Pre-Operational Validation of a Deviation-Ready QMS for Source Plasma Centers: Readiness Metrics and Hematology Supply Implications PDF Free Download
1 / 24
/24
100%
Academic Editor: Weiyong Liu
and Glen L. Hortin
Received: 1 November 2025
Revised: 8 December 2025
Accepted: 30 December 2025
Published: 14 January 2026
Copyright: © 2026 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license.
Article
Pre-Operational Validation of a Deviation-Ready QMS for
Source Plasma Centers: Readiness Metrics and Hematology
Supply Implications
Ankush U. Patel 1,*, Ryan McDougall 2and Samir Atiya 3
1Department of Pathology, Wexner Medical Center, The Ohio State University, Columbus, OH 43210, USA
2Superhero Biologics, Santa Monica, CA 90401, USA; rmcdougall1987@gmail.com
3Department of Pathology, The University of Chicago Medical Center, Chicago, IL 60637, USA;
samir.atiya@uchicagomedicine.org
*Correspondence: ankush@digitalpathomics.com
Abstract
Source plasma centers sustain hematology therapeutics by safeguarding testing, traceability,
and cold-chain integrity before fractionation. Despite regulatory requirements (21 CFR
606/640; EU Directive 2005/62/EC), published pre-operational validation frameworks
demonstrating deviation-readiness before first collections remain sparse. We conducted a
simulation-based pre-operational validation of an electronic quality management system
(eQMS) with an Incident
→
Deviation
→
Corrective Action and Preventive Action (CAPA)
pathway at a new source plasma center, performing 20 chairside mock runs, 3 freezer-alarm
drills, and a document-control stress test. Primary endpoints were anomaly rate, alarm-
response time relative to a 15 min service-level agreement (SLA), and deviation-closure SLA
compliance. Analyses were descriptive and designed to demonstrate system functionality,
not long-term process stability. Minor anomalies occurred in 6/20 mock runs (30.0%; 95% CI
11.9–54.3); no major/critical events were observed (0/20; 95% CI 0–16.8). Deviation-closure
SLAs were met in 6/6 tests (100%; 95% CI 54.1–100). Alarm-response times averaged
7.0 min
(SD 1.0; range 6–8 min; 95% CI 4.5–9.5), and all drills met the 15 min vendor SLA,
illustrating a preliminary readiness margin (Cpu
≈
2.7) rather than a statistically stable
capability estimate. Simulation-based pre-operational validation produced inspection-
ready documentation and quantitative acceptance criteria aligned to U.S./EU expectations,
supporting reproducible multi-site deployment. By protecting cold-chain integrity and
traceability before first collections, the validated QMS helps preserve supply reliability for
plasma-derived therapeutics central to hematology care and establishes the measurement
infrastructure for post-operational performance validation.
Keywords: source plasma; hematology; quality management system (QMS); eQMS; CAPA;
BPDR; 21 CFR 606/640; EU 2005/62/EC; EDQM Blood Guide; EU GMP Annex 11;
ALCOA+; simulation; SPC; process capability; pre-operational validation
1. Introduction
1.1. Rationale and Gap
Quality oversight for source plasma centers is codified in 21 CFR Parts 606/640,
with deviation reporting under 606.171 (see Table 1for regulatory crosswalk). European
frameworks similarly mandate a documented quality system and technical standards via
Commission Directive 2005/62/EC and the EDQM Blood Guide (22nd ed., 2025) [
1
]. Yet
LabMed 2026,3, 2 https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 2 of 24
peer-reviewed, pre-operational validation models that demonstrate “deviation-readiness”
before first collections are sparse [
2
]. This pre-operational validation demonstrates that
quality system components function correctly under controlled simulation conditions
(framework validation). It does not establish sustained operational performance, which
requires longitudinal monitoring after launch (process validation). This framework opera-
tionalizes requirements derived from published statutes, regulations, and guidance, and is
intended as an operational readiness and inspection-preparation tool.
Table 1. Regulatory crosswalk of key quality system elements.
Control Element U.S. Alignment EU Alignment
Document/Change Control 606.100 (b) SOPs; 606.160 records 2005/62/EC QMS; EDQM quality
system
Incident →Deviation →
CAPA
606.100 (b) investigations; 606.171 BPDR
(Form 3486; ≤45 days)
2005/62/EC
non-conformance/self-inspection;
EDQM QMS
Donor eligibility/screening 21 CFR Part 630 (e.g., 630.10, 630.15); 640.65 EDQM donor selection;
2005/62/EC personnel/procedures
Sampling and traceability 606.160/165 records/distribution EDQM traceability
Packing/release/shipping
606.100; 606.120–122 labeling; 640.72; 600.15
(shipping temperatures); 640.76 (unacceptable
temperatures); 606.165 records
EDQM storage/distribution
Cold-chain alarms/excursions
606.60–65; 606.100
EDQM storage temperature control
eQMS and data integrity 21 CFR Part 11 expectations EU GMP Annex 11 (computerized
systems)
EDQM: European Directorate for the Quality of Medicines and HealthCare; BPDR: biological product
deviation report.
1.2. Hematology Supply-Chain Context
Source plasma is the starting material for plasma-derived therapeutics, including
FVIII, FIX, PCC, and IVIG. Over the past decade, global immunoglobulin (Ig) demand has
increased by roughly 6–8% annually, driven by broader indications and access expansion,
emphasizing the need to prevent avoidable losses from cold-chain excursions, labeling
errors, or breaks in traceability [
3
]. This study presents a reproducible, evidence-generating
validation framework (including procedures, acceptance criteria, and audit-ready artifacts)
designed to protect supply integrity prior to first collections.
Recent cold-chain failures in research and clinical contexts demonstrate that alarms and
traceability are decisive controls, and that rehearsed protocols can prevent large-
scale loss
:
•
The Harvard Brain Tissue Resource Center (McLean Hospital, Belmont, Massachusetts)
lost 147 samples in 2012, when ‘two separate [freezer] alarms that should have alerted
staff to the problem failed to sound’. As
1/3
of these samples represented a substantial
collection dedicated to autism research, researchers predicted the loss would ‘slow
autism research by a decade.’ [4–6].
•
A 2023 liquid-nitrogen failure at Karolinska Institutet (Stockholm, Sweden) destroyed
decades of leukemia research valued at approximately £37 million across 16 cryogenic
tanks [7].
By validating deviation-ready controls for alarms, traceability, and documentation
prior to first collections, this QMS supports cold-chain integrity and regulatory storage
expectations for plasma components and helps sustain continuity of plasma-derived FVIII,
FIX, and IVIG and related plasma-derived medicinal products (PDMPs) amid sustained
global Ig demand growth (~6–8% CAGR).
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 3 of 24
1.3. EU Alignment
To harmonize with fractionator expectations and potential EU-directed shipments, U.S.
controls are aligned with Directive 2005/62/EC and the EDQM Blood Guide (
22nd ed.
).
EU GMP Annex 11 principles are applied to computerized systems. Current EU SoHO
legislation (Reg. 2024/1938) maintains these directives in force during transition, so this
framework remains aligned with the applicable requirements at the time of validation.
1.4. Objective and Hypothesis
Objective: Describe a practical, pre-operational validation framework that (1) aligns
with U.S./EU expectations, (2) uses simulation-based metrics, and (3) yields artifacts
suitable for inspection and multi-center scale-up.
Hypothesis: A simulation-anchored QMS can produce quantifiable pre-operational
indicators of deviation-readiness (e.g., anomaly rate, response time, SLA compliance)
with documented acceptance criteria and eQMS traceability. This study validates the pre-
operational framework and system functionality (e.g., alarm routing, documentation, CAPA
linkages) rather than long-term process stability or capability, as such require post-launch
operational data and are outside the scope of this report.
