Bow and arrow emerge from the syringe
Two dimensional drawing of a syringe on a blackboard emerges in to three dimensions as colorful dots, with widespread impact

At the Clinical Biomarkers and World CDx Conference, held recently in Boston, industry and academic scientists convened to address approaches to enhancing and validating the next generation of predictive biomarkers and promoting the adoption of high-value precision medicines and diagnostic testing worldwide.

Basing Prognoses on Measures of Immune Cell Morphology

Joseph Krueger, Ph.D., CSO of Flagship Biosciences, discussed the creation of predictive diagnostics for measuring tumor-infiltrating lymphocytes (TILs) and a more contextual approach to marker assessments.

Studies in multiple tumor types have documented that the presence and the quantity of TILs strongly correlate with increased survival. But methodological challenges in obtaining morphological measures have limited the usefulness of TILs in standard prognostic panels and patient profiles.

Presently, TIL counts are primarily measured by immunohistochemistry (IHC). A major limitation of IHC is the 2D nature of stained slides. Assaying a 2D cross-section of a 3D tumor severely limits the information sampled and increases variation. A second key shortcoming of IHC is the difficulty of standardization among laboratories. Staining intensity and specificity vary significantly from lab to lab even when the same antibodies are used with a shared operating procedure.

Flagship Biosciences’ Computational Tissue Analysis (cTA) platform provides an alternative approach for solving the key problems of variability, lack of precision, and insufficient clarity in the relationship between test results and clinical outcomes. Dr. Krueger explained that Flagship developed the cTA platform to simplify tissue context biomarker analysis and to determine the cutoff values that define a positive result. “The cTA platform uses proprietary computer algorithms to quantify IHC-based biomarker content from whole-slide images of patient biopsies. By combining traditional IHC methods and stains with digital pathology approaches it can provide more precise data that can better predict contextual relationships between biomarker expression and treatment outcomes,” says Dr. Krueger.

“The software identifies every single cell in the tissue and its location,” he continues. “It uses AI to identify each cell as a tumor cell or a leukocyte, using the morphological characteristics of each cell from the hematoxylin counterstain typical of an IHC slide to classify what type of cell it is. Determination of what specific type of leukocyte a cell is requires an IHC biomarker stain, such as CD8 (for T cells) or CD68 (for macrophages).” When this combination of morphological and staining characteristics is used, each tumor cell and each TIL is identified in its location in the tumor biopsy, and the TILs are reported as being infiltrative in the tumor epithelium or in the tumor-associated stroma.

Currently, pathologists can only estimate the number of TILs in a tumor biopsy through an error-prone, laborious procedure requiring a survey of several high-powered fields across the slide and counting the TILs in the tumor epithelium and adjacent stroma separately. “This tedious exercise requires more than a half hour, and ultimately yields report qualitative values such as high, medium, or low,” Dr. Krueger complains.

“Our technology allows [users] to access quantitative values for these endpoints, which are determined by cell-by-cell analysis, and because the areas measured include the whole tissue, they are not susceptible to the errors caused by variability and heterogeneity in the tissue.” Furthermore, this rich data profile can be used to create more comprehensive endpoints, such as ratios of different types of TILs, proximity to each other or tumor cells, and organizational patterns that truly capture the immune phenotype of the tissue, says Dr. Krueger.

Biomarkers for Cystic Fibrosis

Cystic fibrosis (CF) is a fatal genetic disease caused by mutations in the gene encoding the CF transmembrane conductance regulator (CFTR). This protein, a protein kinase A (PKA)-activated epithelial anion channel, participates in salt and fluid transport in multiple organs. The protein functions at the cell surface and contributes to the regulation of absorption and secretion of salt and water in tissues including the lung, sweat glands, pancreas, and gastrointestinal tract.

Most CF mutations either reduce the number of CFTR channels at the cell surface, due to synthesis or processing mutations, or impair channel function through gating or conductance mutations, or both. Approximately 4–5% of patients with CF have the G551D mutation on at least one allele, resulting in a dysfunctional ion gating function.

In vitro models of CF using transfected cells carrying CF-associated gene defects have greatly advanced drug discovery, notes Fred Van Goor, Ph.D., head of CF research at Vertex Pharmaceuticals. These models recapitulate disease markers found in patients’ cells.

Ivacaftor (Kalydeco®; Vertex), a small molecule treatment for CF, acts as a CFTR “potentiator” indicated for the treatment of CF in patients six years and older who have a G551D mutation in the CFTR gene, in which the amino acid glycine (G) in position 551 is replaced with aspartic acid with the effect of producing a defect in the CFTR ion gating function.

The orally available small molecule drug works by keeping the CFTR gate open longer at the cell surface, facilitating the transport of salt and water through cells to improve hydration and mucus clearance. In 2012, ivacaftor was initially approved in patients two years and older with at least one mutation in their CF gene that is responsive the drug.

In 2017, the FDA expanded approval of ivacaftor for the treatment of additional mutations, a decision, Dr. Van Goor says, not based on clinical studies but on the in vitro studies using biomarkers that established improvement in the protein’s ion channel function in cell lines carrying other mutations. The compound is currently approved in the United States for the treatment of 38 types of CF mutations in patients two years and older.

