Pivotal Phase III Comparison of an Oral Chemotherapeutic Against the Leading Metastatic Breast Cancer Therapy

Challenge

This pivotal Phase III trial compared the safety, tolerability and tumor response of a new oral chemotherapeutic against the leading therapy for metastatic breast cancer. Because the comparator was an established standard, the margin the sponsor needed to demonstrate was narrow, and any systematic bias or noise in central tumor assessment could obscure a real difference between arms.

With more than 5,500 radiological reads across a mixed population of measurable soft-tissue disease and non-measurable bone disease, the sponsor required blinded reads that were insensitive to treatment arm, consistent handling of bone-only progression, and a review process whose documentation would stand up to regulatory inspection of a comparative efficacy claim.

Intrinsic Imaging’s Solution

Intrinsic Imaging developed the imaging charter with acquisition standards for CT, MRI and bone scintigraphy, qualified sites before enrollment and trained site staff on timepoint scheduling and the submission requirements for correlative imaging of bone lesions. Image transfer and processing were centralized with quality control at receipt.

Oncology radiologists from the company's 12-member oncology team performed blinded independent double reads under RECIST 1.1, with adjudication on discordant response categories and progression dates. The charter specified how new bone-scan lesions would be confirmed on CT or MRI, how flare would be distinguished from progression, and how non-target disease would be assessed for unequivocal progression.

Reader blinding to treatment arm was protected through image de-identification and randomized case presentation, and reader agreement was monitored throughout, with recalibration where needed. The locked imaging dataset was reconciled with the sponsor's clinical database before each analysis.

Intrinsic Imaging’s Results

• More than 5,500 blinded independent reads completed across 95 sites in 17 countries with adjudication on every discordant assessment
• Bone-only progression handled under a documented confirmation rule, with roughly 18% of progression events requiring correlative confirmation
• Reader agreement on overall response held above 85% across the trial with two recalibration sessions documented
• Imaging dataset reconciled with the clinical database and locked on schedule for the primary analysis under an FDA-audited quality system