2. Materials and Methods
2.1. Setting and Design
Pre-operational validation was conducted at a single U.S. source plasma center in
an urban setting. The facility operates on utility power without an on-site generator. A
documented dry-ice contingency supports temporary backup storage and transfer during
outages. The cold chain includes a
−
30
◦
C class plasma freezer (Helmer iPF125) configured
with an alarm setpoint of
−
32
◦
C, which served as the platform for alarm and excursion
simulations. Each cold-storage unit is uniquely labeled in the equipment inventory. Two
automated Fresenius-Kabi 6R4601 apheresis devices (‘Aurora Plasmapheresis System’)
were installed, and high-priority equipment (e.g., bench-top centrifuge (Spectrafuge 6C or
equivalent), donor scale, BP device, tachometer, check weights) held current calibration
and qualification at the time of validation with International Organization for Standards
(ISO)-traceable certificates on file. A validated electronic QMS (eQMS) enforced document
control, training gates, and Deviation-to-Corrective Action and Preventive Action (CAPA)
workflows, and a donor-management/cross-donation system (v6.0.1) was configured for
site operations. No donors or biological materials were used in simulations, and no private
health information (PHI)/personally identifiable information (PII) was generated. Log
captures used for evidence collection excluded any identifiers.
2.2. Regulatory Alignment
Framework development addressed 21 CFR Parts 606/640 (and related 610.40 testing),
606.171 (BPDR), and EU quality standards (2005/62/EC, EDQM Blood Guide, 22nd ed.).
Clause-level mapping is provided in Table 1(Regulatory Crosswalk) [2,8–12].
2.3. QMS Framework and Computerized Systems
Quality system scope. The operational control framework comprised ten core ele-
ments: Document/Change Control; Management Review and Internal Audit; Incident →
Deviation
→
CAPA; Donor Eligibility; Collection; Sampling and Traceability; Packing and
Release; Emergency Shipments; Cold-Chain Alarm/Excursion Response; and Training and
Records. These elements align with 21 CFR Part 606 (current GMP for blood components)
and European Directive 2005/62/EC quality system requirements.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 4 of 24
Electronic QMS (eQMS) validation. To ensure data integrity and regulatory compli-
ance, the eQMS underwent formal validation following 21 CFR Part 11 (electronic records
and signatures) and EU GMP Annex 11 (computerized systems). The validation scope
was defined using risk-based principles per ICH Q9(R1), with particular emphasis on
GMP-critical functions: version control, electronic approvals with timestamps, training
gates that prevent operations on outdated procedures, and automatic record linkages
(deviation←→CAPA←→attachments).
Data integrity framework (ALCOA+). System validation verified that all electronic
records meet ALCOA+ principles (the international standard for data integrity in regulated
environments). ALCOA+ requires that records be Attributable (traceable to the individual
who created them), Legible (readable throughout the record lifecycle), Contemporaneous
(recorded in real time), Original (first capture or certified copy), Accurate (error-free),
Complete (all data present), Consistent (timestamps follow sequence), Enduring (durable
throughout retention), and Available (retrievable for review). These attributes were verified
during Installation Qualification (system installed correctly), Operational Qualification
(functions perform as specified), and Performance Qualification (system performs reliably
under real-world conditions).
Change control and record retention. Records are retained per 21 CFR 606.160 (mini-
mum 10 years, or 6 months beyond expiration date, whichever is longer. For source plasma
with no assigned expiry, this corresponds to at least a 10-year retention period. Changes
to GMP-critical system functions (audit trails, approval routing, training gates) trigger
mandatory re-validation to ensure continued compliance. See Supplementary S3 for the
eQMS validation summary, drill work instructions, and the change-control matrix.
2.4. BPDR Pathway (Biological Product Deviation Reporting)
Regulatory requirement. Under 21 CFR 606.171, manufacturers must report to the
FDA any deviation that affects or has the potential to affect the safety, purity, or potency
of distributed product within 45 calendar days of detection. Reports are submitted using
Form FDA 3486 (available electronically or on paper). Importantly, deviations involving
a product that remains in quarantine (never released/distributed) do not require BPDR
submission but still undergo full internal investigation and CAPA.
eQMS workflow for BPDR assessment. The eQMS automatically flags whether
the affected product has been distributed by checking release and shipment timestamps.
For distributed-product deviations, the system routes to corporate quality assurance
(QA)/regulatory affairs (RA) for rapid regulatory assessment:
•Internal escalation: ≤24 h (local QA-to-Corporate QA/RA).
•BPDR determination: ≤72 h (Corporate QA/RA decision on reportability).
•FDA submission: Within 45 calendar days (if reportable).
These internal timelines ensure the 45-day regulatory deadline is met with time to
spare. The essential escalation steps are summarized in Table 2.
Table 2. Internal escalation timelinhe for BPDR assessment and reporting.
Step Responsible Target Timeline
Local deviation detection and escalation Site QA ≤24 h from event detection
BPDR reportability decision Corporate QA/RA ≤72 h from escalation
FDA BPDR submission (if reportable) Corporate QA/RA ≤
45 days from detection (21 CFR 606.171)
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 5 of 24
2.5. Simulation Design and Fidelity
Scenario development. Simulation scenarios were derived from a rapid risk assess-
ment combining Failure Mode and Effects Analysis (FMEA) and Hazard and Operability
Study (HAZOP) methods. This assessment mapped the collection
→
labeling
→
sampling
→
packing/release workflow to identify high-likelihood and high-impact failure modes—
particularly those involving handoffs between staff, out-of-specification (OOS) equipment,
and alarm conditions.
Each scenario was designed to generate a complete evidence chain: Alarm Drill
Log →Alarm
Investigation Report
→
Excursion Investigation (with temperature graph)
→
Product Movement Record
→
Deviation Report
→
CAPA
→
QA Review. This chain
mirrors what inspectors would expect to see during a regulatory audit. Representative
templates for these evidence steps are included in Supplementary S3 and are available from
the authors upon request.
Fidelity framework. To ensure simulations reflected real-world conditions, we priori-
tized three types of fidelity:
1.
Functional fidelity: System behaviors replicate reality (e.g., vendor call/ticket genera-
tion, QA sign-off with approval authority).
2.
Physical/environmental fidelity: Actual equipment and documentation (temperature
monitors, reference thermometers, eQMS forms).
3.
Psychological fidelity: Realistic stressors (time pressure, role assignments, account-
ability for outcomes).
Drill checklists rated fidelity on a 1–5 scale across five dimensions (environment,
instruments, vendor engagement, QA review, evidence completeness). Failure on any
element invalidated the drill and was labeled a “fail.”
2.6. Simulation Protocols
Three simulation types were conducted:
1.
Chairside mock runs (n= 20): Complete end-to-end walkthrough of the donor
collection process, from registration through sample labeling and documentation.
Any observed anomaly, no matter how minor, triggered a formal Deviation report
and, if warranted, a CAPA.
2.
Cold-chain alarm drills (n= 3): Full execution of the freezer alarm response protocol,
generating the complete evidence chain described above (Alarm Drill Log through
final QA review). For each alarm drill, the protocol followed the following steps:
i. Trigger or simulate a freezer alarm at the −30 ◦C plasma freezer.
ii. Record alarm time and the initiating staff member in the Alarm Drill Log.
iii. Verify temperature with an independent calibrated reference thermometer.
iv.
Implement interim control (secure product, prepare backup freezer or dry-ice
transfer if needed).
v. Contact vendor service and document the call and ticket number.
vi.
Complete alarm and excursion investigation forms, including
temperature graph
.
vii.
Route the deviation and associated CAPA to QA for review and elec-
tronic approval.
3.
Document-control stress test: A complete change-control cycle testing the eQMS
workflow: Change Request
→
multi-level approvals
→
version release
→
training
gate enforcement (staff cannot proceed until attestation is complete)
→
effective
date →archival of superseded version.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 6 of 24
2.7. Definitions and Acceptance Criteria
Severity classification. All deviations are classified using a three-tier system (consis-
tent with FDA guidance on CGMP deviation handling) that determines escalation, CAPA
requirements, and potential BPDR reporting. This three-tier schema is consistent with
risk-based categorization in ICH Q9(R1) and with internal severity tiers commonly used to
support FDA BPDR determinations under 21 CFR 606.171:
•
Minor: No impact on product safety, purity, or potency; promptly correctable with
simple intervention; containment achieved ≤24 h; does not trigger BPDR reporting.
•
Major: Could affect product safety, purity, potency, or traceability if not promptly
controlled; requires formal CAPA; product cannot be distributed until risk is mitigated
and QA releases.
