Introduction Despite the success of combination antiretroviral therapy, a significant proportion of patients experience a loss of virologic, immunologic, or clinical benefit from their current regimens. Frequently, these patients have limited options for alternative treatment regimens. Developing safe and effective therapies for treatment-experienced patients, particularly for those with documented three- or four-class antiretroviral drug resistance is a public health priority [1] that poses significant challenges. Causes of treatment failure are diverse as are target patient populations with respect to the type and duration of treatment experience. These aspects of patient care necessitate individualized treatment plans and make standardization of objectives and study designs for development of new therapies difficult. Issues in study design include the choice of adequate control arms, appropriate endpoints, and reasonable expectations for duration of response and safety considerations. The Salvage Therapy II workshop was held 16–17 April 2004, to discuss these issues with HIV-treating clinicians, clinical research investigators, pharmaceutical industry representatives, regulatory authorities and interested HIV-community leaders. The workshop focused on issues concerning heavily treatment-experienced patients, specifically patients who are three- or four-drug classes experienced with limited or no options for suppressive antiviral regimens. This review summarizes the proceedings from this meeting and represents both a consensus of the views expressed during the workshop as determined by the authors as well as the views of the authors themselves. The review also highlights the discussions on challenges that surround new therapy development and clinical management of treatment-experienced patients [2]. Drug development issues for treatment-experienced patients Historical perspective The introduction of suppressive (triple-combination) antiretroviral therapy led to dramatic decreases in morbidity and mortality of HIV-infected individuals during the 1990s [3–6]. Unfortunately, limitations accompanied these therapies, especially for patients with previous exposure to antiretroviral therapy. Issues of drug resistance, cross-resistance to multiple drugs from a single class, regimen complexity, frequency of dosing, variable plasma concentrations and drug toxicities all contributed to poor antiviral responses or failure of initially successful regimens. By the late 1990s, an increasing number of patients receiving triple combination therapy experienced treatment failure, leaving them with few or no new therapeutic options. Virologic failure rates over 60% after 1 year on treatment were observed in some cohort studies [7,8]. ‘Salvage therapy’ was now required for those who not only failed earlier mono and dual combination therapies, but also later suppressive triple combination regimens. Limited treatment guidance was available to help physicians and patients select an optimal regimen as few post-triple combination ‘salvage therapy’ trials were being conducted and most of these included small sample sizes. Furthermore, unknown drug–drug interactions resulting in reduced drug exposure also contributed to poor virologic responses seen in studies and in the clinic [9]. Nevertheless, several expert groups formulated treatment guidelines and provided initial guidance for the selection of regimens with optimal activity in treatment-experienced patients. In the absence of evidence from clinical trials, the recommended regimens for treatment-experienced patients were based on the expert opinion of the Department of Health and Human Services Treatment Guidelines Committee [10]. The recommendation that ‘Optimally and when possible, the regimen should be changed entirely to drugs that have not been taken previously … at least two and preferably three new drugs should be selected that are not anticipated to be cross-resistant to drugs given previously …’ was difficult to implement in the absence of available new drugs. The guidelines stressed the need for high-level expertise and recommended assistance through consultation with more experienced clinicians [10]. To further illustrate the state of uncertainty regarding clinical management of treatment failure, the guideline committee noted that these recommendations were a ‘work in progress’ and that more clinical trials were urgently needed [10]. Early studies investigating potential regimens in treatment-experienced patients were designed simply by selecting drugs that the patient had not previously taken. These studies yielded disappointing results, probably due to high degrees of cross-resistance between drugs. Studies of protease inhibitor (PI)-experienced patients showed poor virologic response rates; only 20 to 50% achieved HIV RNA levels below 400 copies/ml at 24 weeks [11–13] Investigators pursued alternate strategies, including multiple-drug rescue therapy (regimens containing up to nine drugs), and structured treatment interruptions (STIs). Results from the multiple-drug rescue studies were variable. Most studies reported virologic suppression rates of only 30 to 50% and participants commonly experienced moderate to severe toxicities resulting in suboptimal adherence, frequent changes in regimens or discontinuation of certain drugs [14,15]. Through retrospective studies, several groups documented shifts from multiple-drug-resistant virus to wild-type virus populations in treatment-experienced patients undergoing an interruption of antiretroviral treatment and suggested investigating STIs as a treatment strategy for this patient population [16,17]. Subsequently, prospective controlled studies of STIs showed more rapid CD4 cell declines, HIV RNA rebounds and clinical events in comparison with continuous treatment [18,19]. Recognizing the lack of treatment options and guidance for heavily treated patients, advocates and researchers focused their efforts by organizing a series of meetings on this issue. A meeting was co-sponsored by The Forum for Collaborative HIV Research in 1999 to discuss the design and implementation of studies in the treatment-experienced population; to present needs, priorities and challenges faced by industry, researchers, regulators and patients; to define treatment failure and success; and to discuss what was necessary and feasible when designing studies of new drugs in this setting [20]. This meeting led to an FDA Antiviral Drugs Advisory Committee Meeting in 