 |
|
|
| |
HIV At ICAAC 2010: Complications & Co-Infections
|
| |
| |
David H Shepp, MD
Associate Professor of Medicine, Hofstra University School of Medicine
Division of Infectious Diseases, North Shore University Hospital-Manhasset
Cardiovascular Disease. Excess cardiovascular disease (CVD) is an important cause of decreased life expectancy for HIV-infected individuals treated with ART. The increase in CVD is multifactorial with social, behavioral, epidemiologic, immunologic and treatment-related factors contributing. Much research and controversy has focused on the potential role of ART as a contributing factor in the pathogenesis of atherosclerotic CVD. Investigators from the SMART study mined their extensive database to examine the association of ART with another form of cardiac disease, conduction disturbances1. In an analysis adjusting for multiple potential confounders, use of boosted or unboosted protease inhibitors (PIs) was associated with a statistically significant 5.1ms lengthening of the PR interval when compared to NNRTI exposure. Effects on QTc were small and less consistent for drugs in these two classes, both individually and combined. This study provides new information on the conduction effects of HIV medications, a topic of current interest to the FDA and other regulatory agencies, but the clinical consequences of the findings are unclear. No data currently exists to suggest clinical events potentially related to cardiac conduction defects occur with increased frequency in patients exposed to HIV PIs. Further studies are planned to identify clinical events such as cardiac syncope, atrial fibrillation, heart block, or sudden death that might be increased by exposure to antiretrovirals.
Intense controversy exists about the role of exposure to specific antiretrovirals in risk for cardiovascular disease. An analysis of the D:A:D cohort demonstrated a strong link between abacavir exposure and MI2. Other analyses have provided conflicting results. None of these studies can be considered definitive or establish a cause-and-effect relationship. An important obstacle to acceptance of the hypothesis that abacavir increases MI risk is the lack of a plausible mechanism. A previous study that examined the function of platelet taken from patients treated with abacavir suggested platelet hyperactivity as possible cause3. To further investigate this possibility, Diallo4 et al studied the addition in vitro of abacavir to platelets from healthy volunteers. No negative effect of abacavir on platelet function was found. This study suggests that abacavir does not have a direct effect on platelet function, but leaves open the possibility that platelet function could be altered indirectly, for example by induction of certain cytokines or prothrombotic factors.
Tuberculosis. Tuberculosis is one of the most common opportunistic infection and a leading cause of death in HIV-infected individuals worldwide. Treatment of TB has not evolved greatly since the introduction of rifampin 47 years ago, but the antibiotic pressure exerted by incorrectly administered or incompletely taken treatment has led to a global epidemic of M. tuberculosis strains classified as multi-drug resistant (MDR) and extensively drug-resistant (XDR). In HIV-negative patients, these infections are difficult to cure with available drug combinations and in HIV-infected patients, they are highly fatal. To address this public health crisis, novel drugs for TB are desperately needed. Few such drugs have entered clinical development, largely because of the lack of commercial opportunity in TB treatment. Therefore, a highlight of the 2010 ICAAC was the late breaker presentation on TMC 2075 (Tibotec Therapeutics), an investigational antituberculous antibiotic with a novel mechanism of action. This small (n=47) phase 2, randomized placebo-controlled trial showed a significantly shorter time to achieving sputum culture negativity when TMC 207 was added to standard therapy for the first 8 weeks only of treatment (11 vs. 18 weeks for TMC 207 vs. placebo, p=0.03). A higher rate of sputum culture conversion at last follow-up (81% vs 57%) was also reported. Importantly, TMC 207 well safe and well tolerated during the relatively short period of administration. Much more work is needed to confirm the safety and efficacy of TMC 207 in larger numbers of patients and over the longer durations of time typically needed to cure TB, but these preliminary results are promising and suggest TMC 207 may prove to be one of several new agents needed to deal with the epidemic of drug-resistant TB.
