DetermaIO algorithm and qPCR assay

DetermaIO algorithm and qPCR assay are used to show an association with molecular features of gastric cancer that have also been previously shown to be associated with response to ICI therapies so this is important because gastric cancer is the fifth most incident cancer worldwide and the third leading cause of cancer-related death worldwide and so unfortunately most gastric cancers are asymptomatic and go undiagnosed until they are in late stage and so as a result the survival outcomes for late-stage gastric cancers are exceedingly poor and so we need better treatments and detection mechanisms for identifying patients who are likely to respond to therapy in these late stages as time is critical so currently the most of the first line treatments involve chemotherapy or surgery and immunotherapies or immune checkpoint inhibitor therapy has been approved in the US for second and third line therapies but recently ICI therapy has also been now included for first line therapy and gastric cancer so currently there’s a variety of biomarkers out there that are used for a prognostic determination of a response to ICI therapy and these include things like PD⁠-⁠L1 one score at a threshold of one percent t or mutational burden EVV status and MSI high tumors so as I mentioned currently ICIs are currently improving the US for advanced gastric cancer in the second line or third line setting but different trials have demonstrated a wide range of response rates in advanced and metastatic gastric cancer in the salvage therapy setting or second and third line setting and with or without PD⁠-⁠L11 positivity and so this opens the door for the opportunity to look for better biomarkers that could be used to predict response to therapy.

So, one of the biomarkers that is of potential use is the DetermaIO biomarker which is what we call the IO score here and the term io is a 27 gene assay that can be used either by real-time qPCR or RNA-seq and it’s used to identify the tumor immune micro environment and identifies patients as a binary result immunoscope positive or negative or pardon IO score positive or negative and those that are IO score positive are likely more likely to respond to immune checkpoint inhibitor therapy compared to those who are IO negative so this assay really seeks to classify the tumor immune microenvironment in a tumor agnostic fashion and we’ve had previous data show that the io score is predictive of response and triple negative breast cancer and is also associated with response in non-small cell lung cancers so in this study this is more of a proof-of-principle small study in which we took data RNA-seq data from either the TCGA database or the Asian cancer research group ACRG database and looked at the association between the IO score and the other molecular features and so what we saw was that indeed the io score was significantly associated with the known molecular features that are have been shown to be prognostic of outcomes in gastric cancer including EBV status MSI high and TMD high tumors as well as PD-L1 expression and so given that we then had shown that in these two RNA-seq databases but the IO score was associated with these known prognostic markers of ICE of response to ICI we then sought to determine whether the IO score was also associated with these biomarkers in a clinical cohort as well as whether or not the IO score was associated with objective response in this clinical cohort so we had an opportunity to pull down RNA-seq data from like I said this clinical cohort of mostly Korean patients who were treated with pembrolizumab and these were all advanced gastric cancer patients and they had known outcome data to ICI therapy and so in this cohort of which we had 59 patients we were able to see that the EVV status and PD⁠-⁠L1 expression were associated with IO score and then importantly the IO score itself was a fact associated with the objective response rate and io score was able to identify 68 of the responders and 70 of the non-responders in this group and   also curiously when we examined the continuous value of the IO score with the molecular subtypes that are classified by TCGA which are EVV high MSI high chromosomal and stable or CIN and gnomically unstable GIN the IO score was correlated with these molecular features such that the TCGA classifier most associated with response which are EBV high and MSI high had the highest IO scores and the TCGA classifiers likely least likely to respond to ICI therapies including CIN and GIN also had the lowest continuous IO score and these scores are actually below our threshold of positivity such that the EBV high and MSI high would were associated with io positives and the GIN CIN molecular classifiers from TCGA were IO negative so all of this data together suggests that the IO score is a composite that’s picking up a lot of these different components from these other biomarkers and that   we’re seeing a lot of the similar data in a singular test and that alone the IO score is also associated with objective response rate in this clinical cohort.

 Common Questions From Your Colleagues? 

So, the most common question we usually get is how does the IO score compare to other biomarkers and so this study really set to address the fact that the IO score was associated with known biomarkers that are prognostic of outcome and that we too showed that the IO score in itself is associated with the objective response rate in this clinical cohort the added benefit of the IO score is that again it’s a tumor agnostic biomarker that assesses the tumor immune microenvironment so due to the fact that gastric cancers have a large amount of tuber heterogeneity this really helps in distilling a lot of the information and that in a singular test we can glean the same information that we would gain from examining a lot of the similar biomarkers individually.

