Results of Antiretroviral Treatment Interruption and Intensification in Advanced Multi-Drug Resistant HIV Infection from the OPTIMA Trial

Background Guidance is needed on best medical management for advanced HIV disease with multidrug resistance (MDR) and limited retreatment options. We assessed two novel antiretroviral (ARV) treatment approaches in this setting. Methods and Findings We conducted a 2×2 factorial randomized open label controlled trial in patients with a CD4 count ≤300 cells/µl who had ARV treatment (ART) failure requiring retreatment, to two options (a) re-treatment with either standard (≤4 ARVs) or intensive (≥5 ARVs) ART and b) either treatment starting immediately or after a 12-week monitored ART interruption. Primary outcome was time to developing a first AIDS-defining event (ADE) or death from any cause. Analysis was by intention to treat. From 2001 to 2006, 368 patients were randomized. At baseline, mean age was 48 years, 2% were women, median CD4 count was 106/µl, mean viral load was 4.74 log10 copies/ml, and 59% had a prior AIDS diagnosis. Median follow-up was 4.0 years in 1249 person-years of observation. There were no statistically significant differences in the primary composite outcome of ADE or death between re-treatment options of standard versus intensive ART (hazard ratio 1.17; CI 0.86–1.59), or between immediate retreatment initiation versus interruption before re-treatment (hazard ratio 0.93; CI 0.68–1.30), or in the rate of non-HIV associated serious adverse events between re-treatment options. Conclusions We did not observe clinical benefit or harm assessed by the primary outcome in this largest and longest trial exploring both ART interruption and intensification in advanced MDR HIV infection with poor retreatment options. Trial Registration Clinicaltrials.gov NCT00050089


ORIGINAL ASSUMPTIONS REVISED ASSUMPTIONS
-Standard-ART cumulative event rate at Year 1 is 13% with a 25% annual increase thereafter until the end of the study -Crossover (Mega to Standard) is 20% in Year 1; decreases 50% every year thereafter -Hazard reduction of 30% with full adherence (unadjusted for losses and crossovers; adjusted hazard reduction 22.7%) -Loss to follow-up at 3.5 years is 10% -Drop-in (Standard to Mega) is 5% Year 1; increases 10% every year thereafter -Alpha-level of 0.05 -2. 5 year accrual; 1 year minimum follow-up -Power of 80% -Standard-ART cumulative event rate at Year 1 is 23% with a 25% annual increase thereafter until the end of the study -Crossover (Mega to Standard) is 5% in Year 1; decreases 50% every year thereafter -Hazard reduction of 30% with full adherence (unadjusted for losses and crossovers; adjusted hazard reductions are shown in table 3) -Loss to follow-up at 5.5 years is 5% anti-retroviral (mega-ART) salvage therapy is tolerable, and that treatment interruption and re-treatment can favorably impact surrogate response markers. [1][2][3] Whether the balance of these responses and toxicities to treatment interruption and intensified drug therapy translate into significant clinical benefit is unknown.
There are no foreseeable new anti-HIV drug treatments that are safe, effective, and would not invoke treatment-emergent drug resistance in a way that would supplant the therapeutic principles tested in OPTIMA. OPTIMA will have lasting relevance to anti-HIV treatment, where there continue to be treatment failures with multi-drug resistant (MDR) HIV infection.
Treatment failures due to drug resistance were the rule before HAART was developed and put into wide use. Unfortunately, despite the significantly positive impact of HAART, today's anti-HIV treatments still fail at a predictable rate due to the limits of tolerance, side effects, variable adherence to these complicated regimens and development of resistance to these medications. As well, there is an increasing emergence of transmitted drug resistance in primary HIV infections, which will more than likely further limit efficacy even of initial therapies. Thus, OPTIMA's target patient population persists and will grow.
In order to address this, clinicians continue to attempt to improve outcomes of HIV therapy among those patients for whom currently recommended treatment regimens have failed by either intensifying therapy (by adding more drugs to existing treatment regimens) and/or by offering treatment interruptions to decrease toxicities and improve adherence and possibly viral responses when treatment is restarted. Opinions about these treatment strategies are diverse, but strongly held. However, the effectiveness of treatment interruption and/or intensification in salvage therapy has not been proven in a controlled clinical endpoint study.
Thus, OPTIMA is poised to define important principles in the management of anti-HIV treatment failures.
OPTIMA continues to have important relevance to practicing clinicians and patients with HIV. The documented increase in levels of drug-resistant HIV and the scientific importance of OPTIMA is growing. Despite the relevance of this study and multiple efforts to improve accrual, OPTIMA has been unable to achieve the accrual targets established for the original sample size of the trial (n=1700).
The primary hypothesized causes for the observed accrual problems are: 1.
The length of time required to obtain IRB approval for local sites to begin active recruitment.

2.
A smaller than anticipated number of patients were eligible for the study because of the restrictive eligibility criteria.

3.
Site specific issues (e.g., lack of clinician and/or patient equipoise; Mega-ART funding) The first issue has been nearly resolved. There is certainly a belief in many quarters that treatment interruptions are not helpful, and this belief may be more strongly held now than when the study was being planned. Unfortunately, this appears to be based on both poorly controlled, observational data and not on reliable evidence. Similarly, there is lack of equipoise about the relative value of standard vs. mega-ART, and this may be present in either direction. In some settings, mega-ART has virtually become standard of care, while in others, mega-ART is not supported by clinicians and/or paying agencies. The Trial Management Committee's understanding that these clinician and patient beliefs are based on insufficient reliable scientific evidence was supported by the Trial Steering Committee.
Lack of equipoise has its greatest effect on accrual in trials when the interventions under study such as mega-ART and drug-free periods can be provided to patients at their direct request, rather than through a randomization process. While the foregoing explains, in part, difficulties with accrual we have experienced thus far, it also underscores how important it is for the study to continue. When therapeutic strategies vault into widespread use based on inadequate or inconclusive data, this is anathema to evidencebased practice. Despite the difficulties that widespread use imposes on trials, it becomes even more imperative to subject the interventions to rigorous scientific evaluation.
The original sample size was calculated based on assumptions that have proven to be very conservative, particularly event and cross-over rates. The actual event rate seen in OPTIMA is similar to that in a comparable trial, the CPCRA MDR-HIV study in patients with similar baseline CD4 count. Thus, the revised sample size of 504 is realistic, likely to achieve the scientific goals of the study and, with an extension of the accrual period, is achievable at the current rate of accrual and at some cost savings.
These proposed changes and the rationale for them have been endorsed by the Trial Steering Committee (TSC) and the Data Safety and Monitoring Board (DSMB) as appropriate, based on sound scientific rationale and correct management of this study.
A realistic estimate of the accrual rate is 9-10 patients per month. During the last 4 months, accrual from the three countries has averaged 11 patients per month.
After evaluating the recently released results from two randomized studies (GIGHAART and the MDR-HIV study) 1,2 and reviewing the accumulated data in OPTIMA, it is apparent that the original sample size estimate was based on too conservative assumptions about the primary event (progression to a new AIDS defining event or death), treatment crossover and loss to follow-up rates. Sample size estimates are largely driven by the assumed primary event rate (the smaller the event rate, the larger is the required sample size), the expected treatment difference in the primary event rate, the treatment crossover rate and (to a lesser extent) the loss to follow-up rate. In light of new data the sample size was revised. Monitoring sample size assumptions is a standard procedure and an integral part of trial management, so that if accumulating data from the trial suggest substantial deviations from the original assumptions, sample size is revised (very often upwards).
However, it is important that such revisions are not based on the observed treatment difference in the primary outcome. The TSC made their decision blinded to treatment differences in OPTIMA, hence no modification to the original expected treatment effect is made.

Revised Sample Size and Impact on Health Economics:
The health economics component of this trial will not be adversely affected by these changes. To the extent that estimates of cost-effectiveness will be based on modeling the event rates and disease progression observed during the trial to extrapolate survival over the patients' life time, we will now have a longer observation period upon which to model future events. With fewer patients in the study, the precision of the cost and outcome measurements will be affected. However, the trial was not initially powered according to cost and quality of life measures but rather according to clinical events. The uncertainty associated with cost-effectiveness results will be handled using methodology that is different from what is typically adapted for efficacy estimates emanating from randomized clinical trials. Specifically, we will adopt a Bayesian approach by reporting the probability that one treatment is the most cost-effective given the data rather than using classical hypothesis tests. We will implement this by estimating cost-effectiveness acceptability curves that show the probability that each management option is the most costeffective conditional on the value the decision-maker places on additional health outcomes.
We feel that this methodology will adequately handle the reduction in sample size by showing the implications of the reduced precision in cost and outcome estimates for the decision uncertainty faced by decision makers.
In summary, after careful consideration of the proposed sample size reduction and potential changes in protocol design, we feel that the economic analysis will still provide the policy decision maker with vital information upon which to base their funding decisions.
The revised sample size is still based on conservative assumptions, given data from similar studies and more importantly, current data from OPTIMA. We have assumed a lower primary event rate than currently observed in the study to allow for possible reduction in this rate in the future. If the current rates are maintained during the study, the required 292 primary events could be achieved over a shorter trial time. These sample size assumptions will continue to be monitored by the TMC, TSC and the DSMB.
A graphical presentation of the projected revised target sample size ( Figure 2) clearly indicates that the projected accrual performance is realistic and achievable given the study's accrual performance to-date.

