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.
Oncology
Rescuing a Global Phase III Solid Tumor Trial
- 70+ sites
- CT, MRI, PET/CT
- RECIST 1.1 BICR
Challenge
Midway through a global Phase III solid tumor program, the sponsor confronted a central review process with substantial operational deficiencies. The incumbent core lab's reads were marked by substantial inter-reader disagreement, and a growing backlog of unresolved queries was compromising the reliability of the imaging dataset. Protocol adherence varied widely across more than 70 sites spanning multiple continents, compounding the inconsistency of the response assessments on which the trial's primary endpoint depended. With the NDA submission timeline already fixed, the sponsor needed a partner capable of assuming control of the program mid-study without sacrificing continuity, data integrity, or regulatory defensibility.
Read the case studyOncology
Phase III Immunotherapy RECIST and iRECIST
- 90+ sites in 18 countries
- CT, MRI
- RECIST 1.1 and iRECIST
Challenge
Immunotherapy trials present a distinctive interpretive challenge: apparent radiographic progression may reflect treatment-related inflammation rather than true tumor growth. The sponsor required central reviewers to distinguish pseudoprogression from genuine disease progression with consistency across more than 90 sites in 18 countries, while applying two response frameworks—RECIST 1.1 and iRECIST—in parallel. Any variability in how unconfirmed progression was identified, confirmed, or dated threatened to distort the trial's efficacy conclusions and invite regulatory scrutiny.
Read the case studyAI/CAD
MRMC Reader Study for AI 510(k) Submission
- Multi-reader multi-case
- Reference-standard panel
- Part 11 reading environment
Challenge
An AI developer preparing a 510(k) submission required a pivotal multi-reader multi-case study to evaluate whether algorithmic assistance improved reader performance. The study's credibility depended in part on its reference standard: every case needed a systematically established reference determination supported by documentation suitable for FDA review. The design also had to control for recall bias and ordering effects across reading sessions, while documenting every step in a manner suitable for regulatory inspection.
Read the case studyCentral Nervous System
Phase II Alzheimer's Amyloid PET and MRI
- 55 sites
- 3T MRI, amyloid PET
- Volumetric and visual PET endpoints
Challenge
In a Phase II Alzheimer's disease program, the trial's biomarker endpoints depended on detecting subtle changes in brain volume and amyloid burden over time. Yet the 55 participating academic and commercial imaging centers operated scanners from multiple vendors, each introducing its own measurement variability into volumetric MRI quantification and amyloid PET assessment. Without adequate control, this cross-site variability could obscure treatment-related changes and compromise interpretation of both the secondary and exploratory biomarker endpoints.
Read the case studyMedical Device
IDE Pivotal Transcatheter Valve Trial
- Structural heart
- CT, fluoroscopy, echocardiography
- Pre- and post-implant assessment
Challenge
A pivotal IDE trial of a transcatheter valve replacement system required standardized imaging for anatomical assessment and longitudinal evaluation of device performance. Precise anatomical sizing before the procedure was essential to device selection and patient safety, while consistent core lab review of post-implant imaging over extended follow-up was required to document device performance, positioning, and durability. Measurements taken at different timepoints and by different readers had to remain directly comparable, and every assessment needed to satisfy the evidentiary standard of a PMA submission.
Read the case studyCardiovascular & Thoracic
Global Cardiovascular Trial With CCTA and MRI
- 60 sites
- CCTA, cardiac MRI
- Plaque, stenosis and LVEF
Challenge
This global cardiovascular outcomes trial required reproducible quantification of coronary plaque burden, stenosis grading, and left ventricular ejection fraction across 60 sites operating a heterogeneous mix of scanner models. Each measurement contributed to endpoint assessment, and even modest inter-reader or inter-site variability could obscure a potential treatment effect. The sponsor required acquisition and interpretation methods that supported measurement comparability across sites and scanner platforms, with technical limitations identified and documented throughout the program.
Read the case studyMusculoskeletal
Phase III Knee Osteoarthritis DMOAD Trial
- 24-month follow-up
- Knee MRI, radiographs
- JSN and WORMS
Challenge
Evaluating disease modification in knee osteoarthritis requires reproducible measurement of submillimeter changes in joint space narrowing over extended follow-up. The sponsor required readings of sufficient precision to register these minute structural changes, together with consistent tracking of WORMS scores across 24 months of serial MRI examinations. Because the primary structural endpoint rested on longitudinal comparison, any drift in reader interpretation or measurement technique over the study's duration would directly compromise the trial's conclusions.
