 |
|
|
| |
Topics from the Drug Therapy in HIV Infection Meeting
|
| |
| |
Glasgow Scotland
November 2008
By Stephen Becker, MD
Lessons From the Developing World
Pharmacology
The Tuberculosis HIV Nexus
The Longer-Term Safety of Maraviroc
Malignancy and Immune Suppression
The Drug Therapy in HIV Infection (HIV9) meeting held in Glasgow is an opportunity to view the treatment imperatives from a non-US perspective. The meeting itself is a combination of synthetic topic overviews and presentation of original data occurring over four days. The sessions are sequential without the concurrent arrangement of the typical scientific meeting. There was considerable representation of African and Asian researchers, investigators and clinicians. Given the breadth of material any summary of the meeting content will necessarily be subjective. The author has selected several topic areas that appeared to provide a unique content and have application to all involved in HIV work. Significant new data has already been reported and posted on the natap site (www.natap.org) in the form of daily reports.
Lessons from the Developing World
Resistance
The rollout of antiretroviral therapy in regions of Sub-Saharan Africa has brought the desperately needed and widely expected clinical benefits first apparent in the developed world, and has revealed important and unique features of the epidemic. Several presentations from African investigators highlighted important clinical, resistance and pharmacologic aspects of disease management. Several contextual features that define the reported clinical experience reported here should be noted:
♦ Despite WHO and national guidelines, ART initiation frequently does not occur until late in disease; for much of southern Africa the median CD4 count at time of ART is less than 100 cells
♦ Most virus in the region is clade C
♦ Relevant major coinfections for the region are tuberculosis and malaria
♦ First line ART consists of stavudine + lamivudine + nevirapine or efavirenz. Zidovudine may be substituted for stavudine, while tenofovir and abacavir are considered 'alternative' NRTIs. Second line ART is a boosted PI + didanosine + whatever NRTIs remain from the initial regimen. There is no routinely available therapy beyond second line at this time
♦ Clinical monitoring - CD4 count, HIV RNA and resistance testing - is rarely available. Transition from first line to subsequent ART occurs at the time of clinical disease progression, rather than laboratory-based virologic or immunologic failure. Thus it is expected that significant first-line NRTI and NNRTI resistance will have developed at the time of clinical disease progression and transition to second-line therapy
♦ At a population level there is no history of prior mono- or dual NRTI therapy that characterized and conditioned drug development and sequencing in much of the developed world
♦ Rates of transmitted drug resistance are currently low, but may be expected to rise in the future
♦ Prevention of mother-to-child-transmission programs are widespread and have utilized both single dose nevirapine as well and combinations of nevirapine + NRTIs. Population viral sequencing has demonstrated nevirapine resistance in 30-40% of treated mothers and approximately 50% of treated infants. Single genome sequencing shows rates of resistance that are considerably higher, approaching 90% for both mother and infant. This has obvious implications if NNRTIs are to remain the central component of a combination antiretroviral treatment approach
A research collaboration between the Desmond Tutu HIV Centre and Harvard University (Orrell, abstract 0113) reported resistance patterns among 230 patients (120 treatment naïve; 110 treatment experienced) receiving care in Cape Town, South Africa. Among the treatment naive individuals transmission of drug resistance was low (estimated at 2.5%). Among those failing first line ART treatment, rates of resistance were high. Failure was defined as HIV RNA > 1000 copies on two successive occasions. Seventy-three percent of patients were receiving an efavirenz-based regimen, and 89% were taking stavudine. Treatment limiting NNRTI mutations (K103N, Y181C, and V106M) were noted in 83%. Most had 2 or fewer NNRTI mutations. NRTI resistance mutations were frequent, with M184V being present in 78%. Surprisingly, and of great concern, the K65R mutation was noted in 9.5% of those failing first line therapy. All had received stavudine, and none had received tenofovir as part of their ART regimen. The selection for K65R in clade C viruses confirms in vitro work and earlier clinical observations, but was here reported at a 2-fold higher frequency than in previous surveys. The high frequency of emergent K65R virus immediately poses the question of the optimal composition of first line regimen used throughout most of the developing world. As currently constructed, with predominant use of stavudine, selection of the K65R mutation will impair not only subsequent tenofovir use, but that of abacavir and didanosine as well. Any second line regimen would therefore constitute only the PI as a fully active agent, thus setting the stage for rapid second-line treatment failure. The extensive knowledge base accumulated through two decades of antiretroviral use in the developed world while of substantial value may not always be relevant in portions of the world with differing viral predominance. The Orrell study while limited in size confirms studies elsewhere and suggests an urgent need to reconsider design of first line therapy. Indeed the selection of stavudine + lamivudine + nevirapine as first line therapy for the developing world was based on issues initially relating more to cost, patent and availability. Few clinicians would suggest these agents as an optimal first regimen given current understanding of ART potency, toxicity and resistance, and fewer would likely do so if nevirapine is to remain as an important component of mother-to-child transmission prevention. There is a growing movement in support of a delineation of drugs used for prevention (perinatal, pre-exposure prophylaxis and post-exposure prophylaxis) and those used for treatment of established infection. The consistently superior response rates of NNRTI-based regimens rightly places these agents as first line therapy in both the developed and developing world. Consideration of newer classes of antiretrovirals, especially the CCR5 antagonists, as agents designated for prevention, and not therapy may be appropriate. This is a subject that will likely generate useful debate in the coming period. Commercial issues aside, with 6 classes and 20+ ART agents, it may be reasoned to consider designation of specific antiretroviral agents for anticipated use as part of larger prevention strategies.
