Bone Marrow as a Tool to Predict Response to Therapy for MM

Kareem Azab, PhD, Associate Professor, Department of Radiation Oncology, Department of Biomedical Engineering, Washington University, St. Louis, Missouri, discusses 3D tissue-engineered bone marrow as a tool to predict patient responses to therapy for multiple myeloma (MM).

Transcript:

Dr Kareem Azab: My name is Kareem Azab, and I'm an associate professor in the Department for Radiation Oncology under the Department of Biomedical Engineering at Washington University in St. Louis. I'm also the founder of Cellatrix LLC, a company in Missouri, that is related to this research as it has an exclusive license to the 3D technology that we will be mentioning later on.

Oncology Learning Network: What existing data led you and your co-investigators to conduct this research?

Dr Azab: Rather than data, it's more a fact that we face in the treatment of cancer. The sad fact is that in the lab, we “cure” cancer every single day. We just put chemotherapies or drugs on cancer cell lines, and these kill the cancer very efficiently. The problem that we want to translate this into patients, many times we fail. The real question here, the more provocative question, if you want to call it like that, why do our lab models fail to represent the actual pathophysiology of the patient? We succeed in the lab, but we fail in the patients. We started thinking on what makes our models less efficient. That's how we started.

OLN: Could you briefly describe the study and its findings?

Dr Azab: Generally, when we try to compare the efficacy of drugs in the lab to the efficacy of drugs in the clinic, we found very little correlation. This can be attributed to many factors. One, that our lab models mainly depend on cell lines, and these do not always represent the actual biology of the patient.

I'll give you an example. In multiple myeloma (MM), we use for example, MM1 cell line. That's a great cell line that everybody uses. It was literally isolated in 1990 from the patient. This is probably the newest cell line that we have. Think about the 40, 50 years of passages that these cell lines are going in the lab, and how would they really correlate to the original disease they were isolated from?

The other big issue with our models is that the cancer cells in the body do not live alone. They live with a micro environment that's made of tens of different types of cells with extracellular metrics, with blood vessels, with bone around in the specific case of MM. Therefore, having a cell in a Petri dish on a plastic, doesn't really represent what's going on in the disease. We and many others have shown that other cells in the tumor micro environment will help the cancer cells to grow and will help them be drug-resistant.

But probably what's the most important part is the huge heterogeneity between patients. Every patient is different. Every patient has different biology, and not only genomics, by the way. The genetics are very variable that's for sure, but not only that. There's the epigenetics that also play a crucial role.

What we came up with is, we developed a 3-dimensional cell culture based on biopsies isolated from patients. We take a biopsy from the patient, we reconstruct it in a tissue engineering manner to represent the way it looked in the bone marrow in the patient, and we call this the 3D tissue-engineered bone marrow. From each patient, we can take a biopsy, reconstruct it in a certain way to very much represent the original disease. This actually covered all the weaknesses that we just said, that we are not depending on cell lines anymore, it has the tumor micro environment with it, and it's also patient-derived. It's not dependent on cell lines anymore. We have patented this technology, and one of the big things about it, is that it allowed us to culture primary patient cells outside the body for weeks, which wasn't possible before.

Now, why this is big news? Because right now, I have a window of about 2 to 3 weeks that I can do biology on the patient's cells. I don't need to use cell lines. Now, again, why is that important? Because now, I can take these cells, which represent the patient, but in the lab. I could, for example, screen drugs on them. For every specific patient, I could treat that patient outside with say, 10 different medications used for that disease. Now, we can say that drug number 2 and 6 and 8 work for this specific patient, forget the protocol right now. Because the protocol just treats everyone the same based on their stage. That's the way it's done right now. So we are trying to change this paradigm completely—that no more protocols. We will take every patient, we will try the drugs on his own cells with his own biology, and try to match the best treatment to that specific individual patient.

Now, before we jump to changing treatment protocols, we wanted first to check if this can work retrospectively. We performed an ex vivo clinical trial. We would take a bone marrow biopsy, we reconstruct it in the lab, the clinical team will tell us what the patient is going to be treated with. We will reproduce the treatment that the patient is getting right now in the clinic, we will reproduce it in the lab. After a week—That's a great also advantage of this technology that within a week—we can tell what the patient responds to the therapy.  We actually kind of mirrored the treatment in the clinic, in the lab. After a week, we come up with the response, if the patient is responsive or nonresponsive, and we will send this data back to the clinicians. The clinicians, after they finish their treatment for the actual patient, they will have also response or no response. Then we will correlate this data together and see if our model could predict the clinical response in the patient.

Luckily, in our study, we found that we could predict 89% of the responses. Whether these were responsive or non-responsive, we could predict both. So why is that big news? It's not only we were going to improve the response, but also save patients from trying non-useful treatments. We will save the patients a lot of health and a lot of time, and also a lot of money spent by the system as a whole on treatments that will not work.

