Mark R. Litzow, MD: Ryan, can you talk a little bit about polymeasure MRD [minimal residual disease], what the approaches are, and maybe some of the pros and cons?

Ryan D. Cassaday, MD: I think the most common tool or method used to measure MRD, at least in the United States, is multiparameter flow cytometry. It’s fast. It’s relatively reliable. The nice thing about lymphoblasts is they tend to have a relatively unique immunophenotype, so it’s relatively easy for our pathologist to pick them out of a bone marrow specimen. The challenge is there is a lot of variability in the quality of the different techniques. So, for example, if one were to try to use the COG [Children’s Oncology Group] experience to follow strictly how the data are done, those assays are generally performed, as I understand it, at either the University of Washington School of Medicine hematopathology lab where I work or at Johns Hopkins Medicine. It’s very, very possible that if you’re in community practice and doing a bone marrow exam and sending it for flow cytometry, then it is not going to be of that same reference lab quality, and the sensitivity of it could be one or maybe even two logs less sensitive. So, an MRD negative result in a patient being tested by that modality is certainly not as reliable a result.

On the flip side, in Europe they base a lot of their MRD assessments on PCR [polymerase chain reaction]. It’s a more labor-intensive strategy, but it tends to be more sensitive when done in proper laboratories. And they worked out a system to standardize and then centralize it that we just don’t have here in the United States. One exception to that is probably the BCR-ABL quantitative PCR that we can use for Ph [Philadelphia chromosome]-positive patients and a few other examples of fusion oncogenes that we can use for RT-PCR [reverse transcriptase PCR] that are probably more reliably sensitive than most standard flow cytometry.

The newest modality that we have, which is starting to gain more traction I think, is using high-throughput sequencing or next-generation sequencing to look for a clone-specific gene sequence in either the immunoglobulin heavy chain or T-cell receptor genes that are unique to the leukemia. And this can often plum up another one or two logs below some of the methods we have right now.

You have to have a specimen from the leukemia to be able to identify the sequence, which is not necessarily an assured thing. And it takes a little bit more of a turnaround time so it’s not a result you can get back as quickly as flow cytometry. Furthermore, there’s still quite a bit of work to be done to understand how that test can really be used to make treatment decisions, at least in my opinion. But I think it’s certainly an important and potentially useful tool that is hopefully going to become more and more integrated into the clinical trials so we can use it a bit more reliably.

Rachel E. Rau, MD: Can I ask something? With the Ph-positive patients, we were somewhat surprised by our result trial that showed the BCR-ABL PCR really had zero correlation with outcome. Even flow cytometry and PCR MRD had pretty poor correlation with outcome, very different than every other patient with B-cell ALL [acute lymphoblastic leukemia]. On the adult side, do you see a good correlation with BCR-ABL PCR and outcome?

Ryan D. Cassaday, MD: I think this is an area where we have some data, but not a lot. My feeling about this, and you all can certainly chime in and correct me if you have a different take or know some data that I’m not thinking of, is the P190 PCR tends to track more closely, particularly our flow cytometry at the University of Washington. There’s comparable sensitivity, at least in my clinical practice. They tend to track pretty reliably. The P210s tend to clear a little bit more slowly. My suspicion for that, and I think there are data that back this statement up and it certainly makes sense biologically, is P210 is a little more common in CML [chronic myeloid leukemia]. And I think a lot of those patients with Ph-positive ALL probably have CML underlying it, and maybe they just came to medical attention seemingly as de novo Ph-positive ALL, and that persistent P210 may just reflect some persistent chronic phase CML. But by flow cytometry, we’re not seeing any lymphoblasts.

So, the disease that’s actually going to get them into trouble faster, the ALL, is actually in a deep remission. But I’ve seen patients with 0.5%, 0.8%, 1% detectable by P210, but their flow cytometry, which is sensitive down to .01%, is negative. I think having an appreciation and understanding of that is important, and it’s certainly for tracking, making sure that you’re ordering the right test. I’ve definitely seen a few patients where that got mixed up.

Mark R. Litzow, MD: That’s been my experience as well and I think you’re spot on with that assessment.

Jae Park, MD: I think the MRD role in Ph-positive ALL is now well defined, to the point where there’s flow cytometry and then the PCR. Now another challenge is in the very, very low levels of the PCR positivity were sometimes 0.0001% at the transcript level. And then the flow cytometry is negative. What do you do for these patients? We just don’t have a lot of data because most of our data is also driven by Ph-negative patients getting a certain type of a therapy with a different type. It’s also that kinetics of the response will be different, the time point will be different. To your point, it’s a little bit of challenging to know which MRD testing to follow. Should I go with the BCR-ABL transcript versus the traditional flow cytometry that’s been validating the other decisions?

Mark R. Litzow, MD: I think one thing that gets under emphasized is the sample and the importance of a good quality sample, trying to get that first pull to get a good marrow sample as opposed to the third pull or the fourth pull when you’re getting a lot of dilution in the peripheral blood. I think we still think that bone marrow is the gold standard for assessing MRD, certainly in ALL. I do want to emphasize that. I think it sounds like you agree with me in getting that good quality sample from the bone marrow.

Jae Park, MD: I think that’s really key. I’m glad you mentioned it because especially at the time point that you are making some decision, I think it’s very, very important for the patient’s treatment too. Because a lot of times in some places you may not be the one who’s getting the bone marrow, so you may not have control over what sample is to be taken where then you really need to inform the performer, whether it’s an interventional radiologist—some places are utilizing them to get the bone marrow—or the technicians or the nurse practitioners or whoever is the APP [advanced practice provider] or doing the bone marrow biopsy. I think that really needs to be well established. And the quality of the sample is just as important because if there is a total dilution of the sample, it is no longer really relevant for the testing. So I think that’s an excellent point.

Transcript Edited for Clarity