Pharmacology
Drug-drug interactions characterize the majority of antiretroviral agents and regularly result in dosage adjustment based on pharmacokinetic studies. Rightful concern about the effect of lowered drug exposures has limited pharmacodynamic evaluation in most cases. Further, among the numerous drug-drug interaction studies the preponderance of participating subjects are of Caucasian descent, and the relevance of PK findings from this population to Africans is not widely understood. The substantial enzyme induction by the NNRTI agents efavirenz and neviripine and the rifamycins used in the treatment of tuberculosis, routinely results in dose adjustment or avoidance of certain combination therapies altogether. Two carefully performed clinical studies examined the effect of rifampicin when dosed with efavirenz (Orrell, abstract 0124) and the tablet form of lopinavir/ritonavir when dosed with either efavirenz or nevirapine (Kityo, abstract 0125). The Cape Town group asked the question whether the recommended efavirenz dose elevation to 800mg, rather than a conventional 600mg dose, was required when combined with rifampicin used as part of an anti-tuberculosis regimen. Numerous PK studies have demonstrated the enzyme induction effect of rifampicin, and proposed the need to up-dose various co-administered medications. To test the clinical application of this approach 60 individuals on efavirenz-based ART with concomitant TB were enrolled. Sixty-three percent were female and the cohorts were well matched for weight, baseline CD4 and HIV RNA. Thirty-one subjects received efavirenz at 800mg/day and 29 at 600mg/day. Upon completion of all TB therapy those dosed with efavirenz 800mg/day received conventional 600mg dosing, while those initially treated with 600mg continued this dose. Two mid-dose samples were obtained from each study participant, the first 4 weeks after beginning co-dosing with rifampicin and the second upon completion of the TB therapy. Efavirenz levels were not statistically (or clinically) different in the two cohorts (3.2mg/ml (IQR 1.5-6.3) vs. 2.85mg/ml (IQR 1.8-5.0) during the co-dosed phase of study. At the completion of anti-tuberculous therapy efavirenz concentrations fell in the group initially dosed at 800mg/day, the level was similar to that of the group who received the 600mg/day dose throughout the study (2.15 mg/ml (IQR 1.38-3.09) vs. 2.4 mg/ml (IQR 1.4-4.3). Importantly there were no differences in clinical outcomes, including CD4 count, HIV RNA, adverse events or abnormal laboratory values. On the basis of this study, it appears that when using rifampicin as part of an antituberculous regimen efavirenz doses do not require an increase from 600mg/day. A limitation to this study is that it may have application to those of South African descent, and thus study in other populations is warranted. Efavirenz, like neviripine, is metabolized by the CYP isoform 2B6. Genetic polymorphisms within this cytochrome vary, but those at position 516 (G>T) are associated with slowed clearance of efavirenz. Both the 516G/T and 516T/T polymorphisms are known to be more frequent among those of African ancestry.