•
Critical: Actual or likely impact on product safety, purity, potency, or traceability; or
indicates systemic control failure; requires immediate containment; triggers BPDR
assessment if any affected product was distributed (per 21 CFR 606.171).
Alarm drill acceptance criteria (all must be met for “pass”):
i. The temperature display versus the reference thermometer agrees within ±1◦C.
ii. Vendor response/callback achieved within ≤15 min.
iii. Interim control measures documented (e.g., product secured, backup contacted).
iv. QA review and validation completed with formal approval.
v. Complete evidence chain attachments present in eQMS.
Service-level agreements (pre-operational targets; risk-based):
•
Containment (interim control):
≤
24 h (limits time window for additional prod-
uct exposure).
•
Vendor callback/engagement:
≤
15 min (ensures rapid technical support for equip-
ment failures).
•
QA review:
≤
72 h from event opening (maintains inspection readiness and pre-
vents backlog).
•
Deviation closure:
≤
30 days (pre-operational target) unless CAPA effectiveness
testing requires a longer timeframe.
Rationale for SLA targets. These timelines minimize cold-chain excursion risk for
plasma destined for fractionation (Per EDQM Blood Guide (22nd ed.), FFP is stored
36 months
at
−
25
◦
C or below, or 3 months at
−
18
◦
C to
−
25
◦
C; freezing should achieve
core
≤ −
25
◦
C within 1 h; maintain frozen during transport) and ensure inspection-ready,
auditable documentation trails. Rapid containment protects both product integrity and
supply chain reliability.
Figure 1illustrates the complete Incident
→
Deviation
→
CAPA workflow enforced
by the eQMS, showing decision points, severity classifications, and the service-level agree-
ments tested during pre-operational validation. The pathway distinguishes between quar-
antined and distributed products, triggers Corporate QA/RA escalation for BPDR as-
sessment (21 CFR 606.171; FDA Form 3486), and enforces the containment SLAs
(≤24 h
interim control) and closure timelines described above. Each box represents a docu-
mented step with eQMS-enforced gates, timestamped audit trails, and role-based routing
to ensure accountability.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 8 of 24
2.8. Failure Response Plan (Pre-Operational)
Structured response to simulation failures. If any acceptance criterion was not met
during a simulation, we implemented a standardized failure response protocol designed to
identify root causes, implement corrective actions, and verify effectiveness before opera-
tional launch. The protocol followed a five-step sequence:
1.
Immediate halt: Stop the affected simulation or drill immediately to prevent normal-
ization of the deviation.
2.
Formal deviation report: Document the failure in the eQMS as a deviation, triggering
the same investigative workflow used for operational deviations.
3.
Interim containment: Implement temporary controls within 24 h to prevent recur-
rence during subsequent simulations (e.g., additional supervision, checklist modifica-
tions, equipment checks).
4.
Root cause analysis: Conduct a structured root cause investigation using two comple-
mentary methods [13]:
#
5 Whys technique [
14
]: Iterative questioning to trace the failure back to its
fundamental cause (e.g., “Why did the label print incorrectly?”
→
“Printer
was misconfigured”
→
“Why was it misconfigured?”
→
continues until root
cause identified).
#
Fishbone (Ishikawa) diagram: Visual mapping of potential contributing fac-
tors across categories (People, Process, Equipment, Materials, Environment,
Measurement) to ensure comprehensive analysis. See Supplementary Mate-
rials for an illustrative fishbone template (Supplementary S1) for a freezer
alarm scenario.
5.
CAPA with verification: Develop a corrective and preventive action plan with specific
effectiveness criteria and target completion date, then repeat the failed simulation
end-to-end to verify that the corrective action resolved the issue (Table 3).
Table 3. Example CAPA record for a minor labeling anomaly observed during mock runs.
Element Example Entry
Deviation summary Sample tube label misaligned. Caught at chairside,
double-check.
Severity Minor
Root cause Label printer template misconfigured for tube diameter.
Corrective action Reconfigure printer template, reprint affected labels.
Preventive action Add pre-shift test print and check. Update SOP with
explicit verification step.
Effectiveness criteria No repeat occurrences in ≥20 subsequent mock runs.
Go/No-Go decision criteria. After completing all simulations and addressing any
failures, we applied explicit criteria to determine operational readiness (Table 4; Figure 2).
Table 4. Go/No-Go decision matrix.
Criteria “Go” Requirement “No-Go” Trigger
Critical-severity findings Zero Any (regardless of closure)
Major-severity findings Zero open (all closed with
verified CAPA)
≥1 open without
completed CAPA
Minor-severity findings All contained ≤24 h with
resolution plans N/A
Drill pass rate 100% <100%
Traceability systems No unresolved gaps Unresolved gaps
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 10 of 24
same Go/No-Go criteria applied to the repeated validation. Figure 2summarizes these
criteria as a Go/No-Go decision flow diagram for operational launch.
Documentation and audit trail. All failure response steps, from initial halt through
CAPA verification, were recorded in the eQMS with timestamped audit trails and electronic
signatures. This creates an inspection-ready record demonstrating that pre-operational
issues were systematically addressed rather than dismissed or deferred.
2.9. Statistical Analysis and Methodological Transparency
Given small pre-operational samples (n= 20 mock runs, n= 3 alarm drills, n= 6
simulated deviations), we used descriptive statistics with exact confidence intervals to
demonstrate system functionality. Detailed sample-size and precision calculations are
provided in Appendix A.
3. Results
3.1. Primary Outcomes
Mock run anomalies. Among 20 chairside mock runs, we observed six minor anoma-
lies (30.0%; 95% CI 11.9–54.3%) and zero major or critical anomalies (0/20;
95% CI 0–16.8%
).
Intercepting only minor, chairside anomalies (with 0/20 major/critical) translates to
avoided pool disqualifications and recall-level investigations, i.e., rework cost and in-
ventory loss preserved; see Supplementary S4 (value-preserved calculations).
Table 5presents the detailed breakdown stratified by severity level.
Table 5. Mock-run anomalies (n= 20) with exact 95% confidence intervals.
Outcome n/N % 95% CI
Any minor anomaly 6/20 30.0 11.9–54.3
Major anomaly 0/20 0.0 0–16.8
Service-level agreement compliance. All six simulated deviations met closure SLA
targets (6/6; 100%; 95% CI 54.1–100%), demonstrating that the eQMS routing, approval
workflows, and documentation requirements functioned as designed under pre-operational
conditions. Meeting the deviation-closure SLA in all simulations limits backlog accumula-
tion and avoids delay-driven overtime, preserving inspection-readiness and throughput.
See Supplementary S4 (value-preserved calculations) for scenario-level value preserved.
Alarm drill performance. Three freezer alarm drills yielded response times of 6, 7, and
8 min (mean 7.0 min; SD 1.0; 95% CI 4.5–9.5), with all three drills achieving “pass” status
(3/3; 100%; 95% CI 29.2–100%). At a 7-min mean response, the modeled excursion-exposure
window is reduced by ~53% versus the 15-min USL, preserving release eligibility during
off-hours. See Supplementary S4 (value-preserved calculations).
A one-sided one-sample t-test versus the 15-min upper specification limit (USL)
yielded p= 0.0026, supporting a margin below the USL (assumes approximate normality).
Table 6summarizes individual drill performance against the pre-specified 15-min
vendor SLA.
Table 6. Alarm drill response times and pass rates.
Drill Time (min) Pass
1 8 Yes
2 6 Yes
3 7 Yes
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 11 of 24
Table 6summary statistics: mean 7.0 min; SD 1.0; 95% CI 4.5–9.5. Pass rate: 3/3 (100%;
95% CI 29.2–100%). Process capability: Cpu = 2.67 versus the upper specification limit of
15 min.
Preliminary readiness-margin indicator (Cpu). Using the 15-min vendor service-
level agreement (SLA) as the upper specification limit (USL), we calculated a preliminary
capability index: Cpu = (USL
−
mean)/(3
×
SD) = (15
−
7)/(3
×
1) = 2.67. Given n= 3
drills/alarm-response observations and a sample SD of 1 min, Cpu is presented only as
an illustrative readiness-margin indicator relative to the 15-min specification and is not
a statistically stable process-capability estimate. Future operational data (
≥
20–30 events)
will be required before computing formal capability indices.
