What Really Matters?

Andrew L. Pecora, MD

Editor-in-Chief

Oncology & Biotech News

Chief Innovations Officer, Professor, and Vice President of Cancer Services John Theurer Cancer Center at Hackensack University Medical Center

Every day around the globe, oncologists and hematologists sit in front of patients with newly diagnosed, life-threatening diseases and must offer compassion and hope. At the same time, they must use the latest data to provide a sense of prognosis and a treatment plan. Good or bad, this process had been something that was fairly straightforward until the development and use of molecular markers that conferred prognostic and therapeutic information to be used by the physician in guiding their patient and choosing the treatment plan.

Recently, it has been getting particularly complicated with acute myelogenous leukemia (AML). Prior to the availability of cytogenetics, treating AML with standard (7+3) induction followed by consolidation resulted in varied and unpredictable outcomes, with some patients cured and others relapsing and dying soon after relapse. In a seminal publication by Slovak et al (Blood. 2000;96[13]:4075- 4083), the proper use of cytogenetics to guide prognosis and treatment showed an impactful benefit to the clinical approach to AML therapy. A subgroup of patients with the so-called “core-binding factor” AML did so well with standard therapy that intense consolidation with allogeneic transplant became contraindicated.

However, careful analysis of even this seemingly good-risk subset revealed a varied outcome. On later inspection, a “subgroup of this subgroup” of patients with AML in fact did quite poorly (those with a mutation in c-KIT), and in this “sub-subgroup,” allogeneic transplant was now considered an appropriate treatment option.

If this was not enough to give practicing physicians a headache, along came new molecular markers, including FLT-3 and NPM-1. FLT-3 is considered particularly bad when the FLT-3 mutation (it comes in two flavors) is ITD. However, even if ITD is present, if NPM-1 is mutated, then FLT-3 may not be quite as bad. Add all of this to c-KIT mutation, and it becomes alphabet soup, with physicians left scratching their head on what to say or recommend.

Well, not so fast. Jourdan and colleagues (Blood. 2013;121[12]:2213-2223) may have brought some sanity back to clinical medicine. They analyzed a total of 198 AML patients with a core-binding factor subtype and looked at what factors predicted outcomes. In univariate analysis it was all of the usual suspects: high WBC, c-KIT mutation, and FLT-3/ITD. In multivariate analysis, only one factor was significant, and thankfully it was the presence or absence of minimal residual disease measured as a 3-log reduction after the first cycle of consolidation.

In other words, do not pay attention to all of the markers up front; instead, just wait and see if the standard induction and one cycle of consolidation rids the patient of the leukemia. If so, complete four cycles of consolidation and call it a day. If not, well then it is appropriate to consider other options including allogeneic transplant. I suspect this roller coaster ride is far from over with NextGen sequencing on the horizon. However, Jourdan and colleagues have given us a brief reprieve where we can sit with a patient and say we’ll see together if the standard will work, and if not, go from there.