The second study, a collaboration between Ugandan and UK investigators examined dosing strategies among African patients receiving the new enhanced bioavailability tablet formulation of lopinavir/ritonavir and either efavirenz or neviripine. Early data from the lopinavir/ritonavir development program suggested the need to increase lopinavir/ritonavir capsules from 3 twice daily to 4 twice daily when co-dosed with a NNRTI. The current co-formulated tablet does not permit dosing at the same levels as with the capsules. When co-dosed with a NNRTI the capsule formulation of lopinavir/ritonavir was dosed at 533/133 twice daily. Using the new formulation 2 tablets yields a dose of 400/100 and 3 tablets 600/150. Investigators were concerned that dosing 2 tabs twice daily might lead to insufficient lopinavir plasma concentrations, while 3 tabs would result in enhanced toxicity. Using the conventional 533/133 twice daily capsule dosing as a referent, they compared the geometric mean ratios when dosing with either 2 or 3 tablets twice daily. Geometric mean ratio (GMR) values less than 1.0 signify concentrations less than the referent, while those greater than 1.0 signal concentrations above the referent. In this elegantly designed study, each subject served as their own control and received the capsule formulation as well as both doses of the tablet form of lopinavir/ritonavir in sequence. The study enrolled 40 subjects, 19 receiving neviripine and 21 efavirenz. Predictably women were under-represented in the efavirenz group, but both groups were otherwise well matched. When compared to the standard 4 capsule dosing, 2 tablets yielded an AUC GMR of 0.82 with efavirenz and 0.90 with neviripine. When dosed with 3 tablets GMR values were 1.40 with efavirenz and 1.66 with neviripine. While the GMR values for 2 tablets are reduced within a range that might be considered clinically insignificant, the inter-subject variability of lopinavir AUC was greater than expected. The authors properly suggest that further study in this population is required. As noted by Dr. Kityo use of two tablets plus _ of the pediatric tablet (100/25) twice daily might provide exposures that fully offset the effect of a co-dosed NNRTI and permit a greater degree of clinical certainty given the substantial variability in lopinavir exposures.
These studies demonstrate the importance of pharmacokinetic - and pharmacodynamic - evaluation within the designated treatment population. Such studies, funded by the manufacturer, should be performed locally at the early stages of antiretroviral rollout. Academic collaborations between clinicians and scientists from the developed world can usefully leverage the substantial therapeutic and pharmacologic knowledge base accumulated during the past two decades, and enhance the impact of ART programs. Putative racial differences in response to antiretroviral therapies have been the subject of research interest for the past 8 to 10 years. To date the data that support clinically meaningful differences have been mixed and often contradictory. Studies within African populations will add meaningfully to the large datasets in those of Caucasian descent. Particularly useful will be the genetic analyses that allow us to go beyond skin color in determining any existing racial differences in ART response.
The Tuberculosis - HIV Nexus
Tuberculosis is a primary co-infection of the HIV epidemic worldwide. Gerald Friedland from Yale University in collaboration with workers from the South African province of KwaZulu-Natal presented a sobering picture of the twined and combined epidemics. The presentation went beyond the numbing statistics and hardships of work in the developing world to provide a glimpse of a deeply committed team of health care providers whose work is in the greatest tradition of public health. Tuberculosis remains the leading cause of death world-wide. Nearly 1/3 of the world's population is infected with TB, and an estimated 12 to 14 million are co-infected with HIV. Regions in sub-Saharan Africa, Southeast Asia, Latin America and Eastern Europe are most affected. The region within KwaZulu-Natal province, the basis of Friedland's report, has the highest rate of extensively drug resistant (XDR) and multi-drug resistant (MDR) TB in the world. (MDR TB is defined as resistance to isoniazide and the rifampin compounds; XDR is MDR plus resistance to the fluroquinolones and second-line agents such as capreomycin and the aminoglycosides). Drug resistance occurs through both an acquired route (inadequate treatment, or treatment failure) and primary resistance from transmission of drug resistant organisms. The latter is the primary mechanism in areas of HIV-TB co-infection. The mortality rates for MDR and XDR TB in HIV co-infected patients in this region of the KwaZulu-Natal province are staggering: 67% and 82% respectively. Yes, 67% and 82% mortality. Most deaths occur within the first month from diagnosis, and the median time from first sputum collection to death among XDR infected individuals is 16 days. The local clinic lacks microbiology facilities which results in frequent empiric treatment use, while the central laboratory facility lacks rapid diagnostic methods. Airborne infection control methods are scarce and the systems of care for TB and HIV are institutionally and culturally vertically partitioned. Among those diagnosed with M/XDR TB 90% are unemployed, 72% live in mud-built housing, 11% have electricity and only 2% a toilet.
The HIV co-infections in the developing world, tuberculosis, malaria and certain helminthic diseases among others will pose extreme challenges. The lessons of co-infection management from the developed world will not be instructive, and novel and systematic solutions are urgently needed. Yet within this intersection of HIV and TB the lessons and gains of ART implementation in the developing world - not ART itself - offer hope and guidance. The creation of medical infrastructures including local and regional diagnostic facilities, early case finding and airborne infection control programs, integrated TB and HIV care services and a cadre of scientists to advance and inform clinical and social policy provide a means forward. Intensive efforts are underway to develop a TB vaccine funded by several US and European government agencies and private foundations. Tibotec has developed the first new anti-tuberculous agent in decades. This agent, TMC 207 has had very encouraging results in early proof of concept studies. Friedland's telling of the TB - HIV interface while agonizing provides a reasoned formula for controlling the 'perfect storm' of HIV, TB and accelerated TB drug resistance.