2001 to discuss clinical trial design issues in heavily treatment-experienced patients and an FDA letter to pharmaceutical companies urging collaboration to allow testing of multiple investigational agents in one study [1,21]. Improvements in the management of treatment-experienced patients resulted from these meetings, including incorporation of novel strategies (e.g., drug-resistance testing) and development of new antiretroviral agents. However, despite some advances, treatment for antiretroviral-experienced patients remains the greatest challenge in the clinical management of HIV-infected individuals. Current standard of care for the treatment-experienced patient Clinical management of treatment-experienced patients has evolved over the years. More than 20 different antiretroviral agents and formulations spanning four drug classes are available for the treatment of HIV infection. Prior to the year 2000, controlled trials were largely conducted in treatment-naive or nucleoside-experienced patients and data regarding dosing, safety and efficacy in treatment-experienced patients were lacking. Today, data exist for several antiretroviral agents in the setting of three-drug class experience. In addition, we have a better understanding of how to use technologies such as resistance testing. Current treatment guidelines suggest several approaches to the management of treatment-experienced HIV-infected patients [22,23]. Strategies for assessment and management of patients with limited, intermediate or extensive prior treatment and/or drug resistance are different [22] and reflect the recognition that different treatment goals may be appropriate for these populations. Although a number of reasons for treatment failure can be identified, medication intolerance and suboptimal adherence are probably the most common, and drug resistance is the final common pathway [24,25]. Patients experiencing failure of their first treatment regimen generally have several treatment options and the goal of therapy remains complete suppression of HIV RNA. In contrast, a patient with failure of multiple prior regimens and significant drug resistance has limited, if any, treatment options; for this patient the goal of therapy is to preserve immune function and prevent clinical progression while waiting for newer strategies or treatments. The evaluation of patients with treatment failure includes assessment of virologic, immunologic and clinical status; determination of the cause of treatment failure; and review of pharmacokinetic parameters. In addition, resistance testing is performed while patients are still taking their failing regimen [21,26] to identify active antiretroviral drugs for subsequent treatment regimen [27–29]. The inclusion of a drug class that is new to each patient, such as a non-nucleoside reverse transcriptase inhibitor (NNRTI) [12,13,30] or HIV entry inhibitor [31,32] provides virologic benefit for many patients. Not surprisingly, the more active drugs (ideally three or more) in a regimen the greater the likelihood of viral suppression [26,27]. However, selection of subsequent regimens can be problematic; studies show expert advice combined with resistance testing provides greater benefit than selection of an antiretroviral regimen based on resistance testing alone [33]. Other strategies to maximize antiviral activity include the use of dual PIs and pharmacokinetic enhancement. Ritonavir, a potent inhibitor of the cytochrome P450 3A hepatic enzyme, increases concentrations of most other PIs resulting in improved antiretroviral activity against partially resistant viral strains [13,34]. In patients experiencing virologic failure on a PI, the use of dual PIs is also associated with improved virologic responses [12,13]. Ritonavir-boosted dual PI regimens have been explored [35] but unexpected drug–drug interactions may occur [36] and this approach has not been shown to be better than unboosted dual PI regimens. Further research is needed before this approach can be integrated into treatment guidelines. Older strategies such as empiric multiple drug selection of regimens have largely been abandoned. More recently, a strategy of partial treatment interruption (i.e. continuing some drugs in a regimen) suggested sustained virologic control and reduced toxicity [37,38] but this strategy awaits further evaluation. In the US, routine use of therapeutic drug monitoring (TDM) is not currently recommended; however, in Europe TDM is frequently used [2]. The clinical utility and optimal use of TDM in treatment-experienced patients is under investigation. In summary, the current standard of care for patients not responding to their present antiretroviral regimen includes the following: identify and address the reason for treatment failure; perform drug-resistance testing; and use drug-resistance testing results to design a new regimen with at least two active antiretroviral drugs (including consideration of investigational agents, if available). The major challenge facing patients and physicians today is the availability of at least two active antiretroviral drugs to construct a new regimen. Addition of one active drug to a failing regimen typically is not recommended because of the risk of developing resistance to that drug. However, in patients at high risk for clinical progression, this strategy may provide some benefit, at least temporarily, while waiting for availability of newer active therapies. Challenges to developing new therapies In 2003, a fourth therapeutic drug class, entry inhibitors, was added to the armamentarium of HIV treatment. Today, development of new molecular drug targets such as HIV entry and viral with development of new agents in drug classes and challenges in trial design and data in treatment-experienced patient populations of study of efficacy and safety due to options for participants who experience virologic failure; and of and The of study participants to challenges in designing trials that are to patients and and reflect the population to use the new therapy in clinical [2]. these challenges is and to this study with standard of including the use of and/or resistance data to construct an antiretroviral regimen are regimens with at least two active antiretroviral drugs are studies should allow the use of and investigational agents. Drug interactions and of the resistance also to the in designing treatment-experienced patients drugs for of or regimens that or dual PI therapy The potential for multiple drug