In a press release, the FDA said, “The approval triples the number of rare gene mutations that the drug can now treat, expanding the indication from the treatment of 10 mutations, to 33. The agency based its decision, in part, on the results of laboratory testing, which it used in conjunction with evidence from earlier human clinical trials. The approach provides a pathway for adding additional, rare mutations of the disease, based on laboratory data.”

Vertex tested a combination of its VX-152/tezacaftor/ivacaftor combination therapy in Phase III trials. Both VX-152 and tezacaftor are correctors of CFTR, small molecules that can enhance movement of the protein into its functional position in the cell membrane.1 The drug is intended to help treat the almost 90% of people with CF carrying F508del, resulting in proteins that do not fold into their proper shape and are targeted for degradation before making it to the cell membrane.

Ongoing triple-combination therapy trials in Phase III development include VX-659 and BX-445, two next-generation correctors. The decision to advance VX-659 and VX-445 into Phase III development was based on initial Phase II data, including new data from ongoing Phase II studies that showed mean absolute improvements in percent predicted forced expiratory volume in one second (ppFEV1) of up to 13.3 and 13.8 percentage points from baseline through four weeks of treatment for the triple-combination regimens with VX-659 (400 mg QD) or VX-445 (200 mg QD), respectively.

The triple combination of these drugs has the potential to benefit a broad range of people with CF with two copies of the F508del mutation (homozygous), and with one copy of F508del mutation and one copy of another mutation that affects CFTR function (heterozygous).

Preclinical testing of the triple combination of VX-152/tezacaftor/ivacaftor human bronchial epithelial (HBE) cells showed an increase in the chloride transport, as well as in cilia beat frequency, when compared to the lumacaftor/ivacaftor combination, suggesting an improvement in the CFTR function.

Dr. Van Goor tells GEN, “CF is a lifelong disease. What we really want to accomplish is to stop the decline in lung function over time due to multiple exacerbations. The disease affects multiple body organs, including the pancreas, and more recently, we have progressed the use of our therapies to younger and younger patients.

“With the triple combinations in late-stage trials, we use two correctors and a potentiator—correctors to get the protein to the cell surface and a potentiator to make it work better.  Data from Phase I and II studies demonstrated the potential to treat the underlying cause of CF in people who have one F508del mutation and one minimal function mutation not responsive to ivacaftor, tezacaftor, or the tezacaftor/ivacaftor combination, that is, in people who have a severe and difficult-to-treat type of the disease,” Dr. Van Goor says. All studies showed statistically significant improvements in lung function (ppFEV1) with a triple-combination regimen in these patients.

“There are still patients for whom we will use biomarkers, for example, the F508L1 mutation,” he notes. “In vitro assays can identify people without the F508del mutation on either allele that could respond.”

Clues within Isolated Immune Cells

GEN spoke with David Messina, Ph.D., COO of Cofactor Genomics, who participated in the panel discussion, “Building the Best Biomarker Driven Precision Medicine Clinical Trials.”

Founded by three former Human Genome Project scientists, Cofactor has built a proprietary platform, its ImmunoPrism assay, that the company says overcomes the chemical and computational challenges of performing complex immune profiling on clinical-grade human samples. The platform, comprising targeted molecular reagents and machine-learning software, accommodates samples from FFPE, FNA, and CNB archives as well as samples with very low sample inputs, and it provides extensive immune characterization beyond current technologies—including differentiating cells such as M1 and M2 macrophages and regulatory T cells.

Dr. Messina says that RNA analysis provides the key to figuring out exactly what immune cells are present in a patient’s tumor, potentially using the information to help predict a patient’s response to immunotherapy.

“One of the critical factors in using novel immunotherapies such as checkpoint inhibitors is understanding how the immune system interacts with the tumor in the microenvironment. The amounts and ratios of immune cells can determine whether a patient will respond to a drug,” says Dr. Messina. For example, he cites findings that show that the use of CD8+/CD4+ TILs ratios in tumor biopsies may predict response to anti-PD1 treatment in metastatic melanoma and non-small cell lung cancer.

“By isolating the pure immune cell subtypes and analyzing their RNA individually, we’ve been able to develop a multidimensional gene expression model for what makes each immune cell unique, and getting a real-time readout of what’s going on in the cell,” he continues. “It’s not possible with current technology to effectively assay every protein or cell surface marker, but we can quantify immune cells in tremendous detail using RNA.

“We are profiling the immune content of the tumor with high sensitivity and we require very little tumor material. We process it and return a fully analyzed report on each individual sample—the assay profiles multiple immune cell types in a tumor sample simultaneously. This enables a better understanding of response to therapies, disease progression, mechanism of action, and what combination of immunotherapies might make sense for a patient.”

“A patient sample contains a mixture of multiple cell types—we compare the material from the sample to what we have in our curated database and can identify the mixture components—say 20% T cells or 5% regulatory T cells,” Dr. Messina details. “This level of resolution has previously not been possible. We can compare the relative amounts of immune cell types within the sample. This is very important where the ratio of immune cells is key to understanding of the patient response to immunotherapies.”

“Success of immunotherapies has led to great efforts to identify new therapies of this class, and we see hundreds of immunotherapies in development,” he notes. “While it is exciting to think about the expansion of these therapies, we know that some of the biomarkers in use today are not sufficiently predictive of patient responses.”

“We believe our RNA-based assays can produce much more refined and predictive information about potential patient responses,” he asserts. “We are eager to demonstrate this in a clinical setting.” 

This site uses Akismet to reduce spam. Learn how your comment data is processed.