Control-charting framework. Control-charting is established as the deployment
framework for post-operational monitoring. With n= 3 alarm drills, we present an illus-
trative I-chart to demonstrate the approach only; inferential limits from small-n are not
claimed. The illustrative I-chart therefore, demonstrates the monitoring structure that will
be used post-launch, not control limits that are statistically justified by the pre-operational
dataset. See Supplementary S2 for the seed dataset used to illustrate control-chart structure.
Value preserved. At a 7-min mean response versus a 15-min USL, the modeled
excursion-exposure window is reduced by ~53%, preserving release eligibility during off-
hours; see Supplementary S4 (value-preserved calculations) for value-preserved formulas
and examples.
Response time distribution. Figure 3displays the distribution of alarm-response times
across the three drills (median 7 min; range 6–8 min; interquartile range
(IQR) ≈1 min
,
Tukey method), with all observations meeting the 15-min vendor SLA, demonstrating readi-
ness for operational cold-chain management. The compact distribution (
IQR ~1.25 min
)
reflects standardized alarm protocols and trained staff response. See Supplementary S2 for
underlying figure datasets and annotated plotting notes.
Figure 3. Distribution of freezer alarm response times from pre-operational drills.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 12 of 24
3.2. Secondary Outcomes
Document control stress test. The change control simulation confirmed that the eQMS
enforced version control, effective-date management, and training gates as specified. En-
forced training gates and certified-copy retrieval mitigate the risk of outdated procedures
entering use, preserving release eligibility and reducing rework from documentation de-
fects (quantitative assumptions in Supplementary Materials (value-preserved calculations;
Supplementary S4)). Staff members could not access or operate under the updated proce-
dure until completing electronic training attestation on the released version, demonstrating
that the system prevents the use of outdated documents.
Release packet readiness. Mock release packets included all required documenta-
tion elements: bill of lading (BOL), testing certificates, shipping documentation, donor
case list, and QA release approval. This confirmed that document templates, automatic
population of metadata (lot numbers, dates, approvals), and packet assembly workflows
functioned correctly.
3.3. Anomaly Characterization and CAPA Prioritization
A Pareto analysis of minor anomalies was conducted to prioritize CAPA focus areas
(Table 7and Figure 4).
Table 7. Minor anomaly categories ranked by frequency (n= 6).
Category Count
Percent
Cumulative %
Labeling callouts (verbal confirmation required) 2 33.3 33.3
Sample tube label catch (misalignment detected) 1 16.7 50.0
Label placement correction 1 16.7 66.7
Weigher QC failure leading to device
out-of-specification (OOS)
1 16.7 83.3
Missed second-person verification check 1 16.7 100.0
Figure 4. Pareto analysis of minor anomalies during pre-operational mock runs.
The six minor anomalies observed during mock runs fell into five distinct categories
(Table 7). Labeling-related issues (verbal callouts requiring confirmation, label placement
corrections) accounted for 50% of observations, identifying this as the highest-priority area
for corrective action.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 13 of 24
Labeling-related observations (callouts and placement corrections) comprised 50% of
all minor anomalies, while equipment, procedural, and verification issues each occurred
once. All six anomalies were contained within 24 h per protocol. The absence of major or
critical findings (0/20; 95% CI 0–16.8%) met pre-defined acceptance criteria. Based on this
Pareto distribution, targeted CAPA focused on labeling procedure clarification and staff
retraining, which was completed before operational launch.
4. Discussion
4.1. Principal Findings
This pre-operational, simulation-based validation produced two categories of deliver-
ables aligned to 21 CFR Parts 606/640 and BPDR reporting requirements (Form
FDA 3486
):
I. quantifiable readiness indicators, including anomaly rates with confidence intervals,
alarm response time distributions, and pass rates.
II. auditable artifacts, including investigation packets, temperature graphs, product
movement logs, and electronic deviation-to-CAPA linkages suitable for
regulatory inspection
.
4.2. Framework Validation vs. Operational Performance
This study validates a methodological framework (documented procedures, quantita-
tive acceptance thresholds, and traceable documentation pathways) rather than long-term
operational outcomes. The framework’s key strength is transferability: standardized drills,
select templates, and acceptance criteria (see Supplementary S1–S4) enable replication at
other plasma centers without site-specific redesign.
What pre-operational validation establishes: The validation confirms that quality
system components function correctly under controlled conditions (alarms trigger and route
properly, deviations follow escalation pathways, CAPA records link to root investigations,
training gates prevent use of superseded procedures, and staff execute protocols as written).
These are necessary preconditions for operational success.
What pre-operational validation does not establish: Simulation cannot predict sus-
tained performance under operational pressures: real donor volume, equipment wear
patterns, staff turnover, seasonal workload variation, or how quickly deviation closure
times stabilize. These operational characteristics require longitudinal monitoring after
launch, which is standard practice in quality system validation: demonstrate that proce-
dures work (pre-operational), then prove they work consistently at scale (operational).
4.3. Implications for Plasma Supply and Inspection Readiness
The validated QMS provides three layers of protection for plasma-derived therapeutic
supply chains:
1.
Traceability assurance. The electronic deviation-to-CAPA workflow with automatic
record linkages protects chain-of-custody documentation required for lot release.
When fractionators receive plasma shipments, they rely on complete traceability
records to pool donations into manufacturing lots. A single broken link (missing sam-
ple documentation, unclear product movement during an excursion) can disqualify
an entire pool. Our validation demonstrated that the eQMS maintains these linkages
even during deviations, protecting product eligibility.
2.
Cold-chain resilience. Rapid alarm response limits freeze–thaw exposure that de-
grades the labile coagulation factors V and VIII, which are particularly sensitive to
temperature instability in plasma. Cold-chain protection also preserves more stable
but clinically critical proteins relevant to plasma-derived medicinal products (PDMPs),
including Factor IX, prothrombin complex concentrate (PCC), and immunoglobu-
lins. The EDQM Blood Guide specifies storage at
≤−
20
◦
C with prompt recovery
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 14 of 24
from excursions; our mean alarm-response time of 7.0 min (95% CI 4.5–9.5) demon-
strated a substantial margin below the 15-min SLA. The validated response protocol
(alarm detection, vendor engagement, interim control, documentation), therefore, pro-
vides a repeatable framework that safeguards both labile plasma factors (V, VIII) and
therapeutic-relevant PDMP proteins (e.g., FIX, IVIG, PCC) during the most vulnerable
operational phases, including overnight and weekend periods
3.
Inspection readiness. The regulatory crosswalk (Table 1) and simulation artifacts
(investigation packets, control charts, CAPA records) provide objective evidence for
FDA and EU inspectors that quality systems were validated before first collections.
Pre-operational validation demonstrates proactive risk management rather than reac-
tive problem-solving, which aligns with both the FDA’s CGMP expectations for blood
(21 CFR 606) and the EU GMP principles of quality risk management (ICH Q9).
4.4. Real-World Validation: Lessons from Cold-Chain and Identification Failures
The operational elements validated in this study, i.e., alarm response protocols (mean
7 min; Table 3), labeling procedures (chairside drills; Table 5), and traceability systems,
address documented failure modes in blood banking and clinical laboratories.
4.4.1. Cold-Chain Resilience
A 2021 freezer malfunction at Kaiser Permanente (Seattle, Washington) documented
successful cold-chain preservation when freezer alarm protocols enabled rapid mobilization
of personnel, resulting in administration of 1600 temperature-sensitive vaccine doses within
the validated storage window and preventing total product loss [
15
–
17
]. This case supports
the protective value of the
≤
15-min vendor response SLA and documented interim control
procedures validated in our alarm drill protocol.
Contrarily, equipment monitoring failures at two U.S. fertility centers in 2018 (Uni-
versity Hospitals, Cleveland, Ohio, and Pacific Fertility Center, San Francisco, Califor-
nia) resulted in undetected liquid-nitrogen temperature excursions, compromising over
7500 cryopreserved
specimens [
18
–
20
]. Subsequent legal proceedings established manufac-
turer liability for inadequate sensor validation and failure to implement known corrective
measures. These failures emphasize the necessity of validating both primary monitor-
ing systems and redundant backup sensors, as demonstrated in our three-drill protocol
(
Table 6
), which verified alarm detection, escalation pathways, and independent tempera-
ture verification using calibrated reference thermometers.