interactions is and can safety or loss of patients are to the risk of new investigational agents before all necessary drug interactions studies are These are not for all treatment-experienced patients and be with for patients with few or no treatment options. of the patient population can also be as a of trials heavily treatment-experienced patients. resistance allow evaluation of a of responses to parameters. to responses by resistance provide evidence of activity and for of the viral population patients also challenges for the development of new agents such as The utility of these drugs may on viral populations and viral Patients may viral strains with dual or of with different The antiviral activity of several in patients with virus was shown in clinical studies however, in these drugs were not active against at this of the activity of against or dual is In addition, progression may when a virus changes from use to use a a viral population or be before treatment with a however, this assessment use of an investigational Current are not not provide a of or and are not to between and dual challenge of antiretroviral drug development in is and of regimens. In treatment-experienced patients with few or no treatment an individualized is an treatment are because due to in or assessment of patients from or patient of the treatment Unfortunately, is not possible, particularly for or drugs with and toxicities or formulations such as is not possible, studies include to and of study designed to include choice of prior to for changes to in all study arms, and for active and control groups A regimen may be an however, this is not feasible for all treatment-experienced patients because of cross-resistance can be used to the that patients are drugs have design challenge is the need for a trial to be to patients clinical care include 1 and designs that provide to the investigational for all patients at some during the designs have been used in trials however, this design challenges for efficacy and safety options to a number of in the control and challenges in safety This design also of the of events for the duration of exposure as of and is and HIV RNA levels below the of may not be feasible for certain treatment-experienced patients include from HIV RNA levels or proportion of with 1 or greater in HIV RNA. Results from the trials, 1 and and the trials, and that a 1 in HIV RNA levels from and HIV RNA 400 copies/ml or copies/ml is for many patients of response rates from the and studies and other studies is for designing trials in patient populations. challenge is the to the of an treatment-experienced a physicians and patients difficult on to or to use new therapies. patients into a clinical trial with one new investigational or for the availability of two or more new new therapies treatment The and of the of new drugs may be in the management of such patients. Furthermore, challenges to exist for the development of new investigational agents in drug and each class of antiretroviral drugs has Although several investigational agents with activity against HIV strains are in benefit from these drugs for heavily treatment-experienced patients may be resistance can with as as one to and new the of active drugs available for combination therapy to the development of activity has been observed with new investigational however, the utility of these agents may be limited in treatment-experienced patients with no options. certain such as reverse transcriptase or with activity against the number and/or type of and may virologic response rates seen with patients with three or more that included at reverse transcriptase or had reduced response rates to patients who not have the or present at II studies with an investigational PI, showed reduced response rates when more than two PI were present at and a this was used to in In addition, determination of the of and for new investigational agents is for patient selection into clinical drug new agents that are not by or PI are Despite these challenges in HIV drug development 1 provides a of new HIV agents in The need to new drugs from drug classes and new antiretroviral drug targets is and collaboration between pharmaceutical clinical researchers and the HIV is to the success of new agents issues in the development of new therapies for treatment-experienced patients Developing new therapies for treatment-experienced patients issues and with the need to data and to new therapies, improved study designs and are needed for treatment-experienced patients. these may allow the to in a Patients who have available therapies need new treatment options to their risk of progression and/or these patients, in controlled trials is not and in safety studies or studies is more and rapid to investigational antiretroviral drugs for patients, clinicians and is an However, at with of trials or with for other In has been that the most rapid to investigational agents may from and of and/or data with the goal of and rapid of for physicians and their also to the of these challenges for studies include the limited availability of on active and drug–drug are many regulatory for to investigational agents in all of drug In the these from in drug development to use of more later in to in the clinic with implementation of these In the challenge is with the least of regulatory and of or use of these to regarding efficacy and safety is not new events be and as a for further to of after initial of data in heavily patients, studies designed to safety or virologic are of the type of the FDA the use of more than one investigational in clinical however, the to use investigational agents with other including other investigational agents, should be in studies prior to typically provides to of patients. between several pharmaceutical companies in the development of can be for more than one investigational available at the to patients who need of successful collaboration was the of for and in these patients to active therapies comparison of the each of a drug drug. Meeting regulatory regulatory for drug and to investigational agents is a to new therapies and the need to how to use a new drug in combination with other antiretroviral are both on dosing, drug interactions and resistance is for effective clinical management of treatment-experienced patients. also need regarding the of resistance on treatment regulatory evidence of safety and efficacy be in adequate and clinical however, studies in patient are not The Committee for and FDA make a