We think right now we have enough solid base to be confident with our system to predict therapy. Now, we are planning to do it prospectively, meaning that if we get a biopsy from the patient, we will screen drugs and then go back to the oncologist and say, "Hey, don't do the standard of care now, let's do, based on our predictions, the combination of these 2 or 3 or 4 drugs that we think are going to work to the patient and compare it to the actual standard of care arm." At this point, we are exactly at this point where we want to start the prospective clinical trial. Hopefully we will see more efficacy.

OLN: Were any of the outcomes particularly surprising?

Dr Azab: We expect it to be successful, but not that successful honestly. When you predict 90% of the response, it means you will be able to have response in 90% in the patients. Now, I have to say that these patients that we worked with in this specific study, were relapse refractory (R/R) MM patients. It means they have tried 2, 3 lines of treatment and they failed. The point there is that the response rate in this specific group is no more than 25%. We would be very happy if we can improve it to 50%, but getting something close to 90% was literally a dream. I think there would be big impact on the field. The other thing that we started to do it with other malignancies right now, so that's where the treatment or the future directions are going, because we don't want to restrict it to myeloma. We have shown that at least ex vivo, this system can work for leukemias, AML (acute myeloid leukemia), CML (chronic myeloid leukemia), CLL (chronic lymphocytic leukemia). It can work for lymphomas as well.

We didn't do the clinical trials yet, but we are in the process of doing the phase 1 clinical trial that's the retrospective one for AML, but at least we have shown a proof of concept that it can work for a vast variety of at least hematologic malignancies. We are trying to expand that to solid tumors who metastasize to the bone marrow, including lung cancer, breast cancer, prostate cancer, because the main problem with these are the metastasis to the bone.

OLN: What are the possible real-world applications of these findings in clinical practice?

Dr Azab: For the clinical practice, we identified an important avenue to improve that treatment. That's why we got this technology patented and we licensed it to a startup company, Cellatrix, that is a spinoff from the lab. Now, a lot of this research is led by Cellatrix rather than my academic lab, but the idea is, we can help on 2 levels for the clinical practice. One, as I just described, is that we will be kind of the resort for the oncologist when they run out of options for a specific patient, they can send us a tumor biopsy, and within a week, before the patient finishes their MRI and PET CTS, and all of the tests, within a week, we will have kind of a heat map what works for that patient and what doesn't work. The oncologist will have ready-to-go information about this specific patient, what they will be able to respond to, which I think is going to first, again, improve the efficacy of our treatment dramatically. But also we work now with many pharmaceutical companies and biotech companies, especially in the immune therapy area, where we could do “clinical trials” ex vivo, before we go to try the things on patient.

How? If we have a new drug right now, that's being developed, before we go on and we spend millions of dollars in a phase 1 or a phase 2 two clinical trial, we could simply take biopsies from 20 patients, 30 patients, and run a “clinical trial”, just not on the patient themselves. If we see harsh toxicities, if we see no efficacy, we could just back up. This happened with one company that we worked with. They had 2 compounds, 2 lead compounds that they were trying to run in clinical trials. In our kind of ex vivo clinical trial, we found that one of them was way much better than the other. And you know what? They stopped that one and they went for one of the compounds for clinical trials. So we saved a lot of trial and error on the patients, that's the most important thing. The other thing that we help this company focus its research and efforts and resources on something that has a higher chance to be successful clinically. It can be also used at a personalized medicine level, for every single patient, but also it can be for drug development to help better drug development.

OLN: Do you and your co-investigators intend to expand upon this research? If so, what will be your next steps?

Dr Azab: The system is actually used in different stages of development.

For MM, we finished the phase 1 retrospective trial, and we are trying now to fundraise and put more resources into the phase 2 prospective trial, to try to improve upon the standard of care.

For leukemia AML, we have proven the concept biologically. We are now conducting the retrospective phase, and we are now in other hematological malignancies, we have proven the concept and we are trying to push it also into solid tumors. The next step will be just proving that we can improve the standard of care, and obtain the FDA approval to become kind of a diagnostic test to improve the efficacy of treatment.

OLN: Is there anything else pertaining to your research and findings that you would like to add?

Dr Azab: We are very excited for both pathways for also improving the clinical response for individual patients or what we call personalized medicine for MM, for AML, for other leukemias and solid tumors, especially in the phases where these patients really run out of options.

We think we will provide a very powerful tool to provide new options for these patients, as well as to improve the efficiency of drug development, and rather just trying it on cell lines and things that do not really represent the actual patients and have more failures in clinical trials, now we provide a tool that will make this process more efficient, and hopefully get more information before we get into the patient's phase.