4.4.2. Identification Integrity
Another 2018 incident was prompted by a fatal transfusion error at Baylor St. Luke’s
Medical Center (Houston, Texas). The tragedy, attributed to specimen mislabeling, doc-
uments systemic vulnerability in sample identification processes. Federal investigation
identified six procedural control points that failed to detect the labeling error before transfu-
sion occurred, tracing the root cause to inadequate handoff verification between collection
and processing personnel.
Incidents such as this illustrate the necessity of the chairside mock-run methodology
employed in our pre-operational framework (Table 5), which systematically tests critical
control points, including labeling handoffs and second-person verification. The observation
that 33% (2/6) of minor anomalies involved labeling-related issues (Table 7, Pareto analysis)
demonstrates that such vulnerabilities manifest even under controlled simulation condi-
tions, providing a rationale for systematic detection and corrective action implementation
before operational deployment.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 15 of 24
4.5. Synthesis: Pre-Operational Validation as Risk Mitigation
These incidents collectively demonstrate the dual nature of quality system perfor-
mance: the Seattle case reflects successful risk mitigation through validated alarm protocols,
while the fertility center and transfusion failures exemplify consequences of inadequate
system validation. The contrasting outcomes support the fundamental premise of pre-
operational validation: simulation-based verification of critical control points (cold-chain
monitoring, identification protocols, documentation traceability) reduces the probability
of system failures during operational deployment. By establishing deviation-readiness
through structured simulation before first collections, this framework operationalizes
preventive risk management rather than reactive failure response.
4.6. Clinical and Regulatory Relevance: Protecting Hematology Therapeutic Supply
Every procedural safeguard at the plasma-collection level ultimately supports thera-
peutic reliability. For example, failsafe freezer alarm protocol enables rapid implementation
of interim control measures (product secured, backup freezer contacted, temperature
logged), and document recovery for the product to remain within validated storage bounds.
This prevents excursion and protects labile coagulation factors that would otherwise de-
grade during freeze–thaw cycles, ensuring that plasma entering the fractionation pool
retains manufacturing viability. As another illustration, identification of a near-miss label
discrepancy before product release should reflexively catalyze a deviation investigation and
CAPA, strengthening the second-person verification process. This traceability protection
prevents pool disqualification at the fractionator, a scenario where even a single mislabeled
unit can render an entire manufacturing lot unusable. Given global immunoglobulin de-
mand growth of 6–8% annually, protecting supply chain integrity at the source plasma level
stabilizes downstream availability of IVIG for hematologic and immunologic indications.
Plasma center QMS safeguards maintain continuity for hematology and immunology
patients who depend on the streamlined facility of two primary pathways for plasma
collection: the fractionation route, whereby plasma is sent to an industrial manufacturer
(fractionator) and processed via cold ethanol or chromatographic fractionation into purified
proteins for PDMPs, including Factors VIII and IX, prothrombin complex concentrate (PCC),
IVIG, albumin, and alpha-1 antitrypsin; and the direct transfusion route, whereby plasma is
transfused without fractionation (as FFP, cryoprecipitate, or pathogen-reduced plasma, i.e.,
‘labile blood components’ not classified as PDMPs) for the treatment of coagulopathy, TTP,
and acute liver failure. Regulatory distinction classifies PDMP and non-PDMP (transfusion
plasma) oversight to each pathway (Table 8).
Table 8. PDMP and non-PDMP oversight (per jurisdiction).
Jurisdiction PDMP Oversight
Non-PDMP (Transfusion Plasma) Oversight
United States (FDA)
Center for Biologics Evaluation and Research
(CBER), Office of Blood Research and Review:
licensed biologics under the Public Health
Service Act (Biologics License Applications).
Blood establishments/transfusion services
under 21 CFR 606 and 640; units labeled as
FFP, PF24, etc.
European Union
(EMA/EDQM)
Medicinal products regulated under Directive
2001/83/EC, manufactured per GMP and
Annex 14.
Blood components regulated under Directive
2002/98/EC and EDQM Blood Guide:
hospital or blood-bank products.
4.7. Adaptation Framework for Multi-Site Deployment
The validation framework is adaptable to facilities with different operational con-
straints. Table 9provides a decision matrix for common variations:
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 16 of 24
Table 9. Adaptation guidance for varied facility configurations.
Facility Characteristic Required Adaptations Technical Rationale
Paper-based QMS (no
electronic system)
Use identical forms as hard copies, implement
a wet-signature issuance log, and maintain a
manual audit trail with date/time/initials for
all approvals and changes
Preserves ALCOA+ principles
(attributability, traceability,
contemporaneous documentation) and
the evidentiary chain without
requiring software infrastructure
Small facility (
≤
3 freezers)
Conduct alarm drills per freezer unit until
accumulating n≥16 total observations to
achieve ±1-min precision (95% CI) assuming
σ≈2 min
Sample size calculation: n≈(1.96 ×
σ/half-width)2= (1.96 ×2/1)2≈15.4,
rounded to 16
International/EU-directed
shipments
Retain U.S. regulatory anchors (21 CFR
606/640) while adding EU Directive
2005/62/EC and EDQM Blood Guide
(22nd edition) crosswalk, implement EU
GMP Annex 11 for computerized systems
Harmonizes quality system to both
domestic and fractionator/Binding
Entity (BE) expectations, enabling
shipments to EU markets
Different alarm vendor
Substitute call-back documentation and
service ticket artifacts from the actual
provider, retain ≤15-min response SLA as
acceptance criterion
SLA threshold is risk-based (rapid
containment protects plasma for
fractionation) rather than
vendor-specific, principle applies
across monitoring systems
4.8. Comparison with Existing Regulatory Frameworks
This validation framework operationalizes multiple regulatory requirements that typi-
cally exist as separate guidance documents: BPDR rules (45-day reporting timeline, Form
FDA 3486 submission), 21 CFR Part 11 electronic records controls, EU GMP Annex 11 com-
puterized system validation, and ALCOA+ data integrity principles. By integrating these
requirements into a unified QMS with pre-operational acceptance criteria, the framework
provides a practical implementation guide.
The approach is complementary to, rather than duplicative of, existing laboratory
standards, including ISO 15189:2022 [
21
] (medical laboratory quality requirements) and
International Council for Standardization in Haematology (ICSH) recommendations for
coagulation sample handling. Our framework addresses the upstream collection and
storage phases that precede analytical testing, filling a gap in pre-operational validation
guidance specific to source plasma operations.
4.9. Resource Requirements and Scalability Economics
Labor and timeline. Table 10 presents estimated staffing requirements for single-
site validation, totaling approximately 252 h over an eight-week timeline for a first-in-
network implementation. Key personnel include QA Specialist (QAS), Center Director
(CD), Information Technology (IT) support, Medical Director, and collection staff. U.S.
Bureau of Labor Statistics median wages (May 2024) provide cost anchors: phlebotomists
USD 43,660/year (approximately $21/h), clinical laboratory technologists/technicians
USD 61,890/year (approximately $30/h), and medical and health services managers USD
117,960/year (approximately $57/h) [22–24].
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 17 of 24
Table 10. Estimated staffing, hours, and cost implications (single site).
Activity Personnel Hours Timeline Illustrative Labor Cost
Document and change-control
setup QA Specialist (QAS) 40 Weeks 1–2 USD 1600
eQMS configuration and
validation
QAS 60, IT 24,
consultant 16 100 Weeks 2–4 USD 4000
SOP training (10 procedures) CD, QAS, 4 staff ~38 Week 5 USD 1520
Chairside mock run (n= 20) QAS + 2 staff 24 Week 6 USD 960
Cold-chain alarm drills (n= 3) QAS, Facilities 6 Weeks 6–7 USD 240
Document-control stress test QAS, CD 8 Week 7 USD 320
Deviation/CAPA documentation QAS 12 Week 7 USD 480
Internal audit (readiness) Corporate QA/3rd party 16 Week 8 USD 640
Management review and report CD, QAS, Medical
Director 8 Week 8 USD 320
TOTAL ~252 ~8 weeks ~ USD 10,080
Labor cost assumptions. Illustrative costs are based on a blended labor rate of approximately USD 40 per hour
across involved roles (QAS, CD, clinical staff, IT support, and management), yielding an estimated direct labor
cost of
≈
USD 10,080 for 252 h. Additional materials, overhead, and any third-party audit expenses account for
the higher total pre-operational investment range of USD 20,000–33,000 discussed in the return-on-investment
scenarios above [25].
