Johanna Kelly

Incidence of the BRAF V600E mutation in chronic lymphocytic leukaemia and prolymphocytic leukaemia

The spectrum of underlying molecular abnormalities of clinically and biologically heterogeneous chronic lymphocytic leukaemia (CLL) and prolymphocytic leukaemia (PLL) has yet to be identified. While whole genome sequencing has identified several genes implicated in the pathogenesis and progression of CLL, the molecular lesions in a substantial proportion of patients remain to be elucidated. The incidence of the BRAF V600E mutation, widely implicated in solid tumours and other B-cell malignancies, was sought in a cohort of patients with CLL and related disorders. One CLL patient and one patient with B-prolymphocytic leukaemia (PLL) were found to harbour this mutation. Although present at a low frequency, the finding of BRAF V600E has biological and clinical implications for CLL and PLL.

Allogeneic Hematopoietic Stem Cell Transplantation for a BCR-FGFR1 Myeloproliferative Neoplasm Presenting as Acute Lymphoblastic Leukemia

Hematopoietic myeloproliferative neoplasms (MPNS) with rearrangements of the receptor tyrosine kinase FGFR1 gene, located on chromosome 8p11, are uncommon and associated with diverse presentations such as atypical chronic myeloid leukemia, acute myeloid leukemia, or an acute T- or B-lymphoblastic leukemia, reflecting the hematopoietic stem cell origin of the disease. A review of MPN patients with the t(8;22) translocation that results in a chimeric BCR-FGFR1 fusion gene reveals that this disease either presents or rapidly transforms into an acute leukemia that is generally unresponsive to currently available chemotherapeutic regimens including tyrosine kinase inhibitors (TKIS). The first case of a rare BCR-FGFR1 MPN presenting in a B-acute lymphoblastic phase who underwent allogeneic hematopoietic stem cell transplantation (HSCT) with a subsequent sustained complete molecular remission is described. Allogeneic HSCT is currently the only available therapy capable of achieving long-term remission in BCR-FGFR1 MPN patients.

Acute Lymphoblastic Leukaemia with an e1a3 BCR-ABL1 Fusion

inter-phase fluorescence in situ hybridization with dual labelled BCR and ABL1 probes detected hybridization signals in 66% of nuclei inspected, consistent with the presence of a t(9; 22). A standard RT-PCR methodology [9] detected the sole presence of e1a3 BCR-ABL1 transcripts. The patient commenced induction therapy according to the UK ALL XII protocol [10] but had 60% residual leukaemic blasts after induction with 54% of inter-phase nuclei examined, containing a BCR-ABL1 hybridization signal. The patient then received FLAG-Ida (fludarabine, cytarabine, granulocyte colony-stimulating factor, idarubucin) and achieved a morphological and cytogenetic remission with e1a3 BCR-ABL1 transcripts still detectable by RT-PCR. She then proceeded to consolidation with three courses of high-dose methotrexate, subsequently receiving continuous imatinib maintenance of 400 mg escalated to 600 mg. The patient relapsed 18 months after presentation with a karyotype of 44,XX, –3,–7,t(9; 22)(q34;q11),add(15)(q22–24) in 6 evaluable metaphases. She received re-induction with vincristine and prednisolone and subsequent imatinib maintenance (600 mg). Her clinical course over the next few months was complicated by Mycobacterium chelonae infection of the skin. The patient had a central nervous system relapse at 27 months after presentation and died soon after. MinApproximately 25–30% of adult patients with acute lymphoblastic leukaemia have evidence of the Philadelphia chromosome (Ph+ ALL). Identification of BCRABL1 transcripts is vital as these patients have a relatively adverse prognosis necessitating intensive therapies including allogeneic transplantation in eligible patients. More recently, it has been demonstrated that incorporation of a tyrosine kinase inhibitor into treatment regimens significantly improves prognosis [1] . Approximately 70% of Ph+ ALL patients express e1a2 BCR-ABL1 transcripts and 25% express either e13a2 or e14a2 BCR-ABL1 transcripts, with a number of other less common, variant transcripts reported that usually involve fusion of alternate exons [2] . The e1a3 BCR-ABL1 variant fusion transcript, lacking ABL1 exon a2 that partially encodes the src homology 3 (SH3) domain, has been described in three cases of chronic myeloid leukaemia (CML), all with a relatively indolent clinical course [3, 4] . In Ph+ ALL, this BCR-ABL1 genotype has only been reported previously in 8 adult patients [2, 5–7] and 1 paediatric case [8] , making any relationships between phenotype and outcome difficult to ascertain. We describe the clinical course of a further case of Ph+ ALL with an e1a3 BCR-ABL1 fusion. A 62-year-old female presented with pre-B ALL by immunophenotype. Although G-banded karyotype failed, Received: June 28, 2011 Accepted after revision: July 20, 2011 Published online: September 16, 2011

