Skip to content

Strategic Timing of Antiretroviral Treatment

Strategic Timing of AntiRetroviral Treatment

Status
Completed
Phases
Phase 4
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT00867048
Acronym
START
Enrollment
4688
Registered
2009-03-23
Start date
2009-04-15
Completion date
2022-07-27
Last updated
2024-03-04

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

HIV Infection

Keywords

highly active antiretroviral therapy (HAART), CD4 Count, Early Intervention, HIV, HIV Infection, HIV Infections, treatment naive

Brief summary

Objectives: * To find out if the chance of developing a serious illness or of getting AIDS is less if patients start taking HIV medicines at a time when their cluster-of-differentiation-4 (CD4)+ cell count is still fairly high, instead of waiting until the CD4+ count is at the level where there is good evidence for starting medicines. * To learn more about how a strategy of starting HIV medicines early might affect other aspects of care, such as the chances of developing other illnesses or resistance to HIV medicines, the frequency of doctor visits, the cost of medical care, and general health and satisfaction.

Detailed description

Background: * Most guidelines agree that if the number of your CD4+ cells (cells in your blood which help fight infection) drops below 350 cells/mm3, or if you have symptoms of AIDS, you should start taking HIV medicines. There are randomized trials that support this recommendation. (Randomized trials are usually considered the strongest form of evidence to support treatment decisions. Other studies, like observational studies, provide evidence too, but the evidence is often considered to be weaker than evidence from randomized trials. A randomized trial gives the most certain information about how well a treatment works because randomization makes sure each group is similar except for the treatment they receive.) Some experts believe that HIV treatment should be started even when the number of CD4+ cells is above 350 cells/mm3. For example, guidelines issued in the US in December 2009 include a new recommendation for starting HIV medicines if your CD4+ cell count is between 350 and 500 cells/mm3. However, this recommendation is based on information from observational studies, not randomized trials. We are doing this study to find out if the chances of getting a serious illness or of getting AIDS are less if people start taking HIV medicines at a time when their CD4+ cell counts are still fairly high, instead of waiting to take HIV medicines at a CD4+ count where there is good evidence for starting medicines. Objectives: * To find out if the chance of developing a serious illness or of getting AIDS is less if patients start taking HIV medicines at a time when their CD4+ cell count is still fairly high, instead of waiting until the CD4+ count is at the level where there is good evidence for starting medicines. * To learn more about how a strategy of starting HIV medicines early might affect other aspects of care, such as the chances of developing other illnesses or resistance to HIV medicines, the frequency of doctor visits, the cost of medical care, and general health and satisfaction. Eligibility: * Patients 18 years of age and older who are infected with HIV, have CD4+ cell counts of greater than 500 cells/mm3, and who have never had antiretroviral therapy to treat HIV. Design: * Initial screening visits (2) to draw blood for CD4+ cell counts and provide a full medical history * Patients will be randomly split into two groups: Early: Patients will begin receiving HIV medications from the start of the study. Deferred: Patients will begin to take HIV medications when the CD4 drops below 350 cells/mm3, or they develop AIDS or other symptoms of HIV infection. * HIV medications for each patient will be determined by the study doctors. * Evaluations during the treatment period: * Physical examination, including vital signs and body weight checks, and pregnancy test for women who can become pregnant. * Questions about daily life, including sexual behaviors. * Blood and urine tests. * Heart tests with electrocardiogram. * Patients will return for evaluations at 1 and 4 months after randomization, and every 4 months thereafter for the duration of the study. Substudies will take advantage of the START randomization to compare outcomes in people starting ART early vs. later. The purpose of this randomized study is to determine whether immediate initiation of antiretroviral treatment (ART) is superior to deferral of ART until the CD4+ declines below 350 cells/mm(3) in terms of morbidity and mortality in HIV-1 (subsequently referred to as HIV) infected persons who are antiretroviral naive with a CD4+ count above 500 cells/mm(3). The study will enroll an estimated 4,000 participants. Participants will be followed for at least 3 years after enrollment, to a common closing date. Substudies will take advantage of the START randomization to compare outcomes in people starting ART early vs. later. These will measure outcomes that do not require the entire sample size of START to determine whether early ART is related to a difference in these outcomes over the course of the study.

Interventions

DRUGAll licensed antiretroviral medications

In both arms, participants may be prescribed any licensed antiretroviral medication, in accordance with national treatment guidelines. The nature of the intervention is the timing of when to begin treatment with these medications, as described in the two treatment arms.

