# Lock the biomarker cut-off before phase 3, and publish it

Source: https://onco.cc/ideas/idea-tr2-cutpoint-lock/  
OnCo record `idea-tr2-cutpoint-lock` (Idea). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Companies sometimes choose the biomarker threshold that makes their trial look best after seeing the data. Requiring the threshold to be fixed and published before the big trial starts prevents this.

## Summary

Cut-points for continuous biomarkers (PD-L1 percentages, HER2-low boundaries, ctDNA thresholds) are frequently selected from phase 2 data with multiple thresholds explored, and sometimes revised during phase 3. Regulators could require that the phase 3 protocol specify the cut-point and the assay version, derived from an independent training set, with the derivation report deposited in the trial registry before enrolment.

## Fields

- Kind: Idea
- Last checked: 2026-09-08
- Hypothesis: Trials with locked and published cut-points will show smaller shrinkage of the biomarker treatment interaction between phase 2 and phase 3 than trials with adaptive or post hoc cut-points.
- Rationale: Optimism bias from cut-point selection is a known statistical phenomenon; pre-registration is the standard remedy for analytic flexibility.
- Proposed test: Audit the last 30 biomarker-restricted approvals for cut-point derivation and revision; implement the requirement and compare interaction estimates in subsequent trials.
- Maturity: early-clinical
- Actor: regulator

## Sources

- Bottleneck evidence (Biomarkers are not validated or standardised): Fernandez et al., Examination of low ERBB2 protein expression in breast cancer tissue (JAMA Oncology 2022): https://doi.org/10.1001/jamaoncol.2021.7239

## Connected records

- ideas: [Test the drug in biomarker-negative patients too, so the biomarker can be validated](https://onco.cc/ideas/idea-tr2-marker-stratified-default/)
- targets: [PD-L1](https://onco.cc/targets/pdl1/)
- terms: [Companion diagnostic](https://onco.cc/terms/companion-diagnostic-term/)
- bottlenecks: [Biomarkers are not validated or standardised](https://onco.cc/bottlenecks/b-biomarker-validation/), [Trial design, endpoints and cost](https://onco.cc/bottlenecks/b-trial-design/)
- key papers: [Examination of Low ERBB2 Protein Expression in Breast Cancer Tissue](https://onco.cc/key-papers/paper-fernandez-jama-oncol/)

---
JSON: https://onco.cc/api/v1/entities/idea-tr2-cutpoint-lock.json