Beating a pediatric target with five sites, by ignoring volume

1. Background

The sponsor commissioned a pediatric serum specimen collection program supporting immunoassay validation for Hepatitis A and Hepatitis B serological assays. The clinical-validation question for both Hepatitis A (HAV) and Hepatitis B (HBV) immunoassays in the pediatric specimen matrix is whether assay performance characteristics validated in adult specimens transfer cleanly to pediatric samples; this requires a representative pediatric specimen library across the relevant age strata.²

The protocol enrolled across two parallel cohorts: HAV-pediatric (target 120 subjects) and HBV-pediatric (target 120 subjects), with a maximum of 15 reimbursable screen failures per cohort. The eligibility envelope used CDC-defined risk-factor lists for HAV and HBV exposure rather than a confirmed-positive filter; this is a population-screening problem rather than a chart-review problem, and the operational implication is that recruitment relies on identifying sites whose patient mix overrepresents CDC risk-factor populations.

Pediatric blood-volume restrictions per IRB guidance shaped specimen logistics. The protocol enforced pediatric-IRB blood-volume maxima — up to 30 mL for subjects 16+ at ≥50 lbs, dropping to 5 mL for subjects 2–12 under 50 lbs — against minimum yield requirements of 3 mL serum for subjects 12–21 and 2 mL for subjects 2–11. The protocol required at least 30% of enrolled samples from subjects under age 11.

2. Constraints encountered

Four constraints shaped the execution plan:

1. Risk-factor-based pediatric eligibility

   The eligibility envelope used CDC-defined risk-factor lists rather than a confirmed-disease filter. For HAV: international travelers, men who have sex with men, people who use or inject drugs, occupational exposure, close contact with an infected person, and others. For HBV: infants born to people with HBV, people born in countries with high HBV prevalence, people on dialysis, people who inject drugs, sex partners of infected individuals, and others. Recruitment is a site-population-mix problem, not a chart-review pull.

2. Pediatric IRB blood-volume restrictions

   Pediatric IRB blood-volume maxima are tighter than adult, scaling with age and weight: 5 mL for under-50 lb subjects ages 2–12, escalating to 30 mL at age 16+ at ≥50 lbs. Specimen yield per subject is constrained at the youngest age strata, and the protocol’s 30%-under-11 distribution requirement compresses the tightest specimen-yield strata into the largest portion of enrolment.

3. Two-cohort parallel enrolment

   Sites are enrolled into both HAV and HBV cohorts simultaneously, with overlapping but distinct CDC risk-factor lists. Subjects qualifying for one cohort may not qualify for the other; subjects qualifying for both can only be assigned to one. Site-level workflow has to track which cohort each subject is assigned to and which risk factors are enforced per cohort, increasing CRA review overhead per enrolled subject.

4. Pediatric specimen collection at primary care

   Most pediatric subjects are seen at primary-care pediatric practices rather than at academic medical centers. Primary-care pediatric sites have shallow research administration (limited dedicated CRC capacity, limited regulatory infrastructure) compared to academic sites. The site network for pediatric specimen collection skews toward sites that have built up research administration capacity at the primary-care level, which is a small and concentrated subset.

3. Methods

RDI executed the program as full-service CRO. The protocol was developed against the sponsor’s sample specifications, submitted to a central IRB, and executed across a 5-site network selected against three independent inputs:

• Patient-population claims data — insurance claims data showing the subset of each candidate site’s pediatric volume aligning with CDC risk-factor populations (international travel patterns, immigrant-population density, occupational exposure profiles, etc.)
• CDC regional risk-factor demographics — geographic distribution of HAV and HBV risk-factor populations, used to weight site selection toward regions with overrepresented risk-factor density
• Prior performance in pediatric specimen collection — documented enrolment rate, screen-failure rate, and protocol-deviation rate from prior RDI programs at candidate sites

Five sites activated for enrolment, distributed across pediatric primary-care, primary-care, and specialty research practices:

• Site A — specialty pediatric research site, lead enroller across both cohorts
• Site C — primary-care site with research administration, late-window enroller
• Site D — pediatric specialty practice
• Site B — multi-practice primary-care group with risk-factor-overrepresented patient mix
• Site E — gastroenterology / hepatology specialty practice

Sites were trained on parallel HAV/HBV enrolment workflows with cohort-specific risk-factor enforcement. Specimen collection at the research visit produced serum aliquots at sponsor-defined volumes per the IRB blood-volume schedule, processed and frozen at site, and shipped weekly on dry ice to RDI for inventory and onward shipment to the sponsor.