Even when M. tuberculosis retains sensitivity to standard antibiotics, a critical barrier to effective therapy in HIV-infected persons is drug-drug interactions between antiretrovirals and the antituberculous drug rifampin. Rifampin is a key component of standard TB therapy, but the optimal ART regimen for use with rifampin is not known. Interactions with HIV PIs make co-administration of rifampin virtually impossible for both safety and efficacy reasons. Rifabutin may be substituted for rifampin when PIs are needed, but still requires dose adjustment, is much more costly and may be less effective. Efavirenz (EFV)-based ART is an alternative, but rifampin reduces efavirenz concentrations moderately. Either a standard dose (600mg daily) or a slightly increased EFV dose (800 mg daily) has recommended. Investigators from the French ANRS reported a non-comparative study using co-formulated EFV/TDF/FTC along with antituberculous therapy in 69 HIV-infected TB patients who were previously ART naive6. 90% were on a rifampin-containing TB regimen and in these patients, an extra 200 mg of EFV was added. After 48 weeks, 57 (83%) were considered to have achieved the combination end-point of both TB cure and HIV RNA <50 copies/mL. Treatment failures were mainly due to incomplete HIV suppression. Only a single patient who was lost to follow-up was classified as a TB treatment failure. Eleven patients on 800 mg EFV with rifampin were found to have elevated plasma EFV levels (>4 mg/L) at week 2; 8 of these 11 had adverse events, suggesting therapeutic drug monitoring and dose adjustment may have a role in this unique clinical setting. Immune reconstitution inflammatory syndrome (IRIS) reactions were common (23 patients) but were considered serious in only 5. While firm conclusions about which ART regimen is optimal cannot be drawn, this non-comparative study does suggest that high rates of treatment success for both HIV and especially for TB are achievable when the fixed-dose combination of TDF/FTC/EFV is combined with rifampin-containing TB treatment. The convenience of the co-formulation also helps address another common treatment barrier, namely regimen complexity and pill burden.
Metabolic Disorders. Abnormal distribution of body fat is a prominent feature of the metabolic disorder that may complicate long-term administration of ART. Its cause is multifactorial. Peripheral lipoatrophy is a toxicity of certain older nucleoside analogs and is now uncommon in countries where these drugs are no longer used. The cause of visceral abdominal fat (VAT) accumulation is less well understood. A direct effect of antiretrovirals has not been clearly shown. Chronic inflammation may play a role. It may also reflect "return to health", or really the transition from the metabolic disturbances seen with untreated HIV to those increasingly common in individuals with a typical Western diet and life style. Beyond diet and exercise, effective treatment has been elusive. Tesamorelin (Theratechnologies) is an investigational analog of growth hormone releasing factor that has been shown in randomized, placebo controlled trials to reduce VAT in HIV-infected patients during 26 weeks of treatment7. Important questions remain about the use of this injectable and possibly costly agent. Does the resulting loss of VAT improve health, for example, by reducing risk of cardiovascular disease or diabetes? Is long-term administration safe? Can it favorably affect a patients body image and quality of life? Since the clinical manifestations of metabolic disorders differ by gender, age and race, do all patients groups derive similar benefits? The latter two concerns were addressed in two poster presentations. Data from 816 patients who participated in phase 3 clinical trials showed that 26 weeks of tesamorelin resulted in a mean waist circumference reduction 2.4 cm (1 inch) greater than in placebo recipients8. Waist reduction correlated statistically with VAT reduction. Tesamorelin patient assessment of belly appearance, general well-being and psychological distress, and physician assessments of belly appearance were also significantly correlated with VAT reduction, while in placebo recipients most assessments did not correlate. Subset analyses showed that all groups (women, older patients, those with hepatitis co-infection) achieved similar VAT reductions with tesamorelin, with the possible exception of non-white patients, in whom sample size was small9. These findings are encouraging and suggest that favorable, perceptible body image changes can be achieved by patients who receive tesamorelin treatment. However, the true usefulness of this new therapy will depend on demonstration of long term safety, reduction in serious medical illness and cost-effectiveness.