2) A 3-month Antiretroviral Drug-Free Period (ARDFP) versus No ARDFP
The OPTIMA trial is the first large-scale, multicenter, randomized controlled trial to compare the relative efficacy of these different therapeutic strategies. The use of multiple settings in different 'therapeutic cultures' will allow for generalizability of the findings and provide evidence that will facilitate management of HIV disease in this group.
A total of 504 patients will be randomized over a period of 4.5 years at 65 medical centers in the three countries. Patients will be followed at regular intervals during the 5.5 year study, until the last patient randomized has been followed for 1 year. The study is designed (taking into consideration drop-in, drop-out and loss to follow-up) to detect a 30% reduction in the relative hazard (under full compliance) between mega and standard-ART (2sided =0.05).
The trial funding will be tripartite: the Department of Veterans Affairs (VA) Cooperative Studies Program will provide funding for the 25  Furthermore, subsequent treatment options are narrowed by the problem of drug resistance, which is complicated by the existence of cross-resistance within each of the three classes of anti-retroviral drugs. 8,9 An effective virologic and immunologic response may be elicited by changing as many drugs of the combination as possible, particularly if a new class of drug is used, and if the virologic breakthrough is of lower degree or shorter duration. In practice, unfortunately, the response to switching therapy to 3 or 4 new drugs after failure on the first treatment, is often transient. Once a prolonged and significant virologic breakthrough occurs on two different HAART regimens that have included all three classes of drugs, few treatment options are available with confidence, and the optimal management of such patients remains unclear.
Failure of successive regimens becomes increasingly likely due to mutational crossresistance between drugs of the same class or decreased adherence caused by intolerance or increased complexity of the remaining effective regimens. The introduction of newer medications may offer a respite but, with current knowledge, is unlikely to offer a solution to these trends. The clinical dilemma posed in selecting treatments for patients facing this situation is the focus of the OPTIMA trial and these treatment strategies will also be evaluated from a societal perspective with cost-effectiveness analyses. including those on which the patient has failed in the past, is given in the hope that their combined anti-HIV activity will lead to maximal virologic suppression 10 , and immune recovery with acceptable tolerance. 11 Third, drug therapy is stopped temporarily (known variously as 'drug-free period', 'drug-holiday', 'structured treatment interruption') in order to reduce the symptoms of toxicity so that the new regimen may be better tolerated and in the hope that more sensitive viruses will re-emerge without continued drug pressure, which may allow superior anti-HIV activity and clinical efficacy of subsequent treatment. The latter phenomenon is biologically complex, and controversial. Information from case-series is gradually emerging, 12 but many questions remain unanswered. 13 The probability of viral response to mega-ART, defined as a decline in viral load to below 500 copies/mL, increased with the length of ARDFP in one study. 14 It is hypothesized that, in the absence of pressure from drugs, most of the virus population returns to wild-type, and is thus sensitive to drugs re-introduced after the ARDFP. Enhanced response to mega-ART has been observed in patients with at least 2 preceding months of ARDFP in spite of not unexpected falls in CD4 counts during the ARDFP. In addition, it is possible that a temporary ARDFP may provide the patient with an improved quality of life without seriously affecting long-term survival, or possibly even improving survival even with a standard-ART regimen because the patients are better able to tolerate the new therapy. The concept of treatment interruptions has been supported by NIH therapeutic guidelines which state -For patients with no rational alternative options who have virologic failure with return of viral load to baseline (pre-treatment levels) and declining CD4 T cell count, there should be consideration for discontinuation of anti-retroviral therapy‖. 15 The optimum duration of an ARDFP , however, has not been determined.
The first question addressed by this trial is whether there is a difference in effectiveness between mega-ART and standard-ART regimens. The use of a large number of anti-retroviral drugs, referred to as 'mega-ART', is an experimental treatment strategy that has met at least some success, as defined by virologic response. 16 Essentially, the strategy is to treat with as many anti-retroviral drugs as possible (defined as 5 or more) and maintain them for as long as possible. In the largest case series to date, investigators reported on 106 heavily pre-treated patients with a median of 43 months of prior anti-retroviral therapy and a median plasma viral load (pVL) as measured by HIV RNA of 62,000 copies/mL. 10 As initially presented, patients received anywhere from 5 to 9 drugs as part of their mega-ART therapy and were followed for a median of 15 months. An on-treatment analysis showed that 48% of 88 subjects had pVL below 400 copies during weeks 47-57. On an intent-to-treat analysis, 40% of patients had pVL values <400 copies/mL between weeks 47 and 57 of follow-up. Among those subjects who achieved suppression below 400 copies/mL on 2 consecutive measurements, the probability of remaining suppressed at 40 weeks of followup was 59%. Lower pVL and being non-nucleoside reverse transcriptase inhibitor (NNRTI)naïve at initiation of mega-ART were predictors of a favorable virologic response. Baseline susceptibility to lamivudine, stavudine, didanosine, and saquinavir were also strongly associated with a favorable virologic response. In this series, 26 (25%) of patients experienced severe laboratory abnormalities or subjective adverse drug effects, of whom 6 discontinued treatment. Toxicity was considered frequent, but generally manageable, by the investigators. 10 Although these virologic data lead to some optimism about the value of this approach, there are inadequate long-term data to indicate whether the virologic response is translated into gains in survival or decreases in AIDS-related complications. There are no randomized controlled trials of mega-ART versus standard-ART and there are no studies evaluating the long-term clinical outcomes comparing these treatment strategies.
The second question addressed by the OPTIMA trial is that of the clinical utility of an anti-retroviral drug-free period (ARDFP). An anti-retroviral drug-free period as proposed in a person whose treatment has failed is different from a structured treatment interruption, i.e. cessation of anti-HIV therapy in the face of successful virologic suppression. For this trial, an ARDFP is a relatively brief period (i.e. 3 months) prior to reintroduction of treatment. When treatment is re-introduced, it is hoped that its benefits may exceed those that would have been achieved without an ARDFP. Declines in CD4 count and increases in pVL during the ARDFP are anticipated, but they are expected to be transient and reversible, without major long-term adverse consequences. All patients will receive optimal prophylaxis against opportunistic infections.
The potential value of an ARDFP, in the presence of multi-drug resistance, is not only the freedom from drug-related toxicity and improved quality of life, but also the possibility that the efficacy of a subsequent HAART regimen may be improved relative to such a regimen initiated without an interruption. At least one well-documented case evaluation demonstrates proof of principle for the concept but is it generalizable? 17 What is the optimal length of time for the ARDFP?
Further support for the concept of renewed treatment effectiveness following treatment interruption comes from a number of observational cohort studies. 14 Evaluation of other cohorts has led some investigators to opposite conclusions. 21 Retrospective evaluation of the non-randomized longitudinal EuroSIDA cohort suggests an accelerated progression akin to the natural history of progressive HIV-associated immune deficiency disease for failing patients having increasing durations of treatment interruptions.
Clearly, disease and death rates in patients who discontinue medical treatments (both anti-HIV and anti-OI) are higher than in those patients who do not. This has been represented as relevant to outcomes of treatment interruption in future patients who wish to subsequently resume medical health maintenance. 22 However, treatment interruptions in this cohort were ad hoc and not linked to a strategy for resuming treatment designed to obtain favorable therapeutic outcomes.
Recent prospective studies have provided some support for this hypothesis.  25 Also, the effort to increase treatment in impoverished countries is likely to amplify the problem of multi-drug resistance because of the inability to sustain and monitor fully effective therapies. These emerging trends in drug resistance are likely to increase with time so that the underlying rationale for OPTIMA, which seeks to find the best approach to salvage, will continue to be of relevance to an increasing number of people.
The rationale for OPTIMA would also be compromised if other studies convincingly answer the questions posed, which might make continuation of the randomization unethical.
The GIGHAART study provides important support for the controversial concept that a treatment interruption can enhance response to subsequent anti-viral treatment. Its design, however, evaluates the limited surrogate outcome of virological response over a short period and does not adequately address the more meaningful impact of treatment interruptions on clinical endpoints. The MDR study was terminated for futility and in part also because of undesirable declines in CD4 lymphocytes and a trend toward increased HIV associated clinical events in the treatment interruption arm. Unfortunately, the majority of these clinical events appear to have been avoidable with assiduous prophylaxis. Neither of these trials has persuasively answered the clinical question of the effect of a drug free period on subsequent treatment in this population, and no randomized trials compare the risks and benefits of Standard-ART vs Mega-ART. Thus, the failure of other studies to answer the questions posed by OPTIMA, coupled with evidence supporting one of the more controversial hypotheses, enhances the ongoing relevance of the trial.

V. Study Objectives
The study will compare different strategies for the management of patients with HIV infection for whom first and second line HAART has failed.
The OPTIMA study aims to evaluate (a) the effect of mega-ART compared to standard-ART and (b) the effect of an intended 3-month ARDFP compared to no ARDFP in the management of patients for whom previous HAART therapy has failed. The impact of mega-ART and ARDFP on cost and Quality Adjusted Life Years will be determined; lifetime costs and outcomes will be modeled to determine whether these strategies are costeffective.

VI. Study Outcome Measures
The primary and secondary endpoints are:

VII. Study Design
The trial is a 2x2 open randomized study of patients with advanced HIV disease in whom ART, including all three classes of anti-HIV drugs, have failed. Randomization will allocate patients (a) either to a ARDFP of at least 3 month duration or to 'No ARDFP' and (b) to either a 'mega-ART'' regimen (5 or more drugs) or to a 'Standard-ART' regimen (up to 4 drugs). Patients will be followed until the last participant enrolled has completed 12 months of follow-up (a median of 3.5 years).
The primary hypothesis is that mega-ART is superior to standard-ART in terms of prolonging life and delaying the occurrence of new or recurrent AIDS events. This is a management trial comparing different strategies for salvage therapy. The first comparison will be between the strategy that incorporates the use of at least 5 anti-HIV drugs (mega-ART) (e.g. 3 reverse transcriptase inhibitors, 1 non-nucleoside reverse transcriptase inhibitor and 2 protease inhibitors used in therapeutic doses) and a conventional strategy of up to 4 anti-HIV drugs (Standard-ART). For the purposes of this trial, dual protease combinations involving ritonavir for pharmacokinetic reasons (in doses of 100-200 mg bid) will be counted as one drug. Hydroxyurea is not regarded as an antiretroviral drug for this study, but may be used at the investigator's discretion. The choice of anti-HIV drugs used during this trial will be at the discretion of the clinician and will be chosen according to the patient's previous drug history, history of drug intolerance and drug susceptibility assessment (determined by genotypic resistance testing).
The second comparison is between an intended '3-month ARDFP' and 'No ARDFP'.
The ARDFP is intended to be a minimum of 3 months in duration. There may be some patients who remain stable and feel well during the ARDFP and who may wish to extend the ARDFP duration. This will be permitted at the clinician's discretion as long as the 2 and 6- week safety assessments after the re-initiation of ART are performed.
Genotypic viral resistance testing will be performed on all subjects during the screening process, prior to randomization. The results will be recorded on a data form provided by the VIRCO laboratory and forwarded to the treating physician or center at the time of treatment initiation. A copy of this form will also be sent to the coordinating center in each country.
A central randomization list will be prepared for the whole trial with variable block size used for individual large sites or groups of smaller sites within each country and stratified by screening CD4 cell count. However, each country will be responsible for its own randomization processes. Patients will be randomized by telephone randomization or fax through the Canadian HIV Trials Network Data Center, the VA Cooperative Studies Program Coordinating Center or the UK-MRC Clinical Trials Unit.
Patients will be randomized to either mega-ART or standard-ART and will start treatment either immediately or following a 3-month ARDFP. For patients randomized to ARDFP, knowledge of their eventual drug allocation could influence the initial phase of the study. Consequently, for this group of patients, assignment to standard-or mega-ART will not be communicated by the data center until the time of drug reintroduction.
The variety of drugs, which might be used in the trial, is too large to allow for the use of a blinded treatment methodology. Objective end-points, especially suitable for open studies, will be used to measure the efficacy of outcomes. The primary end-point is the incidence of a new or recurrent AIDS-defining illness or death, and will be adjudicated by a blinded Endpoint Review Committee.
This is a clinical management trial comparing strategies involving ARDFPs and numerous drugs; as such, this trial is not intended to test the efficacy of individual drugs (none of which alone are likely to have considerable effects). In this context, the overriding need is for individualization rather than consistency of re-treatment regimens. Factors that influence and guide regimen decisions include treatment history, drug intolerance, available choices and their expected toxicities, and measures of HIV susceptibility. Although standardization of the process of selecting treatment regimens would be ideal across the trial, this will not be possible even though each clinician will likely be consistent in their selection of Mega or Standard-ART regimens. Changes in treatment strategy will not be recommended prior to patients reaching a trial end-point.
This is an open randomized study using drugs that are available in routine care or through compassionate use programs or other open-label access to new HIV drugs. There is obviously a potential risk of poor compliance to the allocated strategy. For instance, patients allocated to ARDFP may resume drug therapy earlier than 3 months because of concern over disease progression and the 6-week visit for ARDFP patients could allow earlier reinitiation of ART in patients who experience a precipitous decline in CD4 counts. Similarly, patients on mega-ART may stop treatment because of presumed drug toxicity. Patient compliance and adherence to anti-HIV drug therapy is a universal problem in therapy and trials. It may be relatively more operative in this trial than in general, since the study population will be selected based on treatment failure. This is often due to past nonadherence to therapies, whether through non-compliance, or drug intolerance, or other factors that may still be present in individual cases. The best way to deal with adherence in good clinical practice is by planning and counseling, and either identifying prompts or eliminating obstacles to adherence. The same standard of good clinical practice will be followed for all study volunteers, regardless of allocation. The DSMB will closely monitor adherence to allocated regimens by treatment groups and advise the Steering Committee as necessary. Since there is no consensus (or guidelines) about the best treatment strategies for such patients, clinicians should not have great difficulty supporting patients to remain on their allocated treatment strategy.

VIII. Patient Population
The intent of the trial is to include patients, male and female, who have advanced HIV disease and in whom regimens that have included all three classes of drugs have failed.
Sixty-five enrolling sites (25 VA, 20 UK and 20 Canada) will participate. The overall anticipated accrual rate is 9-10 patients per month.

a. Inclusion Criteria:
Patients will be eligible for enrollment in OPTIMA if they have:

IX. Trial Procedures and Assessments
a. Screening The following evaluations/procedures will be performed at the time of screening.
Results from tests done within 30 days will be acceptable for screening. The site coordinator will review eligibility and complete the baseline evaluations and procedures. Then he/she will telephone or fax the coordinating center to obtain the randomization allocation. The assignment will then be confirmed by letter or fax.

d. Follow-up evaluations
After randomization (Time 0) all patients will have visits at weeks 6, 12, 24 and then every 12 weeks.
At 2 and 6 weeks after beginning ART in the study all patients will have 'safety' visits; the timing (since randomization) of these visits may vary from patient to patient depending on the length of the ARDFP period which is intended to be 3 months. The treating physician, in consultation with the patient, may consider prolonging or shortening the ARDFP as deemed clinically appropriate. All other visits throughout follow-up will be based on the time of randomization.
The following will be recorded on each visit or as specified in the table of trial assessments ( All patients will be seen by their primary physician and by the site coordinator at the pre-specified intervals. There will be a window for completing follow-up visits: +/-1 week for WEEK 2 and 14 visits; +/-2 weeks for the WEEK 6, 12 and 18 visits; and +/-4 weeks for all other visits. In the event of inability to complete a scheduled clinic visit due to hospitalization, intercurrent illness, or logistical constraints, the patient or family will be contacted by phone. When necessary, the patient's primary care physician will be contacted Vital status and date of death of patients will be verified by regular follow-up or by using national HIV databases. AIDS-defining events will be assessed clinically and adjudicated using well-established evaluation guidelines (Appendix 2) by an independent Endpoints Review Committee that will be blinded to the patients' treatment allocation.
Site coordinators will follow each randomized patient at their study center; missed follow-up visits will be noted.

e. Assessment of Compliance
Responses to standardized questions will be reported at each follow-up visit to assess patient compliance with ART.

f. Reporting of Death, HIV Disease Progression, Adverse Events
If the patient has died, experienced any new or recurrent AIDS-defining illness or experienced an adverse event (serious, Grade 3 or 4, or one leading to a modification of ART; see Appendix 3) since the last visit, the investigator/study coordinator will complete specific case report forms.  This is a clinical management trial comparing strategies involving ARDFPs and drug combinations, rather than testing the efficacy of individual drugs and therefore the overriding need is for individualization rather than consistency of treatment regimens. Baseline genotypic testing is incorporated in the study and individualization of the regimen is left to the treating physician.

