Treatment response in AML: Understanding available molecular diagnostic data
“Doctor, am I in remission?”
The honest current answer is often: “It depends what you mean by remission.”
A 54-year-old patient with FLT3 internal tandem duplication (ITD) and WT1-mutated acute myeloid leukemia recently posed a similar question. She initially received induction 7 + 3 chemotherapy and midostaurin (Rydapt, Novartis) and later re-induction with mitoxantrone, etoposide and cytarabine, or MEC.
Five months after diagnosis, she relapsed and received azacitidine and sorafenib (Nexavar, Bayer) with incomplete hematologic recovery (CRi). Marrow after four cycles demonstrated a morphological disease-free state with myeloid differentiation. She was thus in complete remission with CRi by European LeukemiaNet (ELN) criteria.
Molecular testing disclosed persistence of the FLT3-ITD (36.6%) and WT1 mutations, consistent with persistence of cells arising from the original FLT3-ITD-mutated leukemia. She then received an allogeneic hematopoietic stem cell transplant, and a 30-day post-transplant marrow disclosed normal morphology and 99% donor chimerism (engraftment), but with FLT3-ITD-mutation positive at minimal residual level. She continued to have pancytopenia and again met criteria for CRi.
Thus, before and after her transplant, honestly answering her question regarding remission was certainly not straightforward.
Value of complete remission
The term complete remission has established prognostic value.
Morphologic complete remission has long been the standard for determining remission in patients with AML. Specifically, this term requires less than 5% blasts in blood and a cellular bone marrow and recovery of granulocytes absolute neutrophil count higher than 1 x 109/L (1,000/mL) and platelet count higher than 100 x 109/L. Remission subsets — including complete remission with incomplete platelet recovery (CRp) and CRi — have been developed to further aid response and prognostic information.
The era of molecular diagnostics and therapy of AML has just begun, but emerging data are clouding the waters of diagnostic certainty.
Morphologic complete remission remains the gold standard for initial therapeutic success. A study by Chen and colleagues demonstrated patients with CRp or CRi had higher rates and levels of minimal residual disease (MRD) than patients with complete remission.
Refinements in testing and monitoring for mutations and the advent of targeted chemotherapeutics have led to an exciting time in the field. However, the effects of these innovative treatments and ongoing controversies surrounding the assessment of MRD pose novel questions and challenge historical paradigms in both diagnosis and therapy.
Targeted therapeutics: in vivo differentiation
In vivo differentiation of leukemic cells has been well described in acute leukemias and is perhaps best described and studied in acute promyelocytic leukemia (APL).
In APL, the PML-RARA oncoprotein confers a block in terminal myeloid differentiation. Pharmacologic doses of all-trans retinoic acid (ATRA) interrupt the transcriptional repression caused by this oncoprotein, thereby allowing in-vivo terminal differentiation of leukemic cells. Arsenic trioxide operates at least in part similarly in promoting myeloid differentiation, and the combination with ATRA prevents emergence of resistance in leukemic cells.
In the conduct of the phase 2 trial of the FLT3 inhibitor quizartinib (Daiichi Sankyo) in patients with AML and FLT3-ITD mutations, Sexauer and colleagues observed that bone marrow myeloblasts in 13 of 14 patients underwent terminal differentiation. This was followed by in vitro testing of blasts that revealed quizartinib caused cell-cycle and arrest and terminal differentiation.
These data, combined with the wealth of information from APL studies, have led to a greater understanding of the role of blocking terminal differentiation in the pathogenesis of AML, as well as an avenue for targeted therapies. In the wake of the development and FDA approvals of new targeted therapies, this phenomenon of in vivo differentiation has also been observed in additional FLT3 inhibitors as well as isocitrate dehydrogenase inhibitors.
In the above case, the finding of FLT3-ITD at a high percentage makes clear that the bulk of the tissue analyzed was clonal or derived directly from leukemia. Thus, although the marrow was not morphologically leukemic, most of the marrow had differentiated directly from the leukemia.
AML and MRD
Monitoring for the presence of residual leukemia has been well-established for APL and pediatric acute lymphoid leukemia for years.
Clinical outcomes for patients with AML vary significantly despite efforts to refine prognostic factors with cytogenetic and molecular testing. In ALL and APL, MRD testing after induction chemotherapy predicts risk for relapse and suggests the value of potential additional treatment. However, MRD monitoring in non-APL forms of AML, although studied intensively recently, has not yet been well integrated into practice.
Techniques for MRD quantitation in AML include multiparameter flow cytometry (commonly performed for Philadelphia chromosome-negative ALL), and assessment of residual mutational burden (by polymerase chain reaction with or without next-generation sequencing) in tissues. These techniques yield very different data, and this and the heterogeneity of AML have generated challenges in applying MRD to clinical decision-making.
The molecular ontogeny of AML includes cases developing from mutations seen in functionally normal hematopoiesis — clonal hematopoiesis of indeterminate potential, or CHIP. After effective AML therapy, many patients retain large burdens of morphologically normal but clonal marrow as identified by mutations of various genes (DNMT3A, TET2, ASXL1, etc). The finding of these mutations in AML remissions has no prognostic implications, whereas MRD identified by other leukemia-associated mutations (NPM1 and FLT3-ITD) predicts incipient relapse.
‘Deluge’ of data
A publication from the ELN Working Party — published last year in Blood — made suggestions for standardizing testing for and clinical use of MRD in AML. This marks the starting point for how to best incorporate this information into prognostication and decision-making for patients with AML that has previously been lacking. There is emerging prognostic data, particularly before and after allogeneic HSCT, but as yet little data exist to suggest how these data should guide therapy.
“Doctor, am I in remission?”
The better question is, “How deep is my remission?” or, “How can we sustain my remission?”
Techniques to answer the first question are increasingly available, but their interpretation requires understanding the biological implications of the data. How to offer optimal therapy for many patients with AML remains unanswered despite the deluge of molecular data and the growing list of therapies to choose from.
Chen X, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.58.3518.
Döhner H, et al. Blood. 2017;doi:10.1182/blood-2016-08-733196.
Gocek E and Marcinkowska E. Cancers (Basel). 2011;doi:10.3390/cancers3022402.
Jongen-Lavrencic M., et al. N Engl J Med. 2018;doi:10.1056/NEJMoa1716863.
Kim T, et al. Blood. 2018;doi:10.1182/blood-2018-04-848028.
Paietta E. Clin Lymphoma Myeloma Leuk. 2015;doi:10.1016/j.clml.2015.02.009.
Schuurhuis GJ, et al. Blood. 2018;doi:10.1182/blood-2017-09-801498.
Sexauer A, et al. Blood. 2012;doi:10.1182/blood-2012-01-402545.
Walter RB, et al. J Clin Oncol. 2010;doi:10.1200/JCO.2009.25.1066.
For more information:
Megan Wheelden, MD, is a hematology-oncology fellow at Penn State Cancer Institute. David F. Claxton, MD, is professor of medicine at Penn State Cancer Institute. They can be reached at Penn State Cancer Institute, 400 University Drive, Hershey, PA 17033.
To contribute to this column or suggest topics, email Wafik S. El-Deiry, MD, PhD, FACP, at firstname.lastname@example.org.
Disclosures: Wheelden reports no relevant financial disclosures. Claxton reports research support from Ambit Biosciences, Astellas Pharma, Celgene, Cyclacel Pharmaceuticals, Daiichi Sankyo, Gilead Sciences, Incyte, MedImmune, Merck and Novartis.