Read the case studyGastroenterology & Metabolic
Phase IIb Inflammatory Bowel Disease Trial
- Crohn's disease and ulcerative colitis
- Endoscopy video, MR enterography
- SES-CD and Mayo Endoscopic Score
Challenge
In inflammatory bowel disease trials, endoscopic and cross-sectional assessments of disease activity are inherently interpretive. When scoring is performed at individual sites, that subjectivity translates directly into variability: the same endoscopic finding may be scored differently by different investigators, and eligibility decisions can diverge accordingly. The sponsor needed to reduce this inconsistency from both the screening process and the SES-CD and Mayo Endoscopic Score assessments that anchored the trial's efficacy evaluations.
Read the case studyOphthalmology
Geographic Atrophy and Neovascular AMD Trial
- SD-OCT, fundus autofluorescence
- Lesion area measurement
- Multi-reader consensus
Challenge
The trial's efficacy assessments depended on precise central measurement of geographic atrophy lesion area and consistent longitudinal tracking of choroidal neovascularization. Lesion boundaries in retinal imaging are frequently indistinct, and small differences in how those boundaries are delineated translate into meaningful differences in measured area—differences large enough to confound endpoint analysis when measurements are compared across timepoints and readers.
Read the case studyOncology
Fast-Track Rare Disease Oncology Trial
- 40 sites in 14 countries
- Fewer than 3 patients per site
- Priority Review program
Challenge
Registration trials in rare diseases face substantial recruitment and operational constraints. With fewer than three patients per site spread across 40 sites in 14 countries, each imaging timepoint contributed materially to the limited dataset, and any delay in central review directly threatened enrollment momentum. The sponsor's expedited development and submission schedule required rapid core lab activation and responsive query resolution without compromising the methodological rigor or documentation of imaging assessments.
Read the case studyCentral Nervous System
Mid-Study Transfer of a Phase III MS MRI
- 80 sites
- Brain MRI
- Gd-enhancing and new T2 lesions
Challenge
A Phase III multiple sclerosis program experienced persistent central review deficiencies under its previous core lab. Lesion assessments demonstrated weak inter-reader agreement, calling into question the reliability of the gadolinium-enhancing and new T2 lesion counts on which the trial's endpoints depended. Compounding the problem, more than 1,800 MRI examinations awaited central review, and the backlog continued to grow. The sponsor needed to transfer the program mid-study while preserving interpretive continuity with the assessments already completed.
Read the case studyMedical Device
Rescuing a Stalled Peripheral Vascular Trial
- 35 sites
- Duplex ultrasound, angiography
- Patency and restenosis
Challenge
A peripheral vascular medical device trial encountered substantial delays attributable to inadequate imaging quality. Substandard duplex ultrasound submissions and incomplete angiographic series had rendered nearly a quarter of follow-up timepoints unevaluable, placing the longitudinal patency and restenosis dataset—and with it the trial's 12-month endpoint—at substantial risk. Restoring endpoint evaluability required correction of site acquisition practices in addition to improvements in central interpretation and submission review.
Read the case studyOncology
Lymphoma Trial Using Lugano Classification
- 45 sites
- FDG PET/CT
- Lugano 2014 and Deauville scoring
Challenge
In lymphoma trials, treatment decisions and endpoint determinations rest on metabolic response classification, yet Deauville scoring can vary materially when performed independently at individual sites. Wide divergence in site-level scoring had made complete metabolic response classifications difficult to trust, threatening the integrity of the trial's response-based endpoints and complicating the planned interim analysis. The sponsor required a central review process that could standardize interpretive conventions and document the resolution of discordant assessments.
Read the case studyOncology
Glioblastoma Trial With RANO Criteria
- 30 sites
- Brain MRI with perfusion
- RANO and mRANO
Challenge
Following chemoradiation, glioblastoma patients frequently exhibit treatment-related imaging changes that mimic tumor progression. Distinguishing pseudoprogression from true progression is among the most difficult judgments in neuro-oncology imaging, and premature or inconsistent progression determinations can distort progression-free survival analyses and lead to inappropriate treatment changes. The trial required central application of RANO and modified RANO criteria rigorous enough to manage this ambiguity uniformly across all 30 sites.
Read the case studyOncology
Metastatic Prostate Cancer With PCWG3 Review
- 65 sites
- Bone scintigraphy, CT
- PCWG3 and RECIST 1.1
Challenge
Assessment of radiographic progression-free survival in metastatic prostate cancer requires consistent application of PCWG3 bone progression conventions, including the 2+2 rule and its specified lesion-count and confirmatory imaging requirements. Across 65 sites, this rule was being applied inconsistently, with the flare phenomenon and unconfirmed lesions variously counted or disregarded. Such inconsistency threatened the reliability of the rPFS endpoint and the reproducibility of progression event dating across the trial.