Serious Adverse Events (SAE):
The rapid process of new drug development challenges HIV treatment studies.
However, it is unlikely in the face of multi-class drug resistance and cross-resistance within drug classes, that new drugs within existing classes (NRTI, PI, and NNRTI) will offer enough anti-retroviral activity to overcome existing virological and immunological failure.
Further, potential new classes of drugs include integrase inhibitors, which may be three years away from broad availability are not likely to be any more effective in rescue than currently available drugs that have failed. Likewise, T20, an injectable oligopeptide fusion inhibitor can produce over a ten-fold reduction in plasma viremia as monotherapy, 12 but resistance rapidly emerges. In addition, immune modulators and therapeutic vaccines may emerge as new treatment options. Nevertheless, if new classes of drugs become available during the course of this study, likely through compassionate access, their benefit will critically depend on the efficacy of concurrent treatment with existing agents.
In patients with multiple drug resistance, the optimum strategy determined by this study is likely to be the best strategy to be used with the new drug. In order to enhance the validity of this trial, the inclusion of new agents into the standard and mega-ART regimens as the agents become available will be encouraged.

b. Opportunistic Infections (OI) Prophylaxis
All patients should receive OI prophylaxis according to established guidelines (Appendix 6) . Declines in CD4 counts may appear at any time in patients in the trial, but can be anticipated in patients on an ARDFP, and therefore, appropriate action should be taken.

c. Treatment Failure/ Guidelines
For the purposes of this study, drug failure will be defined in accordance with current guidelines.

d. Maintenance of Allocated Treatment Strategy and Modification
If possible, investigators are urged to keep patients on their assigned strategy in the absence of a new or recurrent AIDS-defining event, or medical contraindications to their assigned therapy. However, if warranted, the study medications should be continued or discontinued according to established guidelines. Since patients are assigned to a strategy, individual drugs may be withdrawn and/or substituted without necessarily changing the allocated strategy.

e. Patient Follow-up
The aim is to follow every patient until death or study closure. However, investigators may withdraw patients from their assigned treatment strategy in the event of intercurrent illness, adverse events, withdrawal of consent, and other compelling clinical conditions. Unnecessary withdrawal of patients from their assigned treatment strategy should be avoided.
Whenever possible patients should continue to be followed according to the study procedures even if they have withdrawn from allocated strategies.

XI Biostatistical Considerations
The sample size estimated for this study is based upon a number of assumptions about the expected rate of outcomes [on the 'standard-ART' arm for comparison with 'mega-ART' arm, and 'No anti-retroviral drug-free period' (No ARDFP) arm for comparison with 'anti-retroviral drug-free period' (ARDFP) arm], crossovers between groups, and losses to follow-up. Computations were made for a number of different scenarios detailed in Appendix 7 (REVISE Appendix). Ultimately, the sample size was set to detect a difference between standard-ART and mega-ART of 63.2% versus 52.9% in the proportion of patients who, by the end of the study, would have had an event with a power of 80% and a two-tailed alpha of 0.05. The expected proportions factor in the expected crossover and loss to follow-up rates. The study has the same power to detect the same difference between No ARDFP and ARDFP arms.
The primary comparisons in the OPTIMA Trial are between Standard-ART and Mega-ART, and ARDFP and No ARDFP. The analysis plan for the trial is described in In each country, data will be sent to the coordinating center, which will then perform data checks and generate data edits. Verified data from each country will be sent to the Canadian coordinating center on a regular basis so that the data from all three countries can be merged. Analysis tables will be generated both for trial monitoring purposes by the Trial

Steering and Trial Management Committees, as well as for the Data Safety Monitoring
Board (see Appendix 9).

XII. Study Administration
The management of the trial will be shared between the three countries. Each country will be responsible for the day-to-day management of the trial within its borders.
This includes handling randomization, orienting and monitoring clinical sites, ensuring quality and timeliness of data and dealing with site-specific issues. Within each country, the trial will be managed by the Canadian HIV Trials Network, the MRC-UK Clinical Trials Unit and the VA Cooperative Studies Program Coordinating Center (CSPCC-West Haven).

a. Organizational Structure
The organizational structure and the functions of the various committees and groups of the OPTIMA Trial are summarized below. OPTIMA committees and groups will meet prior to the beginning of patient intake and as required. i.

Trial Steering Committee
The Trial Steering Committee (TSC) will provide general oversight of the trial conduct, accrual, retention, quality, data systems and management in accordance with the protocol and in light of reports or recommendations from any of the trial committees. The trial committees are: the Data Safety and Monitoring Board, the Endpoint Review Committee and the Trial Management Committee.
At regular intervals the TSC will also review and consider relevant information or developments in the field of HIV therapy and in legal, statutory, or regulatory requirements.
The TSC will maintain regular contact with the 3 national trial organizations and will provide a forum for the exchange of information about the trial. The TSC will meet twice a year, with additional meetings and teleconferences arranged when necessary. All decisions will be made by a simple majority vote. In the event of a no majority vote, the Chairperson for the meeting will have a deciding vote.
Membership of the TSC: ii.

Trial Management Committee
A Trial Management Committee (TMC) will oversee the day-to-day management of the trial, resolving questions about eligibility, enrollment, randomization, regimen determination, length of drug free periods, grading and disposition of toxicities and adverse events, and determination of endpoint questions that need resolution by the ERC.
Leadership will consist of the 5 Clinical PIs, chairing in rotation. Each subgroup will be represented on the TMC with at least one member from each country. The TMC will operate by consensus, not by majority vote.

Statistics and Data Management Subgroup (OPTIMA Biostatisticians and Information Technology Expert)
The Statistics and Data Management Subgroup will advise on protocol development, determine database structure, and develop data management and data capture procedures. The Canadian statistics/data management team will perform the 3country data merging functions.

Investigators; and an ad-hoc member)
The Sub studies Subgroup will determine the mechanisms for sub study concepts to be reviewed and approved, and will make recommendations to the TSC for prioritization of sub studies.

Public Relations and Publications Subgroup (Representatives of each of the other subgroups plus a co-chair)
The Public Relations and Publications Subgroup will manage and coordinate publicity, presentations and preparation of reports. All reports, abstracts, or other public disclosures on the trial prepared by clinical investigators, principal investigators, or any of the 3 national trial organizations will be assigned or reviewed by the PRP group, which will report to the TSC.

Endpoint Review Committee
The Endpoint Review Committee (ERC) will determine the validity of potential endpoints that fail to meet standard criteria, as defined by the protocol. The TMC will forward such potential endpoints to the ERC for review. Two members, who are not from the country where the potential endpoint occurred, will review and make the determination on such an endpoint. The DSMB members will be completely independent of the trial and will advise the TSC if any interim results or information from other sources indicate that the trial should be modified or discontinued.
The Trial Biostatisticians will provide information and statistical reports for meetings of the DSMB and will attend meetings to present reports and answer any

iv. Secretariat
A Secretariat will be appointed which will reside outside of the various committee structures. The Secretariat will relate to all of the committees and will serve as the main coordinator of trial committee activities, including planning of meetings and teleconferences, development of meeting and teleconference agenda and minutes, and maintenance of committee memberships. The functions of the Secretariat may be shared between 2 or 3 countries, as needed.
The UK will perform Secretariat functions for the DSMB and Trial Steering Committees; the US will perform Secretariat functions for the Trial Management Committee and Endpoints Review Committee.
Note: For the VA only. The Human Rights Committee at the Coordinating Center will review the study annually to ensure proper protection of patients' rights and safety. A report about the progress of the study and ethical issues relevant to these bodies will be presented at regular intervals. In the interim, these bodies may be asked to convene if there is any serious adverse event requiring its attention. Members of these bodies could attend each meeting of the DSMB and will site-visit enrolling sites annually to determine that the patients' rights and safety are being properly protected.

b. Monitoring Patient Intake/Probation, Termination of Participating Sites
The TMC and TSC will monitor the intake rate and operational aspects of this study.
Participating medical centers will continue to recruit in the study only if adequate patient intake is maintained. Actions leading to discontinuation of patient enrollment at a center will only be taken by the TSC with the concurrence of the DSMB and the leadership of the three participating agencies.
If recruitment is not proceeding at an appropriate rate, the TSC will scrutinize the reasons. Based on this information this committee may choose, with the approval of the Data and Safety Monitoring Board (and Director/Chief of agency-for VA only), to stop further patient enrollment at existing centers, add additional centers, or make modifications to the inclusion/exclusion criteria. In the VA, participating sites (except for small centers where accrual rate is anticipated to be slower) that do not enroll at least 1 patient in 3 months will be given an opportunity to improve within a reasonable period. If there is no improvement in accrual after the probation period, the site may be subject to reduced funding or possible termination as a study site.

c. Monitoring Clinical Site Performance
Strict adherence to the protocol will be expected of every participating center and monitored by the TMC, the TSC and the DSMB. Documentation of protocol breaches will be required and any site with repeated major protocol violations may be considered for termination. If a participating site investigator feels that adherence to the protocol will in any way be detrimental to a particular subject's health or well-being, the interest of the patient must take precedence. By agreeing to participate in this study, the clinical site delegates responsibility for global monitoring of the ongoing study to the TSC, the DSMB and the Coordinating Center (and the Human Rights Committee for VA). However, local committees and subcommittees involved in research at the enrolling medical centers may require the participating investigators to submit annual reports concerning the status of the study at the medical center for local monitoring purposes.

d. Site Visits
Site visits will not be performed on a pre-determined and regular basis. However, guidelines for site monitoring exist for each country and are described in the Good Clinical Practice section of the Trial Operations Manual.

e. Data Safety Monitoring
The OPTIMA Trial will be overseen by an independent DSMB, which in addition to seeing that the trial is being conducted in the manner proposed, will be responsible for overseeing the ethical conduct of the study.
At each of its meetings during the study period, the DSMB will review the recruitment rate and assess the difference between the actual and the projected rate, as well as the impact of these assessments on overall trial size. If accrual is inadequate, inclusion and exclusion criteria may be scrutinized and revised if appropriate. An assessment of whether the trial should be continued will be made followed by a recommendation, as appropriate. Serious adverse events will be reported to the DSMB.
Since this is a 2X2 factorial design, there is a possibility that one or both of the treatment comparisons could be stopped early for efficacy and/or safety reasons. If one treatment comparison is stopped early, continuation or termination of the other comparison would be a consideration.
At each meeting of the DSMB, the accumulating information on endpoints will be reviewed. At its first meeting, the DSMB will decide on the method they will use to adjust for the interim look at the primary endpoints (after all the patients have been followed for six months).

f. Endpoint Review
The Endpoint Review Committee (ERC) will meet on a regular basis to adjudicate endpoints, blinded to treatment allocation. A Table of Presumptive and Definitive Criteria for AIDS events provided in Appendix 2 will assist in adjudication of endpoints.
The ERC will determine the validity of potential endpoints that fail to meet standard

XIII. Publications a. Publication Policy
The TMC will develop guidelines for the preparation of papers (abstracts) for presentation at national and international meetings, as well as the preparation of manuscripts for peer-reviewed publication. Any publication or presentation during the active phase of the study must have prior approval of the TMC and TSC.
All publications are to be approved by the TMC and TSC before submission for publication. The TMC will develop guidelines to protect patient confidentiality, to prevent unwarranted release of study information, and to prevent conflict of interest. The TMC and TSC will resolve problems of authorship and maintain the quality of publications. All publications will acknowledge appropriate funding sources. The presentation or publication of any data collected by the participating investigators on patients entered into the OPTIMA study is under the direct control of the study's TMC and TSC. This is true whether the publication or presentation is concerned with the results of principal undertaking or is associated with the study in some other way.
Although individual participating investigators will not have any inherent right to perform analyses or interpretations or to make public presentations or seek publication of any of the data other than under the auspices of and with the approval of the TMC and TSC, they will be encouraged to develop substudies or propose analyses subject to the approval by the TSC and TMC.
Outcome data by treatment group will not be revealed to the participating investigators until the data collection phase of the study has been completed. This policy safeguards against possible biases affecting the data collection. The DSMB (and the Human Rights Committee in the VA) will be monitoring the outcome results and may recommend that the study be stopped for safety reasons or if a definitive answer is reached earlier than the scheduled end of the study.

