Leukaemic transformation of donor cells in a patient receiving a second allogeneic bone marrow transplant for severe aplastic anaemia

Allogeneic blood or bone marrow transplantation is a successful treatment for leukaemia and severe aplastic anaemia (SAA). Graft rejection following transplantation for leukaemia is a rare event but leukaemic relapse may occur at varying rates, depending upon the stage of leukaemia at which the transplant was undertaken and the type of leukaemia. Relapse is generally assumed to occur in residual host cells, which are refractory to, or escape from the myeloablative conditioning therapy. Rare cases have been described, however, in which the leukaemia recurs in cells of donor origin. Lack of a successful outcome of blood or bone marrow transplantation for severe aplastic anaemia (SAA), however, is due to late graft rejection or graft-versus-host disease. Leukaemia in cells of donor origin has rarely been reported in patients following allogeneic bone marrow transplantation for SAA. This report describes leukaemic transformation in donor cells following a second allogeneic BMT for severe aplastic anaemia. PCR of short tandem repeats in bone marrow aspirates and in colonies derived from BFUE and CFU-GM indicated the donor origin of leukaemia. Donor leukaemia is a rare event following transplantation for severe aplastic anaemia but may represent the persistence or perturbation of a stromal defect in these patients inducing leukaemic change in donor haemopoietic stem cells.

Persistent donor chimaerism is consistent with disease-free survival following BMT for chronic myeloid leukaemia

Chronic myeloid leukaemia (CML) can be treated successfully with allogeneic bone marrow transplantation (BMT) leading to long-term disease-free survival. Leukemia relapse, however, remains a significant clinical problem. Relapse following BMT presumably results from the expansion of small numbers of recipient leukaemic cells which have survived the conditioning therapy. In order to define patients who are at a high risk of leukaemia relapse, a variety of techniques have been employed to detect persistence of host haemopoiesis (mixed chimaerism, MC) or residual leukaemia (minimal residual disease, MRD). However, the precise relationship between the detection of MC and MRD post-BMT is unknown. We have investigated chimaerism and MRD status in 22 patients who were in clinical and haematological remission post-allogeneic BMT for chronic phase CML. Chimaerism was assessed using short tandem repeat PCR (STR-PCR) while BCR-ABL mRNA detection using reverse transcriptase polymerase chain reaction (RT-PCR) was performed to detect the presence of MRD. Seventeen patients received unmanipulated marrow (non-TCD) while in five patients a T cell-depleted transplant (TCD) was performed as additional GVHD prophylaxis. Chimaerism was evaluated in 18 patients (14 non-TCD, four TCD). Mixed chimaerism was an uncommon finding in recipients of unmanipulated BMT (21%) when compared to TCD BMT (100%). No evidence of MRD, as identified using the BCR-ABL mRNA RT-PCR assay, was detected in those patients who were donor chimaeras. Early and transient MC and MRD was detected in four patients (two non-TCD, two TCD) who have subsequently converted to a donor profile. One patient has stable low-level MC but remains MRD negative 4 years post-BMT. Late MC and MRD was observed in two patients who relapsed >6 years after TCD BMT for CML. We conclude that mixed chimaerism is a rare event in recipients of unmanipulated BMT and that donor chimaerism as detected by STR-PCR assay is consistent with disease-free survival and identifies patients with a low risk of leukaemic relapse post-BMT for CML.

Recurrence of Philadelphia chromosome-positive leukemia in donor cells after bone marrow transplantation for chronic granulocytic leukemia

Allogeneic bone marrow transplantation has been shown to be a very effective therapy for Chronic Granulocytic Leukemia with long term disease free survivals in excess of 60%. Relapse rates remain low at 15% following histocompatible sibling transplants and lower rates following matched unrelated donor grafts. Relapse rates however, are higher if BMT is carried out in transformation or blast crisis. Leukemic relapse in donor cells following transplantation for CGL is a rare event. The occurrence of donor leukemia however, may be under reported as accurate and sensitive investigation of the origin of relapsed leukemia following BMT requires DNA based technologies. A possible mechanism of donor leukemia in CGL is transfection of donor cells with the chimeric gene which is unique to this disease. It is possible that the malignant cells found in transformed or blast crisis of CGL may have a greater potential to transfect donor haematopoietic material. Careful evaluation of the incidence of donor leukemia using molecular biology methods may elucidate the frequency of this event following BMT for CGL.

Model selection for prognostic time-to-event gene signature discovery with applications in early breast cancer data

Model selection between competing models is a key consideration in the discovery of prognostic multigene signatures. The use of appropriate statistical performance measures as well as verification of biological significance of the signatures is imperative to maximise the chance of external validation of the generated signatures. Current approaches in time-to-event studies often use only a single measure of performance in model selection, such as logrank test p-values, or dichotomise the follow-up times at some phase of the study to facilitate signature discovery. In this study we improve the prognostic signature discovery process through the application of the multivariate partial Cox model combined with the concordance index, hazard ratio of predictions, independence from available clinical covariates and biological enrichment as measures of signature performance. The proposed framework was applied to discover prognostic multigene signatures from early breast cancer data. The partial Cox model combined with the multiple performance measures were used in both guiding the selection of the optimal panel of prognostic genes and prediction of risk within cross validation without dichotomising the follow-up times at any stage. The signatures were successfully externally cross validated in independent breast cancer datasets, yielding a hazard ratio of 2.55 [1.44, 4.51] for the top ranking signature.