Sponsors

National Institute of Allergy and Infectious Diseases (NIAID)
CollaboratorNIH
Copenhagen HIV Programme (CHIP) -- Copenhagen, Denmark
CollaboratorUNKNOWN
Medical Research Council
CollaboratorOTHER_GOV
Kirby Institute
CollaboratorOTHER_GOV
Washington D.C. Veterans Affairs Medical Center
CollaboratorFED
ANRS, Emerging Infectious Diseases
CollaboratorOTHER_GOV
German Federal Ministry of Education and Research
CollaboratorOTHER_GOV
NEAT - European AIDS Treatment Network
CollaboratorOTHER
National Health and Medical Research Council, Australia
CollaboratorOTHER
National Institutes of Health Clinical Center (CC)
CollaboratorNIH
National Cancer Institute (NCI)
CollaboratorNIH
National Heart, Lung, and Blood Institute (NHLBI)
CollaboratorNIH
National Institute of Mental Health (NIMH)
CollaboratorNIH
National Institute of Neurological Disorders and Stroke (NINDS)
CollaboratorNIH
National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS)
CollaboratorNIH
Abbott
CollaboratorINDUSTRY
Bristol-Myers Squibb
CollaboratorINDUSTRY
Gilead Sciences
CollaboratorINDUSTRY
GlaxoSmithKline
CollaboratorINDUSTRY
Merck Sharp & Dohme LLC
CollaboratorINDUSTRY
Tibotec Pharmaceutical Limited
CollaboratorINDUSTRY
University of Minnesota
Lead SponsorOTHER

Study design

Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
NONE

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No

Inclusion criteria

* Signed informed consent * HIV infection documented by a plasma HIV RNA viral load, rapid HIV test or any licensed\* ELISA test; and confirmed by another test using a different method including but not limited to a rapid HIV test, Western Blot, HIV culture, HIV antigen, or HIV pro-viral DNA at any time prior to study entry. * Age greater than or equal to 18 years * Karnofsky performance score greater than or equal to 80 (an indication that the participant can perform normal activities) * Perceived life expectancy of at least 6 months * For women of child-bearing potential, willingness to use contraceptives as described in the product information of the ART drugs they are prescribed * Two CD4+ cell counts greater than 500 cells/mm(3) at least 2 weeks apart within 60 days before randomization * The term licensed refers to an FDA-approved kit or, for sites located in countries other than the United States, a kit that has been certified or licensed by an oversight body within that country. Confirmation of the initial test result must use a test method that is different than the one used for the initial assessment.

Exclusion criteria

* Any previous use of ART or interleukin-2 (IL-2) * Diagnosis of any clinical AIDS event before randomization (including esophageal candidiasis and chronic Herpes simplex infection) * Presence of HIV progression such as oral thrush, unexplained weight loss, or unexplained fever * Cardiovascular event (myocardial infarction, angioplasty, coronary-artery bypass grafting, stroke) within 6 months before randomization * Non-AIDS-defining cancer, excluding basal and squamous cell skin cancer, within 6 months before randomization * Dialysis within 6 months before randomization * Diagnosis of decompensated liver disease before randomization * Current imprisonment, or compulsory detention (involuntary incarceration) for treatment of a psychiatric or physical illness * Current pregnancy or breastfeeding (a negative serum or urine pregnancy test is required within 14 days before randomization for women of child-bearing potential)

Design outcomes

Primary

MeasureTime frame
Composite Endpoint of AIDS, Serious Non-AIDS Diagnoses, and All-cause Mortalityfull follow-up, 9.3 years