Bone Disease. Osteopenia and osteoporosis are very common in HIV-infected patients. The causes are incompletely understood. Compared to the general population, certain risk factors such low body-mass index (BMI), tobacco, opiate and alcohol use are more often present. Fractures also occur more often in HIV-infected10. Both low bone mineral density (BMD) and increased risk of falls and trauma are contributing factors. Antiretroviral therapy in general causes further BMD declines, probably due to complex alterations in cytokines and hormones involved in bone turnover. Specific antiretrovirals such as tenofovir and perhaps protease inhibitors are associated with additional BMD loss. Recent research has focused on another risk factor for low BMD that is extremely prevalent in both the HIV-infected and uninfected population - vitamin D deficiency. Using slightly different definitions, two cross-sectional studies of HIV-infected patients from Madrid, Spain (n=147) and Lille, France (n=395) demonstrated 48% and 41%, respectively, had vitamin D deficiency11,12. Previous studies have associated EFV use with lower vitamin D levels. The Spanish study did not confirm this association but surprisingly found nevirapine use was associated with a protective effect. The French study did not suggest a lower risk with nevirapine but found NNRTI use as a class was independently associated with vitamin D deficiency. A third study retrospectively reviewed 391 patients from Barcelona who had at least two DEXA scans between 2000 and 200913. The median age was 43 years, 73% were men and the average duration of ART was 8 years. At a median of 2.5 years between scans, the proportion with osteoporosis increased from 22 to 27%. Twenty-eight percent had progression from normal to osteopenia/osteoporosis or from osteopenia to osteoporosis. In a multivariable analysis, older age, lower BMI, male gender, duration of PI and duration of tenofovir use were all independent risk factors for progression of bone loss.
These studies highlight the importance of bone disease in HIV-infected patients and the high prevalence of a readily reversible risk factor, vitamin D deficiency. The high prevalence of vitamin D deficiency in Spanish patients is not surprising. Despite its milder climate Madrid, lies at a the same latitude as New York City, making adequate sun exposure for vitamin D synthesis difficult to achieve during winter months. The finding of protection from vitamin D deficiency with nevirapine exposure is intriguing, especially since a study of a new, extended release preparation presented during ICAAC 201014 help make nevirapine a more attractive treatment option. However, the finding is too preliminary and needs confirmation in additional studies. It is possible EFV and nevirapine differ in their effect on the vitamin D hydroxylating and metabolizing enzymes. Further research should address this possibility.
Liver Disease. Many recent studies examining mortality in HIV-infected patients show liver disease to be a leading non-AIDS defining cause of death. Most of this liver disease is due to chronic hepatitis C virus (HCV) infection. Although treatment of HCV is poised for progress with the anticipated introduction of directly acting anti-HCV (antivirals), current treatment with the combination of ribavirin and pegylated interferon is unsatisfactory, especially in HIV co-infected patients. Patients who achieve a sustained virologic response (SVR) benefit from large decreases in liver disease-related events and improved survival. However, only a minority of treated patients achieve an SVR. A multicenter study of 1428 co-infected patients examined if treatment outcomes other than SVR conferred any health benefit15. The median duration of follow-up was 46-49 months. In a multivariable analysis adjusting for potentially confounders, patients who achieved an undetectable HCV RNA after 48 weeks of treatment (end of treatment response) but subsequently relapsed, experienced a 60% reduction in liver-disease related events when compared to those with no response. By comparison, those with an SVR had a 92% reduction.
Among the most important discoveries in HCV research in the past few years was the identification of a specific single nucleotide polymorphism (SNP) in a regulatory region of the IL-28B gene, encoding interferon lambda, as a determinant of responsiveness to ribavirin/interferon treatment of HCV. Barreiro et al. retrospectively reviewed the correlation between this SNP, designated rs12979860 CC and the presence of cirrhosis in 304 Spanish HIV/HCV co-infected patients who were evaluated pre-treatment by either elastometry, biopsy or both16. Cirrhosis was significantly more common in those with the CC SNP (24% vs 13% than in those with CT or TT, p=0.01). Similar findings were seen in the subset of patients with HCV genotype 1 or genotype 3. ALT also was higher and among those with a liver biopsy, inflammation scores were higher in the presence of the CC SNP. In multivariable analysis, the CC SNP was the only risk factor that remained statistically significantly associated with cirrhosis.
These two studies provide support for continuing effort to treat HCV in HIV co-infected patients. The first study demonstrates that important benefits may accrue to some patients who relapse after discontinuation of therapy. The second supports a role for the IL-28B gene in control of the natural host immunologic response to HCV. The IL-28B CC SNP identifies not only patients with the best chance of treatment success but perhaps also those with the greatest risk of progression to cirrhosis. Therefore, co-infected patients with the IL-28B rs12979860 CC SNP should be prioritized for treatment.
