Risk Factors For Relapse In Children Undergoing Preemptive Immunomodulatory Therapy After Allogeneic Hematopoietic Stem Cell (HSCT) For Acute Leukemia

Relapse of leukemia following allogeneic HSCT is associated with a low survival rate and remains a major cause of transplant failure. To reduce risk of relapse after transplant, our institution introduced preemptive immunomodulatory therapy (IT) consisting of fast withdrawal of immunosuppression (FWI) and escalating doses of donor lymphocyte infusions (DLI). We performed a retrospective analysis of risk factors for relapse in children treated with preemptive post-transplant IT.

Sixty-nine children (33 AML, 33 ALL, and 3 biphenotypic leukemia) underwent myeloablative HSCT followed by IT between 2005 and 2012. 61% were male. Stem cell sources included unmodified bone marrow (BM) in 26 (38%) or peripheral blood stem cells (PBSC) in 43 (62%) patients. Conditioning regimens were TBI-based in 41 (59%) and chemotherapy based in 28 (41%) patients. Donors were matched related in 29 patients (42%), 10/10 HLA allele-matched unrelated in 22 (32%) patients, and mismatched unrelated in 18 (26%) patients. Thirteen patients (19%) had evidence of disease at the time of transplant.

58% of patients were enrolled in one of 2 prospective studies of IT open at our institution, while 42% of patients received IT as standard of care.

Chimerism analysis was done at day 30±7 on whole PB or BM and cell subsets (CD3, CD14/15, CD19), using a semi-quantitative PCR-based method with a sensitivity of 1%. Confirmatory testing was done 2 weeks later. If residual host cells (mixed chimerism – MC), or if positive disease was found on post-transplant testing, FWI over 2-4 weeks was done. If MC persisted following FWI, DLI was initiated and given every 2-3 months until full donor chimerism (FDC) was documented or graft versus host disease (GVHD) developed. Patients who showed FDC in all subsets, or who developed GVHD prior to IT, became the observation group.

36 patients (52%) had either FDC or acute GVHD and did not undergo IT; 15 (22%) patients underwent FWI and 18 (26%) underwent FWI followed by DLI. In patients undergoing IT, mean±SD time to FWI was 50.2±11 days post transplant, and mean time to first DLI was 135±62 days. Mean follow-up of living patients was 42±26 months. Kaplan Meier (KM) estimated event-free survival for the entire group was 72±5% at 24 months post transplant and 49±9% at 60 months post transplant. Events included relapse (N=20, 29%), death due to multi-organ failure (N=2, 3%), or death due to GVHD (N=4, 6%, only one of which was in the IT group). Out of 20 relapses, 6 were in IT group and 14 were in patients not receiving IT. Acute GVHD developed in 5 (15%) patients undergoing IT and in 17 (47%) patients not receiving IT. Chronic GVHD developed in 8 (24%) patients in IT group and 13 (36%) patients not receiving IT.

Univariate KM analysis of risk factors for relapse in the entire cohort indicated that use of post-transplant IT (log rank p=0.047), pre-transplant serotherapy (p=0.04), negative disease prior to transplant (p=0.04), and cGVHD (p=0.002) we related to reduced risk of relapse. In multivariate Cox Regression analysis, lack of cGVHD was the only risk factor for relapse (p=0.004). Within a group of 33 patients who underwent post-transplant IT therapy, only the presence of disease prior to transplant was related to the increased risk of relapse (p=0.005).

In a cohort of children receiving preemptive post-transplant IT, the relapse rate is low (18%) and toxic death due to GVHD is low (3%). These results suggest that post-transplant IT is a safe and effective approach to decreasing the risk of relapse in pediatric patients. Additional studies are needed to optimize the use of post-transplant immunomodulatory therapy in patients with high-risk disease.

No relevant conflicts of interest to declare.