These case studies illustrate Intrinsic Imaging’s approach to complex imaging challenges in clinical development. Each examines the sponsor’s objectives, the central review strategy and the resulting contribution to study execution.
Sponsor identities remain confidential. The studies describe work conducted within our quality management framework, supported by three FDA inspections with zero Form 483 observations and five ISO certifications.
Study Rescue
Rescuing a Global Phase III Solid Tumor Trial
70+ sites
CT, MRI, PET/CT
RECIST 1.1 BICR
Challenge
The incumbent core laboratory exhibited substantial inter-reader discordance, an accumulating query backlog and inconsistent protocol adherence across more than 70 global sites. Together, these deficiencies threatened the sponsor's NDA submission timeline.
Our Solution
Intrinsic Imaging revised the blinded independent central review charter to establish dual independent reads and adjudication. More than 4200 subject scan series were transferred to our validated imaging platform, accompanied by targeted site retraining in image acquisition and submission.
Results
Outstanding imaging queries resolved within 45 days
98.4% reader concordance after transition
Database locked two weeks ahead of the revised deadline
Oncology
Phase III Immunotherapy Trial Using RECIST 1.1 and iRECIST
90+ sites in 18 countries
CT, MRI
RECIST 1.1 and iRECIST
Challenge
The sponsor required consistent differentiation of pseudoprogression from true disease progression across more than 90 sites, with concurrent assessment under two response frameworks.
Our Solution
Subspecialty oncologic radiologists conducted dual independent assessments, supported by automated discordance triggers that routed unconfirmed progression findings for adjudication.
Results
Consistent central assessment of 12500+ timepoints
Full audit trail supported the BLA submission
Zero Form 483 observations during the sponsor inspection
AI/CAD
Pivotal MRMC Reader Study for an AI 510(k) Submission
Multi-reader multi-case
Reference-standard panel
Part 11 reading environment
Challenge
An AI developer required a pivotal multi-reader multi-case study supported by a rigorously established reference standard to evaluate reader performance with and without algorithmic assistance.
Our Solution
Intrinsic Imaging convened a consensus reference-standard panel of fellowship-trained subspecialists. Reading sessions were randomized and counterbalanced, with washout periods incorporated into the workflow within our Part 11 compliant reading environment.
Results
Audit-ready reference-standard datasets delivered
510(k) clearance received
Reader study design reviewed without FDA questions
Central Nervous System
Phase II Alzheimer's Disease Volumetric MRI and Amyloid PET Study
55 sites
3T MRI, amyloid PET
Volumetric and visual PET endpoints
Challenge
Differences among scanner vendors introduced variability into volumetric MRI measurements and amyloid PET assessments across 55 academic and commercial imaging centers.
Our Solution
Intrinsic Imaging implemented phantom-based site qualification and automated verification of acquisition parameters at submission. Central assessment combined quantitative volumetric analysis of atrophy with standardized visual PET scoring.
Results
Cross-site measurement variation reduced by 38%
Reliable secondary and exploratory biomarker endpoints
Comparable imaging data across scanner vendors
Medical Device
IDE Pivotal Transcatheter Valve Trial
Structural heart
CT, fluoroscopy, echocardiography
Pre- and post-implant assessment
Challenge
The transcatheter valve replacement protocol required precise preprocedural anatomical sizing and consistent longitudinal core laboratory assessment following implantation.
Our Solution
Cardiovascular radiologists and imaging engineers applied 3D multiplanar reconstruction and calibrated geometric measurement techniques to the assessment of each implant.
Results
100% technical protocol compliance across implant sites
Core lab data included in the PMA submission
Core laboratory assessments accepted by FDA reviewers
Cardiology
Global Cardiovascular Outcomes Trial With CCTA and Cardiac MRI
60 sites
CCTA, cardiac MRI
Plaque, stenosis and LVEF
Challenge
The trial required reproducible quantitative plaque characterization, stenosis grading and ejection fraction measurement across 60 sites operating different scanner models.
Our Solution
Intrinsic Imaging established standardized acquisition guidelines and a dedicated quality control gateway to identify scans unsuitable for interpretation before they entered central review.
Results
99.1% evaluable scan rate across 3400+ patients
Inter-reader variability maintained below 4%
Ongoing reader calibration throughout the trial
Musculoskeletal
Phase III Knee Osteoarthritis DMOAD Trial
24-month follow-up
Knee MRI, radiographs
JSN and WORMS
Challenge
The sponsor required reproducible assessment of submillimeter joint space narrowing and longitudinal changes in WORMS scores over a 24-month observation period.
Our Solution
Fellowship-trained musculoskeletal radiologists conducted paired longitudinal assessments of high-resolution knee MRI and radiographs, with readers blinded to timepoint order.
Results
Reader intraclass correlation above 0.94
Central imaging assessments supported the primary structural endpoint
Consistent longitudinal scoring across readers
Gastroenterology
Phase IIb Inflammatory Bowel Disease Trial
Crohn's disease and ulcerative colitis
Endoscopy video, MR enterography
SES-CD and Mayo Endoscopic Score
Challenge
Site-level interpretation of endoscopic video and MR enterography introduced substantial subjectivity into assessments of disease activity.