Molecular response to imatinib in chronic myeloid leukaemia with a variant e13a3 BCR-ABL1 fusion

Abstract The majority of chronic myeloid leukaemia (CML) patients express either e13a2 or e14a2 BCR–ABL1 transcripts. Variant fusion genes can arise, usually due to alternative splicing of either BCR or ABL1 exons, with molecular monitoring by quantitative PCR (qPCR) in response to tyrosine kinase inhibitor therapy rarely reported in such cases. A case of CML is described in which an e13a3 BCR–ABL1 fusion was characterised. A qPCR methodology was developed and applied prospectively to demonstrate a favourable molecular response to imatinib treatment. This case serves to highlight the requirement for molecular monitoring of those CML patients harbouring the e13a3 and other variant BCR–ABL1 transcripts.

No prognostic impact of P2RY8-CRLF2 fusion in intermediate cytogenetic risk childhood B-cell acute lymphoblastic leukaemia.

A subset of childhood B-cell precursor acute lymphoblastic leukaemia (BCP-ALL) patients have recently been found to harbour rearrangements of the cytokine receptor-like factor 2 gene (CRLF2) (Mullighan et al, 2009; Russell et al, 2009). A number of molecular lesions result in deregulated overexpression of CRLF2 (CRLF2-d); the two most common being interstitial deletions of the pseudoautosomal region PAR1 (Xp22/Yp11) that results in a P2RY8-CRLF2 fusion with CRLF2 expression driven by the P2RY8 promoter, or alternatively, translocation of CRLF2 to the IGH@ locus (14q32) leading to deregulated CRLF2 expression via IGH@ enhancer elements – IGH@-CRLF2 (Mullighan et al, 2009; Russell et al, 2009). CLRF2-d has been reported in 5–15% of childhood and adult BCP-ALL (Haslam et al, 2011; Mullighan et al, 2009; Russell et al, 2009) and is particularly more frequent in Down Syndrome (DS)-ALL patients, where its presence is associated with mutations of JAK2, postulated to cooperate in propagating the leukaemia in such instances (Harvey et al, 2010; Chen, et al 2012). The prognostic significance of this abnormality remains unclear in childhood BCP-ALL with studies assigning CRLF2-d as either an adverseor intermediate-risk prognostic factor (Cario et al, 2010; Palmi et al, 2012; Ensor et al, 2011; Harvey et al, 2010). The aim of our study was to determine the incidence and prognostic impact of CRLF2-d in diagnostic material from normal karyotype BCP-ALL patients treated according to the UK ALL 2003 protocol (ClinicalTrials.gov NCT00222612). Forty-two patients were selected from the total cohort of 238 children diagnosed with BCP-ALL and treated with UK ALL 2003 protocol between 2004 and 2010 in our institution. The selection process was based on the availability of a good quality stored diagnostic bone marrow sample for patients with intermediate cytogenetic risk, as defined by absence of the following: high hyperdiploid karyotype (chromosome N > 50), recognized translocations (ETV6-RUNX1, BCR-ABL1, MLL-AFF1 and IGH@-MYC) and constitutional trisomy 21 (DS). Cytogenetic and fluorescence in situ hybridization (FISH) data were available for all patients. CRLF2 over-expression was determined by real-time quantitative polymerase chain reaction (RQ-PCR) with underlying genomic rearrangements detected by either reverse transcription polymerase chain reaction (RT-PCR) (P2RY8-CRLF2 fusion transcripts) or FISH (break-apart IGH@; Abbott Laboratories, Abbott Park, IL, USA) as previously described (Haslam et al, 2011). Data were correlated with clinical risk factors, minimal residual disease (MRD) status and clinical outcome and analysed using Statistica (Statsoft, Milton Keynes, UK). The parents or guardians consented according to the Declaration of Helsinki for all participants. Overall, eight out of 42 patients (19%) were positive for the P2RY8-CRLF2 fusion, four of whom had high CRLF2 over-expression. No IGH@ translocations were detected. There was no significant difference in National Cancer Institute (NCI) risk status between CRLF2-d and non-CRLF2-d groups, although there was a tendency for lower blast cell count at presentation (10 0 vs. 39 0, P = 0 3) in the CRLF2-d group. No statistical significance was observed between day-28 MRD status and CRLF2-d. Four relapses were observed (two non-CRLF2-d (5 8%) and two CRLF2-d (25%). No statistical difference in overall and progression-free survival (PFS) was observed (PFS, 93 2% non-CRLF2-d versus 86% CRLF2-d, P = 0 11). There was also no clinical or laboratory differences between those patients with high and low CRLF2 expression. Genomic alterations of CRLF2 are emerging as additional high-risk molecular/cytogenetic features in some childhood ALL treatment trials. As reported by Attarbaschi et al (2012), we too did not observe a higher relapse-risk associated with the P2RY8-CRLF2 fusion regardless of any association with high levels of CLRF2 expression, suggesting that these patients are most likely to be of intermediate protocol-independent risk and therefore should not be considered at the present time for more intensive therapy. Given the deficiencies of our study, namely; small sample size and retrospective design, a larger study using multiple regression analysis would be needed to confirm the true prognostic impact of CRLF2-d in BCP-ALL. In addition, further assessment is required to define the underlying molecular mechanisms responsible for the absence of high CRLF2 expression in those patients with a P2RY8-CRLF2 translocation. Future risk algorithms are needed that will incorporate CRLF2-d status in prospective clinical trials in order to determine its true prognostic significance, as well

Chronic Myeloid Leukemia with e19a2 BCR-ABL1 Transcripts and Marked Thrombocytosis: The Role of Molecular Monitoring

While most patients with chronic myeloid leukemia (CML) express either e13a2 or e14a2 BCR-ABL1 transcripts, a significant minority expresses variant transcripts, of which e19a2 is the most common. Although considered to have a relatively favourable outcome, reported responses to tyrosine kinase inhibitor (TKI) therapy are variable with molecular monitoring in CML patients with e19a2 BCR-ABL1 transcripts rarely reported. A case of e19a2 BCR-ABL1 CML with marked thrombocytosis is described in which the value of molecular monitoring is emphasised during treatment interruptions, dose reductions, and changes. This case serves to demonstrate the requirement for prospective real-time quantitative PCR (RQ-PCR) assays for patients with variant BCR-ABL1 transcript types and standardisation of such assays to enable modern patient management.

Chronic myeloid leukaemia presenting post-radiotherapy for prostate cancer: further evidence for an immunosurveillance effect.

