The Relationship of Acute Gvhd and Pre- and Post-Transplant Flow-MRD to the Incidence and Timing of Relapse in Children Undergoing Allogeneic Transplantation for High Risk ALL: Defining a Target Population and Window for Immunological Intervention to Prevent Relapse

Abstract ⁴⁷⁰

We previously reported lower rates of grade II-IV aGVHD, but no improvement in survival with the addition of sirolimus to tacrolimus/methotrexate GVHD prophylaxis in a randomized phase III Children's Oncology Group/Pediatric Blood and Marrow Transplant Consortium trial, ASCT 0431. We performed further analysis of this cohort to test whether the presence or absence of aGVHD combined with detection of pre- and/or post-HCT minimal residual disease (MRD) would allow definition of high-risk populations amenable to early intervention to prevent relapse. MRD was measured on patient bone marrow samples with multi-channel flow cytometry at a single reference laboratory (detection threshold 0.01%). The trial included children ages 1–21yrs with high-risk CR1 and CR2 ALL receiving related and unrelated donor TBI-based allogeneic HCT. The analysis included 144 eligible patients with an estimated median follow up of 757d (interquartile range 558–1022d).

Both the absence of grade II-IV aGVHD at day 100 and positive MRD pre-and/or post-HCT were correlated with relapse. To illustrate the effect of aGVHD on relapse, Table 1 shows the cumulative incidence of TRM, relapse, and EFS at 100 days, 1 yr, and 2 yrs after HCT in patients with or without aGVHD.

Thirty eight percent of patients developed grade II-IV aGVHD (most (89%) by day +50). Analysis of patients who did or did not experience aGVHD by day 100 showed that patients who developed aGVHD rarely experienced TRM or relapse from that time forward (7.0 and 6.4% with TRM and relapse at 2 yrs post-HCT, respectively). Conversely, patients who did not develop aGVHD had a small risk of early relapse (3.7% at d100) but a steadily increasing rate of relapse starting day 200 post-HCT. Patients who did not develop grade II-IV aGVHD relapsed nearly 5 times more than those with aGVHD (30% vs. 6.4 at 2 yrs), with the result that EFS at 2 years was 19% (17/89) in those with no aGVHD vs. 35% (19/55) in those with aGVHD (p=0.001). It is important to note that the occurrence of aGVHD did not impact the rate of TRM.

To assess the effect of the MRD status on relapse, we measured MRD by flow cytometry of BM pre-HCT, within 2 wks of engraftment, and at 3, 6, 9, or 12m after HCT and correlated the presence of MRD in these samples with relapse rates. More than 400 samples were analyzed. For patients positive for pre-HCT MRD, the rate ratio (RR) for relapse was 3.7 (1.6–8.2) with an estimated two-year relapse risk of 27.4% (17.4–43.1) and 70.8% (50.4–99.4) in pre-HCT MRD negative and positive patients, respectively. For patients positive for post-HCT MRD, we constructed a Cox regression model to assess the effect of positive post-HCT MRD in the context of other transplant-related variables. In the post-HCT period, an MRD positive result was associated with a 14-fold increase in relapse rate over an MRD negative result (RR=14.3, 95% CI:6.1–33.6). These findings did not change after controlling for risk group, immunophenotype, donor type and pre-HCT MRD status.

Finally, we did an analysis of the combined effect of aGVHD and MRD status. Patients who developed grade II-IV aGVHD had low rates of post-HCT MRD and relapse, and a combined predictive effect of MRD status in patients with aGVHD could not be assessed. However, among patients who did not develop aGVHD, pre- and post-HCT MRD statuses were found to be independent risk factors for relapse (Table 2).

In summary, measurement of flow MRD on BM samples pre- and early post-HCT is feasible and positive results along with absence of grade II-IV aGVHD is highly predictive of relapse. There is a brief “window of opportunity” after day 50 and before day 200 post-HCT when most patients who will get aGVHD have experienced it, and the large majority of high-risk patients identified by MRD and aGHVD status have yet to relapse. It is imperative that novel post-HCT interventions be developed and given during this window in order to decrease relapse in the very high-risk patients we have defined.