Every sponsor running a glucagon-like peptide-1 (GLP-1) program has the same problem: more sites want in than the protocol can support, and the first screening call is usually short. In our experience, sites expanding from adjacent specialties — orthopedics, rheumatology, general cardiology — often lose that call early. Not because they lack ability, but because they answer feasibility questions the way sites answered them a decade ago: with enthusiasm, an equipment list, and a claim about patient access that nobody can verify.
Prior GLP-1 experience helps, but it is not the only evidence sponsors accept. What wins the call is transferable evidence — documented proof from the work you already do that maps directly onto what a GLP-1 trial demands. This article lays out what that evidence is, what carries no weight, and how to assemble it in about two weeks.
Why Sponsors Screen GLP-1 Feasibility Differently
GLP-1 trials are not ordinary outpatient studies, and experienced sponsors know where they fail:
- Demand cuts both ways. Public use of these drugs is already widespread: in a KFF poll fielded October 27 to November 2, 2025, 18% of US adults said they had ever taken a GLP-1 drug and 12% said they were taking one at the time. Recruitment interest is rarely the problem. The work is screening out volunteers who do not meet eligibility criteria (check each protocol’s rules on prior or current GLP-1 use) and managing the disappointment of those randomized to placebo.
- Retention is the real risk. These protocols run long. FDA’s January 2025 draft guidance on weight-reduction drugs describes long-term effect as at least one year on the maintenance dose. The STEP 1 trial of semaglutide 2.4 mg ran 68 weeks, with dose escalation every four weeks up to week 16. Gastrointestinal side effects cluster early: in a pooled analysis of STEP 1–3, they occurred most often during or shortly after dose escalation, and nausea, diarrhea and vomiting peaked around week 20. A site that enrolls fast and loses patients by month three is worse than a site that enrolls steadily and keeps them.
- Injection training is a protocol task, not a detail. In injectable programs, participants usually self-inject at home between visits. The Wegovy prescribing information, for example, tells prescribers to train patients on injection technique before the first dose and to delay escalation when a dose is not tolerated. Sites must teach technique, document training, and handle missed-dose and storage questions for the full study duration. Oral programs change the details, not the adherence burden.
- Metabolic monitoring must be exact. Weight, body mass index (BMI), hemoglobin A1c (HbA1c), and laboratory values are primary data, not background information. In FDA’s weight-reduction draft guidance, the primary efficacy endpoint is the mean percentage change in body weight; in FDA’s 2023 draft guidance on diabetes efficacy endpoints, HbA1c remains an acceptable primary endpoint for glycemic control. Sloppy measurement discipline shows up directly in the endpoint data.
Every feasibility question a sponsor asks is really one of these four risks in disguise. Answer the risk, not the question.
The Transferable Evidence That Actually Carries Weight
1. An adjacent cardio-metabolic enrollment record. You do not need a prior GLP-1 study. You need documented enrollment in conditions that share the same patient: type 2 diabetes, hypertension, dyslipidemia, cardiovascular risk-reduction studies. Name the therapeutic area, the enrollment target, what you actually randomized, and your screen-failure rate. Specifics from adjacent work beat adjectives about the new indication every time.
2. A verifiable count of your own eligible patients. Run a query against your electronic health record (EHR) system using the protocol’s headline criteria — age range, BMI threshold, HbA1c range, key exclusions — and report only the aggregate, de-identified count. Under the HIPAA Privacy Rule, reviewing records to prepare for research is a defined pathway with conditions, including that no protected health information leaves the covered entity (45 CFR 164.512(i)(1)(ii); see also HHS research guidance), so involve your privacy officer. For example, a sentence such as “214 active patients in our practice met the published inclusion criteria as of this month” (an illustrative number) is something a sponsor can act on. It proves the patients exist, that your team can operate its own records, and that your recruitment estimate has a denominator.
3. Principal investigator oversight in metabolic care. The principal investigator (PI) does not need a GLP-1 publication. The PI needs a credible story of managing metabolic patients over time: titration decisions, comorbidity management, and longitudinal follow-up. In our experience, sponsors are screening for clinical judgment across a long trial, not for a bibliography in one molecule class.
4. Documented startup speed. Give dates, not promises: your average time from contract execution to site activation, and from activation to first patient in (FPI), across your last several studies. For example, a site that can show a consistent 45-day path to activation (an illustrative figure), with the paperwork to prove it, is a lower-risk choice than a famous site with no timeline data.
