Every FDA-approved drug has been through a multi-phase clinical trial program. The phases test safety, efficacy, and dosing in progressively larger populations, ending with trials that randomize thousands of patients to drug or placebo. Reading these trials is the most reliable way to understand what a medication does and does not do.<\/p>\n
The GLP-1 class has been studied more extensively than almost any other drug category in modern medicine. STEP for semaglutide, SURMOUNT for tirzepatide, SUSTAIN for diabetes outcomes, SELECT for cardiovascular, FLOW for kidney, and many more. The trial names follow a sponsor convention but each refers to a specific protocol and patient population.<\/p>\n
Knowing how to read these trials lets you compare drugs, understand what to expect from treatment, and recognize when claims are not backed by data.<\/p>\n
At TrimRx, we believe that understanding your options is the first step toward a more manageable health journey. You can take the free assessment quiz if you’re ready to see whether a personalized program is a fit for you.<\/p>\n
Phase 0 is a small pilot study, sometimes called microdosing, in 10 to 15 volunteers to confirm pharmacokinetics.<\/strong> Not all drugs go through phase 0.<\/p>\n Quick Answer: Phase 1 tests safety in small groups, phase 2 tests efficacy in hundreds, phase 3 in thousands<\/p>\n Phase 1 tests safety and dosing in 20 to 100 healthy volunteers or patients. The main goal is to identify the maximum tolerated dose and basic pharmacology. Phase 1 takes about 6 months to a year.<\/p>\n Phase 2 tests efficacy in 100 to 500 patients with the target condition. This is where the drug starts showing whether it works for its intended use. Phase 2 also refines dose selection. Trials typically last 1 to 2 years.<\/p>\n Phase 3 is the key efficacy testing, with 1,000 to 5,000 or more patients across multiple sites and often multiple countries. Phase 3 trials are the basis for FDA approval and typically run 2 to 4 years.<\/p>\n Phase 4 is post-approval surveillance, where the drug continues to be studied in larger populations to detect rare adverse events and confirm real-world effectiveness.<\/p>\n A randomized controlled trial (RCT) randomly assigns patients to receive either the experimental treatment or a control (placebo or standard care).<\/strong> Randomization balances known and unknown patient characteristics across groups, which lets the trial attribute outcome differences to the treatment.<\/p>\n Double-blinding means neither patients nor researchers know who got which treatment until the data is locked. This prevents bias in outcome assessment.<\/p>\n STEP 1 was a double-blind, placebo-controlled phase 3 RCT that randomized 1,961 adults with BMI 30 or higher (or 27 with comorbidities) to semaglutide 2.4 mg or placebo for 68 weeks. The randomization, blinding, and large sample size make the results highly reliable.<\/p>\n Every trial specifies a primary endpoint before the trial starts.<\/strong> This is the main outcome the study is designed to detect. In STEP 1, the primary endpoint was percent change in body weight at 68 weeks.<\/p>\n Secondary endpoints are additional outcomes specified in advance. In STEP 1, secondary endpoints included the proportion of patients achieving 5%, 10%, and 15% weight loss, changes in cardiovascular risk factors, and changes in physical function.<\/p>\n Pre-specifying endpoints prevents “fishing” through the data for whatever happens to be significant. Trials that succeed on the primary endpoint are considered positive. Trials that miss the primary but find something interesting in secondary endpoints are usually hypothesis-generating, not definitive.<\/p>\n Wilding et al. published STEP 1 in NEJM in 2021. The headline result was that semaglutide 2.4 mg produced 14.9% weight loss at 68 weeks, compared to 2.4% with placebo. The difference was highly statistically significant.<\/p>\n Secondary endpoints were also strong. 86% of patients on semaglutide lost at least 5% of body weight, 69% lost 10%, 50% lost 15%, and 32% lost 20%. Cardiovascular risk factors including blood pressure, lipids, and inflammatory markers improved significantly.<\/p>\n The trial defined a new standard for pharmacologic weight loss. Before STEP 1, the best non-surgical results were typically 5 to 8% with older drugs like phentermine-topiramate or naltrexone-bupropion. STEP 1 doubled or tripled those numbers.<\/p>\n Jastreboff et al. published SURMOUNT-1 in NEJM in 2022. The trial randomized 2,539 adults with obesity to tirzepatide 5, 10, or 15 mg, or placebo, for 72 weeks.<\/p>\n Weight loss was dose-dependent. The 5 mg dose produced 15.0% loss, the 10 mg dose 19.5%, and the 15 mg dose 20.9%, all compared to 3.1% in placebo. The proportion of patients achieving 20% weight loss reached 57% at the highest dose.<\/p>\n This made tirzepatide the most effective pharmacologic weight loss treatment to that point, exceeding semaglutide by 5 to 6 percentage points at the highest dose. The dual GLP-1\/GIP receptor agonism appeared to provide meaningful additional benefit.<\/p>\n SELECT (Lincoff et al.<\/strong> 2023, NEJM) was a cardiovascular outcomes trial of semaglutide 2.4 mg in 17,604 adults with established cardiovascular disease and BMI 27 or higher, but without diabetes. Follow-up averaged about 40 months.