Read the case studyOncology
Hepatocellular Carcinoma Trial Using mRECIST
- 50 sites in 12 countries
- Multiphase CT, MRI
- mRECIST and LI-RADS
Challenge
Assessment of hepatocellular carcinoma under mRECIST relies on measuring viable, arterially enhancing tumor rather than overall lesion size, making appropriate arterial-phase timing essential for reliable measurement. Across 50 sites in 12 countries, variation in multiphase acquisition timing had compromised the reliability of viable tumor measurements, since a poorly timed arterial phase can render enhancement unassessable. The trial's response-based endpoints required that every examination capture the contrast phases necessary for valid mRECIST assessment.
Read the case studyAI/CAD
Ground Truth for a Breast Cancer Detection AI
- Digital mammography, tomosynthesis
- Lesion-level annotation
- Pathology-confirmed truth
Challenge
An AI developer required a large, independent test dataset to support a standalone performance study of a breast cancer detection algorithm. The evidentiary requirements for such studies are substantial: the reference standard must be established independently of the algorithm, truth determinations must be defensible at the lesion level, and the dataset must remain independently sequestered from any use in algorithm training. Assembling thousands of cases meeting these criteria demanded both scale and methodological discipline.
Read the case studyAI/CAD
Chest CT Lung Nodule MRMC Reader Study
- Low-dose chest CT
- 18 readers
- Aided and unaided reads
Challenge
Evaluating the effect of a lung nodule detection tool on reader performance required a multi-reader multi-case study comparing reader performance with and without algorithmic assistance. The design demanded readers spanning a range of experience levels to support generalizable conclusions, together with a reading workflow that would prevent recall of prior interpretations from contaminating the comparison between aided and unaided sessions. Every element of the design had to satisfy the evidentiary expectations of a 510(k) submission.
Read the case studyMedical Device
Spinal Fusion Device Trial With Fusion Review
- 28 sites
- CT, flexion-extension radiographs
- 24-month fusion endpoint
Challenge
Evaluating the effectiveness of a spinal fusion medical device requires reproducible imaging evidence of fusion and implant stability across longitudinal follow-up. Fusion status and subsidence had to be assessed consistently across 24 months of post-surgical follow-up, with measurements sufficiently reproducible to support the trial's fusion endpoint. Given the well-documented subjectivity of radiographic fusion assessment, the sponsor needed a central methodology that could withstand scrutiny in an IDE final report.
Read the case studyCentral Nervous System
Parkinson's Trial With DaT-SPECT Imaging
- 40 sites
- DaT-SPECT, 3T MRI
- Striatal binding ratios
Challenge
Quantitative striatal binding ratios derived from DaT-SPECT imaging served both as an enrollment screening tool and as a longitudinal biomarker in this Parkinson's disease trial. However, differences in imaging systems and reconstruction practices across 40 sites introduced variability into the measurements used for screening and longitudinal assessment. Without harmonization, apparent differences in binding ratios risked reflecting scanner characteristics rather than underlying biology, undermining both screening accuracy and biomarker reliability.
Read the case studyCardiovascular & Thoracic
Heart Failure Trial Echocardiography Core Lab
- 75 sites
- Transthoracic echocardiography
- LVEF, GLS and LV volumes
Challenge
In heart failure trials, ejection fraction and myocardial strain measurements frequently determine both eligibility and efficacy, yet echocardiography is among the most operator-dependent of imaging modalities. Across 75 sites, differences in image acquisition quality and measurement technique placed the LVEF, global longitudinal strain, and ventricular volume endpoints at risk. The sponsor required central assessment capable of delivering consistent, reproducible quantification from imaging acquired under real-world site conditions.
Read the case studyMusculoskeletal
Rheumatoid Arthritis Modified Sharp Scoring
- 60 sites
- Hand and foot radiographs
- mTSS erosion and JSN
Challenge
Structural progression in rheumatoid arthritis is measured in small increments of erosion and joint space narrowing, and detecting change over 52 weeks demands a scoring methodology of exceptional repeatability. Modified total Sharp scoring is labor-intensive and interpretive, and without rigorous central control, reader drift and inter-reader variability can easily exceed the magnitude of the treatment effect the trial seeks to demonstrate. The sponsor needed central assessment disciplined enough to report the smallest detectable change with confidence.
