Delivering on the promise of precision medicine in oncology depends on the predictive performance of biomarker tests. Immuno-oncology treatments can be lifesaving – but only for a fraction of patients. Put simply, the predictive biomarkers available today don’t always accurately identify the patients who will respond, leading to higher incidences of adverse events and treatment resistance. Conventional immunohistochemistry (IHC) techniques assessing PD-L1 expression and, to a lesser degree, microsatellite instability status are currently the only clinically validated predictive biomarkers for checkpoint inhibitors. However, most experts agree that neither assay is sufficient; some lead to false positives or false negatives, whereas others address only a small fraction of the patient population. There is much room for improvement.
It is evident that our current repertoire of immuno-oncology biomarkers does not adequately capture the root-cause biological mechanisms that correlate with response. New evidence suggests that a critical source of information is revealed in the cell-to-cell biology of the tumor microenvironment (TME). But most efforts to identify clinically useful biomarkers do not preserve spatial information related to the arrangement of cell types in TME and are thus blind to cellular and tissue cartographic coordinates. When it comes to cellular interactions and spatial clustering, many of which are critical for cancer development and progression, we are left in the dark.
A recent study underscores the need for spatial information to accompany immunotherapy-related biomarkers (1). This meta-analysis of more than 50 studies – spanning more than 10 solid-tumor cancers and over 8,000 patients – found that spatial characterization significantly improved the predictive power of biomarker assays. The authors reviewed traditional biomarker assays – including gene expression profiling, tumor mutational burden assessment, and immunohistochemistry – as well as two new types of assays that incorporate spatial data: multiplex IHC (mIHC) and multiplex immunofluorescence (mIF). The latter is a new type of biomarker assay that makes it possible to quantify multiple immune and tumor markers in a tissue section while preserving and imaging the spatial architecture of the tumor microenvironment.
The authors found that traditional biomarker types performed comparably in distinguishing between responders and non-responders to anti-PD-1/PD-L1 immunotherapy – but the spatially resolved biomarker assay types performed significantly better. These assays had fewer false positives; that is, they were less likely to predict positive response for patients who would not benefit from checkpoint inhibitor immunotherapy. Moreover, the results from this study showed that combining traditional biomarker assays improved predictive performance, but even this additive effect paled in comparison to the singular performance of mIF and mIHC assays.
mIF and mIHC assays represent robust and standardized tools that can generate spatially resolved molecular maps of biopsy tissue sections, allowing us to see the cell-to-cell interactions and spatial biology occurring in the TME. Multidimensional examination of the TME should play an integral role in immuno-oncology research and in the search for better predictive biomarkers. From predicting immunotherapy treatment response to characterizing tumor heterogeneity, we now have access to a new dimension of information to benefit patients and help us expedite the battle against cancer.
Newsletters
Receive the latest analytical scientist news, personalities, education, and career development – weekly to your inbox.
References
- Steve Lu et al., “Comparison of biomarker modalities for predicting response to PD-1/PD-L1 checkpoint blockade”, JAMA Oncol, [Epub ahead of print] (2019). PMID: 31318407.