Kidney Disease. According to current HIV treatment guidelines for developed countries, tenofovir DF (TDF) along with its co-formulated companion emtricitabine (FTC) form the nucleoside backbone for all preferred initial ART regimens. A large number of studies conducted over the past 10 years have demonstrated that TDF has a unique combination of efficacy, tolerability, ease of administration and safety. However, clinically overt renal tubular toxicity occurs in a small number of patients, manifesting either as acute tubular necrosis or a subacute presentation of Fanconi syndrome, usually with diminished glomerular filtration rate (GFR). Both syndromes are largely reversible after discontinuation. Several studies have suggested that low grade tubular dysfunction, with or without small declines in GFR, occurs in additional patients who otherwise appear to be tolerating long-term TDF. Tomas-Jimenez, et al reported a longitudinal analysis of tubular function in 79 patients receiving TDF-based ART17. Patients were monitored at study entry and at 12 months, but pre-TDF levels were not obtained. The median duration of treatment at entry was 14 months. Estimated GFR declined by 8-10 mL/min depending on calculation method, similar to changes found in many other studies. Glucosuria, phosphaturia, urinary n-acetyl glucosamine were measured in 24 hour urines. All were found to increase significantly, findings consistent with tubular dysfunction. An unexplained observation was a significant increase in albumin excretion, a manifestation of glomerular injury, suggesting the presence of other underlying kidney disease in the study population (patients with diabetes and hypertension were excluded) or a previously undescribed effect of TDF. TDF has been associated with low molecular weight proteinuria, a more typical sign of tubular disease. Although the findings of subclinical tubular dysfunction are consistent with those of several previous studies, there are limitations to the current study, including the lack of pre-TDF evaluation and no non-TDF comparator group. Clinical renal or bone disease events were not described. Clinicians need to know if subclinical tubule dysfunction identifies patients at risk for progression to clinically overt Fanconi syndrome, chronic kidney disease or osteomalacia, or if it is clinically benign. Extensive clinical experience with TDF over the past 9 years reinforces its overall safety, but further studies with clinical correlation could determine if screening TDF-treated patients for tubular dysfunction is clinical useful.
Use of HIV protease inhibitors (PI) was revolutionized when it was recognized that their unfavorable pharmacokinetic (PK) properties could be dramatically improved ("boosted") by co-administration of low-dose ritonavir (RTV). Using RTV as a booster for other drugs with fragile PK is an attractive idea. While low doses are usually well tolerated, RTV has certain undesirable properties, including gastrointestinal toxicity, hypertriglyceridemia, possible induction of drug-resistant HIV, off target effects on enzymes not involved with PI metabolism and high cost. Cobicistat (Gilead Sciences) is an investigational PK enhancer being developed as an alternative to RTV for use with either the investigational integrase inhibitor elvitegravir (Gilead Sciences) or the HIV PI atazanavir (ATV). Elion et al presented 48 week results from two small, randomized, placebo-controlled, phase 2 studies of cobicistat (COBI)18. The first compared the combination of TDF, FTC, and elvitegravir (EVG) boosted by COBI to the standard preferred first line regimen of TDF, FTC and EFV. Each multi-drug regimens was co-formulated into a single tablet so pill burden remained low. The second study directly compared COBI and RTV as enhancers for ATV, with TDF/FTC forming the nucleoside backbone for each regimen. High rates of viral suppression (82-90% with HIV RNA <50 copies/mL) were achieved with all regimens and no dramatic differences in safety or tolerability were noted. Trends favoring fewer CNS adverse events and fewer discontinuations were seen for EVG/COBI vs EFV. At 48 weeks, larger increases from baseline in serum creatinine occurred with EVG/COBI vs. EFV (+0.17 vs +0.06 mg/dL), as did corresponding larger declines in estimated GFR (-19.7 vs -5.5 mL/min). In the study comparing COBI to RTV to boost ATV, changes in creatinine and estimated GFR similar to those seen in with EVG/COBI occurred in both arms. This result is consistent with recent studies suggesting that RTV-boosted PIs have independent renal effects19 which may prove to be similar with COBI. In a presentation at CROI 2010, Gilead scientists suggested the elevations in creatinine seen with COBI were artifactual, because 7 days of COBI administered to healthy volunteers produced similar changes in creatinine but no change in true GFR as measured by iohexol clearance20. They hypothesized the elevations in creatinine may be due to inhibition of tubular secretion of creatinine, an effect seen with other widely used drugs such as trimethoprim and cimetidine.