4. Results

Cumulative enrolment trajectory was approximately linear from program open to peak ramp month, with a closeout tail following targets achieved (Fig. 1). Monthly enrolment was: November 2024 = 13, December 2024 = 23, January 2025 = 37, February 2025 = 43, March 2025 = 58, April 2025 = 76 (peak ramp month), and May 2025 = 10 (closeout tail). Cumulative enrolment crossed 36 by end-December 2024 (15% of target in the first two months), 116 by end-February 2025 (48% of target in 4 months), and 250 by end-April 2025 (104% of target in 6 months).

The April 2025 ramp month carried 76 enrolments — 29% of total program volume in a single calendar month — and reflects the convergence of three sites (Site A, Site C, Site B) running at peak weekly cadence simultaneously. May 2025 closeout activity captured residual scheduled subjects across 4 of the 5 sites; closeout shipment activities and final data transfer continued into June 2025.

Site-level enrolment was concentrated: Site A carried 106 of 260 subjects (41%), Site B 90 (35%), Site C 34 (13%), Site D 16 (6%), and Site E 14 (5%). The two lead sites (Site A + Site B) carried 75% of program volume; the three remaining sites carried the remaining 25% across a combined 64 subjects (Table 1). The lead-site concentration is consistent with patient-mix density — sites whose pediatric volume overrepresents CDC risk-factor populations enrol disproportionately faster than sites with broader pediatric patient mixes.

The program closed at 260 enrolled / 240 target (+20 subjects, +8.3%), with no protocol amendments required for either cohort. The 7-month enrolment window from first to last enrolment matches the SOW projection of T+4 weeks first SIV to T+7 months last enrolment.

5. Discussion

Three operational points are consequential for sponsors planning analogous pediatric specimen collection programs against CDC risk-factor eligibility envelopes.

First, site selection is the dominant lever. Risk-factor-based eligibility shifts the recruitment problem from chart review (find the documented-positive subjects) to population screening (find the subjects whose presenting demographics align with CDC risk factors). Sites whose pediatric patient mix overrepresents CDC risk populations enrol disproportionately faster than sites with broader patient mixes; the 41% Site A lead site contribution is consistent with this. Sponsors should expect a 2–3 lead site distribution at well-selected networks rather than a balanced spread, and should plan the site list against patient-mix density inputs (claims data, demographic overlays) at the program design phase.

Second, the pediatric blood-volume restriction interacts with the under-11 distribution requirement. The protocol’s 30%-under-age-11 floor lands the largest subject share at the lowest specimen-yield strata (5 mL maximum draw for under-50 lb subjects). Specimen logistics — SST tube selection, processing volume, freezer cycle — have to be designed against the lowest-yield strata as the modal subject, not against the best-case adult-equivalent volume. Plan the specimen pipeline against the pediatric floor, not the adolescent ceiling.

Third, two-cohort parallel enrolment is a tractable operational pattern when the eligibility envelopes overlap meaningfully. The HAV and HBV CDC risk-factor lists overlap on injection drug use, occupational exposure, sexual contact, and immigrant origin; subjects qualifying for one cohort frequently qualify for the other, and site workflows can be designed to assign subjects to whichever cohort is further from target at the moment of enrolment. This is more efficient than running two single-cohort programs sequentially.

6. Conclusion

Pediatric Hepatitis A and Hepatitis B specimen collection at scale is achievable through a small site network selected for patient-mix overlap with CDC risk-factor populations rather than for broad pediatric volume. The program closed at 260 enrolled / 240 target (+8.3%) within the originally projected 7-month enrolment window, with no protocol amendments required for either cohort. The dominant operational lesson — site selection against patient-mix density determines yield more than site count or per-site enrolment-rate optimisation — is consistent with prior RDI engagements against narrow-prevalence eligibility envelopes.