Patient Tumor Cell Assays Guiding Personalized Therapy

Jeffrey Tyner, PhD

Today, ahead of the American Society of Hematology meeting, being held in Orlando, Florida from December 5th to the 8th, we are speaking with Dr. Jeffrey Tyner, an assistant professor of cell, developmental, and cancer biology at the Oregon Health & Science University in Portland. Professor Tyner who will present research from his lab on a way to screen tumor cells derived from patients for sensitivity to available anticancer drugs to guide treatment decisions. His talk, "Functional Screening to Guide Personalized Cancer Therapy" will be on Saturday, December 5th at the ASH meeting.¹

-Interviewed by Anna Azvolinsky, PhD

OncoTherapy Network: Whole genome sequencing is becoming more readily available and cheaper. Can you talk about the way that sequencing of tumor samples is currently being used to guide treatment for cancer patients? Is this approach still only being applied in a clinical trial and research setting?

Dr. Tyner: Sure, great question. You’re absolutely right that whole genome sequencing and deep sequencing in general is becoming much cheaper and faster, and more readily available. There are at least a couple of different ways that it is being applied clinically right now. Of course there are certain genetic tests that have been around for a long time, for decades even, for specific genes that have been known about early on, that correlated with a particular treatment that was effective.

The great example of course being, is BCR-ABL, detected with a basic karyotype and led to the development of very effective therapies to combat the disease that harbors that genetic mutation, CML [chronic myeloid leukemia]. And now as we get more broad whole genome sequencing or even whole exome or more targeted sequencing approaches that cover many more genes, the amount of information you get is really incredible. And there are at least two different areas to which this is being applied, certainly in a clinical trial setting.

There are trials underway, some that have been going on for some time now where patients are enrolled onto different arms, different therapeutic arms of the trial based on different mutational subtypes of disease. So if you have a mutation in gene A that predicts a response to a particularly drug, you get that drug, and if you get a mutation in a different gene you get a different drug. So certainly clinical trials are one major area where this is being applied, but there is also a more ad hoc way of doing this where the sequencing ongoing now is performed in a CLIA approved setting, meaning it is approved in a format that is compatible with clinical decision making.

So it is now possible to send tumor DNA out for this clinically approved sequencing and then it is really up to the treating clinician as to how to use that information, and whether that information is useful in providing a therapeutic option for a patient. And there certainly are great examples of mutations that do provide an obvious therapy that is likely to positively influence the disease course of a patient. Unfortunately, there are a whole lot of mutations that we don't yet know what to do with yet from a therapeutic perspective, there really aren’t therapies that are immediately obvious that should be applied, so as a biologist we really have a lot of work ahead to be able to make connections between mutational events and the drugs that are likely to impact those mutations.

OncoTherapy Network: You mentioned one of the limitations of whole genome sequencing as far as applications in cancer patients. Are there other limitations as far as identifying potential target somatic mutations through whole genome sequencing?

Dr. Tyner: Whole genome sequencing or deep sequencing in general are awfully good at picking up somatic point mutations, where there is just a single nucleotide that is changed, the most simple type of mutation that occurs and these occur commonly in cancer, and deep sequencing is good at picking these up even at low levels. And the technology is getting better and better. There are certainly types of mutations that are more complex that involve more than one nucleotide, insertion or deletion of multiple nucleotides. And certainly the algorithms that sift through the large amount of data generated by whole genome sequencing and make the calls of whether a mutation is present or not, are not as successful at picking up these more complex mutations. That said, deep sequencing is good at detecting the mutations that are present in a tumor. To my mind, the main limitation is what we were just talking about, which is right now, it is very common to see mutations occur in tumors for which we really have no notion of what to do from a therapeutic perspective.

So we really have a gap that has emerged between our ability to call these somatic mutations and identify them, and then actually translate that information into clinically meaningful therapies for patients. There really is a biological gap that has emerged and we have a lot of work to do to resolve that separation.

OncoTherapy Network: So towards that end, what is the approach that your lab is taking to develop tools to guide treatment decisions for hematological malignancies?

Dr. Tyner: So in my lab, and there are several other labs around the country and around the world that are taking a similar approach, we call it a functional screening approach. The idea is that we can take primary tumor cells from the tumor of a patient and we can culture these cells in a petri dish in vitro for a short period of time, for a few days, and we can culture them against panels of libraries of cancer drugs and we can over that short culture in vitro, understand which of the drugs are able to effectively kill the tumor cells and which of the drugs are not so good at killing the tumor cells. And therefore, on an individualized basis, we can understand, for each individual tumor and patient, what are the drugs that are most likely to be able to wipe out the tumor cells, the cancer cells from that particular patient. And this approach, we have applied on over a thousand patient specimens. And my particular focus is hematological malignancies, so these have been leukemias and lymphomas, and certainly there are others around the country and the world who are adopting similar lines of inquiry in the solid tumor space. So my feeling is that certainly until we can understand these complex connections between genotype and drug response, this is really an important way that allows us to understand what therapies might be most effective for individual patients. 

OncoTherapy Network: Whatis the time frame in which this patient-specific assay can be completed, and then how do you envision this screening tool being used in the clinical setting in the long term?

Dr. Tyner: So, the assay time frame is really just 3 days, and sometimes it stretches to 4 days. So really a few days which is clinical relevant in all settings. And this assay is already being used in early phases in the clinical setting. We actually have some clinical trials that are running right now that are using this assay as a tool to select individualized therapies for patients in a similar way as we discussed at the beginning, of mutations being used, we can also use this functional assay to select therapies for patients. And we are in the early stages,but I do expect and hope that this assay will grow and will continue to be used in this clinical setting much more broadly in the near future.

OncoTherapy Network: Thank you so much for joining us today, Dr. Tyner.

Dr. Tyner: Thank you very much for having me.

References:

• American Society of Hematology. (2015). Functional Screening to Guide Personalized Cancer Therapy. American Society of Hematology's 57th Annual Meeting & Exposition, December 5-8, 2015.