The Longer-Term Safety of Maraviroc
Safety issues have surrounded the class of CCR5 antagonists for the past several years. The well known hepatotoxicity that ended development of alplaviroc and the occurrence of an unexpected (and still unexplained) number of malignancies in the vicriviroc program have conditioned a greater concern about safety within this class. CCR5 antagonists are of course the first class of agents directed at a host rather than a viral target. CCR5 is known to be an immune modulatory receptor, though full knowledge of its downstream signaling remains a work in progress. In humans the natural deletion of CCR5 (the observation that launched CCR5 antagonist programs) appears to have minimal impact on overall survival, though certain infectious diseases may occur with greater frequency or severity. It is within this context that longer-term pooled safety data from the maraviroc (MVC) phase III studies can be viewed. Reporting for the MOTIVATE team David Hardy (abstract 0425) presented general safety information on 1075 study participants who completed 96 weeks of observation. The study's primary endpoint was 48 weeks, at which point subjects in the MVC once daily arm were switched to open-label twice daily MVC. Subjects who maintained virologic suppression on the placebo arm were offered the opportunity to add MVC or remain on their OBT. All subjects were permitted to re-optimize therapy at week 48. Thus duration of MVC exposure was different among the originally randomized arms. The adjustment for duration of exposure is an important element of this study. The table below summarizes selected safety measures as reported by Hardy:
It is reassuring to note a lack of a general safety signal to 96 weeks in a population of 1000+ advanced patients. The absence of significant MVC-associated hepatotoxicity was understood by the time of 48 week data reporting. The fundamental question surrounding CCR5 antagonist safety, if one exists, surrounds immune dysfunction. This may not be manifest by traditional measures such as SAEs, category C events or even malignancies over a 96 week period. Should immune modulatory dysfunction result from CCR5 pharmacologic blockade, this may require a longer period of clinical observation and biomarkers, such as auto-antibodies, not planned as part of initial safety assessment. Pfizer has committed to a long-term follow up (5 years) for those enrolled in MVC clinical trials. Lacking a matched control group and the logistical difficulties of following subjects for such a period of time, it is unclear the eventual value of this information. To date, however, MVC appears safe and well tolerated by traditional measures. Its effectiveness in those harboring R5 tropic virus is clear for both treatment experienced and treatment naïve individuals. Vicriviroc appears on a development path leading to regulatory approval. What additional clinical benefit is provided by this agent and how successfully it will compete in the CCR5 market is unclear at this time. Also uncertain is the future of the several CCR5 antagonists in earlier development stage.
Malignancy and Immune Suppression
HIV has been associated with a number of malignant conditions from the earliest days of the clinical recognition of the disease. The reason for the increased incidence of malignant diseases was assumed to be a direct effect of immune deficiency, though proof of this has been lacking. Shared risk confounding, for example the high rate of smoking among those with HIV infection, and exposure to numerous oncogenic viruses, are cited as alternative or confounding hypotheses. Building on work published last year, Andrew Grulich and colleagues from New South Wales, Australia (abstract 0321) have extended their comparative analysis of cancers arising in a population of solid organ transplant recipients and those with HIV. Their evaluation revealed that there were remarkable similarities in the types of cancer seen in these two populations, thus supporting the notion of immune deficiency as the proximate cause of HIV associated cancers. The value of their approach lies in the fact that transplant recipients share few of the lifestyle risks for cancer with the HIV-infected population. Further the iatrogenic nature of immune deficiency among transplant recipients may provide information on the extent of immune suppression - or recovery - associated with increased malignant risk.
The majority of cancers seen in both the solid organ transplant and HIV populations are those associated with oncogenic viruses rather than the more typical epithelial malignancies such as breast, prostate, colorectal and ovary, seen in the general population. Among the oncogenic viruses, Epstein-Barr, human herpes virus 8, human papilloma virus and hepatitis B and C are associated with the greatest incidence of malignancies. The table below lists selected oncogenic infections and associated cancers seen in both the HIV and solid organ transplant populations.
The level of immune deficiency associated with specific malignancies remains uncertain. In studying a group of renal transplant recipients whose graft failed and who were returned to dialysis (with cessation of all immunosuppressive therapy) their elevated risk of cancer rapidly returned toward normal. Effective antiretroviral therapy is associated with a dramatic reduction of some, but not all cancers and rates even in those with full and prolonged virologic suppression remain elevated above the HIV seronegative population.
The value of this ongoing work lies in the establishment of immune deficiency as the cause of HIV associated cancers. This has bearing on the question of when to initiate antiretroviral therapy, as certain cancers show an increased incidence among those with HIV even at modest levels of immune deficiency. Further this data may also inform the goals for optimized immune recovery.
|
| |
|
|
|
|
|