XIV. SUB-STUDIES
Ancillary Studies (or "sub-studies") may divert resources from or interfere with the conduct of the primary study. However, since this study may provide a unique opportunity to examine other questions of scientific interest, investigators will be encouraged to undertake substudies and sub-analyses with the approval of the TMC and TSC.
The following guidelines are proposed for reviewing studies related to the OPTIMA Trial.
1. Any study specifically involving study patients will be considered an ancillary study, even if it is limited to one site. Ancillary studies will normally involve collection of additional data. Many -ancillary‖ analyses of currently collected data are already planned as part of the primary study.

Requests to perform ancillary studies will be accepted only from Participating
Investigators at any of the designated study sites. Requests should be submitted initially as a letter of intent addressed to the TMC. The letter should specify goal and general design of the proposed research, the proposed number of subjects and study sites, and an estimate of the funding, if any, that will be required.
3. Letters may be submitted at any time, but the additional workload from a proposed substudy should be carefully considered in order to ensure that recruitment in the primary study is not hindered.
4. Letters of intent will be reviewed by the TMC. Although the scientific merit of the study will be considered, the primary purpose of this initial review is to establish that the proposal in no way conflicts with the conduct of the primary study. Recruitment success and the overall performance of the proponent's site (and any proposed collaborating sites) will be one of the factors considered in this review. Sites that are struggling to meet recruitment goals for the primary study may be considered poor candidates for ancillary studies.
5. If the proposal is acceptable, the proponents will be asked to submit a formal study protocol (including a human consent form) and a budget. This proposal should be 4-6 pages in length, including references, and be in a format appropriate for submission to a research committee.
6. The protocol will be reviewed by the TMC and possibly by one or two additional reviewers with expertise in the area of interest. The purpose of this review is to determine the scientific merit of the study and to determine if the proposed ancillary study conflicts with the goals and/or conduct of the primary study. Depending on the nature of the ancillary study, it may require approval by the DSMB, (and HRC in the VA), local IRB and the three funding agencies. Based on the required reviews, proposals will either be approved or disapproved.
7. Any analysis which needs unblinding of treatment groups can ONLY be undertaken and published before results of the trial are reported if the TMC, TSC and DSMB agree.
8. Locating funding for approved studies is the responsibility of the proponents.
Any publications (including abstracts) resulting from ancillary studies must conform to publication policies of the three involved agencies, as specifically outlined for OPTIMA.
This includes statistical review by the three Coordinating Center and adherence to authorship policies. The proponent of the ancillary study will normally serve as principal author of any resulting manuscripts.

XVI. Health Economics Component a. Health Economics Evaluation
The health economics analysis will evaluate the cost effectiveness of the following strategies for managing the care of patients for whom previous standard-ART therapy has failed: Mega-ART (compared to standard-ART) An antiretroviral drug free period (ARDFP) followed by standard ART or Mega-ART (compared to No ARDFP) These strategies will be evaluated using standard methods for cost-effectiveness analysis. The strategies must be evaluated relative to each other, to rule out any therapy that is dominated by another (e.g., strategies that are more costly and less effective than available alternatives). For strategies that are not dominated, an incremental cost-effectiveness ratio will be determined. Stochastic analysis will be undertaken using cost-effectiveness acceptability curves to express the uncertainty around cost-effectiveness based on mean costs and outcomes. 1,2,3

b. Methods
The cost-effectiveness of the interventions will be estimated using the patient's life span as the time horizon for the analysis. Standard methods will be used. 4,5 The societal perspective will be adopted so that all costs, including patient-incurred costs, as well as the direct costs of health care costs will be considered. The cost of each resource will be estimated using unit costs and practices appropriate to each country.
Outcomes will be expressed in Quality-Adjusted Life Years (QALYs). In the primary analysis, quality adjustments will be based on periodic measures of utilities using the Health Utilities Index (HUI) and the EQ-5D, standardized multi-attribute utility instruments that will be used in all countries. We will also assess utilities of a random sample of U.S. participants will using a computer-based interview that uses time-tradeoff and standard gamble approaches to measuring utility. Decision analytic modeling will extrapolate from trial-based results to project lifetime health care cost and QALYs for each treatment group.
Sites in all three countries will gather a common set of utilization and outcomes data.
A single model will be developed to find trial participants' lifetime cost and outcomes.
These common data sources will be used to create a separate economic analysis for each country. Each analysis will reflect the country's guidelines for accepted practices for performing for cost-effectiveness studies, sources of costs for pharmaceuticals and health services, factors affecting lifetime survival, and discount rate. In addition to country specific analysis, we will employ statistical methods to investigate the extent to which costeffectiveness differs between countries 6 .
Costs will be expressed in current year dollars. In the U.S., costs will be adjusted by the Consumer Price Index or other appropriate indexes for all urban consumers. In the U.K., the Health Service Cost Index, published by the NHS Executive, Leeds, U.K. will be used to adjust for costs of health care services and a range of other indices, which would cover other items (e.g., wages). In Canada, appropriate indexes will be used to adjust for costs of health care and other services. Cost and outcomes will be discounted. For the country-specific study, each country will use its own standard discount rate. For the joint study, we will present cost-effectiveness results using a range of alternative rates.

c. Sample Size Considerations:
The proposed pharmacoeconomic component of this trial can easily accommodate changes in sample size and duration of follow-up . To the extent that estimates of costeffectiveness will be based on modeling the event rates and disease progression observed during the trial to extrapolate survival over the patients' life time, increased follow-up time will provide a longer observation period upon which to model future events. With fewer patients in the study, the precision of the cost and outcome measurements will be affected.
However, the trial is not powered according to cost and quality of life measures but rather according to clinical events. The uncertainty associated with cost-effectiveness results will be handled using methodology that is different from what is typically adapted for efficacy estimates emanating from randomized clinical trials. Specifically, we will adopt a Bayesian approach by reporting the probability that one treatment is the most cost-effective given the data rather than using classical hypothesis tests. We will implement this by estimating costeffectiveness acceptability curves that show the probability that each management option is the most cost-effective conditional on the value the decision-maker places on additional health outcomes. We feel that this methodology will adequately handle a reduction in sample size by showing the implications of the reduced precision in cost and outcome estimates for the decision uncertainty faced by decision makers.

d. Estimating Indirect Health-Care costs
Cost-effectiveness analysis that adopts the perspective of society requires consideration of costs borne by patients. These costs include the cost of securing medical care, such as travel expense and lost time from work or leisure activities.
The baseline survey form will obtain information on the distance the participant ordinarily travels to receive care. Travel expenses will be estimated as the product of distance traveled and the per-mile travel expense for private automobiles. In the U.S., we will use the rate allowed as a tax-deduction by the U.S. Internal Revenue Service. In the U.K., we will use the rate published by the U.K. Automobile Association. In Canada, the standard cost per kilometer will be used for estimating patients' travel expenses.
We will also estimate the time spent in obtaining health care, using a count of visits and days of hospital stay. This time will be valued using mean national wage rates adjusted by age and gender. For retirees, unemployed, and disabled, we will select an appropriate measure of the value of time to reflect their average opportunity cost.
Information about each patient's caregiver, such as their usual activities, will be collected at the baseline interview. The number of days caregivers have been taken away from their usual occupation due to patients' illness between two follow-up visits will be collected. The average market payment rate for home health aides performing chore services will be used as the unit cost for caregiver.

e. Estimating Direct Health-Care Costs
Key health care resource measures will be obtained-hospital care, medication, outpatient visits, and long-term care. This set of core resource use data will be collected in all countries. Additional database available in each country may be used for validation. For each hospital stay, we will determine the length of stay and the number of days of stay spent in intensive care. For each medication taken by patients, we will determine the medication name, the total daily dose, and the number of days it was taken. For outpatient visits, we will identify the type of practitioner visited, and whether the visit involved a therapeutic procedure. For long-term and rehabilitation care in an institution, we will determine the length of stay. Table B provides an overview of the three-country costing methods.

U.S. health care cost methods
We have designed a cost-finding method that takes advantage of the available data and complies with guidelines for cost-effectiveness research. The U.S. Public Health Task Force panel on cost-effectiveness analysis recommends that both micro-cost and gross-cost methods be used to estimate health care costs. 7 The more labor-intensive micro-cost method is reserved for determining the cost of the care that is central to the analysis; gross costing is used when a less precise estimate can be used. This same hybrid method has been recommended to conduct cost-effectiveness research in the VA health care system. 8 We will use two independent methods of finding VA health care costs. The primary set of VA cost estimates will be derived using national unit costs. Micro-cost methods will be used to find the cost of pharmaceuticals, and a gross-cost method to find the cost of other health care utilization. The micro-cost method will use data on every prescription received by trial participants obtained from the Pharmacy Benefits Management database. The cost of other care will be found by obtaining detailed utilization data from the centralized VA databases and estimating unit costs using gross-costing methods based on department level costs in the Cost Distribution Report (CDR), as described below.
The second set of estimates will use site-specific unit costs. The cost of VA health services will be extracted from the Decision Support System (DSS). DSS is a detailed cost allocation system that has been implemented throughout VA. Although DSS is a sophisticated tool that taps VA cost and utilization databases, it has not been previously used by researchers, nor has its cost estimates been independently validated. Because it is newly implemented, its accuracy is uncertain. 8 For this reason, we will use DSS as a secondary source of cost and utilization data.

Micro-Costing-Pharmacy Cost.
We will obtain information on the use of prescription drugs from the VA National Pharmacy Benefits Management data. This is a new data source that now covers all VA health care systems and medical centers. We will request all data for trial participants from the National Pharmacy Benefits Management Program, in Hines, IL, including the name of the drug, dose, count of pills, the cost to the VA, and the NDC code. We will obtain the cost of each prescription from the national VA pharmacy database, and add an appropriate dispensing fee that reflects labor and supply costs for the source of the prescription, either the VA study sites or the VA centralized mail order pharmacy. We will develop an alternate cost of prescriptions based on the average wholesale price in the Medical Economics Red Book. This will allow us to prepare an estimate that reflects the societal cost for prescription drugs, as VA pays less for drugs than other U.S. health care payers.
Data on VA health care utilization. We will use centralized databases for information on the health care used by trial participants. From the database of VA hospital stays, we will obtain the Patient Treatment File (PTF), the date of discharge, days of stay in each ward (bed section), including the number of days in intensive care, and the ICD-9 diagnoses (up to 10 codes) assigned to the stay. Then we will obtain the outpatient care file (OPC), the date of the visit, the location of care (stop code), and CPT codes assigned to the visit, and the type of provider from the database of VA outpatient visits.

Decision Support System (DSS).
We will obtain information on the cost of all VA health care received by each trial participant from the national extract files from the Decision Support System (DSS). This system includes the cost of all inpatient and outpatient cares, including the cost of outpatient pharmacy. The national extract files do not provide detailed information on the specific cost of each type of pharmacy utilization; for this reason, we will seek permission from each site to extract data on the outpatient prescriptions filled for each participant, as well as its cost.
Cost of non-VA care. We will determine the cost of care that participants incur outside of the VA using provider and payer databases, and participant interviews. Followup interviews will include questions about care received from non-VA providers.
Participants will be asked to identify all non-VA hospital stays, long-term care stays, and the hospital where they received care. Participants will be asked for the number of outpatient visits they have made since the last follow-up interview.
We will extract the cost of care provided under contract to VA from the Fee Basis files. Individuals who are 65 years old or disabled are eligible for health care from Medicare. We will estimate the cost of Medicare services used by participants using claims data from the Health Care Financing Administration. We hope to take advantage of the veterans' Medicare finder file being created by the Health Care Financing Administration (HCFA) for the VA Information Resource Center (VIREC). This file will provide the HCFA identification number of study subjects and will allow us to identify participants' Medicare claims. If the HCFA-VIREC finder file is not available, we will obtained Medicare identifiers and Medicare claims data directly by providing HCFA with each subject's name, birth date, and social security number so that the agency can provide us with each subjects' HCFA identification number. We will then use this identification number to extract claims data from the Medicare Provider Analysis and Review (MEDPAR) and Medicare Part B Provider files. These files provide information about the services utilized, charges, and reimbursement. Hospital charges will be adjusted by the cost-to-charge ratio found in the hospital's Medicare cost report.
When Medicare cost data are not available, we will estimate the cost of non-VA health care using the mean cost of stays in which we have cost data. The cost of ambulatory care will be estimated by multiplying the number of units of service by the mean cost of comparable VA care. This estimate assumes that VA and non-VA hospitals have the same costs. A recent review found no evidence of difference between the cost of VA and non-VA care. 11 We will also look at data available from the HIV Cost and Services Utilization Study (HCSUS) to estimate costs for hospital health care services. 12,13

Canadian Health Care Cost Methods
A resource utilization profile will be developed for each patient based on the minimal amount of prospective data collected from the trial. This will include hospitalization data, all drug therapy related costs, and all clinic visit and other physician visitation costs. Cost borne directly by the patient or their caregivers will also be estimated.  15,16,17 Nonetheless, sensitivity analysis using unit costs from Calgary and Toronto hospital unit costs will also be performed.
Non Health Care System costs. These costs will be estimated from a variety of sources. Home care costs will be according to the Victorian Order of Nurses rate for home care nursing. The costs of travel will be according to cost per kilometer used in other recent studies. Statistics Canada average earning schedule will be used to calculate productivity losses for time-off work. Alternatively, the amount welfare payments received when attributable to HIV/AIDS will be used as proxy or patient reports of lost earnings (when available trial case report form). Finally, patient reports of the cost specialized equipment, special nutritional supplements, vitamins and other costs will be incorporated according to the market prices of these items.