Secondary

MeasureTime frameDescription
Specific Non-AIDS Diagnosesfull follow-up, 9.3 years
Death, All-cause Mortalityfull follow-up, 9.3 yearsCount of participants
Quality of Life- Mean Change From Baseline in a Visual Analog Scale (VAS) for Perceived Current Health4.5 yearsMean change from baseline of the VAS. This was a single data item where participants self-reported their perceived current state of health on a scale of 0-100 (0=worst possible and 100=best possible).
Transmission Risk Behavior Outcome 112 monthsProportion of participants identifying as men who have sex with men (MSM) engaging in condomless sex with HIV serodifferent partners (CLS-D) at study month 12. Participant completed a self-reported questionnaire on transmission risk behavior during the past two months at the month 12 visit following randomization.
Change in Neurocognitive Function (in a Subset of Participants)4.5 yearsMean change from baseline of the QNPZ-8 score. In the Neurology substudy of START, participants were administered a neuropsychological test battery of 8 tests (grooved peg board, finger tapping, Color Trails 1 and 2, Semantic Verbal Fluency, WAIS III Digit Symbol, HVLT-R Learning, HVLT-R Delayed Recall). Test scores were standardized to z-scores using baseline values as reference such that baseline values had a mean of 0 and standard deviation of 1. Test scores were standardized to z-scores using baseline values as reference such that baseline values had a mean of 0 and standard deviation of 1. Z-scores \> 0 indicate improvement over baseline levels. The quantitative neuropsychological performance z-score (QNPZ-8) was the mean of the z-scores across the the 8 tests. Analyses were by intention-to-treat principles.
Large Artery Elasticity (in a Subset of Participants)4.5 yearsMean change from baseline in large arterial elasticity. In the Arterial Elasticity substudy of START, participants had non-invasive measurements of radial artery blood pressure waveforms recorded at baseline, study months 4, 8, and 12, and then annually. Small arterial elasticity (SAE) and large arterial elasticity (LAE) were derived from analysis of the diastolic pulse waveform. Analyses were by intention-to-treat principles.
AIDs or AIDs Related Deathfull follow-up, 9.3 yearsparticipant count
Changes in Bone Mineral Density (in a Subset of Participants) Measure 14.5 yearsMean percent change from baseline in bone mineral density at the spine. In the bone mineral density (BMD) substudy of START, participants underwent dual-enery x-ray absorptionmetry (DXA) to measure BMD at the spine and hip at baseline and annually. Analyses were by intention-to-treat principles.
Transmission Risk Behavior Outcome 212 month visitPercentage of participants identifying as hetrosexual engaging in condomless sex with HIV serodifferent partners (CLS-D) at study month 12. Participant completed a self-reported questionnaire on transmission risk behavior during the past two months at the month 12 visit following randomization.
Small Artery Elasticity (in a Subset of Participants)4.5 yearsMean change from baseline in small arterial elasticity. In the Arterial Elasticity substudy of START, participants had non-invasive measurements of radial artery blood pressure waveforms recorded at baseline, study months 4, 8, and 12, and then annually. Small arterial elasticity (SAE) and large arterial elasticity (LAE) were derived from analysis of the diastolic pulse waveform. Analyses were by intention-to-treat principles.
Rate of Lung Function Decline (in a Subset of Participants) Among Non-smokers4.5 yearsRate of lung function decline (slope of forced expiratory volume (FEV1)) over follow-up among self-reported non-smokers at study entry. In the pulmonary substudy of START, participants had post-bronchodilator spirometry measures collected at baseline and annually. Analyses were by intention-to-treat principles.
Changes in Bone Mineral Density (in a Subset of Participants) Measure 24.5 yearsMean percent change from baseline in bone mineral density at the spine. In the bone mineral density (BMD) substudy of START, participants underwent dual-enery x-ray absorptionmetry (DXA) to measure BMD at the spine and hip at baseline and annually. Analyses were by intention-to-treat principles.
Rate of Lung Function Decline (in a Subset of Participants) Among4.5 yearsRate of lung function decline (slope of forced expiratory volume (FEV1)) over follow-up among self-reported smokers at study entry. In the pulmonary substudy of START, participants had post-bronchodilator spirometry measures collected at baseline and annually. Analyses were by intention-to-treat principles.

Countries

Argentina, Australia, Austria, Belgium, Brazil, Chile, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, India, Ireland, Israel, Italy, Luxembourg, Malaysia, Mali, Mexico, Morocco, Nigeria, Norway, Peru, Poland, Portugal, Puerto Rico, South Africa, Spain, Sweden, Switzerland, Thailand, Uganda, United Kingdom, United States

Participant flow

Participants by arm

ArmCount
Early ART
Initiate ART immediately following randomization All licensed antiretroviral medications: In both arms, participants may be prescribed any licensed antiretroviral medication, in accordance with national treatment guidelines. The nature of the intervention is the timing of when to begin treatment with these medications, as described in the two treatment arms.
2,326
Deferred ART
Defer ART until the CD4+ count declines to \<350 cells/cu mm or AIDS develops All licensed antiretroviral medications: In both arms, participants may be prescribed any licensed antiretroviral medication, in accordance with national treatment guidelines. The nature of the intervention is the timing of when to begin treatment with these medications, as described in the two treatment arms.
2,362
Total4,688

