Transcript:Mark R. Litzow, MD: Bijal, our pediatric colleagues have been using minimal residual disease testing for many years. It’s a relatively new development for us in the adult world. What do you see as the role of MRD testing in our patients?

Bijal D. Shah, MD: I think it’s critical. And I think, as we’ve observed in pediatrics, it’s going to help us delineate high-risk from low-risk. It’s going to help us delineate the high-risk; for example, Ph [Philadelphia chromosome]-like, where we may be defining them now not by gene expression profiling but just by looking at cytogenetics or translocations, as ferreting out those who may be higher risk from those that may be low risk. At a minimum, it’s ferreting our higher-risk patients.

Now the real question is, how do we discern MRD? Because there are multiple approaches now. There’s 6-color flow, 8-color flow, even 10-color flow cytometry being done across different institutions. And there’s now next-generation sequencing [NGS] approaches that are being used.

My approach at Moffitt [H. Lee Moffitt Cancer Center & Research Institute] is to use the most sensitive approach I can. And that for me is going to be the next-generation sequencing approach, because it more reliably gets me down to the 1 in 1 million cell thresholds. We can ask very critical questions about what MRD means below the 1 in 10,000 thresholds. I can tell you anecdotally what we’ve seen is it’s without question meaningful. We’re not seeing patients who were below that threshold do well over time.

Mark R. Litzow, MD: And you have NGS at your center?

Bijal D. Shah, MD: We do, and we’re now using it routinely in all of our patients. The harder question then becomes, what do we do with flow? Should we still be doing multiparameter flow on these patients? Is there a benefit to getting both? I don’t think so. But I do think that when you have MRD, being able to discern whether it’s 19- or 22-positive, if you can with your multiparameter flow, is therapeutically relevant. If you’re thinking about blinatumomab for someone who has very low MRD, you want to know. You want to know the answers to 19-expression, or 22-expression in the case of inotuzumab.

So there’s a lot that I can’t really answer in terms of how we spend money. The other problem with the next-generation sequencing approach is when you’re using it for MRD, you have to make the decision before you have the data. Meaning you order it off the aspirate. It’s got to be a fresh sample that’s sent off. And so you don’t necessarily know what the marrow is going to show by immunohistochemistry or flow before you make the decision to send it off.

It’s not easy to do it reflexively because unless your path lab [pathology laboratory] can give you multiparameter flow in 24 hours, which ours cannot, it gets very difficult to know when to send it, when not to send it. So our approach now is we send it. Timing is controversial. There are some who believe end of induction MRD is the most therapeutically relevant time point. There are some who argue end of consolidation, some who are doing both. And then as we move beyond consolidation, time points for subsequent assessment also become even more controversial. Do you check it every cycle? Do you take a different approach based on what you see post-consolidation?

Let’s say you do see 5 in 1 million cells. Is that someone you continue on therapy? Is that someone taking blinatumomab? And regardless of your therapeutic approach, when do you next check your MRD assessment?

As we move forward, in adults where we do, what we can and do see, a pattern of late relapse, how does that inform the frequency with which we monitor for MRD at the end of therapy? So after delayed intensification, in the context of maintenance, or even post-maintenance. And all I’ll say is it’s very controversial, and I don’t think any of us has the answer.

Mark R. Litzow, MD: And do you do it on marrow or blood or both?

Bijal D. Shah, MD: That is a fascinating question. And we are moving towards peripheral blood monitoring. And the reason for that is 2-fold. We have seen a healthy number of extramedullary relapses where I wonder if we may have better sensitivity getting this from the blood rather than the marrow. And the second reason is it’s really easy to get it from the blood. You’re going to lose a little bit of sensitivity. You’re going to lose probably 1 or 2 logs [logarithms], but we’ve been doing this with PCR [pathologic complete response] for people in 90 and P210 in the Ph-positives for the better part of the last two years. And with very good sensitivity and with enough leverage to do something about it before we see morphologically lapse in the marrow.

Transcript edited for clarity.