United Kingdom Health Care Cost Methods
This section provides details of how the resource use data collected in OPTIMA will be valued in monetary terms for the U.K. analysis. Given that changes are likely to take place over the next few years in sources of cost data, the methods described here may change depending on the point of analysis undertaken. For some unit costs required for the analysis, national statistics are available which are usually updated annually. As shown in However, it is recognized that some unit costs will be unavailable from these sources. For these resources, a specially-designed sub-study will be organized in the UK to estimate relevant unit costs in a sample of trial centers. For this exercise, a common costing protocol will be designed that will ensure, as far as possible, both that costs are estimated consistently in the trial centers in the sub-study and that these are consistent with costing methods used for this study in the US and Canada. In addition to providing a quantified expression of how patients or the public value health outcomes, certain preference-based measures have an axiomatic basis that makes them useful in formal decision analyses and cost-effectiveness analyses.
In our primary analysis, we will estimate quality weights for QALYs using two preference-based health status measures: the HUI-III and EQ-5D. The HUI is a generic multi-attribute utility instrument. 26,27 Patients rate functional status on 6 subscales, which are then combined into a summary utility score based on an empirically derived mapping algorithm based on public preferences. Similarly, the EQ-5D is also a generic multidimensional measure of health-related quality of life. 28 It contains two parts: a health status profile and a visual analog scale to rate global health-related quality of life. Patients' health status is measured in five dimensions, defining 245 health states. A single score is available for each health state using public preferences. Using the self-rated visual analog scale scores, however, the EQ-5D can also give an indication of the respondent's personal value for that health state.
Thus, both HUI and EQ-5D are measures of quality weights for QALYs: the patient supplies information about their health state at intervals and these are ‗valued' on a 0 (equivalent to death) and 1 (equivalent to good health) based on the preferences of a representative sample of the public. The advantage of HUI and EQ-5D is that they are easy and inexpensive to administer. Because the EQ-5D has limited use in Canada and U.S. and HUI is not commonly used in the U.K., we will use both methods and report overall results using both instruments.
As a secondary analysis in a sample of US patients, we will also assess patient preferences for health states using the time-tradeoff and standard gamble which are standard approaches for global patient utilities for health states. To perform this assessment, we will use a computer-based instrument, such as U-titer, to assess patient utilities for health states.
These assessments will be conducted at a sample of U.S. sites. This assessment differs from that of the HUI and EQ-5D in two important respects. First, these assessments are global measures of the utility of the health state, rather than multidimensional assessments.
Second, we will assess patients' preferences with this instrument; the HUI and EQ-5D rely on preferences assessed from members of the public. Although the questions have been used and validated previously, we will pilot test and refine the questions if needed.
The results from these analyses will be used in the modeling part of the study, as appropriate. Changes in health-related quality of life over the time of the trial will be compared across the study arms using standard analysis approaches. A set of utility-type weights for each patient in each time period will be calculated from the scores on the HUI or the EQ-5D instrument, and the computer-based instrument.

d. Analysis
Cost and health care utilization. We will examine whether the intervention has a significant effect on the annual health care costs incurred by participants. Average utilization patterns for each type of resource will be described for each country. Overall utilization patterns for each treatment arm in the trial will also be compared using multivariate analysis controlling for a range of covariates including country and baseline value for surrogate markers, such as CD4 count, viral load, HIV disease stage, and physiological status expressed as a Justice Score summary value. 29 Using similar methods, we will analyze total costs, including costs incurred by patients; we will also undertake separate analyses adopting the perspective of the health care payer alone.
The confidence region surrounding the cost-effectiveness ratio will be estimated using appropriate statistical methods, including bootstrap and Monte-Carlo analyses. 30 This stochastic analysis will generate cost-effectiveness acceptability curves. These curves represent a plot of the probability of one management strategy being more cost-effective than the other two at each possible critical cost-effectiveness threshold, that is, under all different assumptions about the threshold cost per QALY which defines cost-effectiveness.

Modeling.
A Markov model or other state transition simulation approaches will be used for this study. The model is a mathematical simulation that evaluates the costs and benefits of a medical intervention in a hypothetical cohort of patients. General statetransition and Markov models are a standard approach for modeling diseases and interventions that affect health or costs over long time horizons. 31,32,33 These models have been used extensively to assess screening and treatment for HIV. The model tracks a cohort of patients over time as the patients move from one health state to the next, according to transition probabilities that define the likelihood of moving from one health state to another during a given period. We will use a stochastic model, perform analyses from the societal perspective, and will discount costs and health outcomes according to current guidelines. It may be useful to perform analyses that are tailored to the costs and utilization patterns of each country, should we find that they differ.
Our decision model will compare the costs and benefits of the alternative strategies Recognizing the potential difference in preferences across three countries as well as the differential sensitivity of specific instruments given the circumstances of the trial, we will use three well-validated instruments. The extent to which the cost-effectiveness of the treatment strategies differs according to the alternative instruments will be assessed. We will compare quality of life measures and examine their correlation between the two instruments in all three countries. For the US patients, we will also compare the assessments performed with a computer-based interview with measures obtained from HUI and EQ-5D.
If differences emerge, we will explicitly report separate results with alternative instruments, together with a discussion regarding their consistency with other outcomes (mainly clinical) collected in the study.

e. Data collection
We will collect economics data within each country to estimate both costs of each of the treatment strategies, both from a health service and a societal perspective. Due to the fact that the resource utilization data routinely available in individual countries differ, the study requires trial centres in all three countries to gather a common set of utilization and outcome data. As noted in Table B, all three countries will use the trial-case report form and baseline survey form to collect the utilization data. The common set utilization data include: hospitalizations, outpatient visits and services (including tests), prescription drug use, overthe-counter medication use, patient transportation time, patient productivity losses, and formal and informal caregiver time. The countries will also collect the costs for each of these elements of utilization. The specific utilization and cost data to be collected within each country, and the sources of this data are summarized in Table B. Our approach aims to ensure that each country has comparable data.
As noted, quality of life is measured by three utility instruments. We will use the computer-based program to assess utilities in a randomly selected sample of 600 U.S.
patients (Table C). X indicates the tool that will be used f.

Time frame
Economics data collection is integrated into the clinical data collection and follows the same schedule summarized in Table A.

g. Data management
Economics data will be managed in the same way as the clinic data. We adopt the same procedures and policies developed in the clinical study.

h. Quality control procedures
Economics study adopts the same strategies for quality control.

i. Organization and Administration
The economics team and study coordination committee The

j. Study administration
The OPTIMA economics project will be directed by a committee that will be cochaired by three individuals, one from each participating country. The committee will meet periodically by conference call, and at least annually in person. It will review study data for accuracy and comparability between sites and countries. It will invite an outside consultant to review study methods, data, and publications. The committee will also review and approve all publications arising from study data.

k. Sub-Studies Policy
The economics study will adopts the sub-studies policy for the OPTIMA project in the Sub-Studies Policy section.

l. Publications
Publication from the economics study follows the same policy for the OPTIMA project.

Sample Informed Consent FULL TITLE A Tri-National (Canada, UK, USA) Randomized Controlled Trial To Determine The Optimal Management Of Patients With HIV Infection For Whom First And Second-Line Highly Active Antiretroviral Therapy Has Failed.
TNT-1: OPTIMA

INTRODUCTION
You are being invited to take part in a research study at the VA(Facility Name). Before you decide to participate, it is important for you to understand why the research is being done and what it will involve. Please take time to read the following information carefully and discuss it with friends, relatives and your doctor if you wish. Ask us if there is anything that is not clear or if you would like more information. Take time to decide whether or not you wish to participate.

PURPOSE
The purpose of this study is to determine the best way to treat HIV infection after treatment with several highly active combinations of HIV drugs has failed to have an effect on your HIV infection. The study is looking for people whose anti-HIV treatment has stopped working and who have a rising level of HIV in their blood. The study asks two questions. Firstly, does a period of time taking no anti-HIV drugs help the next treatment you take work better than not stopping anti-HIV drugs for a drug-free period? Secondly, will a new (salvage) combination treatment with many drugs lead to a longer improvement in health than a combination containing fewer drugs? The study will also address the safety of these two different strategies.
You are being asked to participate in this study because you are HIV positive, you have taken at least two initially effective combinations of anti-HIV medications, and your current treatment for HIV is no longer working (suppressing your virus). Also, your CD4 (T-Helper) cell count is less than 300 so that you are at risk for opportunistic infections and other HIV related conditions. Approximately 500 persons in the US, Canada, and the United Kingdom will be in the study.

DESCRIPTION OF THE STUDY
This study compares the use of 5 or more anti-HIV drugs (Mega-Antiretroviral Therapy or Mega-ART) to 4 or fewer drugs (Standard-Antiretroviral Therapy or Standard-ART). The study also compares stopping HIV treatment for three months to not stopping treatment. The current standard of care uses 3 drugs from 2 classes of medications. The medicines used within the study are all either licensed drugs or are those available through Expanded Access Programs.
No new agents are being tested in this study although the way they are being used is experimental. It is a trial to define strategy for treatment and to determine which is the best approach to treatment after anti-retroviral treatment failure. You may enroll in studies of other investigational drugs and remain on this study as long as you continue to follow the rules outlined in this form.

PROCEDURES
If you decide you want to be in this study you will be asked to come into the clinic for a visit to determine if you are eligible to participate. At this visit you will be asked to read and sign this consent form. After signing, medical tests will be completed to help determine if you meet the requirements to be in the study. These are called -Screening‖ tests.
The screening tests you will be required to complete will include: HIV antibody test (if this has not been done at your hospital). Blood test to determine HIV RNA blood levels Blood test to measure your immune system (CD4 and CD8) Pregnancy test (for women only) Blood draw for possible viral drug resistance testing at a later date Blood draw for safety tests Any medications you are currently taking will be recorded and you will be asked if you have taken your currently prescribed medications as directed.
For the screening, the amount of blood drawn will be 4 tablespoons (60 mL). All results of the blood tests will remain private. If you meet the study criteria and choose to participate, you will need to return for study enrollment (Baseline/Day 1).
The following will tests/procedures will be performed at the Baseline visit: The same laboratory tests drawn at screening Medical history evaluation, including a review of medications taken to treat HIV A physical examination Questionnaires about how you feel and about your current level of health.
You will then be assigned randomly, like flipping a coin, to one of four treatment groups. The treatment groups are as follows: Anti-retroviral drug free period of 3 months followed by Standard-ART or Anti-retroviral drug free period of 3 months followed by Mega-ART or Immediate start of Standard-ART (4 anti-retroviral drugs or less) or Immediate start of Mega-ART (5 or more anti-retroviral drugs) There is an equal chance that you will be randomized to any one of the four groups.
This is an open study meaning that, once you are on treatment, both you and the study personnel will know what group you are assigned to. But, if you are randomized to have an anti-retroviral drug free period, neither you nor the study personnel will be told whether you are going to later be taking Mega-ART or Standard-ART until after the treatment interruption is completed.
Before you begin treatment in the study, your doctor will be given results of the resistance test taken at screening to help guide the selection of a treatment regimen. When it is time for you to begin treatment, your doctor will recommend what drugs, in either a Mega-ART or Standard-ART regimen, are most likely to benefit you.
Blood will be taken at each visit during the study to conduct safety tests, to see if the treatment option is working by measuring the amount of HIV in the blood (viral load), and to check your immune system (CD4 cells). Blood will also be drawn and stored for future tests such as changes in resistance of the virus to the drugs. An average of 4 tablespoons (60mL) of blood will be drawn at each study visit. You will be given a brief physical examination at each visit if you have noticed any symptoms or side effects. Women will also be given regular pregnancy tests as the drugs could harm an unborn child.
At each visit, you will be asked about any medications you might have taken and any illness that might have occurred since the last visit. Any changes in your medication and/or your health will be recorded and you will be asked direct questions about how you feel and be asked to complete a "quality of life" questionnaire.
You will be followed on the study between 1 year and 4 years, depending upon when you enrolled. If you are randomly assigned to an anti-retroviral drug free period, you will be seen 6 weeks after starting the anti-retroviral drug-free period for blood tests and a symptom directed examination. If your CD4 cell count has fallen by 50% or more since the start of the study, you may decide with your doctor to begin treatment before completing the 3 month anti-retroviral drug free period. You will be seen 2 and 6 weeks after starting Mega-or Standard-ART, then every 3 months afterwards. This should be similar to what you would normally expect in ordinary treatment. Your doctor may wish to see you more frequently for other reasons.
Each regular follow-up visit will take about 1 hour. The first and final visits will take about 1½ hours.
While you are participating in this study you should not use any medications (over-thecounter, prescription, or street drugs) without telling your study doctor. This is important as these drugs may react badly with the drugs you are taking.

