A blinded central radiology review is not the absolute gold standard for assessing patient progression. Expert clinicians argue their holistic assessment, incorporating the patient's clinical status and other biomarkers alongside scans, provides critical context that a disconnected reviewer lacks.

While smaller trials like KEYNOTE-905 can show dramatic results, they are subject to more statistical noise. Larger, thousand-patient studies like B15 and Niagara, with narrower confidence intervals, are considered closer to the true effect size and provide a more stable foundation for establishing the standard of care.

The PROTEUS trial used two pathologic endpoints. The investigator suggests Residual Cancer Burden (RCB), which measures cellularity, is a more meaningful reflection of response than just residual tumor size. The RCB endpoint showed a much larger treatment effect (30% vs. 11%) compared to the tumor size endpoint (9% vs. 1%).

The lack of a placebo arm in some adjuvant trials is not necessarily a fatal flaw. One expert view is that it mirrors real-world practice where treatments are known. This perspective places trust in the investigators' ability to assess disease progression accurately without blinding.

As an open-label trial, investigators' knowledge of treatment arms could introduce bias. Clinicians might give treated patients the "benefit of the doubt" on scans, artificially improving Disease-Free Survival (DFS). This potential bias, which wouldn't affect the harder endpoint of Overall Survival (OS), offers a plausible explanation for the discordance between the two.

Real-world data for pemigatinib in cholangiocarcinoma showed a higher response rate (59%) than the pivotal FITE-202 trial (36%). This discrepancy likely stems from the lack of standardized, centrally reviewed imaging in real-world settings, which can inflate perceived response. Comparable progression-free survival across both settings supports this interpretation.

The CREST trial's positive primary endpoint, assessed by investigators in an open-label setting, was rendered negative upon review by a blinded independent committee. This highlights the critical risk of confirmation bias and the immense weight regulators place on blinded data to determine a drug's true efficacy, especially when endpoints are subjective.

The successful KEYNOTE-564 trial intentionally used a pragmatic patient selection model based on universally available pathology data like TNM stage and grade. This approach avoids complex, inconsistently applied nomograms, ensuring broader real-world applicability and potentially smoother trial execution compared to studies relying on more niche scoring systems.

In the ASCENT-07 trial, blinded central review showed no benefit for sacituzumab, while treating investigators saw a clear benefit. This discrepancy arose because clinicians acted on new lesions or effusions that central reviewers deemed "unclear," showing how rigid trial criteria can miss nuanced clinical signals.