Donor Lymphocyte Infusion Is an Effective Management of Increasing Mixed Chimerism, and High Risk Disease with Minimal Side Effects.

The increasing use of donor lymphocyte infusion (DLI) in both non-malignant and malignant diseases has demonstrated its potential use in cases with either increasing mixed chimerism (i-MC) or minimal residual disease (MRD) post transplant. Little data is available on the efficacy of this therapy in children. We performed a retrospective analysis of 31 consecutive patients who received DLI at Birmingham Children’s Hospital from 1998–2006. 18 patients (58%) had malignant illnesses (5 ALL, 6 AML, 1CML, 4JMML, 1 Hodgkin’s disease, 1 MDS). 13 patients (42%) had non-malignant conditions (3 aplastic anaemia, 1 glycogen storage disorder, 2 FHLH, 2 osteopetrosis, 5 thalassemia major). There were 19 haploidentical (61%) (6 fully matched), 8 unrelated (26%) and 4 sibling (13%) transplants. 15 (48%) were female and 16 (52%) were male. Mean age was 5 years (range 2 months to 16 years). Radiotherapy and non-radiotherapy based myeloablative conditioning regimes were used. 27/31 (87%) grafts were T cell depleted by CD34 selection and were given a T-cell dose between 1×10⁴–10⁵/kg CD3 cells. The mean CD34 cell dose for the haploidentical transplants was 14× 10⁶/kg (range3.63–36.6× 10⁶/kg) and for the allografts was 5.75 ×10⁶/kg (range 1–15 ×10⁶/kg). 19 (61%) patients were at risk of CMV reactivation.

The main indications for DLI in the malignant cohort were

1. post-transplant or extramedullary relapse and

2. identification of locally-determined pre-transplant high risk features.

14/31 (45%) patients with i-MC, were treated with graded increments of DLI ranging from 5× ×103/kg–1×108/kg, delivered in 1–7 infusions, majority receiving 2 or 3. 10/14 (71%) had complete donor chimerism post DLI. Though 4 (29%) patients failed to show improvement in their chimerism, 2/4 had non malignant diseases and have stable mixed chimerism and 2/4 who had JMML had full autologous reconstitution with normal haemopoiesis.

In the whole cohort, 87% had Grade 1–2 skin GVHD or no GVHD. Only 4/31 (13%) patients had severe morbidity due to Grade 3–4 GVHD and all responded to conventional treatment. There were no cases with CMV reactivation. One patient had graft aplasia from which he recovered after temporary cessation of DLI. There were 7 deaths. 6 patients died from relapsed disease. 1 patient given DLI for i-MC, died of streptococcal pneumonia due to non-compliance with penicillin V.

Our experience in this age group demonstrates two main points

1. DLI can be delivered safely in the setting of high risk malignant disease post transplant but further work is needed to define efficacy.

2. The correction of i-MC with DLI can be achieved with minimal morbidity and mortality.

In addition, the use of fludarabine enhances the impact of DLI but carries a theoretical risk of precipitating severe GVHD. No child died as a result of GVHD and to date 24/31 (77%) of the group survive disease free.