DISCOMFORTS AND RISKS
For everyone entering this study there is no course of treatment that we know will halt or slow progression of HIV disease. You may therefore experience complications that result from the progression of HIV during the course of this study. However, your doctor and the study team will make every effort to identify, prevent or treat any complications that may arise.
During the trial your doctor will be following your progress very closely and will explain your treatment options to you if it becomes necessary to consider a change of therapy. There may be specific risks associated with each of the treatment arms to which you may be assigned. These are described below.

RISKS OF CHANGING FROM A FAILING REGIMEN
There may be risks associated with changing from your current regimen even though it is failing. Your current medical regimen may be providing some protection against more rapid progression of your HIV disease. Your study physician is recommending a change in your anti-HIV treatment in the hope that further progression can be slowed or halted, but it is not known if this will occur. All anti-HIV medications have been associated with side-effects and toxicities to varying degrees. Your doctor can review the details of these possible reactions when your treatment regimen is decided upon, but will also answer questions about these medications before you sign this form, if you wish. In addition, there is a risk that you will develop further resistance to the drugs in this study.

RISKS OF AN ANTI-VIRAL DRUG FREE PERIOD
An aim of this study is to determine whether a 3 month interruption in anti-HIV treatment will lead to an improved sense of well-being and to a better long-term response. We don't know the answer to these questions. The amount of HIV in your blood may increase after medications (including those that are failing) are withdrawn. The CD4 count may decrease rapidly when medications (including those that are failing) are withdrawn. Consequently, having a break in treatment may mean that your disease will progress more rapidly. You will have an extra monitoring visit 6 weeks after stopping treatment in order to minimize this risk if you are assigned to this arm.

RISKS OF MEGA-ART
There is no clear information regarding the tolerability of these complex regimens, the potential for toxicities, or their long-term effectiveness. In small studies, up to 40% of persons taking a Mega-ART regimen may achieve an undetectable HIV viral load on this treatment, at least for a while. However, it is anticipated that some persons may experience too many side effects to remain on the drugs. Your doctor can review the details of these possible reactions when you decide on your treatment regimen, and now, if you wish.

RISKS OF STANDARD-ART
The risks associated with a change from one Standard-ART regimen to another are similar to those that may occur with any change in regimen. It is not anticipated that a change in Standard-ART will lead to full suppression of the HIV virus that you are infected with. Your doctor will use the results of HIV resistance testing to select a regimen that should be more effective than the regimen you were on before the study. It may help slow the progression of HIV but this is another question that we don't have the answer to. The virus may develop resistance to drugs included in the regimen that you have not previously taken. You may experience drug-related side effects or toxicities from the medicines in the new regimen. Your doctor can review the details of these possible reactions when your treatment regimen is decided upon, and now, if you wish.

RISKS AND DISCOMFORTS OF STUDY PROCEDURES
Using a needle and syringe to remove blood from the vein is called -a blood draw‖. It may be necessary to insert a needle into your vein more than once if blood does not come out the first time. At each scheduled blood draw, a new needle will be used. During blood draws you may have pain and/or bruising at the place on your arm where blood is taken. Blood clots can form and infections may occur, but these are very rare. If you feel faint during or after a blood draw, you should lie down right away to avoid falling down, then you should notify one of the study staff. Your doctor will be following your progress very closely and will explain your treatment options to you if it becomes necessary.

RISKS DURING PREGNANCY
It is possible that some of the treatments used in HIV disease when given to pregnant women may harm the unborn child. For female subjects, it is therefore very important to not become pregnant during this study. If you are a woman who is able to become pregnant, and choose to have sex during this study, you must agree to use an acceptable method of birth control throughout the study. Your doctor can tell you which brands of contraceptives are acceptable for you to use. Acceptable birth control methods include:

Condom and spermicide Diaphragm and spermicide
Even if you use an acceptable birth control method, you could still become pregnant. You are aware that not having sex is the only certain way to prevent pregnancy. There is a slight chance that a pregnancy test could be wrong. If the pregnancy test is wrong, and you receive the study medication while pregnant, the study drug may harm an unborn baby.

ALTERNATIVE THERAPY
If you do not wish to participate in this study, the alternative therapy could potentially include all or any possible combinations of medications that are used in this study. You can receive this alternative therapy outside the study at your VA facility. Further, if you choose not to participate in this study, you will continue to receive the medical care which your doctor feels is most appropriate and your decision will not affect the level of medical care that you are eligible to receive.

WITHDRAWAL FROM THE STUDY
Your participation in this study is entirely voluntary. You may withdraw from this study at any time without penalty or loss of benefits and you will be treated with the best available options by your doctor. In addition, your participation in the study could be ended if continuation would not be safe for you or if one of the review committees overseeing the study recommends discontinuation of the study or your medical center's participation. In the event that this occurs, you will continue to receive the standard medical care at your medical facility.

BENEFITS
You may not personally benefit from taking part in this study. You may benefit if the medications prescribed successfully prevent progression of your underlying HIV infection but this may not occur. Your participation may lead to knowledge that will help others. All anti-retroviral medications used in this study will be provided free of charge to the study participants.

COSTS AND PAYMENTS
There will be no costs to you as a result of the study. In the event that an injury occurs resulting from your participation in this study, medical treatment will be available at your VA Medical Center at no cost to you.

1.
The answers to the study questions that you provide will be kept in a locked office at (name of Medical Center) and at the VA Cooperative Studies Coordinating Office in West Haven, CT. Confidentiality of this study's records identifying you will be maintained within these offices.

2.
If results of this study are reported in medical journals or at meetings, you will not be identified by name, by recognizable photograph, or by any other means without your specific consent. Your medical records will be maintained according to this medical center's requirements. However, there is a possibility that the Veterans' Affairs Headquarters for Research and Development or its designees may inspect the records relating to this study.

BLOOD SAMPLES
Some of the blood that is being drawn will be frozen for future tests of viral resistance and other studies. This blood will be stored without any personal identifiers at a centralized laboratory. You understand that by signing this consent form you are authorizing the use of this blood for these future analyses and studies.

SPECIAL INFORMATION
1. You are not required to take part in this study: your participation is entirely voluntary 2. Your decision to participate will not affect your benefits or eligibility for medical care 3. As a veteran-subject you will not be required to pay for any treatment received as a research subject that is being done solely for the purpose of this research study. However, your insurance carrier will be billed for all routine care and clinical procedures outside the protocol, if applicable. If you are in the 'discretionary work load' category, you are subject to making a copayment as indicated by a means test (defined in M-1,pt. 9. At the end of the study, the doctor treating you will continue to provide care and use the appropriate medications in your regimen. The costs of your medical care after the study is over will depend on the level of coverage to which you are entitled under VA regulations.
10. At the conclusion of the trial, you will be sent a letter thanking you for participation and describing the study findings.
RESEARCH SUBJECTS' RIGHTS: I have read or have had read to me all of the above. has explained the study to me and answered all of my questions. I have been told of the risks or discomforts and possible benefits of the study. I have been told of other choices of treatment available to me. I understand that I do not have to take part in this study, and my refusal to participate will involve no penalty or loss of rights to which I am entitled. I may withdraw from this study at any time without penalty or loss of VA or other benefits to which I am entitled.
The results of this study may be published, but my records will not be revealed unless required by law.
In case there are medical problems or questions, I have been told I can call Dr.___________ at__________during the day and Dr. at after hours. If any medical problems occur in connection with this study the VA will provide emergency care.

NIAID Standardized Toxicity Grading System
Item Grade 1 Toxicity

ICH Harmonised Tripartite Guideline
Having reached Step 4 of the ICH Process at the ICH Steering Committee meeting on 27 October 1994, this guideline is recommended for adoption to the three regulatory parties to ICH

I. INTRODUCTION
It is important to harmonise the way to gather and, if necessary, to take action on important clinical safety information arising during clinical development. Thus, agreed definitions and terminology, as well as procedures, will ensure uniform Good Clinical Practice standards in this area. The initiatives already undertaken for marketed medicines through the CIOMS-1 and CIOMS-2 Working Groups on expedited (alert) reports and periodic safety update reporting, respectively, are important precedents and models. However, there are special circumstances involving medicinal products under development, especially in the early stages and before any marketing experience is available. Conversely, it must be recognised that a medicinal product will be under various stages of development and/or marketing in different countries, and safety data from marketing experience will ordinarily be of interest to regulators in countries where the medicinal product is still under investigational-only (Phase 1, 2, or 3) status. For this reason, it is both practical and well-advised to regard premarketing and post-marketing clinical safety reporting concepts and practices as interdependent, while recognising that responsibility for clinical safety within regulatory bodies and companies may reside with different departments, depending on the status of the product (investigational vs. marketed). There are two issues within the broad subject of clinical safety data management that are appropriate for harmonisation at this time: (1) the development of standard definitions and terminology for key aspects of clinical safety reporting, and (2) the appropriate mechanism for handling expedited (rapid) reporting, in the investigational (i.e., pre-approval) phase.
The provisions of this guideline should be used in conjunction with other ICH Good Clinical Practice guidelines.

A. Basic Terms
Definitions for the terms adverse event (or experience), adverse reaction, and unexpected adverse reaction have previously been agreed to by consensus of the more than 30

Adverse Event (or Adverse Experience) Any untoward medical occurrence in a patient or clinical investigation subject administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment.
An adverse event (AE) can therefore be any unfavourable and unintended sign (including an abnormal laboratory finding, for example), symptom, or disease temporally associated with the use of a medicinal product, whether or not considered related to the medicinal product.

Adverse Drug Reaction (ADR)
In the pre-approval clinical experience with a new medicinal product or its new usages, particularly as the therapeutic dose(s) may not be established: all noxious and unintended responses to a medicinal product related to any dose should be considered adverse drug reactions. The phrase "responses to a medicinal products" means that a causal relationship between a medicinal product and an adverse event is at least a reasonable possibility, i.e., the relationship cannot be ruled out.
Regarding marketed medicinal products, a well-accepted definition of an adverse drug reaction in the post-marketing setting is found in WHO Technical Report 498 [1972] and reads as follows: A response to a drug which is noxious and unintended and which occurs at doses normally used in man for prophylaxis, diagnosis, or therapy of disease or for modification of physiological function. The old term "side effect" has been used in various ways in the past, usually to describe negative (unfavourable) effects, but also positive (favourable) effects. It is recommended that this term no longer be used and particularly should not be regarded as synonymous with adverse event or adverse reaction.

Unexpected Adverse Drug Reaction
An adverse reaction, the nature or severity of which is not consistent with the applicable product information (e.g., Investigator's Brochure for an unapproved investigational medicinal product). (See section III.C.)

