Minimal Residual Disease By PCR Testing Is a Significant Predictor of Disease Relapse in Patients with FLT3 Positive AML after Hematopoietic Stem Cell Transplantation

Introduction:

FLT3 mutations are present in up to 30 % of AML pts with diploid cytogenetics. Allogeneic hematopoietic stem cell transplantation (AlloSCT) has emerged as the preferred therapy for this high risk group, but these pts remain at high risk for disease relapse. The significance of FLT3 detection by PCR, as a minimal residual disease marker, at the time of AlloSCT remains unclear and little is known about the risk factors associated with risk of relapse among FLT3 AML pts.

Methods:

We reviewed 1255 AML pts who underwent first AlloSCT between 2000 and 2014, and identified 200 adult pts with a FLT3-ITD or FLT3-TKD mutation found at diagnosis (Table1). All donor sources were included: 63 matched sibling (MSD), 95 10/10 MUD, 11 9/10 MUD, 18 cord blood (CB), and 13 haploidentical (Haplo).

Results:

The majority of pts received myeloablative conditioning (n=170; busulfan-based n=148). Disease status at AlloSCT was: CR1 (n=99), CR2 (n=20), CR with incomplete count recovery or hypoplastic marrow or CR3 (CRi; n=31), and active disease (AD; n=50). Thirty percent of pts received a FLT3 inhibitor before AlloSCT. After a median follow-up of 27 mo, the 2-year (2Y) OS and PFS rates were 43% and 41%, respectively. Relapse-related mortality was the main cause of death (68%). Remission and FLT3 status at the time of AlloSCT were significantly associated with PFS. Compared to CR1 pts, PFS was not different in CR2 [HR 1.5, p=0.3], but it was worse in CRi (HR 3.9, p<0.001) and AD pts (HR 5.9, p<0.001). The 2Y PFS was highest in CR1/2 FLT3–ve pts (62%) compared to the other groups: CR1/2 FLT3+ve (41%, p=0.3); CRi FLT3–ve (33%, p=0.01); CRi FLT3+ve (0%, p<0.001); AD (8% p<0.001) (Figure1). On multivariate analysis, independent predictors of worse PFS included: AD [HR 4.5, p<0.001], CRi FLT3+ve [HR 7.2, p<0.001], KPS≤80 [HR2.1, p<0.001], HCT-CI>4 [HR 1.6, p=0.05], and unrelated donor source [HR 1.6, p=0.05]. No difference in PFS was found between Haplo and MSD donor sources (HR=0.9, p=0.9). No significant association was identified by age, conditioning intensity or use of a FLT3 inhibitor prior to transplant. Post-transplant FLT3 status was available for 105 of 190 evaluable pts by day 30. Among them, 7pts had detectable FLT3 PCR and 5/7 pts (71%) relapsed.

Conclusion:

Morphologic remission and FLT3 PCR status at the time of transplant are key predictors of relapse risk in pts with FLT3+ AML. Patients who achieved a morphologic CR with an undetectable FLT3 by PCR at the time of SCT had the best outcomes and should undergo an AlloSCT without delay. Although the experience is limited, Haplo transplants had similar outcomes to MSD transplants. Prospective clinical trials with FLT3 inhibitors post-transplant are warranted at least for pts with AD or persistent FLT3+ on PCR at transplant.

Table 1.

Sex (M/F)	98/102
Median age	51 (18-72)
Age > 50	53%
KPS > 80%	64%
HCT-CI Score >4	19%