<\/p>\n The primary endpoint was a composite of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. Semaglutide reduced this composite by 20% compared to placebo. The benefit was driven by reductions in MI and CV death, with smaller effects on stroke.<\/p>\n SELECT was important because it showed cardiovascular benefit in non-diabetic patients with obesity. Previously, GLP-1 cardiovascular outcomes data came from diabetes populations. SELECT extended the benefit to a broader obesity treatment indication.<\/p>\n Intent-to-treat (ITT) analysis includes all randomized patients in the analysis, regardless of whether they completed treatment.<\/strong> If a patient drops out, their last observation is carried forward or imputed by statistical methods.<\/p>\n This is important because dropouts in trials are often not random. Patients who experience side effects or feel they are not benefiting may stop, and excluding them would bias results favorably for the drug. ITT analysis prevents this bias.<\/p>\n Per-protocol analysis only includes patients who completed treatment as assigned. This sometimes shows larger drug effects but is more prone to bias.<\/p>\n STEP 1 reported both ITT and on-treatment results. The 14.9% weight loss was the on-treatment number; the ITT result, which included dropouts, was slightly lower but still highly significant.<\/p>\n Statistical significance refers to the probability that an observed difference could have occurred by chance alone.<\/strong> A p-value below 0.05 conventionally means the result is statistically significant.<\/p>\n But p-values are not effect size. A trial can have a statistically significant result with a small clinical effect, especially if the sample size is large. STEP 1 reported both p-values (extremely small, indicating high statistical confidence) and effect sizes (large weight loss differences).<\/p>\n Clinical significance is separate from statistical. A 1% weight loss difference might be statistically significant in a 10,000-patient trial but clinically meaningless. A 15% weight loss difference is both statistically and clinically significant.<\/p>\n Key Takeaway: SELECT (Lincoff et al. 2023 NEJM) showed 20% MACE reduction with semaglutide<\/p>\n Trials track and report all adverse events that occur during the study, organized by frequency and severity.<\/strong> Common adverse events for GLP-1 drugs include nausea (about 44% in STEP 1), diarrhea (30%), and constipation (24%).<\/p>\n Serious adverse events are tracked separately. These are events that result in hospitalization, disability, or death. In STEP 1, serious adverse events occurred in 9.8% of semaglutide patients and 6.4% of placebo patients.<\/p>\n Discontinuation due to adverse events is another key metric. In STEP 1, 7.0% of semaglutide patients discontinued due to side effects compared to 3.1% on placebo. This is meaningful for understanding tolerability.<\/p>\n A meta-analysis combines results from multiple trials to produce a pooled estimate of treatment effect.<\/strong> When individual trials are too small to detect a difference or when results vary between studies, meta-analyses can provide more reliable conclusions.<\/p>\n The GLP-1 cardiovascular meta-analyses have pooled SELECT, SUSTAIN-6, LEADER, REWIND, and other trials to estimate the overall cardiovascular benefit of the class. The pooled estimate is approximately 14% reduction in MACE across the GLP-1 receptor agonist class.<\/p>\n Meta-analyses are only as good as the trials they include. Combining well-designed RCTs gives reliable conclusions. Combining observational studies or poorly designed trials can amplify biases.<\/p>\n TrimRx clinical protocols follow the dosing schedules and titration patterns established in the STEP and SURMOUNT trials.<\/strong> The trial data guides starting doses, escalation timelines, and expectations for response.<\/p>\n A personalized treatment plan accounts for individual response variability that the trials show. About 32% of STEP 1 patients lost 20% or more, while a smaller percentage lost less than 5%. The clinical team adjusts based on what trials predict and what the patient experiences.<\/p>\n A free assessment quiz starts the process. Clinical review applies the trial-based evidence to the individual patient profile.<\/p>\n Investigator-initiated trials are studies designed and run by academic investigators rather than pharmaceutical sponsors.<\/strong> These trials often explore questions that sponsors do not prioritize, such as direct comparisons between competing drugs or use in special populations.<\/p>\n The STEP and SURMOUNT trials were sponsor-initiated. Comparative effectiveness studies of semaglutide versus tirzepatide have largely been investigator-initiated or sponsored by health systems. SURPASS-2 is one of the few head-to-head trials with sponsor involvement.<\/p>\n Investigator-initiated trials typically have smaller budgets and sample sizes than sponsor trials. The methodology can be rigorous but power may be limited for detecting smaller effects. The trials add important complementary evidence to the sponsor-controlled trial database.<\/p>\n Real-world evidence comes from observational data including electronic health records, insurance claims, and registries.<\/strong> These data sources include much larger and more diverse populations than RCTs but lack randomization.