T-lymphocytes are known to have a fundamental role in the immunological control of chronic myeloid leukaemia (CML). Experience from stem cell transplantation has shown that recipients of T cell-depleted grafts have a higher incidence of relapse, and in those patients who relapse post-transplant, remissions can be induced by donor lymphocyte infusions. Sizeable T cell clones including specific cytotoxic T cells (CTL) that recognize leukaemia-associated antigens are present in a significant number of CML patients at diagnosis (Kreutzman et al, 2010), which may afford an opportunity for targeted immunotherapy (Riley et al, 2009). According to the immunosurveillance hypothesis, expansion of an occult malignant clone requires escape from immune recognition whereby effector cells are either deleted or rendered anergic: a function that is predominantly performed by the natural killer cells of the innate immune system (Waldhauer & Steinle, 2008). A number of cancer treatments, such as chemotherapy, radiotherapy, surgery and hormonal regulation, can all result in suppression of this innate cytotoxicity. Evidence for an immunosurveillance effect that prohibits development of CML derives from the observations of low levels of BCR-ABL1 transcripts present in normal, healthy individuals (Biernaux et al, 1995; Bose et al, 1998) and from an observed increase in the frequency of CML in longterm immunosuppressed recipients of organ transplants (le Coutre et al, 2010). A case is described in which CML manifested rapidly after radical radiotherapy for prostate cancer and which provides further evidence for an immunosurveillance mechanism, that when deregulated, resulted in clonal expansion and subsequent development of overt CML. A 62-year-old male was referred to the haematology service for investigation of a modest eosinophilia (maximally 0 9 9 10/l) and progressive basophilia (0 2–0 7 9 10/l) with the remainder of his complete blood count (CBC) normal. Retrospective CBCs for the previous 3 years were available (Fig 1). His only notable symptom was mild pruritus of the lower limbs. He was diagnosed with prostate cancer 3 years previously for which he underwent a radical prostactectomy. His prostate-specific antigen (PSA) had remained stable for 2 years post-surgery but had begun to rise; hence he received radical radiotherapy (66 6 Gy) to the prostate bed with a subsequent return to normal of the PSA. Approximately 3 months post-radiotherapy he was noted to have a leucocytosis (22 3 9 10/l) including eosinophilia