Return on investment (illustrative scenarios). Pre-operational validation costs must
be weighed against operational revenue and launch delay risks. Monthly contribution
margin can be estimated as follows:
Contribution margin = (donors per day) ×(operating days per month) ×(margin per donation)
1.
Scenario A (smaller center): 50 donors/day
×
20 days
×
USD 40–50 margin = USD
40,000–50,000 per month.
2.
Scenario B (larger center): 150 donors/day
×
22 days
×
USD 50 margin
≈
USD
165,000 per month.
If pre-operational validation investment totals USD 20,000–33,000 (labor, materials,
third-party audit), avoiding even 1–2 months of launch delay due to post-launch quality
system failures recovers this investment. Subsequent sites in a multi-center network
realize
≥
40% cost savings through economies of scale like template reuse, standardized
eQMS configuration, and established training materials, improving the business case for
systematic pre-operational validation across the network.
Scalability advantage. Once validated at a pilot site, the framework’s standardized
procedures, acceptance criteria, and documentation templates transfer directly to additional
centers with minimal site-specific customization. This contrasts with reactive quality system
development (building procedures after launch in response to findings), which requires
custom solutions at each site and carries higher operational risk.
5. Limitations
This pre-operational validation has five principal limitations that constrain the inter-
pretation of results:
1. Small sample sizes and statistical precision. With n= 20 mock runs and n= 3 alarm
drills, confidence intervals are necessarily wide. The observed minor anomaly rate of 30.0%
has a 95% CI spanning 11.9–54.3%, meaning the true pre-operational rate could plausibly
range from approximately one in eight runs to one in two runs. Similarly, alarm response
time estimates (mean 7.0 min, 95% CI 4.5–9.5 min) reflect substantial uncertainty. Process
capability indices and control limits calculated from n= 3 observations are inherently
unstable and should be interpreted as demonstrations of measurement system functionality
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 18 of 24
rather than precise population parameters. These sample sizes were constrained by pre-
operational feasibility, adequate for detecting major system failures (which were absent),
but insufficient for narrow precision on secondary metrics.
2. Single-observer assessment. Pre-operational drills used single observers to classify
anomaly severity and assess drill pass/fail criteria, precluding inter-rater reliability estima-
tion. This introduces potential for classification bias: different observers might categorize
the same event differently (e.g., Minor vs. Major severity), particularly for borderline cases.
The impact on our results is likely limited. We observed zero Major/Critical events, so mis-
classification would only affect distribution within the Minor category, but the consistency
of severity classification remains unvalidated. Dual-observer protocols with inter-rater
reliability analysis (Cohen’s kappa) were not feasible during pre-operational simulations
but represent an important operational validation step.
3. Simulation fidelity and Hawthorne effects. Despite prioritizing functional, phys-
ical, and psychological fidelity, simulations conducted in pre-operational contexts may
not predict operational performance. Staff awareness of observation and assessment may
have produced optimal “best behavior” responses (Hawthorne effect), yielding faster
alarm response times, heightened attention to labeling, and more careful documentation
than would occur under routine operational pressures. Additionally, simulations cannot
replicate cumulative fatigue, competing priorities during high donor volume, equipment
degradation over time, or staff turnover effects. Our observed performance (zero Criti-
cal/Major anomalies, 7-min mean alarm response) represents controlled conditions, not
sustained operational reality.
4. Inherent limitations of pre-operational validation. Simulations fundamentally
cannot answer questions that require longitudinal operational data: How quickly are
deviations closed in practice when CAPA investigations compete with daily operations?
Do alarm response times remain stable over months, or do they drift as urgency perception
fades? Do different staff members classify deviation severity consistently when facing
actual product-impacting events? Pre-operational validation demonstrates that system
components function correctly (procedures execute, approvals route, records link), but
cannot prove that these systems will perform reliably at scale over extended periods. This
distinction between system validation (demonstrated here) and process validation (requires
operational data) is fundamental to quality system maturation.
5. Single-site generalizability. This validation was conducted at one center with
specific characteristics: small-scale operation (two apheresis devices, one freezer), urban
location with rapid vendor access, and staff experienced in clinical laboratory practices.
Performance at centers with different scales (high-throughput operations), geographic
contexts (rural locations with delayed vendor response), or staffing profiles (newly hired
teams) remains unvalidated. The adaptation guidance (Table 6) provides a framework for
addressing these variations, but actual multi-site performance has not been demonstrated.
6. Conclusions
This study presents the first published pre-operational validation of a deviation-ready
quality management system for a source plasma center, using simulation-based testing to
demonstrate system functionality before first donor collections. The validation confirmed
that critical quality system components (document control, deviation escalation pathways,
CAPA linkages, alarm response protocols, cold-chain documentation, and training gates)
function correctly under controlled conditions and produce inspection-ready audit trails
aligned with 21 CFR Parts 606/640, BPDR reporting requirements, and EU Directive
2005/62/EC.
Key achievements include the following:
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 19 of 24
I.
Quantifiable readiness metrics with confidence intervals (30% minor anomaly
rate, zero major/critical findings, seven-minute mean alarm response) that met
pre-defined acceptance criteria.
II.
Complete evidence chains from alarm detection through QA review demonstrating
end-to-end traceability.
III.
Standardized procedures and acceptance criteria enabling replication at other sites,
and (iv) validated eQMS workflows enforcing ALCOA+ data integrity principles
and Part 11 electronic records requirements.
The framework addresses documented failure modes in biological cold-chain manage-
ment and sample identification, areas where real-world incidents (Seattle vaccine rescue,
fertility clinic storage failures, Houston transfusion fatality) demonstrate the stakes of
quality system robustness. By protecting cold-chain integrity and traceability at the source
plasma level, the validated QMS safeguards downstream availability of plasma-derived
coagulation factors (Factor VIII, Factor IX, prothrombin complex concentrate) and im-
munoglobulins essential for hematology and immunology care.
The validation methodology is transferable and scalable: standardized templates,
acceptance criteria, and drill protocols (provided in Supplementary Materials) enable de-
ployment across multi-center networks with an estimated 40% cost reduction at subsequent
sites through eQMS configuration reuse. The framework is adaptable to varied operational
contexts (paper-based systems, small facilities, international shipments, alternative alarm
vendors) through the decision matrix provided in Table 6.
This work establishes pre-operational validation as a systematic, evidence-generating
approach to quality system readiness in source plasma operations. By demonstrating
deviation-readiness before first collections, the framework enables proactive risk mitigation
rather than reactive problem-solving, providing regulatory inspectors with objective evi-
dence of quality system capability and offering the plasma center community a replicable
model for launch preparation. While operational validation remains necessary to confirm
sustained performance under real-world conditions, this pre-operational framework pro-
vides the measurement infrastructure, acceptance thresholds, and documentation standards
that govern subsequent operational monitoring.
Ethical and Regulatory Considerations
This simulation-based pre-operational validation involved no human participants or
identifiable data. Quality-system design and acceptance criteria conformed to U.S./EU
requirements referenced in the manuscript (21 CFR 606/640; 21 CFR 606.171 Biological
Product Deviation Reporting; Commission Directive 2005/62/EC; EDQM Blood Guide,
22nd ed.), with computerized-system controls aligned to 21 CFR Part 11 and EU GMP
Annex 11 [10,25].
Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/labmed3010002/s1, S1—Fishbone (Ishikawa) template for freezer-
alarm root-cause analysis. S2—Figure datasets (alarm response times: individuals control chart
and box-plot; Pareto counts of anomaly distribution); S3—eQMS validation summary (risk-based),
drill work instructions, change-control matrix, and CAPA effectiveness plan. S4—Value-preserved
calculators and ROI worksheet (assumptions + formulas).