Read the case studyGastroenterology & Metabolic
NASH Trial With MRI-PDFF and Liver Stiffness
- 50 sites
- MRI-PDFF, MR elastography
- Liver fat and stiffness
Challenge
Quantitative liver imaging biomarkers are only as reliable as the scanners that produce them. In this NASH program, vendor differences in MRI-PDFF and MR elastography implementation threatened the comparability of liver fat fraction and stiffness measurements across 50 sites. Because these quantitative endpoints would inform Phase IIb dose selection, even modest cross-vendor bias could influence the interpretation of biomarker differences relevant to subsequent development.
Read the case studyOphthalmology
Diabetic Macular Edema Trial OCT Endpoints
- 70 sites
- SD-OCT, fluorescein angiography
- Central subfield thickness
Challenge
Central subfield thickness measured by OCT anchored this diabetic macular edema trial's primary endpoint, yet automated retinal layer segmentation can introduce platform-dependent boundary errors that compromise measurement reliability. With 70 sites using multiple OCT platforms, uncorrected segmentation errors threatened the comparability of thickness measurements on which the efficacy analysis depended.
Read the case studyOncology
Rapid Pediatric Neuroblastoma Trial Startup
- 25 sites in 9 countries
- MIBG scintigraphy, CT, MRI
- Curie score and INRC
Challenge
Pediatric oncology trials often enroll small, geographically dispersed populations, making timely study activation and screening assessment important operational priorities. The sponsor required central review capability—encompassing MIBG scintigraphy, CT, and MRI with Curie scoring and INRC response assessment—to be fully operational within weeks, across 25 sites in 9 countries. Eligibility determinations had to be returned promptly so that central imaging review could support enrollment decisions within the required timelines.
Read the case studyGastroenterology & Metabolic
Obesity Trial With DXA Body Composition and MRI Fat Quantification
- 90 sites in 14 countries
- DXA, abdominal MRI, MRI-PDFF
- Lean mass, visceral fat and liver fat
Challenge
Weight loss alone no longer satisfies regulators, payers or clinicians evaluating an incretin-based obesity therapy. The sponsor of this Phase III program needed to show not only how much weight participants lost but what that weight consisted of: whether fat mass fell while lean mass was preserved, and whether visceral adipose tissue and liver fat, the depots most closely tied to cardiometabolic risk, changed in parallel. Those questions placed dual-energy X-ray absorptiometry and quantitative MRI at the center of the trial's key secondary endpoints.
Read the case studyOncology
Imaging Core Lab Support Through NDA Submission and FDA Inspection
- 55 sites
- CT, MRI
- RECIST 1.1 BICR and FDA inspection
Challenge
Progression-free survival by blinded independent central review was the primary endpoint of this Phase III solid tumor trial, which meant that the imaging core lab's work would be examined by the FDA as closely as the sponsor's own clinical data. The sponsor planned to file a New Drug Application within months of database lock and expected both information requests during review and a pre-approval inspection of the imaging vendor.
Read the case studyOncology
Multiple Myeloma Trial With IMWG Imaging Criteria
- 120 sites in 22 countries
- Whole-body low-dose CT, FDG PET/CT, MRI
- IMWG response and extramedullary disease
Challenge
Response in multiple myeloma is defined largely by laboratory measures, which can lead sponsors to underestimate the imaging work a global trial requires. In this Phase III study the imaging component was decisive in three places: documenting bone disease and extramedullary plasmacytomas at baseline, confirming progression when new bone lesions or soft-tissue masses appeared, and assessing metabolic response on FDG PET/CT in the subset of patients with imaging-evaluable disease.
Read the case studyCentral Nervous System
ARIA Safety Monitoring in an Anti-Amyloid Alzheimer's Trial
- 110 sites
- 3T MRI with FLAIR and T2*-GRE/SWI
- ARIA-E and ARIA-H severity grading
Challenge
Amyloid-related imaging abnormalities are the defining safety signal of anti-amyloid antibody therapy. ARIA with edema or effusion and ARIA with hemosiderin deposition are usually asymptomatic but can be serious, and dosing decisions at each visit depend on knowing whether they are present and how severe they are. In this Phase III trial, protocol-specified safety MRI scans were scheduled before selected infusions, and the investigator could not proceed with dosing until the central read was available.