These small phase 2 studies were not powered to establish non-inferiority or to provide definitive answers on safety, but they do provide sufficient encouragement for larger phase 3 studies that are now underway. In these trials, the clinical significance of creatinine elevations and estimated GFR declines will need to be carefully defined. Clinicians cannot easily differentiate estimated GFR and true GFR. Because COBI elevates creatinine and lowers estimated GFR, use of this drug may create a clinical management dilemma, particularly when combined with other drugs that have renal effects such as TDF and ATV. Additional data is needed to show that true GFR is not impacted to a greater degree when COBI is combined with TDF and/or ATV in HIV-infected patients who receive these drugs over longer periods of time.
Women & Infants. Use of ART during all or most of pregnancy, combined with additional peripartum regimens for mother and infant, has almost completely eliminated perinatal transmission of HIV. Current guidelines for the use of ART during pregnancy in developed countries recommend older antiretroviral combinations that are no longer considered to be preferred for the non-pregnant patient, because data on the pharmacokinetics (PK), safety and efficacy of the newer antiretrovirals in pregnancy is slow to accumulate. Small incremental data on the use of two ARVs that are preferred in non-pregnant but not pregnant patients were presented. Best et al. studied HIV-infected pregnant women and reported that the PK of raltegravir did not obviously vary when 3rd trimester and post-partum measurements were compared21. As in non-pregnant patients, considerable interpatient variability was seen. Raltegravir crossed the placenta well (mean cord blood/maternal plasma ratio 1.2). Atazanavir (ATV) is another preferred ARV for non-pregnant patients. Since it can cause asymptomatic hyperbilirubinemia, there is a safety concern for the neonate if ATV is used late in pregnancy. Marx et al. retrospectively reviewed 52 infants with in utero exposure to ATV, other PIs or NNRTIs22. No large differences were seen among the 3 exposure groups in clinically documented jaundice or use of phototherapy. Neonatal bilirubin levels were obtained within the first 72 hours in only a minority of cases (and few without ATV exposure), but levels >5 mg/dL were more common in the ATV group.
Both of these studies move toward establishing the safety of these two agents in pregnancy, but are too small and too preliminary to alter prescription practice in HIV-infected women. Further studies in larger numbers of women and a greater accumulation of clinical experience will be required before these agents can be considered preferred in the pregnant patient.
1. Soliman EZ, et al. 50th ICAAC, Boston, MA, Sept. 12-15, 2010; abst. H-218
2. Westring Worm S. et al. J Infect Dis 2010;201:318-30
3. Satchell CS, et al. 16th CROI; Montreal, Canada; Feb 8-11, 2009. Abst. 151LB
4. Diallo YL, et al. 50th ICAAC 2010, abstr. H-230a.
5. Diacon AH et al 50th ICAAC 2010, abstr. L1-521a
6. Lortholary O, et al. 50th ICAAC 2010, abstr. H-232
7. Falutz J, et al. J Clin Endocrinol Metab. 2010;95:4291-304.
8. Falutz J et al 50th ICAAC 2010, abstr. H-227
9. Zoltowska M. et al 50th ICAAC 2010, abstr. H-228
10. Triant V et al. J Clin Endocrinol Metab 2008;93:3499-504.
11. Cervero M, et al. 50th ICAAC 2010, abstr. H-230
12. Pasquet A, et al. 50th ICAAC 2010, abstr. H-225
13. Bonjoch A, et al. 50th ICAAC 2010, abstr. H-226
14. Gathe J. et al. 50th ICAAC 2010, abstr. H-1808
15. Berenguer J et al. 50th ICAAC 2010, abstr. V-1784.
16. Barreiro P, et al. 50th ICAAC 2010, abstr. H-1169
17. Tomas-Jimenez et al. 50th ICAAC 2010, abstr. H-229
18. Elion R, et al. 50th ICAAC 2010, abstr. H-938b
19. Mocroft A et al. AIDS 2010;24:1667-1678
20. Cohen C, et al. 17th CROI, San Francisco, CA, Feb 16-19, 2010, Abstract 58LB
21. Best BM, et al. 50th ICAAC 2010, abstr. H-1668a
22. Marx AH, et al. 50th ICAAC 2010, abstr. H-1664
|
| |
|
|
|
|
|