Baseline characteristics

CharacteristicEarly ARTDeferred ARTTotal
Age, Categorical
<=18 years
0 Participants0 Participants0 Participants
Age, Categorical
>=65 years
18 Participants17 Participants35 Participants
Age, Categorical
Between 18 and 65 years
2308 Participants2345 Participants4653 Participants
Age, Continuous36.8 years
STANDARD_DEVIATION 10.2
36.8 years
STANDARD_DEVIATION 10.1
36.8 years
STANDARD_DEVIATION 10.2
Race (NIH/OMB)
American Indian or Alaska Native
1 Participants1 Participants2 Participants
Race (NIH/OMB)
Asian
197 Participants190 Participants387 Participants
Race (NIH/OMB)
Black or African American
1025 Participants1084 Participants2109 Participants
Race (NIH/OMB)
More than one race
308 Participants303 Participants611 Participants
Race (NIH/OMB)
Native Hawaiian or Other Pacific Islander
701 Participants704 Participants1405 Participants
Race (NIH/OMB)
Unknown or Not Reported
91 Participants74 Participants165 Participants
Race (NIH/OMB)
White
3 Participants6 Participants9 Participants
Sex: Female, Male
Female
624 Participants635 Participants1259 Participants
Sex: Female, Male
Male
1702 Participants1727 Participants3429 Participants

Adverse events

Event typeEG000
affected / at risk
EG001
affected / at risk
deaths
Total, all-cause mortality
66 / 2,32585 / 2,359
other
Total, other adverse events
0 / 00 / 0
serious
Total, serious adverse events
428 / 2,325500 / 2,359

Outcome results

Primary

Composite Endpoint of AIDS, Serious Non-AIDS Diagnoses, and All-cause Mortality

Time frame: full follow-up, 9.3 years

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
Early ARTComposite Endpoint of AIDS, Serious Non-AIDS Diagnoses, and All-cause Mortality133 Participants
Deferred ARTComposite Endpoint of AIDS, Serious Non-AIDS Diagnoses, and All-cause Mortality215 Participants
Secondary

AIDs or AIDs Related Death

participant count

Time frame: full follow-up, 9.3 years

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
Early ARTAIDs or AIDs Related Death31 Participants
Deferred ARTAIDs or AIDs Related Death83 Participants
Secondary

Change in Neurocognitive Function (in a Subset of Participants)

Mean change from baseline of the QNPZ-8 score. In the Neurology substudy of START, participants were administered a neuropsychological test battery of 8 tests (grooved peg board, finger tapping, Color Trails 1 and 2, Semantic Verbal Fluency, WAIS III Digit Symbol, HVLT-R Learning, HVLT-R Delayed Recall). Test scores were standardized to z-scores using baseline values as reference such that baseline values had a mean of 0 and standard deviation of 1. Test scores were standardized to z-scores using baseline values as reference such that baseline values had a mean of 0 and standard deviation of 1. Z-scores \> 0 indicate improvement over baseline levels. The quantitative neuropsychological performance z-score (QNPZ-8) was the mean of the z-scores across the the 8 tests. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTChange in Neurocognitive Function (in a Subset of Participants)0.19 units on a scale
Deferred ARTChange in Neurocognitive Function (in a Subset of Participants)0.21 units on a scale
Secondary

Changes in Bone Mineral Density (in a Subset of Participants) Measure 1

Mean percent change from baseline in bone mineral density at the spine. In the bone mineral density (BMD) substudy of START, participants underwent dual-enery x-ray absorptionmetry (DXA) to measure BMD at the spine and hip at baseline and annually. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTChanges in Bone Mineral Density (in a Subset of Participants) Measure 1-1.8 percent
Deferred ARTChanges in Bone Mineral Density (in a Subset of Participants) Measure 1-0.9 percent
Secondary

Changes in Bone Mineral Density (in a Subset of Participants) Measure 2

Mean percent change from baseline in bone mineral density at the spine. In the bone mineral density (BMD) substudy of START, participants underwent dual-enery x-ray absorptionmetry (DXA) to measure BMD at the spine and hip at baseline and annually. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTChanges in Bone Mineral Density (in a Subset of Participants) Measure 2-2.8 percent
Deferred ARTChanges in Bone Mineral Density (in a Subset of Participants) Measure 2-1.7 percent
Secondary

Death, All-cause Mortality

Count of participants

Time frame: full follow-up, 9.3 years

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
Early ARTDeath, All-cause Mortality66 Participants
Deferred ARTDeath, All-cause Mortality85 Participants
Secondary