B. Serious Adverse Event or Adverse Drug Reaction
During clinical investigations, adverse events may occur which, if suspected to be medicinal product-related (adverse drug reactions), might be significant enough to lead to important changes in the way the medicinal product is developed (e.g., change in dose, population, needed monitoring, consent forms). This is particularly true for reactions which, in their most severe forms, threaten life or function. Such reactions should be reported promptly to regulators. Therefore, special medical or administrative criteria are needed to define reactions that, either due to their nature ("serious") or due to the significant, unexpected information they provide, justify expedited reporting. To ensure no confusion or misunderstanding of the difference between the terms "serious" and "severe," which are not synonymous, the following note of clarification is provided: The term "severe" is often used to describe the intensity (severity) of a specific event (as in mild, moderate, or severe myocardial infarction); the event itself, however, may be of relatively minor medical significance (such as severe headache). This is not the same as "serious," which is based on patient/event outcome or action criteria usually associated with events that pose a threat to a patient's life or functioning. Seriousness (not severity) serves as a guide for defining regulatory reporting obligations. After reviewing the various regulatory and other definitions in use or under discussion elsewhere, the following definition is believed to encompass the spirit and meaning of them all: A serious adverse event (experience) or reaction is any untoward medical occurrence that at any dose: * results in death, * is life-threatening, NOTE: The term "life-threatening" in the definition of "serious" refers to an event in which the patient was at risk of death at the time of the event; it does not refer to an event which hypothetically might have caused death if it were more severe. * requires inpatient hospitalisation or prolongation of existing hospitalisation, * results in persistent or significant disability/incapacity, or * is a congenital anomaly/birth defect. Medical and scientific judgment should be exercised in deciding whether expedited reporting is appropriate in other situations, such as important medical events that may not be immediately life-threatening or result in death or hospitalisation but may jeopardise the patient or may require intervention to prevent one of the other outcomes listed in the definition above. These should also usually be considered serious.
Examples of such events are intensive treatment in an emergency room or at home for allergic bronchospasm; blood dyscrasias or convulsions that do not result in hospitalisation; or development of drug dependency or drug abuse.

C. Expectedness of an Adverse Drug Reaction
The purpose of expedited reporting is to make regulators, investigators, and other appropriate people aware of new, important information on serious reactions. Therefore, such reporting will generally involve events previously unobserved or undocumented, and a guideline is needed on how to define an event as "unexpected" or "expected" (expected/unexpected from the perspective of previously observed, not on the basis of what might be anticipated from the pharmacological properties of a medicinal product). As stated in the definition (II.A.3.), an "unexpected" adverse reaction is one, the nature or severity of which is not consistent with information in the relevant source document(s). Until source documents are amended, expedited reporting is required for additional occurrences of the reaction. The following documents or circumstances will be used to determine whether an adverse event/reaction is expected: 1. For a medicinal product not yet approved for marketing in a country, a company's Investigator's Brochure will serve as the source document in that country. (See section III.F. and ICH Guideline for the Investigator's Brochure.) 2. Reports which add significant information on specificity or severity of a known, already documented serious ADR constitute unexpected events. For example, an event more specific or more severe than described in the Investigator's Brochure would be considered "unexpected". Specific examples would be (a) acute renal failure as a labeled ADR with a subsequent new report of interstitial nephritis and (b) hepatitis with a first report of fulminant hepatitis.

III. STANDARDS FOR EXPEDITED REPORTING
A. What Should be Reported?

Single Cases of Serious, Unexpected ADRs
All adverse drug reactions (ADRs) that are both serious and unexpected are subject to expedited reporting. This applies to reports from spontaneous sources and from any type of clinical or epidemiological investigation, independent of design or purpose. It also applies to cases not reported directly to a sponsor or manufacturer (for example, those found in regulatory authority-generated ADR registries or in publications). The source of a report (investigation, spontaneous, other) should always be specified. Expedited reporting of reactions which are serious but expected will ordinarily be inappropriate. Expedited reporting is also inappropriate for serious events from clinical investigations that are considered not related to study product, whether the event is expected or not. Similarly, non-serious adverse reactions, whether expected or not, will ordinarily not be subject to expedited reporting. Information obtained by a sponsor or manufacturer on serious, unexpected reports from any source should be submitted on an expedited basis to appropriate regulatory authorities if the minimum criteria for expedited reporting can be met. See section III.B. Causality assessment is required for clinical investigation cases. All cases judged by either the reporting health care professional or the sponsor as having a reasonable suspected causal relationship to the medicinal product qualify as ADRs. For purposes of reporting, adverse event reports associated with marketed drugs (spontaneous reports) usually imply causality. Many terms and scales are in use to describe the degree of causality (attributability) between a medicinal product and an event, such as certainly, definitely, probably, possibly or likely related or not related. Phrases such as "plausible relationship," "suspected causality," or "causal relationship cannot be ruled out" are also invoked to describe cause and effect. However, there is currently no standard international nomenclature. The expression "reasonable causal relationship" is meant to convey in general that there are facts (evidence) or arguments to suggest a causal relationship.

Other Observations
There are situations in addition to single case reports of "serious" adverse events or reactions that may necessitate rapid communication to regulatory authorities; appropriate medical and scientific judgment should be applied for each situation. In general, information that might materially influence the benefit-risk assessment of a medicinal product or that would be sufficient to consider changes in medicinal product administration or in the overall conduct of a clinical investigation represents such situations. Examples include: a. For an "expected," serious ADR, an increase in the rate of occurrence which is judged to be clinically important. b. A significant hazard to the patient population, such as lack of efficacy with a medicinal product used in treating life-threatening disease. c. A major safety finding from a newly completed animal study (such as carcinogenicity).

Fatal or Life-Threatening Unexpected ADRs
Certain ADRs may be sufficiently alarming so as to require very rapid notification to regulators in countries where the medicinal product or indication, formulation, or population for the medicinal product are still not approved for marketing, because such reports may lead to consideration of suspension of, or other limitations to, a clinical investigations program. Fatal or life-threatening, unexpected ADRs occurring in clinical investigations qualify for very rapid reporting. Regulatory agencies should be notified (e.g., by telephone, facsimile transmission, or in writing) as soon as possible but no later than 7 calendar days after first knowledge by the sponsor that a case qualifies, followed by as complete a report as possible within 8 additional calendar days. This report must include an assessment of the importance and implication of the findings, including relevant previous experience with the same or similar medicinal products.

All Other Serious, Unexpected ADRs
Serious, unexpected reactions (ADRs) that are not fatal or life-threatening must be filed as soon as possible but no later than 15 calendar days after first knowledge by the sponsor that the case meets the minimum criteria for expedited reporting.

Minimum criteria for reporting
Information for final description and evaluation of a case report may not be available within the required time frames for reporting outlined above. Nevertheless, for regulatory purposes, initial reports should be submitted within the prescribed time as long as the following minimum criteria are met: an identifiable patient; a suspect medicinal product; an identifiable reporting source; and an event or outcome that can be identified as serious and unexpected, and for which, in clinical investigation cases, there is a reasonable suspected causal relationship. Follow-up information should be actively sought and submitted as it becomes available.

C. How to Report
The CIOMS-I form has been a widely accepted standard for expedited adverse event reporting. However, no matter what the form or format used, it is important that certain basic information/data elements, when available, be included with any expedited report, whether in a tabular or narrative presentation. The listing in Attachment 1 addresses those data elements regarded as desirable; if all are not available at the time of expedited reporting, efforts should be made to obtain them. (See section III.B.) All reports must be sent to those regulators or other official parties requiring them (as appropriate for the local situation) in countries where the drug is under development.

D. Managing Blinded Therapy Cases
When the sponsor and investigator are blinded to individual patient treatment (as in a double-blind study), the occurrence of a serious event requires a decision on whether to open (break) the code for the specific patient. If the investigator breaks the blind, then it is assumed the sponsor will also know the assigned treatment for that patient. Although it is advantageous to retain the blind for all patients prior to final study analysis, when a serious adverse reaction is judged reportable on an expedited basis, it is recommended that the blind be broken only for that specific patient by the sponsor even if the investigator has not broken the blind. It is also recommended that, when possible and appropriate, the blind be maintained for those persons, such as biometrics personnel, responsible for analysis and interpretation of results at the study's conclusion.
There are several disadvantages to maintaining the blind under the circumstances described which outweigh the advantages. By retaining the blind, placebo and comparator (usually a marketed product) cases are filed unnecessarily. When the blind is eventually opened, which may be many weeks or months after reporting to regulators, it must be ensured that company and regulatory data bases are revised. If the event is serious, new, and possibly related to the medicinal product, then if the Investigator's Brochure is updated, notifying relevant parties of the new information in a blinded fashion is inappropriate and possibly misleading. Moreover, breaking the blind for a single patient usually has little or no significant implications for the conduct of the clinical investigation or on the analysis of the final clinical investigation data. However, when a fatal or other "serious" outcome is the primary efficacy endpoint in a clinical investigation, the integrity of the clinical investigation may be compromised if the blind is broken. Under these and similar circumstances, it may be appropriate to reach agreement with regulatory authorities in advance concerning serious events that would be treated as disease-related and not subject to routine expedited reporting.

Reactions Associated with Active Comparator or Placebo Treatment
It is the sponsor's responsibility to decide whether active comparator drug reactions should be reported to the other manufacturer and/or directly to appropriate regulatory agencies. Sponsors must report such events to either the manufacturer of the active control or to appropriate regulatory agencies. Events associated with placebo will usually not satisfy the criteria for an ADR and, therefore, for expedited reporting.

Products with More than one Presentation or Use
To avoid ambiguities and uncertainties, an ADR that qualifies for expedited reporting with one presentation of a product (e.g., a dosage form, formulation, delivery system) or product use (e.g., for an indication or population), should be reported or referenced to regulatory filings across other product presentations and uses. It is not uncommon that more than one dosage form, formulation, or delivery system (oral, IM, IV, topical, etc.) of the pharmacologically active compound(s) is under study or marketed; for these different presentations there may be some marked differences in the clinical safety profile. The same may apply for a given product used in different indications or populations (single dose vs. chronic administration, for example). Thus, "expectedness" may be product or product-use specific, and separate Investigator's Brochures may be used accordingly. However, such documents are expected to cover ADR information that applies to all affected product presentations and uses. When relevant, separate discussions of pertinent product-specific or use-specific safety information will also be included. It is recommended that any adverse drug reactions that qualify for expedited reporting observed with one product dosage form or use be cross referenced to regulatory records for all other dosage forms and uses for that product. This may result in a certain amount of overreporting or unnecessary reporting in obvious situations (for example, a report of phlebitis on IV injection sent to authorities in a country where only an oral dosage form is studied or marketed). However, underreporting is completely avoided.

Post-study Events
Although such information is not routinely sought or collected by the sponsor, serious adverse events that occurred after the patient had completed a clinical study (including any protocol-required post-treatment follow-up) will possibly be reported by an investigator to the sponsor. Such cases should be regarded for expedited reporting purposes as though they were study reports. Therefore, a causality assessment and determination of expectedness are needed for a decision on whether or not expedited reporting is required.

F. INFORMING INVESTIGATORS AND ETHICS COMMITTEES/INSTITUTIONAL REVIEW BOARDS OF NEW SAFETY INFORMATION
International standards regarding such communication are discussed within the ICH GCP Guidelines, including the addendum on "Guideline for the Investigator's Brochure." In general, the sponsor of a study should amend the Investigator's Brochure as needed, and in accord with any local regulatory requirements, so as to keep the description of safety information updated.

Attachment 1 KEY DATA ELEMENTS FOR INCLUSION IN EXPEDITED REPORTS OF SERIOUS ADVERSE DRUG REACTIONS
The following list of items has its foundation in several established precedents, including those of CIOMS-I, the WHO International Drug Monitoring Centre, and various regulatory authority forms and guidelines. Some items may not be relevant depending on the circumstances. The minimum information required for expedited reporting purposes is: an identifiable patient, the name of a suspect medicinal product, an identifiable reporting source, and an event or outcome that can be identified as serious and unexpected and for which, in clinical investigation cases, there is a reasonable suspected causal relationship. Attempts should be made to obtain follow-up information on as many other listed items pertinent to the case.

Other Treatment(s)
For concomitant medicinal products (including non-prescription/OTC medicinal products) and non-medicinal product therapies, provide the same information as for the suspected product.

Details of Suspected Adverse Drug Reaction(s)
Full description of reaction(s) including body site and severity, as well as the criterion (or criteria) for regarding the report as serious should be given. In addition to a description of the reported signs and symptoms, whenever possible, attempts should be made to establish a specific diagnosis for the reaction. Start date (and time) of onset of reaction Stop date (and time) or duration of reaction Dechallenge and rechallenge information Setting (e.g., hospital, out-patient clinic, home, nursing home) Outcome: information on recovery and any sequelae; what specific tests and/or treatment may have been required and their results; for a fatal outcome, cause of death and a comment on its possible relationship to the suspected reaction should be provided. Any autopsy or other post-mortem findings (including a coroner's report) should also be provided when available. Other information: anything relevant to facilitate assessment of the case, such as medical history including allergy, drug or alcohol abuse; family history; findings from special investigations.

