Roberta Lorenzatti

Clinical significance of TFR2 and EPOR expression in bone marrow cells in myelodysplastic syndromes

Myelodysplastic syndromes (MDS) are heterogeneous haematopoietic disorders characterized by bone marrow failure, cytopenias and a tendency to transform into acute myeloid leukaemia (AML) (Tefferi & Vardiman, 2009; Vardiman et al, 2009). Scoring systems are used to predict the prognosis of MDS (Greenberg et al, 2012). Percentage of blasts (>10%) and unfavourable cytogenetic abnormalities are the strongest predictors for poor outcome and are associated with high risk or disease progression to AML. Chronic anaemia is the main clinical problem in lowerrisk MDS but responds in 30–50% of cases to erythropoiesisstimulating agents (ESAs). Some prognostic factors of response to ESAs have been identified, with better response rates in patients with no or limited red blood cell transfusion requirement, low baseline serum erythropoietin (EPO) level and no-aberrant myeloid blasts by flow cytometry (Park et al, 2008). Nevertheless, the clinical care of MDS patients is still challenging due to lack of well-established markers that effectively monitor MDS natural history. Therefore, predicting, at diagnosis, those patients with risk of treatment failure is pivotal to personalize treatments in order to improve quality of life and prolong survival. Transferrin receptor 2 (TFR2), homologous to TFR1 (also termed TFRC), is a protein that is mutated in haemochromatosis-type-3 and contributes to the regulation of hepcidin in the liver (Ramos et al, 2011). It is also expressed in erythroid cells and in malignant myeloid disorders (Kawabata et al, 2001). TFR2 associates with erythropoietin receptor (EPOR) (Forejtnikova et al, 2010) and is required for efficient erythropoiesis (Nai et al, 2015). The TFR2 gene gives rise to two isoforms: the full-length TFR2a and the shorter TFR2b (Kawabata et al, 1999). We retrospectively investigated whether TFR2 isoforms and EPOR are differentially expressed in MDS patients and whether the expression is associated with patients’ clinical outcomes. RNA was obtained from the total bone marrow (BM) cells of individuals with non-malignant hematological disorders and in 46 treatment-naive patients at the diagnosis of MDS. After informed consent and ethical approval, TFR2a, TFR2b and EPOR expression was quantified by real-time polymerase chain reaction. Statistical analyses were performed using GraphPad-Prism software (GraphPad Software Inc., La Jolla, CA, USA). Comparison between groups was performed by Mann–Whitney U-test (nonparametric analysis). P values < 0 05 indicated a significant difference. In MDS patients, TFR2a and TFR2b showed higher variability in expression (TFR2a 6 92 5 45; TFR2b 3 16 1 61) than in non-malignant BM cells (TFR2a 8 23 2 97; TFR2b 3 68 0 47). We evaluated TFR2a and TFR2b expression in the different World Health Organization subgroups (Fig 1A–B). TFR2a and TFR2b expression was significantly lower in refractory anaemia with excess, blasts type 2 (RAEB2) (TFR2a 4 44 2 11; TFR2b 2 15 0 59; P < 0 05 and P < 0 01, respectively). A similar expression level was also seen in refractory anaemia with ringed siderobalsts. TFR2a and TFR2b expression was not correlated with total white blood cell, neutrophil and platelet counts, age at diagnosis and no significant differences were observed between sexes (data not shown). We next compared TFR2 expression with that of EPOR. Similarly to TFR2, EPOR expression varied more widely in MDS patients (18 45 18 34) than in non-malignant individuals (17 31 3 97) and was statistically lower in RAEB2 (8 18 2 62, P < 0 005) (Fig 1C). In addition, we found that TFR2a and TFR2b expression was positively correlated with that of EPOR (Fig 1D–E). To assess the clinical implication of this, we analysed the erythroid response in the patients that underwent EPO treatment (Table I). Patients with an increase in haemoglobin of ≥15 g/l after 12 weeks of treatment had TFR2 and/or EPOR levels comparable to normal controls (Fig 1F–G). In contrast, all non-responders had either very high (n = 2) or low (n = 10) TFR2 and EPOR levels. Statistical analysis performed on the 10 TFR2/EPOR low expressing patients showed that the TFR2 and EPOR products were significantly lower compared to EPO responding or non-malignant individuals (Table I). We then tested the effects of TFR2a, TFR2b and EPOR expression on survival in RAEB1-2 and refractory cytopenia with multilineage dysplasia (Fig 1H-J). In the first year of follow-up, patients with very low/low TFR2a or TFR2b expression levels had a significantly worse overall survival than those with normal/high TFR2 expression. Patients in the very low/low EPOR expression group also showed a tendency for poorer survival. Ever since it was demonstrated that TFR2 is a component of the EPOR complex (Forejtnikova et al, 2010), there has been interest in understanding its extra-hepatic function. TFR2 erythroid function has been recently described in normal erythropoiesis in mouse models lacking systemic or Correspondence

Sensitive Replicate Real-Time Quantitative PCR of BCR-ABL Shows Deep Molecular Responses in Long-Term Post–Allogeneic Stem Cell Transplantation Chronic Myeloid Leukemia Patients

Real-time quantitative PCR (RT-qPCR) is commonly used for follow-up of chronic myeloid leukemia (CML) patients treated with tyrosine kinase inhibitors, but its current sensitivity does not allow detection of very low BCR-ABL levels. Therefore RT-qPCR negativity is not synonymous with complete molecular response. Replicate RT-qPCR had shown increased sensitivity in tyrosine kinase inhibitor-treated patients and was, therefore, used here to evaluate whether RT-qPCR-negative post-allogeneic stem cell transplantation (SCT) patients harbor detectable disease. Samples from 12 patients were tested at 2 time points using 82 replicates of BCR-ABL RT-qPCR. One patient (38 months after SCT) had detectable transcripts at baseline and none at the follow-up test, done at a median of 107 months after SCT. This suggests cure from CML in the majority of allogeneic SCT patients who have no transcripts detectable by replicate RT-qPCR for BCR-ABL.