Using Minimal Residual Disease to Improve Treatment Response Definitions and Hematopoietic Cell Transplantation Strategy in Acute Leukemia

Abstract

Acute leukemia is a most challenging illness for the oncohematologist because of the high risk of life-threatening complications and the complexity of treatment protocols as well as the uncertain therapeutic outcome, which depends on several host-, diseaseand treatment-related factors. When cure is the goal, as is normally the case for fit patients younger than 60 to 65 years with either acute myelogenous leukemia (AML) or acute lymphoblastic leukemia (ALL), the general plan is to deliver an intensive multidrug induction and postinduction treatment sequence, in which the greatest intensity, whenever required, is provided by an allogeneic hematopoietic cell transplantation (HCT). For the assessment of treatment response, a set of standard criteria was developed several years ago by a panel of experts, well before the current explosion of sophisticated technologies to assess residual disease status. In accordance, the bonemarrowmorphology is checked after induction chemotherapy to confirm the achievement of complete remission (CR), which is defined by a marrow blast cell content of less than 5%, together with the recovery of normal blood cell counts. Whereas CR is the mandatory first step of the process leading to cure, more recent technology allows the identification of submicroscopic amounts of leukemic cells in the bonemarrowof patients inCR, referred to as minimal residual disease (MRD), which goes undetected by light microscopy (one leukemic cell out of 10,000 to 100,000 cells) but nevertheless exerts a strong and independent prognostic impact, greatly increasing the risk of a patient experiencing an early relapse. HCT is a powerful therapeutic modality commonly prescribed to younger patients in CR with intermediateand high-risk characteristics. HCT is favored because of its superior activity compared with non-HCToptions, both within and outside clinical trials, and it has had little change in treatment strategy over the past decades. Having become standard practice in many developed countries, the number of HCTs performed has grown steadily, as demonstrated by the more than 40,000 and 20,000 HCTs performed worldwide in patients with AML and ALL, respectively, between 2006 and 2014. This treatment burden is unavoidable and reflects established indications and our wish to cure as many patients as possible; yet, the economic costs generated by the procedure itself as well as the management of the many related acute and chronic toxicities—and above all the associated risk of transplant-related mortality (TRM), averaging slightly below 10%, but higher in many published series—are great. To emphasize this point, a minimum of 6,000 TRMs are expected from a total of 60,000 HCTs, which is clearly an issue for the transplant team and any public health system, given that the estimated average cost of the procedure exceeds $200,000US at day 100, only to rise progressively. Apart from these concerns, HCTstill is, and will long remain, necessary to cure many patients who cannot be cured with chemotherapy, until the promises of novel, targeted therapies are fulfilled, which, to date, has occurred only in acute promyelocytic leukemia. Therefore, the clarification of any issue associated with the risk of transplantation failure, that is, recurrence or TRM, can only improve transplantation strategy and use of available resources, leading eventually to an overall therapeutic benefit. The article by Araki et al that accompanies this editorial is another relevant contribution from the Fred Hutchinson Cancer Research Center (Seattle, WA) on the relationship between CR status, reassessed through MRD analysis, and the outcome of 359 patients with AML who received myeloablative HCTs. With the MRD study performed prior to HCT by using a 10-color, multiparametric flow cytometry, Araki et al demonstrate that not all CRs are the same, and that the likelihood of post-transplantation success varies significantly by function of CR status and MRD as well. In fact, among patients in CR, the relapse rate was only 22% in the MRD-negative group compared with 67% in MRD-positive patients in CR and 65% in patients not in CR, and an MRD positivity of 0.1% or greater (or lower in some patients) retained a high, independent predictive power for relapse in multivariable comparisons with all other risk factors. The direct relationship between MRD and relapse risk was clearly evident in the large, well-performed clinical study by Araki et al. For several reasons, it is becoming difficult to ignore MRD in patients in CR with a risk profile that requires an HCT to reach cure, and these include the fact that MRD is a sensitive, patientspecific risk indicator, it is also rather inexpensive and relatively easy to assess, and it provides added information beyond standard cytogenetic risk class and underlying genetic abnormalities. In addition to improving the definition and the clinico-prognostic meaning of a CR status, now labeled as either MRD negative or MRD positive, these results prompt completely new clinical studies and, perhaps, a new transplantation strategy, as clearly stated by the authors: “patients with active AML could be