Optimizing Therapy for Hairy Cell Leukemia - Episode 13

The Role of MRD in Treatment of HCL


Farhad Ravandi-Kashani, MD: Clearly, measurable or minimal residual disease [MRD] is something that you look for in your studies, but I think it’s important to remember that the definition of complete remission is still a morphological one and does not rely on flow cytometry. Then the question becomes is measurable residual disease an important marker or is achieving minimal residual disease negativity an important thing to look for or try to achieve in treating patients with hairy cell leukemia?

Robert J. Kreitman, MD: That’s a very good question. Again, the definition of complete remission requires that we don’t see hairy cells by the nonimmunologic stains, the stains like H&E, hematoxylin and eosin, or Wright stain of the blood. We shouldn’t see clumps of hairy cell in the bone marrow or hairy cells in the peripheral blood or the bone marrow aspirate. We shouldn’t feel an enlarged spleen anymore. That should have resolved. If there are lymph nodes, they should be nonpalpable. Those are the standard definitions of complete remission, and they haven’t changed for the last 50 to 60 years.

However, the new kid on the block is minimal residual disease, and this is something that’s well known in other leukemias. The tests that are used for assessing minimal residual disease vary a little bit between institutions, but in general, they include immunohistochemistry of the bone marrow biopsy, flow cytometry of the blood, as well as the bone marrow aspirate. In our hands, for the bone marrow aspiration, flow cytometry tends to be the most sensitive. I think you found that in your studies as well that you’ve published. Part of that might be that we both have very excellent flow cytometry people at our institutions. I think that’s a great thing. But in general, many institutions are now doing high quality flow cytometry and that tends to be most sensitive.

Bone marrow biopsy immunohistochemistry tends to be useful in multicenter clinical trials when you can take the bone marrow biopsy core, which is saved in paraffin, and you can cut new slides. You can do this MRD test in a central lab. Then you can compare patients on a multicenter clinical trial. So, that’s also very useful.

Farhad Ravandi-Kashani, MD: I think it’s also important to remember that assessing MRD is assessing the response to therapy. So when we have a very effective treatment for hairy cell, which produces very high response rates and is durable in many patients, clearly anything that increases the depth of response is likely to be associated with prolongation of response and prolongation of survival. This is particularly important in acute leukemias where there are patients who do achieve response but then relapse within a few months or few years after. In hairy cell, it has been reported that patients survive and do well many years after diagnosis and after therapy, despite having even a few detectable, morphologically detectable cells in their bone marrow.

So although, in my opinion, MRD is important and academically highly important, in practice it’s important if we are looking for a strategy that will result in the patient being essentially disease-free long term and cured. It is important to remember that a persistence of a few cells or a positivity by flow does not indicate that the patient cannot continue to have a good outcome and essentially do well for many years after initiation of therapy.

Robert J. Kreitman, MD: Right. The one thing I would throw in is that the trials that have been done in long-term follow-up in hairy cell, very few of them will go more than a median of 15, 16 years. Now with the life expectancy being in the 80s and the diagnosis of hairy cell being around the 50s, it may not be possible to know yet what is the final outcome in patients. I think it’s important for trials to keep following patients for as long as possible because we may miss the fact that some patients need salvage treatment because we expect that after first-line treatment. For example, the median relapse if you check blood counts would be in excess of maybe about 15, 16 years. A trial that would follow patients for 10 years maximum, like ours has so far, really wouldn’t be adequate to determine if patients might relapse during their lifetime to the point where they might need more treatment. This is an ongoing process of following patients who have been on these studies.

Transcript Edited for Clarity