5. A retention system, described concretely. Explain your reminder cadence, how you handle visit windows when a participant travels, who on your staff owns lifestyle and adherence support between visits, and your historical retention rate on your longest completed study. Retention is where GLP-1 trials are won or lost, and in our experience many sites improvise it.
What Does Not Work
- “Access to thousands of patients.” Without a query, a date, and criteria, this is noise — and reviewers hear it constantly.
- Equipment lists. Every credible site has the equipment. It is a threshold requirement, not evidence of capability.
- Generic enthusiasm for the indication. Interest in weight management is assumed. It differentiates no one.
- Borrowed credentials. Listing affiliated physicians who will not actually see study patients surfaces quickly — usually at the site selection visit — and can end the conversation for good.
A Two-Week Checklist for Sites Without GLP-1 Experience
- ☐ Pull your enrollment history for cardio-metabolic studies: targets, actual randomized numbers, screen-failure rates.
- ☐ Run one de-identified EHR query against a current GLP-1 protocol’s published criteria (with your privacy officer’s sign-off); date and save the aggregate output only.
- ☐ Write a one-paragraph PI summary focused on longitudinal metabolic care, not publications.
- ☐ Compile activation and FPI dates from your last three to five studies into a single timeline.
- ☐ Document your retention workflow: reminder cadence, visit-window handling, injection-training documentation, and the staff role responsible for between-visit support.
- ☐ Record your retention rate on your longest completed trial.
- ☐ Assemble these into a two-page capability brief you can send within one business day of any feasibility request.
Our Perspective
At bioaccess®, the first-in-human clinical research organization (CRO) for the Americas — the US, Canada and Latin America — we qualify sites and investigators for sponsors’ early-stage programs. Our sister company, Amavita Research Services, is a cardiovascular clinical research site in Miami, so we also answer these questionnaires from the site side. Working both sides taught us the lesson this article is built on: sponsors do not fund potential. They fund evidence. Sites that learn to present the evidence they already have — precisely, verifiably, and matched to the protocol’s real risks — get shortlisted in indications they have never run before. For what feasibility numbers look like on the recruitment side, see our article on patient recruitment feasibility for first-in-human trials.
FAQ
Can a site with no GLP-1 trial experience get selected for a GLP-1 study?
In our experience, yes. Sponsors do select sites without indication-specific experience when the site presents transferable evidence: documented enrollment in adjacent cardio-metabolic conditions, a verifiable count of eligible patients from its own records, and a concrete retention system.
What patient number should a site quote in a GLP-1 feasibility response?
An aggregate, de-identified count from the site’s own electronic health record (EHR) query, run against the protocol’s actual inclusion and exclusion criteria and dated. Keep the query within HIPAA’s preparatory-to-research conditions. An unverifiable estimate of “thousands of patients” carries no weight with reviewers.
What matters more in GLP-1 feasibility: enrollment speed or retention?
Retention, in our view. GLP-1 protocols are long — STEP 1 ran 68 weeks — and gastrointestinal side effects cluster during dose escalation. Sponsors screen for reminder systems, visit-window discipline, and between-visit support because a patient lost at month three costs more than a patient enrolled a month later.
How long does it take to build a credible GLP-1 feasibility package?
About two weeks for a site with organized records: one EHR query, an enrollment history summary, activation and first-patient-in timelines from recent studies, and a written retention workflow.
Sources: KFF Health Tracking Poll (fielded October 27 to November 2, 2025); FDA, Obesity and Overweight: Developing Drugs and Biological Products for Weight Reduction, draft guidance (January 2025); Wilding JPH et al., Once-Weekly Semaglutide in Adults with Overweight or Obesity, N Engl J Med 2021;384:989-1002 (STEP 1); Wharton S et al., Gastrointestinal tolerability of once-weekly semaglutide 2.4 mg in adults with overweight or obesity, Diabetes Obes Metab 2022 (pooled STEP 1–3); WEGOVY prescribing information (FDA, 2025); FDA, Diabetes Mellitus: Efficacy Endpoints for Clinical Trials Investigating Antidiabetic Drugs and Biological Products, draft guidance (May 2023); 45 CFR 164.512(i)(1)(ii); HHS, Research (HIPAA Privacy Rule guidance).