Author Contributions: Conceptualization: A.U.P.; methodology: A.U.P.; software: R.M.; validation:
S.A.; formal analysis: A.U.P.; investigation: A.U.P.; resources: R.M.; data curation: R.M.; writing—
original draft preparation: A.U.P.; writing—review and editing: A.U.P. and S.A.; visualization: A.U.P.,
R.M. and S.A.; supervision: R.M.; project administration: R.M.; funding acquisition: not applicable.
All authors have read and agreed to the published version of the manuscript.
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 20 of 24
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: All data supporting the findings are contained within the article and its
Supplementary Materials (S1–S4). De-identified templates and figure datasets are provided to enable
replication (e.g., fishbone/Ishikawa root-cause template [S1], eQMS validation summary and drill
work instructions [S3], figure datasets and illustrative I-chart [S2], and value-preserved calculation
worksheets [S4]). Additional operational artifacts described in the paper may be available from the
corresponding author upon reasonable request and subject to confidentiality.
Acknowledgments: The authors thank the participating site for assistance with pre-operational
simulations and documentation checks. During the preparation of this manuscript, the authors used
ChatGPT-5.1 (OpenAI; October 2025) for editorial assistance limited to language polishing, section
ordering, and reference formatting. The authors reviewed and edited all AI-assisted output and take
full responsibility for the content of this publication.
Conflicts of Interest: R.M. is a founder of Superhero Biologics. This work describes pre-operational
quality-system validation relevant to a center to be operated by Superhero Biologics. The other
authors declare no conflicts of interest. No funder had any role in the design of the study: in the
collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to
publish the results.
Abbreviations
The following abbreviations are used in this manuscript:
ALCOA+ Attributable, Legible, Contemporaneous, Original, Accurate, Complete,
Consistent, Enduring, Available (data-integrity principles)
ASQ American Society for Quality
BE Binding Entity (fractionator contracting/receiving entity)
BLS Bureau of Labor Statistics
BOL Bill of Lading
BPDR Biological Product Deviation Reporting
CAPA Corrective and Preventive Action
CBER Center for Biologics Evaluation and Research (FDA)
CD Center Director
CGMP Current Good Manufacturing Practice
CFR Code of Federal Regulations
CI Confidence Interval
Cpk Process Capability Index (centered)
Cpu Process Capability Index, Upper-Specification (upper)
COI Conflict of Interest
eCFR Electronic Code of Federal Regulations
EDQM European Directorate for the Quality of Medicines & HealthCare
EMA European Medicines Agency
eQMS Electronic Quality Management System
EU European Union
EUR-Lex European Union law database
FDA U.S. Food and Drug Administration
FFP Fresh Frozen Plasma
FMEA Failure Mode and Effects Analysis
FIX Factor IX
FVIII Factor VIII
GMP Good Manufacturing Practice
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 21 of 24
HAZOP Hazard and Operability Study
I-chart Individual chart (SPC)
ICH Q9(R1) International Council for Harmonisation Guideline Q9
(Quality Risk Management), Revision 1
ICSH International Council for Standardization in Haematology
Ig Immunoglobulin
IQR Interquartile Range
IRB Institutional Review Board
ISO International Organization for Standardization
IT Information Technology
IVIG Intravenous Immunoglobulin
K-M Kaplan–Meier (survival analysis)
MHRA Medicines and Healthcare products Regulatory Agency (UK)
MRB Marketing Research Bureau
NIST National Institute of Standards and Technology
OOS Out-of-Specification
PCC Prothrombin Complex Concentrate
PDMP(s) Plasma-Derived Medicinal Product(s)
PF24 Plasma Frozen within 24 h of phlebotomy
PHI Protected Health Information
PII Personally Identifiable Information
PMC PubMed Central
QA Quality Assurance
QAS Quality Assurance Specialist
QMS Quality Management System
RA Regulatory Affairs
ROI Return on Investment
SD Standard Deviation
SLA Service-Level Agreement
SOP Standard Operating Procedure
SPC Statistical Process Control
TTP Thrombotic Thrombocytopenic Purpura
USL Upper Specification Limit
X-bar Mean Chart (SPC)
R chart Range chart (SPC)
Appendix A
Descriptive methods for pre-operational data. Given the small sample sizes and
simulation context of this pre-operational validation, we employed descriptive statistics
with exact confidence intervals. For binomial proportions (e.g., anomaly rates, pass rates),
we calculated exact 95% confidence intervals using the Clopper–Pearson method, which
provides accurate coverage even with small samples and rare events. For continuous
measures (e.g., alarm response times), we calculated means with 95% confidence intervals
using the t-distribution to account for small sample uncertainty.
Rationale for method selection. We selected statistical methods appropriate for the
pre-operational validation context:
Statistical Process Control (SPC) charts: SPC methods (e.g., Individual charts, X-bar,
and R charts) are designed to monitor ongoing processes and detect shifts or trends over
time using consecutive operational data. These techniques require an “in-control” baseline
established from sequential observations (typically
≥
20–30 points from a stable process).
Our n= 3 alarm drills were conducted as discrete simulations rather than sequential oper-
ations. We offer a preliminary control chart framework (Supplementary S2) to illustrate
the monitoring structure that was validated during pre-operational testing, with the un-
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 22 of 24
derstanding that control limits shown are templates for future use rather than statistically
established boundaries.
Preliminary readiness margin indicators (Cpu): Capability indices quantify how well a
process meets specifications relative to its natural variation. We calculated Cpu for the
n= 3
alarm drills to demonstrate preliminary margin relative to the 15-min upper specification
limit. The resulting Cpu of 2.67 is presented as a preliminary readiness margin indicator
rather than a statistically valid process capability index (which requires
≥
30 observations
from a stable process). This value illustrates that observed response times remained well
below the 15-min specification threshold during pre-operational testing, but should not be
interpreted as establishing statistical process capability.
Kaplan–Meier survival curves: These time-to-event curves analyze closure times
while accounting for censored observations (cases still open at analysis time). In our
pre-operational simulations, all deviations were closed by design (100% closure rate) as
part of the validation protocol, resulting in no censoring. Since Kaplan–Meier methods
require a mix of closed events and censored cases, these curves were not applicable to the
pre-operational dataset and were therefore not generated.
Transparency and appropriate inference: Our statistical approach emphasizes ac-
curate representation of pre-operational simulation results. We report point estimates
with confidence intervals that appropriately reflect small sample uncertainty, and we
present preliminary process monitoring frameworks (control charts, capability indices) as
demonstrations of validated measurement systems rather than as operational performance
benchmarks. Accordingly, p-values are reported only where a prespecified hypothesis
test is meaningful (e.g., alarm-response mean versus the 15-min USL); for small-n bino-
mial endpoints with acceptance criteria, exact confidence intervals are presented without
p-values.
For proportions (e.g., anomaly rate; SLA compliance), 95% confidence intervals are ex-
act (Clopper–Pearson). For alarm-response time, we report mean and range for
n= 3 drills
;
control-chart limits and capability indices are validated as a monitoring framework for
post-launch use rather than inferential estimates from small-n.
Sample-Size and Precision Plan (A Priori)
Pre-operational sample sizes. This validation employed
n= 20
chairside mock runs,
n= 3
freezer alarm drills, and n= 6 simulated deviation scenarios. These samples reflect the
pragmatic constraints of pre-operational validation—balancing the need for quantitative
evidence with the reality that no actual donors or product are available prior to launch.
Mock runs (n= 20). For the primary endpoint (minor anomaly rate), we anticipated
a rate of 20–30% based on pilot observations. Achieving a 95% CI half-width of
±
10%
(e.g., an observed 30% rate with bounds of 20–40%) would require n
≈
80 mock runs using
standard binomial precision calculations. However, n= 20 provides a 95% CI half-width
of approximately ±20%, which we judged sufficient to detect systematic problems (e.g., a
true rate exceeding 50% would be evident) while remaining feasible within pre-operational
timelines. The observed rate of 30% (95% CI: 11.9–54.3%) met acceptance criteria (no
major/critical events) and triggered targeted CAPA on labeling procedures.