Our Solution
Intrinsic Imaging established a blinded dual-reader central review workflow incorporating anonymized video, standardized ulcer measurement and quantitative assessment of bowel wall thickness.
Results
Inter-reader scoring discordance reduced by 42%
Reduced incidence of inappropriate screening exclusions
Consistent eligibility decisions across sites
Ophthalmology
Phase II Geographic Atrophy and Neovascular AMD Trial
SD-OCT, fundus autofluorescence
Lesion area measurement
Multi-reader consensus
Challenge
The trial required precise central measurement of geographic atrophy area and consistent longitudinal assessment of choroidal neovascularization.
Our Solution
The ophthalmic reading team performed lesion segmentation, resolving ambiguous lesion boundaries through multi-reader consensus.
A geographically dispersed trial in an orphan indication required an accelerated startup strategy and timely resolution of imaging queries.
Our Solution
Intrinsic Imaging finalized the imaging charter in under three weeks and established electronic image submission with automated DICOM quality checks. Dedicated project management and adjudication resources supported coordinated execution across sites.
Results
Initial image assessment completed within 48 hours of acquisition
Database locked 10 days after last patient out
Approval granted under Priority Review
Study Rescue
Mid-Study Transfer of a Phase III Multiple Sclerosis MRI Program
80 sites
Brain MRI
Gd-enhancing and new T2 lesions
Challenge
The previous core laboratory's lesion assessments exhibited inconsistent inter-reader agreement, while more than 1800 MRI examinations remained pending central review.
Our Solution
Intrinsic Imaging transferred the complete image archive and reconciled lesion-counting conventions within a harmonized charter. Before review resumed, neuroradiologists received focused training in the prior assessment conventions to support continuity of interpretation.
Results
Read backlog cleared within 60 days
Lesion-count agreement exceeded 95% in repeat-read samples
No change to the primary endpoint timeline
Study Rescue
Rescue of a Stalled Peripheral Vascular Device Trial
35 sites
Duplex ultrasound, angiography
Patency and restenosis
Challenge
Inadequate ultrasound submissions and incomplete angiographic series rendered nearly a quarter of follow-up timepoints unevaluable, compromising the longitudinal imaging dataset.
Our Solution
Intrinsic Imaging revised acquisition guidance and delivered targeted site retraining. Quality control checks at submission, coupled with same-week feedback, enabled sites to address image deficiencies promptly.
Results
Unevaluable timepoints reduced from 23% to below 5%
Follow-up data recovered for the 12-month endpoint
Core lab report delivered for the PMA submission
Oncology
Phase II Lymphoma Trial Using Lugano Classification
45 sites
FDG PET/CT
Lugano 2014 and Deauville scoring
Challenge
Substantial variation in site-level Deauville scoring undermined confidence in the classification of complete metabolic response.
Our Solution
Nuclear medicine physicians conducted blinded dual assessments, incorporating SUV calibration checks and adjudication of discordant Deauville scores.
Results
Consistent response classification across participating sites
Adjudication rate held below 12%
Reconciled central review dataset delivered for interim analysis
Oncology
Glioblastoma Trial Assessed With RANO Criteria
30 sites
Brain MRI with perfusion
RANO and mRANO
Challenge
Following chemoradiation, treatment-related imaging changes complicated the differentiation of pseudoprogression from true tumor progression.
Our Solution
Neuroradiologists applied RANO criteria with confirmatory imaging and incorporated perfusion assessments to inform the interpretation of suspected pseudoprogression.
Results
Reduced incidence of premature progression determinations
Consistent confirmation workflow across sites
Inspection-ready read audit trail
Oncology
Metastatic Prostate Cancer Trial With PCWG3 Bone Scan Review
65 sites
Bone scintigraphy, CT
PCWG3 and RECIST 1.1
Challenge
Inconsistent site-level application of the 2+2 bone scan progression rule threatened the accuracy of radiographic progression-free survival assessments.
Our Solution
Intrinsic Imaging integrated PCWG3 decision logic into the central reading workflow, ensuring that newly identified bone lesions underwent confirmation on subsequent scans.
Results
Central assessment of 2900+ bone scans
Consistent rPFS event dating
No outstanding progression-logic queries at database lock
Oncology
Hepatocellular Carcinoma Trial Using mRECIST
50 sites in 12 countries
Multiphase CT, MRI
mRECIST and LI-RADS
Challenge
Variation in arterial-phase acquisition timing across sites compromised the reliability of viable tumor measurements.
Our Solution
Intrinsic Imaging standardized multiphase acquisition protocols and verified phase timing for every submission before central interpretation by abdominal radiologists.
