Companion Diagnostics in Immuno-Oncology: The Next Regulatory and Commercial Frontier

Why Immuno-Oncology CDx Is Different

Companion diagnostics in immuno-oncology occupy a uniquely complex position in the precision medicine landscape. Unlike the relatively straightforward CDx model for targeted therapies — where a specific mutation or amplification predicts response to a targeted agent — the predictive biomarker landscape for immunotherapy is multidimensional, incompletely understood, and evolving faster than the regulatory frameworks designed to govern it.

PD-L1 expression. Tumor mutational burden. Microsatellite instability. Tumor-infiltrating lymphocyte density. Emerging composite biomarkers that combine immune and genomic features. The immuno-oncology CDx space is not converging on a single predictive biomarker the way that HER2 amplification converged as the CDx for trastuzumab. It is expanding.

This expansion creates both commercial opportunity and regulatory complexity that the field has not yet fully resolved.

The PD-L1 Problem as a Case Study

The PD-L1 diagnostic landscape illustrates the complexity clearly. Multiple PD-L1 immunohistochemistry assays have received FDA approval as companion or complementary diagnostics for multiple PD-1/PD-L1 checkpoint inhibitors. Each assay uses a different antibody clone, a different staining protocol, and a different scoring algorithm. The clinical cutoffs defining treatment eligibility differ across assays and across indications.

The practical consequence for oncology practice is significant: a patient’s PD-L1 status — and therefore their eligibility for specific immunotherapy regimens — can differ depending on which assay is used to test their tumor. This is not a hypothetical concern. Studies examining concordance across PD-L1 assays have consistently demonstrated meaningful discordance at clinically relevant cutoffs.

The regulatory framework has addressed this partially through the approval of interchangeability studies and through FDA guidance on the analytical comparability of PD-L1 assays. But the fundamental problem — that the approved assays measure the same analyte with different methods and different clinical cutoffs — has not been resolved at the regulatory level. It has been managed.

The PD-L1 situation is instructive for understanding where immuno-oncology CDx development is heading. As composite biomarkers — combining PD-L1, TMB, MSI, and immune gene expression signatures — move into clinical development, the analytical and regulatory complexity of the CDx programs co-developed with them will be substantially greater than the PD-L1 landscape that preceded them.

Tumor Mutational Burden: Lessons from a Broad Indication Approval

The FDA’s 2020 approval of pembrolizumab for TMB-high solid tumors — the first tissue-agnostic approval for an immunotherapy biomarker — with the FoundationOne CDx as the companion diagnostic, represented a significant regulatory milestone and a significant commercial challenge simultaneously.

The approval demonstrated that TMB is a clinically actionable biomarker across tumor types. It also demonstrated the limitations of a single CDx platform as the sole approved diagnostic for a broad tumor-agnostic indication.

FoundationOne CDx is a comprehensive genomic profiling test available through a centralized laboratory. It is not available at the point of care. The tissue requirements, the turnaround time, and the cost create access barriers for patients who would benefit from pembrolizumab based on TMB status but whose oncology care is delivered in settings without straightforward access to comprehensive genomic profiling.

The regulatory approval solved one problem — establishing TMB as a validated biomarker for pembrolizumab eligibility — and created another: a single approved CDx platform that the entire eligible patient population must access to receive a tissue-agnostic approved therapy.

The commercial and patient access implications of single-platform CDx approvals for broad indications will be an increasingly important consideration as tissue-agnostic approvals expand.

The Emerging Composite Biomarker Challenge

The next generation of immuno-oncology CDx programs will not be single-analyte tests. They will be composite biomarker systems that integrate multiple data types — genomic, proteomic, transcriptomic, and potentially imaging — to generate multidimensional predictions of immunotherapy response.

Several programs currently in clinical development are built around AI-powered composite biomarkers that combine conventional genomic features with immune gene expression signatures and computational pathology outputs derived from histology images. The predictive performance of these composite systems exceeds that of single-analyte biomarkers in early clinical data. The regulatory pathway for approving them as companion diagnostics is significantly more complex.

Three regulatory challenges are specific to AI-powered composite CDx systems in immuno-oncology.

The analytical validation framework for composite biomarkers is not standardized. Each component of the composite must be analytically validated independently, and the composite must be validated as a system. The interaction effects between components — how the composite output changes when individual component values are near their measurement boundaries — require validation approaches that are not established in current FDA guidance.

The clinical validation dataset requirements for tissue-agnostic composite biomarkers are substantial. Demonstrating predictive validity across the full range of tumor types in the intended use population requires a clinical dataset with adequate representation across indications, stages, and patient demographic characteristics. This requirement creates significant development cost and timeline implications that differ fundamentally from single-tumor-type CDx programs.

The post-market surveillance obligations for AI-powered composite CDx systems include monitoring for model performance drift as the patient population accessing the test evolves. This is a genuinely novel regulatory requirement for the diagnostics industry, and the infrastructure for meeting it — continuous performance monitoring, pre-specified drift thresholds, change control protocols — needs to be built into the development program from the earliest stages.

The Strategic Opportunity

The immuno-oncology CDx market is at an inflection point. The first-generation PD-L1 and TMB CDx programs established the commercial infrastructure and regulatory precedents. The second generation — composite biomarker systems powered by AI and multi-omic data integration — is moving through clinical development pipelines now.

The organizations that will define the next generation of immuno-oncology CDx are the ones that build regulatory strategy into composite biomarker development from the architecture stage, invest in the clinical datasets required for broad indication validation, and design post-market surveillance infrastructure for adaptive AI systems before they seek regulatory approval.

The technical capability to build these systems exists. The regulatory strategy to approve and sustain them at scale is the competitive differentiator.

#CompanionDiagnostics, #ImmunoOncology, #Precision Medicine, #FDARegulation, #Oncology

About the Author

Ashish Yadav is a Life Sciences AI, Regulatory Strategy, and Enterprise Technology executive with 20+ years of experience across pharma, biotech, MedTech, diagnostics, and digital health. He is the founder of PrecisionPulse Consulting LLC and the creator of the ARIA™ AI Governance Framework.

Connect on LinkedIn