between treatment-experienced patients with options and those with no available options investigational therapies. regulatory authorities state in guidance that the population may not be appropriate for type studies typically used to previously these patients may be better for single or comparison studies in one or more new investigational agents. such as pharmacokinetic virologic response to or and frequency of events be in these These data are not for but are and into for heavily treatment-experienced patients. a the of these trials changes from regulatory studies to study designs that provide for the heavily Developing study designs designing studies, the from studies is many patients multiple drug resistance after receiving (i.e. the introduction of new agents one by one over study designs that treatment options are In studies a new investigational or patients are at risk of developing resistance to the one new their therapeutic options or The risk of resistance on the activity from the options. have trial designs one investigational added to is to Patients who can to for new drugs to available not into clinical trials for of developing viral resistance to the investigational and to a new class of drug. Patients who to (e.g., those with CD4 cell and extensive drug are to trials of however, these patients may have a poor response to therapy or a limited response due to the development of resistance to the investigational from studies and illustrate options for trial designs for the heavily exist to the trial designs in participants are to an investigational or options are for while a fourth is more for but may also address clinical study designs an investigational to when added to is an for certain treatment-experienced patients. in these studies are for treatment to the investigational for not or a virologic noted with the development a with this design is the potential for specifically the exposure of the control in comparison with the new drug Patients who from the control to the investigational may be those who are more and not responding to therapy. that events occur with frequency in patients with CD4 cell in treatment may cause To for in the duration of treatment regimen one approach is to rates designs including trials with multiple investigational agents is However, these designs the availability of several study drugs at the of particularly for drug and is to the success of these In the absence of complete are an in pharmacokinetic or TDM data are and in a limited number of patients before of the design options. current current regimen. at and are added to all treatment studies are designed to the and of a new drug. The study is controlled for a all participants the investigational for the duration of the The controlled of the study for the evaluation of activity while the risk of resistance to the new drug or This design is a when only one new drug is The appropriate duration for these studies with to efficacy and safety is still an however, for safety the feasible duration is Virologic response may after to the of response may not be arms, for are in of studies, including the use of multiple investigational agents are particularly designs for patients who are not for controlled clinical if the goal is to provide to an investigational for a is to this use trials with more monitoring or study in with use studies or one of the A is the need for to several new active drugs collaboration from several different pharmaceutical companies and/or is In the this has been exist studies of multiple investigational agents were no feasible due to safety or with one of the new drugs. include in unexpected drug interactions reduced to this challenge exist other than The failure of one should not research is to with multiple pharmaceutical companies and such studies should be many patients in the triple experienced are those who antiretroviral therapy at a when suboptimal treatment regimens or dual were the only The of patients therapy today are unknown and no are available to make these triple combination therapy to current treatment guidelines the of patients who have available treatment options. Nevertheless, drug resistance, and adherence issues to frequent changes in antiretroviral therapies. a the heavily treatment-experienced population to more regimens with toxicities available for or experienced patients. In the the development of effective strategies for the management of HIV-infected antiretroviral treatment-experienced patients is a public health The population of treatment-experienced individuals is and the management of these patients is individualized and a number of antiretroviral agents with activity against virus are in However, challenges in investigational agents and in designing optimal clinical The current goal of therapy is to provide treatment-experienced patients with at least two and preferably three active agents, but this is not patients can in clinical studies with novel designs such as or designs address clinical and clinical development and regulatory of newer antiretroviral drugs. However, not all patients are for clinical trials, and for the use of multiple new are options for certain patients. The from studies help to study designs for patients most in need of new therapies. Challenges in the design and of trials in treatment-experienced patients and collaboration between pharmaceutical clinicians, clinical researchers, and the HIV-infected are for the successful development of new therapies and effective treatment This is based on and discussions during a ‘Salvage Therapy by the Forum for Collaborative HIV Research and The for The workshop was in of of The for The workshop an of and drug HIV clinical researchers, HIV care industry and patient The Forum is a from and industry as well as from the patient care and The workshop was by the Forum and The for with from the and all of the committee of Forum for Collaborative HIV The of The for Health of Forum for Collaborative HIV The for Drug and and Drug for Drug and and Drug for their and guidance in the as well as all workshop participants for their active in the the expert management from and for assistance with and for A review based on a Forum for Collaborative HIV Research in collaboration with The for The Forum for Collaborative HIV Research is through from and pharmaceutical for this workshop were from the and
No takes yet. Share an insight, caveat, or question.
Struble et al. (2005) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: