ABSTRACT
Introduction: Companion diagnostic tests (CDXs) are considered mandatory for decision-making for treatment with targeted therapies in thoracic oncology. The emergence of immunotherapy has also given rise to the development of CDXs. Some CDXs, in particular PD-L1 immunohistochemistry tests, have been questioned and re-examined for use with new combination therapies that are being evaluated in clinical trials. Current questions include: Can we establish therapeutic indications in thoracis oncology without CDXs? Would the addition of new tests benefit patient outcome?
Areas covered: This review covers the use of CDXs for decision-making in the treatment of lung cancer but also covers the limits of certain tests. It discusses the major challenges for present and future development of CDXs in daily practice.
Expert opinion: CDXs can predict the efficacy of drugs if crucial steps in development and validation are fully controlled. Future development of CDXs must consider the detection of biomarkers of resistance and toxicity that are complementary to CDXs predicting therapeutic drug efficacy. Certain CDXs that have already been developed may be of interest for new indications in the field of thoracic oncology.
Article Highlights
In the domain of thoracic oncology, CDX are a guarantee of high-quality tests associated with a therapeutic molecule, but quality also requires performance of the tests in certified laboratories.
In Europe, predictive biological tests for biomarkers do not need to be approved by the FDA but detection of these biomarkers must be done with validated tests.
In the field of thoracic oncology, the number of CDX has progressively increased. The detection of several biomarkers (activating mutations, PD-L1, TMB, to name a few) must be done in an acceptable timeframe to provide optimal treatment to patients.
Combined CDX may be developed from biomarkers with different purposes (predictive response, toxicity, resistance or hyper progression).
The complexity of CDX shall increase with the increase in therapeutic combinations.
A number of new treatments or new therapeutic associations (for example, the association of chemotherapy and immunotherapy) could be administrated without a biomarker predictive of response.
The development of artificial intelligence should help clinicians and biologist to optimize decisions into therapeutic use through algorithms obtained from complex biological analyses.
Acknowledgments
We would like to thank Ms Brahimi-Horn Christiane for editing assistance.
Declaration of interest
PH received honoraria for consultancy activities from Roche, AstraZeneca, Bristol-Myers Squibb, Biocartis, Qiagen, Thermofisher, MSD, Merck, Pfizer, Novartis. FB received honoraria for consultancy activities from AstraZeneca, Bristol-Myers Squibb, Boehringer–Ingelheim, Eli Lilly Oncology, F. Hoffmann–La Roche Ltd, Novartis, Merck, MSD, Pierre Fabre, Pfizer, and Takeda. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Reviewers Disclosure
Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.