Details on Reporter of Event (Suspected ADR)
Name Address Telephone number Profession (specialty)

Administrative and Sponsor/Company Details
Source of report: was it spontaneous, from a clinical investigation (provide details), from the literature (provide copy), other? Date event report was first received by sponsor/manufacturer Country in which event occurred Type of report filed to authorities: initial or follow-up (first, second, etc.) Name and address of sponsor/manufacturer/company Name, address, telephone number, and FAX number of contact person in reporting company or institution Identifying regulatory code or number for marketing authorisation dossier or clinical investigation process for the suspected product (for example IND or CTX number, NDA number) Sponsor/manufacturer's identification number for the case (this number must be the same for the initial and follow-up reports on the same case). Crossover between the two arms during follow-up

SAMPLE SIZE/POWER CALCULATIONS/LEVEL OF SIGNIFICANCE
The rate of crossover from mega-ART to standard-ART (expected to be largely due to toxicity) is likely to be higher initially and to decrease during follow-up. In the Multi-Drug Rescue Therapy cohort, the proportion of patients reducing from 5 or more drugs to 4 or less was 23% by one year. 2 The proportion reducing to three or less would be smaller.
Data from other cohorts showed a lower rate of crossover. Although 25% of 106 patients on mega-ART with a median follow-up of 15 months experienced severe laboratory abnormalities, only 6 (5%) discontinued treatment. 3 The crossover rate from Mega-ART to Standard-ART in OPTIMA was originally assumed to be 20% in Year 1 with a 50% reduction every year thereafter. In the GIGHAART study, only 8% receiving giga-HAART reduced the number of drugs in their regimen to < 6 drugs. The observed crossover rate in OPTIMA from mega-ART to standard ART is less than 1% in Year 1. It is anticipated that such crossovers will occur early in the follow-up since the primary reason for reducing the number of drugs would be toxicity. The drop-in rate from Standard-ART to Mega-ART was initially assumed to be 5% in the first year, but no drop-ins in the first year have been observed in the study.

Loss to follow-up
The three Clinical Trial organizations have an established record of following the vast majority of patients in clinical trials. Linkages with national AIDS & death registries will assist by providing mechanisms for ancillary follow-up.
The loss to follow-up was initially assumed to be 10% by the end of the study. This follow-up refers to inability to assess primary endpoints (death and/or AIDS events). All three co-coordinating centers have been involved in multi-center trials for a number of years and the track record of tracking such endpoints has been excellent. Tracking is aided by the existence of databases where study endpoints can be detected and verified, meaning that loss to follow-up for events is expected to be smaller than originally assumed (10%). To maintain the conservative nature of our assumptions, a 5% loss to follow-up was incorporated into the revised sample size estimation.
All the sample size and power calculations have been done according to the methods proposed by Lakatos, allowing for staggered entry and assumes an exponential distribution of the event rate. 4,5,6 In calculating sample size, the following assumptions were made: -  With the above assumptions and using a log rank test with intention-to-treat, a trial of 503 patients (with 292 expected events) will be sufficient to detect a difference between treatment strategy groups (power 0.80; type I error 0.05).

Statistical Analysis Plan Interim Monitoring and Analysis
Interim monitoring will focus on patient intake--overall and within country, adherence to the protocol, adverse events, baseline comparability of treatment groups, and completeness of data retrieval. Prototype sets of tables and figures for trial monitoring by (a) TMC and TSC, and (b) DSMB are provided in Appendix 9.

(a) TMC and TSC Monitoring
The number of patients enrolled (overall and by country) will be tabulated. If possible, the combined as well as country-specific proportion of enrolled/eligible patients will also be examined. Patients enrolled in OPTIMA by month and hospital within each country, along with the number of minor and major protocol violations will be tabulated.
The number of patients randomized in each CD4 stratum will also be provided (IN Tables 1-4, Appendix 9) Baseline characteristics will be tabulated: demographic information, medical history, HIV-stage antiretroviral therapy, opportunistic infections medication, risk factors, quality of life, virological markers, hematology, biochemistry, and physical examination information.
This will be a descriptive analysis with no formal test or p-values for differences between any comparison groups. Antiretroviral therapy history, current anti-HIV medications, and baseline virological and immunological markers, will be tabulated by country (Base Tables   5-8, Appendix 9).
At follow-up visits data will be collected to examine the following variables: the composition of the first ART regimen on-study and the follow-up ART regimens, virological and immunological markers, as well as process results. Tables and plots describing these variables will be generated for the trial overall and by country (Fables 9-11, Appendix 9).
Compliance to the allocated strategy will be provided in C Figures 3a-3c, Appendix 9.
Grade 3 and 4 Adverse Events and Serious Adverse Events will be tabulated by country and overall (AE Tables 12-13, Appendix 9).
Outcomes will be provided by country and overall in OUT Table 14, Appendix 9.
(b) DSMB Monitoring At every DSMB meeting, the number of patients enrolled (overall, by country and by treatment strategy) will be tabulated. The number of minor and major protocol violations will be tabulated by country as well as the number of patients randomized in each CD4 stratum (DSMB Monitoring Tables, Section A, Tables 1-4, Appendix 9).
Baseline characteristics will be tabulated by treatment strategy: demographic information, medical history, HIV-stage antiretroviral therapy, opportunistic infections medication, risk factors, quality of life, virological markers, hematology, biochemistry, and physical examination information. This will be a descriptive analysis with no formal test or p-values for differences between any comparison groups. Antiretroviral therapy history, current anti-HIV medications, and baseline virological and immunological markers, will be tabulated by country (DSMB Monitoring Tables, Section A, Tables 4-7, Appendix 9).
At follow-up visits data will be collected to examine the following variables: the composition of the first ART regimen on-study and the follow-up ART regimens for each treatment strategy (DSMB Monitoring Tables, Section A, After Randomization/Follow-up,   Tables 1a-1d, and Tables 4a-4d, Appendix 9).
Grade 3 and 4 Adverse Events and Serious Adverse Events will be tabulated by country and overall (DSMB Monitoring Tables, Section A, After Randomization/Follow-up,   Tables 2a-3, Appendix 9).
Compliance will be assessed for each of the four treatment strategies (DSMB Monitoring Tables, Section A ii.

Participant Disposition
The number of patients enrolled (overall, by country and by month, and by screening CD4 count stratum) will be tabulated and presented (Tables B1 and B4 / Figures B1). Study subjects who withdrew prior to study termination will be discussed in the clinical study report.

Eligibility Violations
Tables B2 and B3 will list all major and minor known eligibility violations.
The tables will provide a descriptive summary of each such violation. Major violations are defined as (a) errors in the process of randomization and (b) violations of inclusion/exclusion criteria, which were not reviewed and allowed/exempted.

Evaluation of Baseline Characteristics
In order to assess the adequacy of randomization, baseline characteristics will be tabulated by allocation (overall and by country) in terms of median, interquartile range, mean and standard deviation where appropriate, and distribution tabulations.
Baseline characteristics include: demographic information (age, gender, race), HIVstage, antiretroviral therapy, opportunistic infections medication, likely mode of infection, virological and immunological markers. This will be a descriptive analysis with no formal test or p-values for differences between the allocation groups.
The classes of medications as well as the number of antiretrovirals that patients are discharged with at study entry will be described for the whole study population (Table F2).

Description of follow-up
The total follow-up time will be calculated in person-years and the number of study participants whose survival status is not known for more than 6 months will be described by allocation group and summarized by treatment strategy. (Table F1)

Compliance to Treatment strategy
The proportion of subjects on assigned treatment strategy will be displayed over time (Figures F1(a)-(d)). Changes in treatment strategy will be recorded, as well as the reasons for these changes.

Evaluation of Primary and Secondary Endpoints
The analysis of the primary endpoint [time to (new or recurrent) AIDSdefining event or death] will be done according to intent-to-treat, that is according to original treatment assignment, regardless of adherence. (Tables O1-O3; Figures O1-O5) The primary comparisons will be the main effects: Standard-ART vs Mega-ART and Anti-retroviral Drug-free Period (ARDFP) vs No Drug-Free Period (No ARDFP). These comparisons will be made using the stratified log-rank test for time to event analysis. 1 There will be a total of 12 strata derived according to country, screening CD4 count and other allocated treatment strategy in patients randomized according to the original 2X2 design. All p-values will be two-tailed and the Type I error of 0.05 will be used as the level of significance.
A test of interaction between the main effects will be undertaken using data from subjects who were enrolled using the 2X2 randomization scheme.
Statistically significant qualitative interactions between the main effects (treatment strategies) are not anticipated (some moderate quantitative interactions are possible and will be assessed in a Cox analysis). In the event such interactions are statistically significant, the analysis of the main effects will be performed according to allocation status. For example, the comparison between Mega vs Standard-ART will be performed separately according to ARDFP status.
Standard failure time methods (Kaplan-Meier plots, log rank test and Cox regression techniques) will be employed to make comparisons of efficacy (primary endpoints: death and AIDS-defining events) and toxicity (secondary endpoints: non HIV-related serious adverse events; Tables O5 and Figures O6) between the treatment strategy groups. 1 The Kaplan-Meier (KM) estimate will be used to display the estimated probability of freedom from events across time. Standard errors of this estimate will be computed using the modified Greenwood formula and will be used to construct confidence intervals around the KM estimates.
If the primary analyses of main effects are significant, secondary analyses will examine the pair wise treatment comparisons. In order to provide some control for multiplicity for these secondary analyses, the Hochberg procedure, a sequentially rejective variation of the Bonferroni procedure, will be used to determine the significance at a Type I error of 0.05 (two-sided). 2 Additional analysis adjusting for baseline covariates will be undertaken using Cox analysis. These covariates are: CD4 count, viral load, previous history of antiretroviral treatment. Standard-ART vs Mega-ART will examine the entire follow-up period; for comparisons of ARDFP vs No ARDFP, analyses will also examine the initial 3 months when these differential strategies are to be applied. A more detailed analysis plan for health economics and quality of life will be prepared separately.

Health Economic Analysis
Resources such as hospital care, medication, outpatient visits, and long-term care will be collected for all patients in each strategy arm and compared using t-tests or Wilcoxon rank sum tests. These outcomes will be compared over subgroups using analysis of variance techniques or the Kruskall-Wallis test. The p-values in subgroups will be adjusted using approaches for multiple hypothesis testing. (for details see Health Economics Component, Section XVI)

Safety
The incidence of any adverse events will be calculated and compared among the treatment strategies using chi-square or Wilcoxon-type statistics, as appropriate.

Other analyses
Further analyses will include: Changes in CD4 cell counts over the follow-up period Changes in plasma HIV RNA (log10 HIV RNA) at follow-up visits and average change from nadir over the duration of follow-up will be assessed both graphically and by applying longitudinal methods where appropriate 16 Comparison of Viral Resistance at 12 months among treatment strategy groups (for patients with viral loads dictated by drug-resistance assays detection limits)

OI Medications for prophylaxis
Anti-PCP/CMV Antibacterial Antifungal

Study Feasibility
The sample size of 504 will be accrued over a period of 4.5 years

VA Patient Accrual
VA HIV population The VA system cares for approximately 19,000 HIV patients who have access to HIV-knowledgeable health care providers, to the full spectrum of approved antiretroviral medications, and to laboratories that perform virologic and immunological testing. Providers from 30 VA medical facilities had originally expressed interest in participating in the study and provided conservative estimates, indicating that the required number of patients should be easily accruable within the VA system.
During the planning process for OPTIMA surveys and questionnaires sent to potential participating sites had indicated that the target accrual rate could be achieved. 3. At least one CD4 count ≤300 cells/mm 3 in a 6-month period prior to the target quarter.
4. At least one HIV viral load >5,000 copies/ml in the 6 months prior to the quarter under review.
The results from these queries are shown in the table below These results suggest that the number of OPTIMA eligible patients is slightly increasing (0.3% increase); the most important finding however is that the pool of unique and eligible patients has increased. It is evident from this analysis that the revised target sample size is feasible.

UK Patient Accrual
In light of poor recruitment in the UK for reasons outlined in other areas of this document, the UK currently expects to recruit one patient per month.

Canada Patient Accrual
The best estimate for the accrual rate in Canada over the next few years is based the accrual experience thus far. Over the last year, Canadian sites have averaged two patients per month. The number per month has been fairly consistent with a monthly maximum of 4 patients and a minimum of zero. Accrual could be augmented slightly by the recent addition of two large sites in Vancouver and Montreal. Eligibility criteria were implemented in order to reflect current clinical practice.
After analysis of data from three cohorts (1 US and 2 European) revealed that the patient pool would increase by 40% using the revised criteria, an assessment of the impact of these modifications on accrual at the OPTIMA sites was undertaken.
In the US (VA), participating sites were asked to compare the total number of patients attending their clinics meeting the new vs. old CD4 and viral load criteria. This survey suggested a 41% increase in the pool of patients who would be considered for participation in OPTIMA. A similar assessment in the UK done using data from 4 clinical centers (representative of the UK sites) and other cohort data suggested a 40% increase in the pool of patients. Analysis of cohort data from Canada revealed an increase of at least 25% in the numbers of patients who would be considered for the trial (this was obtained using CD4 <=250 cells/mm 3 as the cutoff level).