Finally, Reliable Drug Screening Platforms for Lung Cancer Biomarkers
This technological advance could help us better select which patients will benefit from immunotherapy or targeted drugs
Non-small cell lung cancer (NSCLC) is still one of the most life-threating diseases worldwide. More people die of it than any other cancer. Despite improved diagnostic tools most of the cases aren’t found until the cancer has metastasized. C. Fortunately the treatment options are expanding beyond chemotherapy. Targeted therapy for tumors harboring specific genetic alterations and cancer immunotherapy, specifically the use of immune checkpoint inhibitors (ICI), have become first-line treatments. Because of these additional treatment modalities, survival rates are improving.
With that said, not every NSCLC patient administered a targeted drug or immune-oncology drug responds favorably even when they meet certain tumor profile characteristics. Several baseline tumor features, including PD-L1 expression, tumor mutational burden (TMB) and CD8+ T cells infiltration have been shown to be associated with response to ICIs, according to recent studies, yet the existing biomarkers are not specific enough to select the patients benefitting the most from ICI treatment. In short, there is an urgent need to define more adequate biomarkers for patient selection.
This is gaining importance with regard to a series of new side effects, the so-called immune-related adverse events (irAEs), which are connected to treatment with ICIs. The irAEs result from an aberrant activation of immune cell-mediated pathway, leading to autoimmunity disorders. Based on the literature, 70% - 90% of patients receiving ICIs will develop irAEs.
In the light of the fact, that large series of clinical studies were not able to shed more light into those questions, it becomes obvious that there is an increasing need to elucidate these issues in a preclinical setting.
The above-mentioned variation in patient responses is mainly based on the heterogeneity of a specific tumor and its microenvironment including the immune cells. The challenge of modeling this in a preclinical setting is two-fold. You must mimic both the complex crosstalk of these different cell types, and the heterogeneity within one tumor as well as a patient cohort. Therefore, continuous development and improvement of immuno-oncology assay methods are essential to advance not only new drugs but as well the related biomarkers to predict the clinical outcome of an individual patient.
In general, preclinical studies can be performed in vitro in a cell culture dish and in vivo, mostly in mice. One advantage of the in vitro approach is that it allows you to evaluate large number of drugs via high throughput screening in a time sensitive manner. The second advantage, and in this context the more important one, is the fact that it is possible to add, remove and modulate different cell types in the complex system of a tumor model. Specifically, for the development of biomarkers this is an advantage as the robustness of the marker can be tested mechanistically. The modulation of specific cell types can elucidate their role in the multi-directional crosstalk within the tumor microenvironment.
Scientists at Cypre, a tumor model company in San Francisco, have developed such an in vitro platform that enables the testing of multiple cell types within one well. The extracellular matrix component is represented by a hydrogel proprietary to Cypre, which can be modified to best mimic the extracellular composition of an individual tumor model. With the possibility to determine the expression of multiple surface markers by immunofluorescence-based image analysis, the validation of biomarkers on the different cell types is enabled.
In a joint pilot study we evaluated the impact of co-culturing fibroblasts together with tumor cells to determine the sensitivity towards immune-oncology drugs as well as cytostatic and targeted drugs in NSCLC PDX derived models. We were able to show that fibroblasts, tumor cells as well as immune cells influenced the activity of the tested drugs. Of note, the sensitivity data of this 3D platform correlated very well with the in vivo data of the corresponding mouse PDX model.
So, the big take home message for oncology researchers and clinicians is that translational data from predictive and clinically relevant preclinical platforms will enable a rational design of clinical trials verifying these results. It will also help develop robust selection criteria for NSCLC patients benefitting from ICI treatment and managing irAEs at an early stage.
In short, we could be on the cusp of figuring out which cancer patients will have durable responses to these game-changing oncology therapies. That is good news for the thousands of lung cancer patients and for the field of personalized medicine.