Alarm drills (n= 3). For mean response time, assuming an operational standard
deviation of
σ≈
2 min, achieving
±
1-min precision at 95% confidence would require
n≈16 drills
, while
±
0.5-min precision would require n
≈
62. With n= 3 drills, the ob-
served mean of 7.0 min has wide confidence bounds (95% CI: 4.5–9.5 min). We accepted
this imprecision because (i) all three drills remained well below the 15-min SLA threshold,
(ii) the
drills confirmed functional readiness (alarm detection, vendor response, documenta-
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 23 of 24
tion pathways) rather than establishing precise population parameters, and (iii) operational
data will refine these estimates post-launch (target: ≥30 events within 6 months).
SLA compliance (n= 6 deviations). The observed 100% compliance (6/6) yields a wide
95% CI (54.1–100%), far from the precision needed to confirm a population compliance rate
above 90% (which would require n
≈
52 events). This reflects the limitation of simulation:
we cannot generate adequate sample sizes for all endpoints without actual operations. The
6/6 result provides qualitative assurance that SLA workflows function as designed, with
quantitative validation deferred to the post-launch monitoring plan.
Precision vs. readiness. These sample sizes prioritize operational readiness (demon-
strating that procedures, alarms, and documentation systems work end-to-end) over statisti-
cal precision. Pre-operational validation cannot substitute for operational data; it establishes
measurement infrastructure, acceptance thresholds, and audit-ready artifacts. Formal sta-
tistical process control (SPC) charts, capability indices, and time-to-closure analyses will be
implemented once ≥30–100 operational events establish baseline performance.
References
1.
EDQM Guide to the Preparation, Use and Quality Assurance of Blood Components, 22nd ed. Available online:
https://www.edqm.eu/en/blood-guide (accessed on 3 September 2025).
2.
21 CFR 606.171—Biological Product Deviation Reporting (BPDR). Available online: https://www.ecfr.gov/current/title-
21/chapter-I/subchapter-F/part-606/subpart-I/section-606.171 (accessed on 9 September 2025).
3.
Prevot, J.; Jolles, S. Global immunoglobulin supply: Steaming towards the iceberg? Curr. Opin. Allergy Clin. Immunol. 2020,20,
557–564. [CrossRef] [PubMed]
4.
Brain Bank at McLean Hospital. Internet. McLean Hospital. Available online: https://www.mcleanhospital.org/research/brain-
bank (accessed on 9 September 2025).
5.
Karen, M. Brain Tissue Used to Study Autism Destroyed by Hospital Freezer Fault. Medical Research. The Guardian. 11 June
2012. Available online: https://www.theguardian.com/science/2012/jun/11/brain-tissue-autism-freezer-fault (accessed on
8 September 2025).
6.
Scott, H. Thaw at Brain Bank Deals Setback to Autism Research. Internet. National Public Radio (NPR). Updated 11 June 2012.
Available online: https://www.npr.org/sections/health-shots/2012/06/11/154775287/thaw-at-brain-bank-deals-setback-to-
autism-research (accessed on 7 September 2025).
7.
Hammarskjöld, A. Investigation into the Christmas freezer failure at Neo. News from Karolinska Institutet. Karolinska
Institutet. 2024. Available online: https://news.ki.se/investigation-into-the-christmas-freezer-failure-at-neo (accessed on
8 September 2025).
8.
‘GxP’ Data Integrity Guidance and Definitions. 2018. Available online: https://assets.publishing.service.gov.uk/media/5aa2b9e
de5274a3e391e37f3/MHRA_GxP_data_integrity_guide_March_edited_Final.pdf (accessed on 5 October 2025).
9.
ICH Harmonised Guideline Q9(R1): Quality Risk Management (Final Version Adopted 18 January 2023). 2023. Available
online: https://database.ich.org/sites/default/files/ICH_Q9%28R1%29_Guideline_Step4_2022_1219.pdf?utm (accessed on
5 October 2025).
10.
EU GMP Annex 11—Computerised Systems. Available online: https://www.gmp-compliance.org/guidelines/gmp-
guideline/eu-gmp-annex-11-computerised-systems (accessed on 5 October 2025).
11.
Commission Directive 2005/62/EC Implementing 2002/98/EC as Regards Community Standards and Specifications
Relating to a Quality System for Blood Establishments. 2005. Available online: https://eur-lex.europa.eu/legal-
content/EN/TXT/?uri=CELEX%3A32005L0062 (accessed on 6 September 2025).
12.
Legal Information Institute; Cornell Law School. 21 CFR Part 606—Current Good Manufacturing Practice for Blood and Blood
Components. Cornell Law School. Updated 2025. Available online: https://www.law.cornell.edu/cfr/text/21/part-606 (accessed
on 6 September 2025).
13.
American Society for Quality (ASQ). Seven Basic Quality Tools. ASQ Quality Resources. 2025. Available online:
https://asq.org/quality-resources/seven-basic-quality-tools?srsltid=AfmBOoqQajMfciDYe8yUkuffCctosVWmhisYmXO9T_cDl
N0yBym79lG4&utm (accessed on 6 September 2025).
14.
American Society for Quality (ASQ). Five Whys and Five Hows. Internet. Updated 2025. Available online:
https://asq.org/quality-resources/five-whys?srsltid=AfmBOorXgfQvqYVgGwEioui8JSNsQcHfBytYkVMHcpQmzRzxSO6tS
39O&utm (accessed on 9 September 2025).
https://doi.org/10.3390/labmed3010002
LabMed 2026,3, 2 24 of 24
15.
Knowles, H. A Failed Freezer Forced an Overnight Dash to Give out More than 1600 Doses of the Coronavirus Vaccine. The
Washington Post Company. 30 January 2021. Available online: https://www.washingtonpost.com/health/2021/01/30/seattle-
coronavirus-vaccine-overnight/ (accessed on 9 September 2025).
16.
KOMO News Staff. Hundreds Line up at Midnight for COVID Shot After Seattle Vaccine Emergency 2021. Sinclair Broadcast
Group. 29 January 2021. Available online: https://komonews.com/news/local/hundreds-line-up-at-midnight-for-covid-shot-
after-vaccine-emergency (accessed on 5 October 2025).
17.
Combos, L. Hundreds Vaccinated in Seattle After Late Night Freezer Failure 2021. Patch Media. 29 January 2021. Avail-
able online: https://patch.com/washington/seattle/hundreds-vaccinated-seattle-after-late-night-freezer-failure (accessed on
29 December 2025).
18.
Letterie, G.; Fox, D. Lawsuit frequency and claims basis over lost, damaged, and destroyed frozen embryos over a 10-year period.
FS Rep. 2020,1, 78–82. [CrossRef]
19.
Moutos, C.P.; Lahham, R.; Phelps, J.Y. Cryostorage failures: A medicolegal review. J. Assist. Reprod. Genet. 2019,36, 1041–1048.
[CrossRef] [PubMed]
20.
In re: Pacific Fertility Center Litigation—Order re Plaintiffs’ Motion for Partial Summary Judgment 2021. Available online:
https://www.govinfo.gov/content/pkg/USCOURTS-cand-3_18-cv-01586/pdf/USCOURTS-cand-3_18-cv-01586-49.pdf (ac-
cessed on 1 November 2025).
21.
ISO 15189:2022; Medical Laboratories—Requirements for Quality and Competence. International Organization for Standardiza-
tion: Geneva, Switzerland, 2022.
22.
Occupational Outlook Handbook: Phlebotomists. 2025. Available online: https://www.bls.gov/ooh/healthcare/phlebotomists.htm
(accessed on 1 November 2025).
23.
Occupational Outlook Handbook: Clinical Laboratory Technologists and Technicians. 2025. Available online:
https://www.bls.gov/ooh/healthcare/clinical-laboratory-technologists-and-technicians.htm (accessed on 1 November 2025).
24.
Occupational Outlook Handbook: Medical and Health Services Managers. 2025. Available online: https://www.bls.gov/ooh/
management/medical-and-health-services-managers.htm (accessed on 1 November 2025).
25.
EudraLex Volume 4, Annex 11: Computerised Systems (Revision 1). 2011. Available online: https://health.ec.europa.eu/system/
files/2016-11/annex11_01-2011_en_0.pdf (accessed on 7 October 2025).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://doi.org/10.3390/labmed3010002