Large Artery Elasticity (in a Subset of Participants)

Mean change from baseline in large arterial elasticity. In the Arterial Elasticity substudy of START, participants had non-invasive measurements of radial artery blood pressure waveforms recorded at baseline, study months 4, 8, and 12, and then annually. Small arterial elasticity (SAE) and large arterial elasticity (LAE) were derived from analysis of the diastolic pulse waveform. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTLarge Artery Elasticity (in a Subset of Participants)0.27 mL/mmHg x 10
Deferred ARTLarge Artery Elasticity (in a Subset of Participants)0.8 mL/mmHg x 10
Secondary

Quality of Life- Mean Change From Baseline in a Visual Analog Scale (VAS) for Perceived Current Health

Mean change from baseline of the VAS. This was a single data item where participants self-reported their perceived current state of health on a scale of 0-100 (0=worst possible and 100=best possible).

Time frame: 4.5 years

ArmMeasureValue (MEAN)Dispersion
Early ARTQuality of Life- Mean Change From Baseline in a Visual Analog Scale (VAS) for Perceived Current Health1.8 units on a scaleStandard Deviation 8.6
Deferred ARTQuality of Life- Mean Change From Baseline in a Visual Analog Scale (VAS) for Perceived Current Health0.22 units on a scaleStandard Deviation 8.7
Secondary

Rate of Lung Function Decline (in a Subset of Participants) Among

Rate of lung function decline (slope of forced expiratory volume (FEV1)) over follow-up among self-reported smokers at study entry. In the pulmonary substudy of START, participants had post-bronchodilator spirometry measures collected at baseline and annually. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTRate of Lung Function Decline (in a Subset of Participants) Among-36.6 mL/yr
Deferred ARTRate of Lung Function Decline (in a Subset of Participants) Among-38.8 mL/yr
Secondary

Rate of Lung Function Decline (in a Subset of Participants) Among Non-smokers

Rate of lung function decline (slope of forced expiratory volume (FEV1)) over follow-up among self-reported non-smokers at study entry. In the pulmonary substudy of START, participants had post-bronchodilator spirometry measures collected at baseline and annually. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTRate of Lung Function Decline (in a Subset of Participants) Among Non-smokers-26.9 mL/yr
Deferred ARTRate of Lung Function Decline (in a Subset of Participants) Among Non-smokers-22.8 mL/yr
Secondary

Small Artery Elasticity (in a Subset of Participants)

Mean change from baseline in small arterial elasticity. In the Arterial Elasticity substudy of START, participants had non-invasive measurements of radial artery blood pressure waveforms recorded at baseline, study months 4, 8, and 12, and then annually. Small arterial elasticity (SAE) and large arterial elasticity (LAE) were derived from analysis of the diastolic pulse waveform. Analyses were by intention-to-treat principles.

Time frame: 4.5 years

ArmMeasureValue (MEAN)
Early ARTSmall Artery Elasticity (in a Subset of Participants)-0.23 mL/mmHg x 10
Deferred ARTSmall Artery Elasticity (in a Subset of Participants)-0.18 mL/mmHg x 10
Secondary

Specific Non-AIDS Diagnoses

Time frame: full follow-up, 9.3 years

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
Early ARTSpecific Non-AIDS Diagnoses107 Participants
Deferred ARTSpecific Non-AIDS Diagnoses137 Participants
Secondary

Transmission Risk Behavior Outcome 1

Proportion of participants identifying as men who have sex with men (MSM) engaging in condomless sex with HIV serodifferent partners (CLS-D) at study month 12. Participant completed a self-reported questionnaire on transmission risk behavior during the past two months at the month 12 visit following randomization.

Time frame: 12 months

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
Early ARTTransmission Risk Behavior Outcome 1156 Participants
Deferred ARTTransmission Risk Behavior Outcome 1155 Participants
Secondary

Transmission Risk Behavior Outcome 2

Percentage of participants identifying as hetrosexual engaging in condomless sex with HIV serodifferent partners (CLS-D) at study month 12. Participant completed a self-reported questionnaire on transmission risk behavior during the past two months at the month 12 visit following randomization.

Time frame: 12 month visit

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
Early ARTTransmission Risk Behavior Outcome 2101 Participants
Deferred ARTTransmission Risk Behavior Outcome 280 Participants

Source: ClinicalTrials.gov · Data processed: Jun 8, 2026