Pre-Hematopoietic Cell Transplant (HCT) Disease Status and Early Post-HCT Bone Marrow (BM) Chimerism Remain Predictors of Leukemia Relapse in Children Who Receive Preemptive Post-HCT Immunomodulatory Therapy (IT)

In 2009, we initiated a prospective study of chimerism-based preemptive IT consisting of fast withdrawal of immunosuppression (FWI) and donor lymphocyte infusions (DLI) in children receiving HCT for leukemia. We enrolled 43 patients. Donors and stem cell sources included matched related (49%), matched unrelated  (30%) and mismatched unrelated donors (21%) of BM (53%) and peripheral blood (PB) (47%). Chimerism analyses on whole PB and BM and subsets were done using a semi-quantitative PCR-based method (1% sensitivity), on day 32.5±SD8 (PB) and day38±10 (BM).  If mixed chimerism (MC) was found in any  PB or BM subset on 2 tests, FWI over 2-4 weeks was initiated, followed by DLI if MC persisted.  The end point for IT was full donor (FD) chimerism in all PB subsets or development of GVHD. Patients with GVHD or FD chimerism in all subsets were assigned to the observation group (N=12); patients with MC in any subset were assigned to the IT group (N=26).  15 patients received FWI and DLI and 11 received FWI only.  Five patients could not be assigned to either group due to early toxic death (N=1) or relapse (N=4). Follow-up of living patients is 28.8±9.7 months. Events included relapse (N=13) and toxic death (N=2), one of which was intervention related, resulting in 4% treatment-related mortality.  Estimated 36-mos EFS is 83±10% for the observation group, 54±14% for the IT group and 56±10% for the entire cohort, including unassigned patients. Acute and/or chronic GVHD developed in 5/26 (19%) patients following IT. Residual disease (RD) prior to HCT remained a predictor of  EFS. Despite IT 7/11 patients with RD and 6/26 without, developed relapse (Log Rank p<0.001). Early post-HCT BM chimerism was better predictor of EFS than PB chimerism (Table 1).  We recommend that post-HCT  IT continues until 100% chimerism in all BM subsets is achieved.

Table 1: Relapse rates in patients with MC and FD chimerism in whole BM, PB, and subsets

Chimerism	N*	Mixed chimerism           (MC)		Full donor chimerism (FDC)		EFS
		N with MC	Relapse rate in pts with  MC	N with FDC	Relapse rate in pts with FDC	Log Rank P value
Whole BM	39	26	35%	13	15%	0.1
BM - CD3 subset	29	24	25%	5	0%	0.09
BM - Leukemia specific subset	33	17	35%	16	6%	0.002
BM- CD34 subset	31	21	33%	10	10%	0.04
All  BM subsets	37	30	33%	7	0%	0.03
Whole  PB	42	22	32%	20	25%	0.8
PB -CD3 subset	42	31	32%	11	18%	0.6
PB - Leukemia specific subset	42	19	26%	23	30%	0.8
All PB subsets	43	35	34%	8	12%	0.14

   *Number of patients who had early post-HCT  chimerism test done on a specific subset.