Read the case studyOncology
Radioligand Therapy Trial With PSMA PET Eligibility and Dosimetry
- 60 sites
- PSMA PET/CT, post-therapy SPECT/CT, CT and bone scan
- Eligibility reads, PCWG3 and dosimetry
Challenge
Radioligand therapy ties patient selection to imaging more tightly than almost any other treatment class. In this Phase III trial of a PSMA-targeted radiopharmaceutical in metastatic castration-resistant prostate cancer, a participant could be randomized only if central review confirmed PSMA-positive disease on PET/CT and the absence of PSMA-negative measurable lesions. Randomization windows were short, so eligibility reads had to be returned quickly and consistently to 60 sites, several of which used different PSMA PET tracers with different normal biodistribution.
Read the case studyCardiovascular & Thoracic
Idiopathic Pulmonary Fibrosis Trial With Quantitative HRCT
- 45 sites
- Thin-section non-contrast HRCT
- UIP pattern confirmation and quantitative fibrosis score
Challenge
Forced vital capacity remains the accepted primary endpoint in idiopathic pulmonary fibrosis, but it is noisy, effort-dependent and slow to show treatment effect. The sponsor of this Phase IIb/III program wanted imaging to do two jobs: confirm at screening that each participant's high-resolution CT showed a usual interstitial pneumonia pattern consistent with the diagnosis, and quantify the extent of fibrotic lung over 52 weeks as a key secondary endpoint that could support the FVC result.
Read the case studyOncology
Phase I/IIa Imaging Endpoint Strategy for a Biotech Go/No-Go Decision
- 12 sites
- FDG PET/CT, contrast CT
- RECIST 1.1, PERCIST and CT volumetrics
Challenge
An emerging biotechnology company was taking its first oncology asset into a Phase I/IIa study and had a single decision to make at the end of dose escalation: whether to open an expansion cohort, a step that would commit most of its remaining capital. The company's board wanted evidence of biological activity by week 8 of treatment in the recommended-dose cohort. Conventional RECIST 1.1 response at that timepoint would almost certainly be stable disease in most patients, which answers nothing.
Read the case studyCardiovascular & Thoracic
Imaging Data Transfer and Query Resolution Integrated With Sponsor EDC
- 140 sites
- Echocardiography, cardiac CT
- EDC-integrated transfers and query management
Challenge
In a large cardiovascular outcomes trial, imaging is one data stream among many, and its value depends on how cleanly it joins the rest. This 140-site study collected echocardiography at four visits and cardiac CT in a sub-study, and the sponsor's data management team had to reconcile every central imaging result with the visit record in its electronic data capture system before each data safety monitoring board meeting and interim analysis.
Read the case studyCardiovascular & Thoracic
Rare Disease Cardiomyopathy Trial With Echo and Cardiac MRI
- 80 sites in 16 countries
- Echocardiography, cardiac MRI
- LVOT gradient, wall thickness and LVEF
Challenge
Obstructive hypertrophic cardiomyopathy is a rare disease in which the imaging measurements are the endpoints. In this Phase III trial, the primary efficacy measure was the change in left ventricular outflow tract gradient at rest and with Valsalva provocation, assessed by echocardiography, and cardiac MRI quantified left ventricular mass and maximal wall thickness in a sub-study. At the same time, the mechanism of the investigational drug carried a known risk of reducing systolic function, so the protocol required that any left ventricular ejection fraction below a defined threshold be reported promptly to guide dose adjustment.
Read the case studyGastroenterology & Metabolic
ADPKD Trial With Total Kidney Volume MRI
- 50 sites
- Non-contrast T2-weighted MRI
- Height-adjusted total kidney volume and Mayo imaging class
Challenge
Autosomal dominant polycystic kidney disease progresses slowly, and kidney function can remain normal for years while the kidneys enlarge. Total kidney volume has been qualified by regulators as a prognostic enrichment biomarker precisely because it identifies the patients most likely to decline during a trial. In this Phase III program, height-adjusted total kidney volume on MRI served two purposes: to classify each candidate at screening into the Mayo imaging class that determined eligibility, and to measure the rate of kidney growth over 24 months as a key secondary endpoint.
Read the case studyMedical Device
Cervical Disc Replacement IDE Trial With Range-of-Motion Analysis
- 25 sites
- Flexion-extension radiographs, CT
- Range of motion, heterotopic ossification and adjacent segments
Challenge
A cervical total disc replacement is intended to do what a fusion cannot: preserve motion at the treated level. The pivotal IDE trial for this device therefore had to demonstrate two things on imaging, that the implant maintained a functional range of motion through 24 months and that it did not produce the complications that can defeat that purpose, chiefly heterotopic ossification that bridges and immobilizes the segment, device migration or subsidence, and accelerated degeneration at the adjacent levels. The device was being compared with anterior cervical discectomy and fusion, so the control arm required fusion assessment as well.
Read the case study