<\/p>\n Real-world studies can detect rare adverse events, long-term outcomes, and effectiveness in patient populations excluded from trials. They can also be confounded by unmeasured differences between treatment groups in ways that RCTs avoid.<\/p>\n The FDA increasingly considers real-world evidence in regulatory decisions, particularly for label expansions and post-approval safety. For GLP-1 medications, real-world data has supported the trial findings and added information about adherence patterns and long-term tolerability.<\/p>\n Patient-reported outcomes (PROs) capture how patients feel and function from their own perspective.<\/strong> PROs supplement clinical measurements like weight and lab values with subjective experience.<\/p>\n For obesity trials, PROs include quality of life scales, hunger ratings, food cravings, body image, and depression symptoms. STEP 1 reported improvements in physical functioning, mental health, and overall quality of life on semaglutide alongside the weight loss data.<\/p>\n PROs are increasingly required for FDA approval and pricing negotiations. They capture aspects of treatment value that clinical measurements miss and help inform real-world clinical decision making.<\/p>\n Bottom line: Intent-to-treat analysis prevents bias from dropouts<\/p>\n Phase 2 is smaller (hundreds of patients) and tests preliminary efficacy. Phase 3 is larger (thousands) and is the key trial for FDA approval.<\/p>\n Placebos control for the natural course of disease, regression to the mean, and patient expectations. Without a placebo arm, you cannot tell if the drug actually worked.<\/p>\n The major GLP-1 trials are large, well-conducted, and published in peer-reviewed journals like NEJM and Lancet. The methodology meets the highest standards in clinical research.<\/p>\n A p-value below 0.05 conventionally means the observed difference is unlikely to be due to chance. It does not mean the difference is clinically important.<\/p>\n Patient populations, doses, durations, and outcome measures differ between trials. Comparing trials directly requires careful attention to these differences.<\/p>\n Usually somewhat lower. Trials enroll selected patients with good adherence, while real-world populations have more variability. The general direction is the same but magnitudes differ.<\/p>\n The full path from phase 1 to approval typically takes 8 to 12 years. Accelerated pathways and priority review can shorten this for drugs treating serious conditions.<\/p>\n Disclaimer:<\/strong> This content is for informational purposes only and does not constitute medical advice. It is not intended to diagnose, treat, cure, or prevent any disease or condition. Individual results may vary. Always consult a qualified healthcare professional before starting any weight loss program or medication.<\/p>\n <\/p>\n Every FDA-approved drug has been through a multi-phase clinical trial program.<\/p>\n","protected":false},"author":11,"featured_media":92662,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"inline_featured_image":false,"_yoast_wpseo_title":"How Clinical Trials Work: Understanding GLP-1 Study Data","_yoast_wpseo_metadesc":"Every FDA-approved drug has been through a multi-phase clinical trial program.","_yoast_wpseo_focuskw":"clinical trials explained","footnotes":"","_flyrank_wpseo_metadesc":""},"categories":[6],"tags":[29,41],"class_list":["post-89267","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-glp-1","tag-glp-1","tag-research"],"_links":{"self":[{"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/posts\/89267","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/users\/11"}],"replies":[{"embeddable":true,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/comments?post=89267"}],"version-history":[{"count":2,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/posts\/89267\/revisions"}],"predecessor-version":[{"id":93663,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/posts\/89267\/revisions\/93663"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/media\/92662"}],"wp:attachment":[{"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/media?parent=89267"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/categories?post=89267"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/trimrx.com\/blog\/wp-json\/wp\/v2\/tags?post=89267"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}What Is a Randomized Controlled Trial?<\/h2>\n
What Are Primary and Secondary Endpoints?<\/h2>\n
What Did STEP 1 Actually Show?<\/h2>\n
What Did SURMOUNT-1 Show?<\/h2>\n
What Did SELECT Show About Cardiovascular Outcomes?<\/h2>\n
What Is Intent-to-treat Analysis?<\/h2>\n
What Is Statistical Significance?<\/h2>\n
How Are Adverse Events Reported?<\/h2>\n
What Is a Meta-analysis?<\/h2>\n
How Does TrimRx Use Trial Data?<\/h2>\n
What Is the Role of Investigator-initiated Trials?<\/h2>\n
How Are Real-world Evidence Studies Different From RCTs?<\/h2>\n
What Is the Role of Patient-reported Outcomes?<\/h2>\n
FAQ<\/h2>\n
What Is the Difference Between Phase 2 and Phase 3 Trials?<\/h3>\n
Why Are Placebos Used?<\/h3>\n
Are GLP-1 Trials Reliable?<\/h3>\n
What Does a P-value Mean?<\/h3>\n
Why Do Effect Sizes Vary Between Similar Trials?<\/h3>\n
Is Real-world Effectiveness the Same as Trial Efficacy?<\/h3>\n
How Long Does FDA Approval Take?<\/h3>\n
Related Articles<\/h2>\n
\n