Screening for CRLF2 overexpression in adult acute lymphoblastic leukemia

Sir, The prognosis of acute lymphoblastic leukemia (ALL) in adults is poor, with a long-term survival rate of <40% in patients younger than 60 years of age. Emerging molecular technologies that allow the interrogation of genomes of patients with leukemia are providing a greater understanding of the underlying disease pathobiology. These approaches can identify novel genetic defects and pathways that may be targeted by specific inhibitors, thus potentially improving the management and outlook of adult patients with ALL (Mullighan, 2008). Several groups have recently identified specific abnormalities that result in overexpression of the cytokine receptor-like factor 2 gene (CRLF2) in pediatric B-ALL (Mullighan et al., 2009; Russell et al., 2009; Chapiro et al., 2010) and in adult B-ALL (Yoda et al., 2010). CRLF2 encodes a subunit of the thymic stromal lymphopoietin receptor and forms a heterodimeric complex with the interleukin-7 receptor alpha chain prior to TSLP ligand binding (He & Geha, 2010). Overexpression of CRLF2 can occur by interstitial deletion of the pseudo-autosomal region of either of the sex chromosomes (Xp22/Yp11), resulting in a P2RY8–CRLF2 fusion, or by the cryptic translocations t(X;14)(p22;q32) and t(Y;14)(p11;q32) that fuse CRLF2 to the immunoglobulin heavy-chain locus (IGH@). Rearrangements of CRLF2 have been reported in approximately 5–10% of pediatric B-ALL (Mullighan et al., 2009; Russell et al., 2009) and also in a high proportion of children with Down syndrome-associated B-ALL with concomitant mutations of JAK2 that result in constitutive activation of the JAK–STAT signaling pathway (Mullighan et al., 2009; Russell et al., 2009; Harvey et al., 2010; Hertzberg et al., 2010). In a cell line model, both CRLF2 and JAK2 mutations contributed to growth factor independence, suggesting that inhibitors of CRLF2/ JAK2 signaling pathways may be of therapeutic benefit in such cases (Yoda et al., 2010). Retrospective studies have shown a correlation between high CRLF2 expression levels and a relatively poor prognosis in pediatric B-ALL (Cario et al., 2010; Harvey et al., 2010). The incidence of CRLF2 overexpression was retrospectively sought in a series of adult patients with ALL referred to our centre for molecular investigation. Seventy patients (46 men, 24 women) comprising T-ALL (n = 2), Philadelphia chromosome (Ph)-positive ALL (n = 26), and B-ALL (n = 42) with a median age of 36 years (range 18–65 years) were analysed. RNA was extracted from unfractionated, presentation bone marrow (n = 41) and peripheral blood (n = 29). CRLF2 and endogenous control GAPDH gene expression levels were calculated by real-time quantitative polymerase chain reaction (PCR) using Applied Biosystems Taqman (Carlsbad, CA, USA) expression assays Hs00845692_m1 (CRLF2) and Hs99999905_m1 (GAPDH). Fold change in CRLF2 levels was expressed relative to an arbitrary calibrator (the mean CRLF2 expression of the peripheral blood of 10 hematologically normal controls) using a comparative Ct analyses (2) after normalization to GAPDH expression. CRLF2 expression was detected in all samples as it is normally expressed in T cells, B cells, mast cells, and monocytes (He & Geha, 2010). The majority of cases (n = 67) showed CRLF2 expression between 0.08-fold less and 88.0-fold more than that of the arbitrary calibrator. The remaining three cases showed significant, aberrant overexpression (846-, 1009and 1086-fold) compared with that of the calibrator (Figure 1a). These three patients all had a pre-B-ALL phenotype with a normal karyotype at diagnosis and represented 7.1% of all adult patients with B-ALL analysed, including those with both normal and abnormal karyotypes. Subsequent characterization of the underlying mechanism of overexpression was examined by reverse transcription-PCR for CRLF2– P2RY8 fusion transcripts (Hertzberg et al., 2010) and fluorescence in-situ hybridization (FISH) with a break-apart probe for IGH@ at 14q32 (Abbott Laboratories, Abbott Park, IL, USA). No evidence of a CRLF2–P2RY8 fusion was found in the three patients with CRLF2 overexpression. However, in one patient where archival material was available, FISH demonstrated a split signal (Figure 1b) indicating the presence of a cryptic IGH@ translocation. Overexpression of CRLF2 was not observed in adult patients with Ph–positive or T-ALL but was limited to B-ALL patients with a normal diagnostic karyotype, LETTER TO THE EDITOR INTERNATIONAL JOURNAL OF LABORATORY HEMATOLOGY

Targeted next-generation sequencing identifies clinically relevant mutations in patients with chronic neutrophilic leukemia at diagnosis and blast crisis

PurposeChronic neutrophilic leukemia is a rare form of myeloproliferative neoplasm characterized by mature neutrophil hyperleukocytosis. The majority of patients harbor somatic mutations of CSF3R gene and are potentially amenable to targeted therapy with JAK inhibitors. The incidence and clinical significance of additional mutations requires clarification.Materials and methodsA next-generation sequencing approach for myeloid malignancy-associated mutations was applied to diagnostic and matched blast crisis samples from four chronic neutrophilic leukemia patients.ResultsNext-generation sequencing confirmed the CSF3R T618I in all patients with identification of concurrent SRSF2, SETBP1, NRAS and CBL mutations at diagnosis. At blast crisis, clonal evolution was evidenced by an increased CSF3R T618I allele frequency and by loss or acquisition of CBL and NRAS mutations.ConclusionThe diagnostic utility of a targeted next-generation sequencing approach was clearly demonstrated with the identification of additional mutations providing the potential for therapeutic stratification of chronic neutrophilic leukemia patients.

A novel, variant BCR-ABL1 transcript not detected by standard real-time quantitative PCR in a patient with chronic myeloid leukaemia.

