilla Hsue*, P Hunt, A Schnell, Y Wu, J Ho, R Hoh, J Martin, P Ganz, and S Deeks Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 709
2. T Cell Senescence and T Cell Activation Predict Carotid Atherosclerosis in HIV-infected Women. Robert Kaplan* , E Sinclair , A Landay , N Lurain , S Gange , R Sharrett , N Xue , P Hunt , H Hodis , and S Deeks . Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 708
3. Earlier Initiation of ART in HIV-infected Individuals Is Associated with Reduced Arterial Stiffness. J Ho, S Deeks, F Hecht, Y Xie, A Schnell, J Martin, P Ganz, and Priscilla Hsue* . Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 707
4. Rapid Progression of Atherosclerosis at the Carotid Bifurcation Is Linked to Inflammation in HIV-infected Patients. Priscilla Hsue*, P Hunt, A Schnell, V Selby, A Bolger, C Kalapus, J Martin, P Ganz, and S Deeks . Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 125
5. Progression of Carotid Intima-media Thickness in a Contemporary HIV Cohort. Jason Baker, K Henry , P Patel , T Bush , L Conley , W Mack , T Overton , M Budoff , H Hodis , J Brooks , and the CDC Study Investigators. . Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 126
6. Rates of Cardiovascular Disease following Smoking Cessation in Patients with HIV Infection: Results from the D:A:D Study. Kathy Petoumenos , S Worm , P Reiss , S De Wit , A d'Arminio Monforte , N Friss-Moller , R Weber , P Mercié , C Pradier , J Lundgren , and the D:A:D Study Group. . Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 124
7. Triglycerides and the Risk of Myocardial Infarction in the D:A:D Study. Signe Worm, A Kamara , W El-Sadr , O Kirk , E Fontas , P Reiss , A Phillips , M Bruyand , J Lundgren , C Sabin , on behalf of the D:A:D Study Group 1 Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 127
8. Assessment of Vitamin D Levels among HIV-infected Persons in the Study to Understand the Natural History of HIV/AIDS in the Era of Effective Therapy: SUN Study. Christine Dao1, P Patel , S Pals , T Bush , F Rhame , T Overton , E Kojic , K Wood , J Brooks , and the SUN Study Investigators. Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 750
9. Prevalence of Hypovitaminosis D among HIV+ Patients Enrolled in a Large Italian Cohort. Marco Borderi1, F Vescini , A Cozzi-Lepri , A Di Caro , I Shlacht , G Cassola , G Pellizzer , J Vecchiet , MC Re , A d'Arminio Monforte1 , for the Icona Fndn Study Group. . Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 751
10. High Prevalence of Severe Vitamin D Deficiency in cARTnaïve and Successfully Treated Swiss HIV Patients. N Mueller , Christoph Fux , B Ledergerber , L Elzi , P Schmid , T Dang , L Magenta , A Calmy , A Vergopoulos , H Bischoff-Ferrari , and Swiss HIV Cohort Study. Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 751
11. Vitamin D and HIV-related Complications and HIV Disease Progression in Women in Tanzania. Saurabh Mehta , D Spiegelman , F Mugusi , E Giovannucci , G Msamanga , and W Fawzi1 Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 753
12. Vitamin D Deficiency and Bacterial Vaginosis among HIVinfected and -uninfected Women in the US. Audrey French , , O Adeyemi , , D Agniel , M Yin , K Anastos , and M Cohen , Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 754
13. Vitamin deficiency among HIV type 1-infected individuals in the Netherlands: effects of antiretroviral therapy. Van Den Bout-Van Den Beukel,C.J et al AIDS Res.Hum.Retroviruses 2008 Nov 24(11):375-1382
14. Fracture Prevalence among HIV-Infected versus Non HIV-Infected Patients in a Large U.S. Healthcare System. Virginia A. Triant, Todd T. Brown, Hang Lee, and Steven K. Grinspoon.. Journal of Clinical Endocrinology & Metabolism, Published July 1, 2008
15. Higher and Increasing Rates of Fracture among HIV-infected Persons in the HIV Outpatient Study Compared to the General US Population, 1994 to 2008. Christine Dao , B Young , , K Buchacz , R Baker , J Brooks , and the HIV Outpatient Study Investigators. Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 128
16. HIV Infection and Fragility Fracture Risk among Male Veterans. Julie Womack , J Goulet , C Gibert , C Brandt , , K Mattocks , , D Rimland , M Rodriguez-Barradas , J Tate , M Yin , J Amy , , and Veterans Aging Cohort Project Team. Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 129
17. Fracture Rates Are Not Increased in Younger HIV+ Women. Michael Yin , Q Shi , D Hoover , K Anastos , A Sharma , M Young , A Levine , M Cohen , E Golub , and P Tien10 Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 130
18. Bone and Limb Fat Outcomes of ACTG A5224s, a Substudy of ACTG A5202: A Prospective, Randomized, Partially Blinded Phase III Trial of ABC/3TC or TDF/ FTC with EFV or ATV/r for Initial Treatment of HIV-1 Infection. Grace McComsey , D Kitch , E Daar , C Tierney , N Jahed , P Tebas , L Myers , P Sax , and AACTG Study A5224. Tien10 Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral 106LB
19. Uridine Supplementation in the Management of HIV Lipoatrophy: Results of ACTG 5229. Grace McComsey , U Walker , C Budhathoki , Z Su , J Currier , L Kosmiski , L Naini , S Charles , K Medvik , J Aberg , and Adult ACTG Study A5229 Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral LB 106
20. Chronic Kidney Disease and Exposure to ART in a Large Cohort with Long-term Follow-up: The EuroSIDA Study. Ole Kirk , A Mocroft , P Reiss , S De Wit , D Sedlacek , M Beniowski , J Gatell , A Phillips , B Ledergerber , J Lundgren , , for the EuroSIDA Study Group Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral LB 107
21. Single-tablet, Fixed-dose Regimen of Elvitegravir/ Emtricitabine/Tenofovir Disoproxil Fumarate/GS-9350 Achieves a High Rate of Virologic Suppression and GS-9350 Is an Effective Booster. Calvin Cohen, D Shamblaw, P Ruane, R Elion, E DeJesus, H Liu, L Zhong, D Warren, B Kearney, and S Chuck. Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Oral LB 58
22. Maraviroc Intensification for Suboptimal CD4+ Cell Response Despite Sustained Virologic Suppression: ACTG 5256. Timothy Wilkin, C Lalama , A Tenorio , A Landay , H Ribaudo , J McKinnon , R Gandhi , J Mellors , J Currier , and R Gulick1 Program of the Conference of Retrovirus and Opportunistic Infections. San Francisco February 16-19, 2010. Poster 285