Will This Effect Clinicians Today?

So, while this data is not immediately actionable these data do suggest that there is a potential biomarker that’s associated that is associated with response to ICI therapy and advanced gastric cancer patients and in the future, the IO score could be a singular more economical test that can be used to aid in the clinical decision making regarding the use of ICI therapies for advanced gastric cancer patients.

 What Are The Next Steps?

So, this was a proof of principle study that was used to investigate the potential utility of the io score known as DetermaIO and its association with response to ICI therapy and so while there are many existing biomarkers they still do a very poor job in identifying likely responders and so our results show that in a singular test determine IO mirrors the results of these well-established biomarkers as well or better than what’s currently available and also utilizes a lot less tissue so future studies will be conducted to examine whether or not the term io score alone or perhaps even as a composite biomarker with these existing factors such as EBV or MSI could provide independent and incremental information   to help predict the response   to ICI therapy and so based on our previous studies and other cancer types we hypothesized that this indeed will be the case and as a result the term io again either a loaners or composite marker may be better able to identify the subpopulations of likely responders and thereby aid in the clinical decision making process for ICI therapy and advanced gastric cancer patients.

Matthew Varga, Ph.D., a multidisciplinary scientist with broad experience in cancer biology, infectious diseases, immunology, molecular biology, and epidemiology, Scientific Affairs at Oncocyte Corporation. In this video, he speaks about the ASCO 2022 Abstract – The 27-gene IO score is associated with molecular features and response to immune checkpoint inhibitors (ICI) in patients with gastric cancer.

Study Details:

Gastric cancer (GC) is the world’s third greatest cause of cancer-related death. Unfortunately, most individuals with stomach cancer are asymptomatic until the cancer has spread to an advanced stage. In a number of malignancies, including GC, ICIs have improved patient outcomes. A number of biomarkers, including high PD-L1 expression, MSI, Epstein Barr virus (EBV) positivity, and TMB, have been used to identify patients who are most likely to benefit from ICI therapy. Despite these possible biomarkers, most advanced GC patients do not react to ICI treatment. As a result, there is still a clinical need for a biomarker that can better predict response to ICI therapy. We show that in a clinical cohort, the 27 gene IO score, a tumor immune microenvironment (TIME) classifier, is related with existing genetic markers of gastric cancer and with objective response to ICI therapy.

Methodology:

RNA-seq expression data from three distinct cohorts were obtained: TCGA (STAD), ACRG (GSE84437, GSE84426), and a clinical cohort including ICI response data (PRJEB25780, PRJEB40416). To generate IO scores, the 27 gene IO algorithm was applied to all available patient data. In each cohort, Fisher’s exact test was performed to assess the relationships between IO score and clinical characteristics and molecular subtypes. R (version 4.1.2) was used to compute ORs with 95% confidence intervals as well as ordinal logistic regression modeling.

Outcomes:

The IO score was linked with the molecular characteristics of EBV, MSI, TMB, and PD-L1 in the TCGA cohort (n = 135, p 0.05 for all). Similarly, the IO score was substantially linked with EBV, MSI, and PD-L1 in the ACRG cohorts (n = 294, p 0.001 for all). We evaluated a cohort of Korean patients with advanced stage GC compiled by Kim et al. to see if the IO score was associated with responsiveness to ICIs. The IO score was related with ICI response in this sample of 59 patients (Fisher’s exact test, p 0.05). The odds ratio for the connection between IO score and response was 5.3 (95 percent CI: 1.3 to 23.92, p = 0.01). The linearity of the IO score’s continuous value indicated a direct association between the IO score and enhanced objective response (ordinal logistic regression, t = 2.59, p 0.01).

Observations:

Because PD-L1 and TMB have demonstrated significant levels of geographical and temporal heterogeneity in GCs, there is a need for a more comprehensive biomarker that can thoroughly measure the TIME. The 27 gene IO score has been linked to numerous existing biomarkers in GC and has recently been linked to ICI responsiveness. As a result, additional research is needed to show that the 27 gene IO score is a more comprehensive biomarker for measuring TIME and provides supplementary data to tumor-specific biomarkers, which could aid in clinical decision making for ICI treatment of GCs.