Sir, Identification of the BCR–ABL1 fusion, the molecular hallmark of chronic myeloid leukaemia (CML), is an essential element of diagnosis. Several established reverse-transcription (RT)-PCR methodologies are able to detect the common e13a2/e14a2 BCR–ABL1 transcripts and the majority of the less frequent, variant transcript types (Cross et al., 1994; van Dongen et al., 1999; Burmeister & Reinhardt, 2008). Characterization of the BCR– ABL1 fusion transcript type at presentation is necessary to allow selection of appropriate primer/probe combinations for real-time quantitative PCR monitoring that provides information on kinetics of response to tyrosine kinase inhibitor therapy, correlates with progression free survival and can detect loss of response (Foroni et al., 2009). A case of CML is described in which discrepant results were observed between qualitative and quantitative PCR methods at presentation prompting the characterization of a novel BCR–ABL1 fusion transcript. A 47-year-old female presented to her general practitioner with a 4-week history of fatigue. On examination, the spleen was palpable 4 cm below the costal margin in the mid-clavicular line. The physical examination was otherwise normal. Full blood count showed a white cell count (WCC) of 56.0 · 10/L, Hb 14.0 g/dL and platelets 366 · 10/L. The WCC differential was neutrophils 40.5 · 10/L, myelocytes 5.07, metamyelocytes 1.69, monocytes 1.13, lymphocytes 6.76, basophils 1.13 and nucleated red blood cells 1/100 white blood cells. Bone marrow aspirate and biopsy were hypercellular with granulocytic hyperplasia consistent with chronic phase CML. Cytogenetics were 46,XX,t(9;22)(q34;q11.2) in 20 metapahases with interphase fluorescent in situ hybridization detecting a concomitant deletion of the 50 ABL1 signal on the der(9) in all cells. A qualitative, standardized RTPCR methodology (van Dongen et al., 1999) detected BCR–ABL1 transcripts consistent with the e13a2 fusion type. Assessment of pretreatment level of disease was performed by a widely used, standardized real-time quantitative PCR assay (Gabert et al., 2003) but failed to detect and quantify BCR–ABL1 transcripts. Direct sequencing of the qualitative RT-PCR product revealed a deletion of 47 nucleotides from the 30 part of BCR exon 13 (nucleotides 3257–3303, NM_004327) that contained the real-time quantitative PCR forward primer (ENF501) annealing site. Inserted into the fusion site were 50 nucleotides from ABL1 intron 1b (nucleotides 73343– 73392, NT_03514.4) (Figure 1). This insertion represents a pseudoexon that maintains the reading frame and results in a novel BCR–ABL1 fusion indistinguishable from e13a2 BCR–ABL1 (by a difference of 3 bp) by the qualitative RT-PCR method and agarose gel electrophoresis used. The patient commenced imatinib 400 mg daily and achieved a hematological response at 4 weeks and a partial cytogenetic response (2/50 Ph+ metaphases) at 6 months. By utilizing primer b2C (van Dongen et al., 1999) as the forward BCR primer for real-time quantitative PCR, a 1.04 log reduction was detected in the peripheral blood BCR–ABL1/ABL1 level at 4 months with a bone marrow BCR–ABL1/ABL1 level of 12.0% detected at 6 months. The patient continues on imatinib 400 mg daily. The phenomenon of ABL1 intron 1b sequence inserted at the e13a2 BCR–ABL1 fusion site was first described in a CML patient because of weak probe hybridization by southern blot analysis (Rubinstein & Purves, 1998). Molecular characterization of these variant e13a2 BCR–ABL1 transcripts in CML reveals concomitant deletions that ablate the real-time quantitative PCR BCR forward primer annealing site are usually evident, with the sequence insertion from ABL1 intron 1b of variable size (Dessars et al., 2006; Matsushita et al., 2009; Jurcek et al., 2010). The nature of the resulting BCR–ABL1 fusion that incorporates intronic sequence is dependent on the size of the insertion with conservation, disruption or restoration of the reading frame all possibilities (Sorel et al., 2010). This case demonstrates that qualitative and real-time quantitative-PCR approaches do not always concur. Although an established qualitative RT-PCR methodology is necessary for the routine detection of the majority of LETTER TO THE EDITOR INTERNATIONAL JOURNAL OF LABORATORY HEMATOLOGY