Early assessment of treatment response in patients with AML using [(18)F]FLT PET imaging

Abstract

Assessment of treatment response in acute leukemia is routinely performed after therapy via bone marrow biopsy. We investigated the use of positron emission tomography (PET) for early assessment of treatment response in patients with acute myeloid leukemia (AML), using the proliferation marker 3'-deoxy-3'-[(18)F]fluoro-l-thymidine (FLT). Eight adult AML patients receiving induction chemotherapy underwent whole-body FLT PET/CT scans acquired at different time points during therapy. Patients who entered complete remission (CR) exhibited significantly lower FLT uptake in bone marrow than those patients with resistant disease (RD). In bone marrow, mean and maximum standardized uptake values were 0.8, 3.6 for CR and 1.6, 11.4 for RD, p<0.001. FLT PET results for CR and RD patients were independent of assessment time point, suggesting that FLT PET scans acquired as early as 2 days after chemotherapy initiation may be predictive of clinical response. This pilot study suggests that FLT PET imaging during induction chemotherapy may serve as an early biomarker of treatment response in AML.

Fig. 1

FLT PET imaging schedule. Blue arrows indicate when each patient was scanned. PET scans were acquired at progressively earlier time points of therapy. Bone marrow aspirate and biopsy were performed after therapy on day 14 (early) and between four and six weeks (follow-up). Patient 5 was not imaged due to development of epistaxis, unrelated to the PET scan.

Fig. 2

FLT PET images of bone marrow of seven AML patients, grouped by clinical response. PET scans were acquired at different time-points of therapy but results were consistent within each clinical response group (CR and RD), independent of time of assessment. RD exhibited elevated uptake (bottom row) while CR displayed low uptake (top and middle rows).

Fig. 3

Comparison of CR and RD patients, pre- and post-treatment. Pre-treatment: FLT PET image of bone marrow of RD patient (D) exhibited markedly greater uptake than CR patient (A). Post-treatment: FLT PET images (B and E) and SUV distributions (C and F) revealed very low uptake for CR patient but substantial uptake for RD patient.

Fig. 4

Overlay of bone marrow SUV distributions of CR and RD. Average SUV distributions (acquired during and after therapy) were determined for each clinical response group (CR and RD). The two distributions overlapped significantly in the lower SUV range. However, the heterogeneous RD distribution (red) with substantial uptake in SUV>2 range could be clearly distinguished from the more homogeneous CR distribution (green) with almost exclusive uptake in SUV<2 range.

Fig. 5

Heterogeneity of bone marrow response in AML patient with resistant disease. (A) Pre- and (B) post-treatment FLT PET/CT images of pelvic bone marrow. (C) Registration of pre- (red) and post-treatment (green) CT scans facilitated co-registration of corresponding FLT PET scans. Well registered regions are in yellow (red + green = yellow). (D) Ratio of post/pre-treatment bone marrow uptake demonstrates the heterogeneous response, with regions of good response (green arrows) and poor response (red arrows).

Fig. 6

FLT PET image of bone marrow of a normal control subject.

Fig. 7

Comparison of bone marrow FLT uptake in normal subjects with that of AML patients pre-treatment. (Top) SUV distribution reveals similarity of normal marrow (blue) and marrow of CR patient pre-treatment (green), with almost all uptake in SUV<8 range. Distribution is quite different for RD patient (red) who displayed significant uptake in the SUV>8 range. (Bottom) At all SUV, the ratio of CR patient’s pre-treatment marrow to normal marrow uptake was approximately unity. At SUV > 8, RD patient’s pre-treatment marrow often displayed uptake over 100 times greater than that of normal marrow.