Results
Evaluable scan rate above 97%
Reliable viable tumor measurements
Central review supported the regulatory filing
AI/CAD
Ground-Truth Dataset for a Breast Cancer Detection Algorithm
Digital mammography, tomosynthesis
Lesion-level annotation
Pathology-confirmed truth
Challenge
An AI developer required a substantial test dataset with an independently established reference standard, aligned with FDA expectations for a standalone performance study.
Our Solution
Fellowship-trained breast radiologists performed lesion-level annotation within our Part 11 reading environment. Pathology correlation informed reference-standard determinations, with expert consensus used to resolve disagreements.
Results
Reference standards established for 5000+ cases with complete audit trails
Dataset sequestered from algorithm training
Standalone performance study completed on schedule
AI/CAD
MRMC Reader Study for a Chest CT Lung Nodule Tool
Low-dose chest CT
18 readers
Aided and unaided reads
Challenge
The study required participation from readers with varied levels of experience and a reading workflow that mitigated recall bias between aided and unaided sessions.
Our Solution
Intrinsic Imaging recruited and trained the reader cohort, implemented randomized reading sequences with a four-week washout period and convened a reference-standard panel of thoracic radiologists.
Results
All reader assessments completed within 10 weeks
Improved reader sensitivity with algorithmic assistance
Study data accepted in the 510(k) submission
Medical Device
Spinal Fusion Device Trial With Radiographic Fusion Assessment
28 sites
CT, flexion-extension radiographs
24-month fusion endpoint
Challenge
Objective, reproducible assessment of fusion status and implant subsidence was essential throughout two years of postoperative follow-up.
Our Solution
Musculoskeletal radiologists used calibrated measurement tools to quantify angular and translational motion, alongside CT assessment of bridging bone formation.
Results
Reader agreement above 0.90 on fusion status
Complete 24-month imaging dataset
Core lab findings included in the IDE final report
Central Nervous System
Parkinson's Disease Trial With DaT-SPECT Imaging
40 sites
DaT-SPECT, 3T MRI
Striatal binding ratios
Challenge
Differences in imaging systems across sites introduced variability into quantitative striatal binding ratio measurements.
Our Solution
Intrinsic Imaging qualified sites through phantom imaging and harmonized reconstruction parameters. Central review integrated quantitative measurements with standardized visual interpretation.
Results
Reduced cross-site signal variation
Accurate imaging-based enrollment screening
Reliable longitudinal biomarker data
Cardiology
Heart Failure Trial With Echocardiography Core Lab
75 sites
Transthoracic echocardiography
LVEF, GLS and LV volumes
Challenge
Variation in echocardiographic image quality and measurement methodology across sites threatened the consistency of LVEF and myocardial strain endpoints.
Our Solution
Intrinsic Imaging trained sonographers in a standardized acquisition protocol and established blinded central assessment of LVEF, global longitudinal strain and ventricular volumes.
Results
Central assessment of 4100+ echocardiographic studies
LVEF reader variability below 5%
Endpoint data delivered on time for the primary analysis
Musculoskeletal
Rheumatoid Arthritis Trial With Modified Sharp Scoring
60 sites
Hand and foot radiographs
mTSS erosion and JSN
Challenge
Detection of subtle changes in erosion scores over 52 weeks required a reproducible central assessment methodology.
Our Solution
Two blinded musculoskeletal radiologists assessed paired radiographs presented in randomized timepoint order, with adjudication used to resolve substantial scoring differences.
Results
Smallest detectable change reported per protocol
Consistent scoring across readers
Central imaging data supported the structural endpoint
Gastroenterology
NASH Trial With MRI-PDFF and Liver Stiffness
50 sites
MRI-PDFF, MR elastography
Liver fat and stiffness
Challenge
Differences among scanner vendors affected the comparability of liver fat fraction and stiffness measurements across participating sites.
Our Solution
Intrinsic Imaging implemented phantom-based scanner qualification, verified sequence parameters at image submission and conducted central quantitative assessment of liver fat and stiffness.
Results
Evaluable scan rate above 96%
Consistent PDFF values across vendors
Reliable data for the Phase IIb dose selection
Ophthalmology
Diabetic Macular Edema Trial With OCT Thickness Endpoints
70 sites
SD-OCT, fluorescein angiography
Central subfield thickness
Challenge
Device-dependent OCT segmentation errors compromised the reliability and comparability of retinal thickness measurements.
Our Solution
Certified ophthalmic graders reviewed automated segmentation, corrected identified errors and assessed angiographic leakage through central interpretation.
Results
Retinal thickness measurements corrected across OCT devices
Grader agreement above 0.92
Primary endpoint data delivered on schedule
Accelerated Programs
Rapid Startup for a Pediatric Neuroblastoma Trial
25 sites in 9 countries
MIBG scintigraphy, CT, MRI
Curie score and INRC
Challenge
The sponsor required central review to become operational within weeks for a trial enrolling a limited pediatric population.
Our Solution
Intrinsic Imaging completed the imaging charter and site training within four weeks. Dedicated nuclear medicine readers supported real-time eligibility assessments to maintain enrollment momentum.
